{"_id":"doc-en-vscode-ba9e610aaad8c4df3322d938125598143ddc1aebd9f3e32f010313018244b9ed","title":"","text":"In VS Code action items inside inputs are vertically aligned to top, in the Chat features they are centered in the middle. This happens when putting in a lot of text or pressing Shift+Enter. Panel: <img width=\"425\" alt=\"image\" src=\"\"Inline: <img width=\"608\" alt=\"image\" src=\"\"Quick: <img width=\"621\" alt=\"image\" src=\"\"For Inline and Quick the action items to the right of the input box should also stick to the top.\nSeems like we have a mix of top-aligned and overflow w/ scroll (see below). But I agree that we should be consistent as much as possible. But it does feel like a chat send button should generally but aligned to the bottom, no? We may even consider rendering the toolbar on its own row once we're into multiple lines. That would ensure the text isn't forced to fill a tiny space now that we have 2+ icons. Here's how Teams does it as an example: ! ! ! !\nWould be awesome if the text could wrap around it, especially in the narrow search widget case, I've always wanted that but I think it would be a pain to implement.\nDon't have a strong opinion about top or bottom, I think each can be justified on a case-by-case basis. but middle looks weird. :) For example, inline Chat may want to do top alignment as the icons would align with those outside the input field (poppet and close) Bottom would help with the wrapping."}
{"_id":"doc-en-vscode-144371b0f775a58af8a6003099a52decf1400fd6b54e6948474c60fe58b2532f","title":"","text":"When you open a recent project from the welcome screen, the last one remains opened in a tab. VSCode Version: 1.9.1 OS Version: macOS Sierra 10.12.3 Steps to Reproduce: a new VSCode window. a recent project from the welcome screen. project is opened and the welcome screen remains opened.\nThe current behavior was chosen in the hope that this would help new users by staying in the same context.\nI agree that the welcome screen should be closed after taking an action from it. Especially in my case, where I prefer to use the setting, having to close the otherwise empty welcome window after opening from the 'Recents' list or via FileOpen feels like a bug.\nYou can just not show it at all. Or do you need any of the actions on it? (I'm trying to preserve the new user experience here which relies much more on guidance.)\nIt's ok showing the welcome screen, but it should be closed after opening a project from there.\nWith we change the Welcome editor to open in preview mode, this way it will close when the next editor opens. (If preview mode is not disabled in settings.)\nThis could be an additional value for the setting.\nHey, I could look into this. Got any tips on where I should start ?\nGreat: Add an additional enum literal here: Check for that literal and the 'open folder' condition here:\nHey , I've ran into some questions which you might be able to help me with: should I name the enum literal? I can't seem to come up with the right name would I check for the 'open folder' condition?\nYou could use . Check for no folder being open: . You can get at the context service by adding it to the constructor of the calling class, similar to:\nHey everything seems to be working now, just one last question which is confusing me: isn't the objective of this issue to NOT show the Welcome Page when opening a new project? So why is your enum suggestion welcomePageInEmptyWorkbench? What is happening right now after my additions is: upon opening a new project, no Welcome Page is shown. Is this right?\nShow it only in the empty workbench. :)\nHey Now I'm getting upon opening a new window (empty workbench). Any ideas? I did the comparison as you said for no folder being open.\nIs the ?"}
{"_id":"doc-en-vscode-2d5cbd7a01430dde044afcb1347dc76a3ff33642e24299bece8d7478d9a20f75","title":"","text":"You can debug it either in VS Code or in the Developer Tools (Help Toggle Developer Tools in the window running from source).\nUpon opening a new window, it isn't undefined (verified by a console log) but then it turns undefined (verified by the same console log). Edit: and immediately after Edit2: Found it! Seems like isWelcomePageEnabled() is called twice, once in the class constructor and the other one in the onReady method. I'll make the PR in a bit"}
{"_id":"doc-en-vscode-6f44398c22ddaca98553f2dc7b6c2397d553ab4bb95e7267f58c07a0db9e3be7","title":"","text":"! is not very discoverable. If it would be a submenu of the context menu, it would be much easier to tweak this setting.\n...\nI'm for making these types of settings more discoverable, however, I do not want to add these types of settings to the contextmenu. We mainly use the context menu to change the visibility of elements/features but we don't really use it to configure them/change appearance. Also, the context menu is already pretty full and I don't want to overload it.\nI'm open to other ideas to make them more discoverable. Maybe a \"configure\" entry in the overflow menu of the toolbar?\nYeah I like the idea of an action in the \"...\" menu to configure all things editor related by narrowing down a settings search in the settings editor.\nverification steps: Right click on the tab bar (not on a tab) At the bottom of the context menu there is an action \"Configure Tabs...\", press it Settings editor should open with tabs related settings"}
{"_id":"doc-en-vscode-53fa2ad8601b15a69fb9ace68b904301a31c30c7e8858ebaad47d7df38b6c2d3","title":"","text":"Testing I can see that the of can be . When should I expect each of these types? I'm not sure I understand what the purpose of is.\nis like the surrounding context lines that you see in the search editor. ! For example, here, lines 1791 and 1793 are context lines because they come directly before and after line 1792. I can try to explain this better in the docs, though!"}
{"_id":"doc-en-vscode-691e8a3f3bfc9e07b9a2f0d7ce8d5963ea1f03d18fafaaa7623d0004a7470dec","title":"","text":"Testing From the API: What is the behavior supposed to be in the following cases: The pattern contains . The pattern contains and is false. The pattern is a regex and is false. The pattern contains whitespace and is true.\nIf regex is not on, it just looks for literally. I believe that this actually acts as if is true (if I'm interpreting code correctly). Any time that you have regex enabled and explicitly match a newline, it will be multi-line. I should probably document this. Any characters in the regex will also match the other case. ie: should also match all characters, not just lowercase letters. It will look for the whitespace, and will look for line boundaries on either side of the search string.\nI can also add extra examples for corner cases so that this is more clear."}
{"_id":"doc-en-vscode-722aca863960362d0731f6eb7130a0082ddbc78473cfe1935ac752f81cbbe0a1","title":"","text":"VSCode Version: Version 1.12.2 (1.12.2) OS Version: OS X E Capitan 10.11.6 Steps to Reproduce: integrated terminal <kbdctrl+`</kbdterminal: key combination <kbdctrl + j</kbdExpected Cursor moves back a word, this the behavior that iTerm and exhibit Actual Newline is sent, essentially the same as using <kbdreturn</kbd\ndoesn't seem to be in the path in for me?\nIt's in zsh which is in mac by default? So just type in\nWorks for me, if you remove the backtick from your command: Not\nOh that was a typo, I am using Hmm that works in the integrated terminal for you? Using zsh without a .zshrc: Expected (in iTerm): ! <kbdctrl + j</kbdcauses the cursor to jump back to the beginning of each word. Actual (in integrated terminal): ! Should have probably left this gif linger a bit longer but at the end it says: As if you had just pressed <kbdreturn</kbd.\nFYI, I just tried rebinding <kbdctrl + j</kbdwith bash's native binding commands: Again this works in iTerm and but not in the integrated terminal. Expected Terminal screen should be cleared Actual Newline is sent, essentially the same as using return\nHm I doesn't seem to be working anymore, maybe I didn't test it right or something. I looked into it a little bit, it seems to work fine in the demo which is the underlying library used. so that's why it's happening. The question is why does it work in the demo but not VS Code. It definitely gets registered within zsh inside VS Code: <img width=\"430\" alt=\"screen shot 2017-05-17 at 9 50 05 pm\" src=\"\"Stepping through the code nothing stuck out as misbehaving through when run inside VS Code; they appear to both be sending 'n'. Nothing in the seems to be wrong either.\nAw I was really of hoping it was something wrong with my local configuration. Also I got the same behavior when trying it on windows + cygwin with a fresh install of vscode, <kbdctrl + j</kbdnot being distinguished from newline.\nHey just wondering if there was any progress on this?"}
{"_id":"doc-en-vscode-230d5dfafcb1596d205bf3baeee10f621af481fcd046c6bf7fa12ff91c30bb9f","title":"","text":"If this is low priority could one of the contributors point me a area/file that might be causing the issue? Thanks!\nI can't look into it any further right now, I'm pretty stumped on where to go from here though. It's a pretty technical problem so it might not be the easiest thing to jump in to. If you want to you can start by cloning and trying to reproduce it using the demo. If you can reproduce it you can check what's sent to the backend by loading up devtools and debugging If works fine, it would be a matter of doing a similar exercise but figuring out why VS Code behaves different by debugging similar code.\n(Copied from ) This was happening because ctrl+j translates to (see ) and since we normalized to all input to the pty, this was getting lost and converted to a new line. The fix here is to never \"sanitize\" data that has already been processed by (ie. key strokes and paste) as knows how to handle them. We only need to process text that we send directly to the terminal such as a script the user is running or text being pasted into vim, etc.\nHmm just updated to 1.14.1 and this still isn't working for me... !\nThis will only be in 1.15.\nOh okay thanks."}
{"_id":"doc-en-vscode-4b4992ec73f5a28ea1fd8fc0b2b75c03fafccbe1cefba3d50627bb81d6ed2298","title":"","text":"Issue Type: <bBug</bI have noticed that every time I try to open an image on vs code, I get the error stated on the title no matter what image file type VS Code version: Code 1.39.2 (, 2019-10-15T15:33:00.827Z) OS version: Darwin x64 18.7.0 <details<summarySystem Info</summary </details<details<summaryExtensions (21)</summaryExtensionVersion ------ html-snippets0.2.1 vscode-browser-preview0.5.9 rails0.13.1 vscode-gemfile0.0.2 solargraph0.21.1 vscode-eslint1.9.1 code-runner0.9.14 vscode-rufo0.0.1 vscode-gutter-preview0.23.0 atom-keybindings3.0.6 vsliveshare1.0.1056 debugger-for-chrome4.12.0 erb0.1.1 LiveServer5.6.1 ActiveFileInStatusBar1.0.3 slim0.1.2 html-preview-vscode0.2.5 highlight-matching-tag0.9.5 simple-ruby-erb0.2.1 quokka-vscode1.0.257 erb-vscode-snippets0.0.4 </details<!-- generated by issue reporter --\nPlease share a screenshot of the error. Also, does this reproduce in the with all ?\nI'm having the same issue on a feshly installed copy.\nFound out the issue to be with the image file name. If the file name has % symbol in it, vscode does not show the image. <img width=\"708\" alt=\"Screen Shot 2019-10-30 at 12 17 40 PM\" src=\"https://user-\"If, i remove the % symbol in my file name, vscode displays the image properly. Tested this both in current stable mac release and the insider build. Do let us know if you observe the same pattern.\nEven i am facing the same problem and my image file name does not contain % character , resolve my issue.\nMy file doesn't contain the % character either and it's not working."}
{"_id":"doc-en-vscode-381d02b60a3175ae92c04e7d84a5fd2e8b1ff2b847c25c53226f4a4785f969c2","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --When folded, folding ranges are displayed in the editor as a single line with darkened background. Hovever (as of ), on the minimap, the background of this line is not darkened. This makes it harder to navigate using the minimap. !\nEither this isn't working or steps are non trivial\nOk, see this working unless I have render characters enabled"}
{"_id":"doc-en-vscode-ad5ebd8b35024a9c6514f008b67b0411c1a7614f1142eb88ed4d3b232ad0610c","title":"","text":"I like the improved \"All done. You can close this tab now.\"-dialog but I think it is missing a \"Close Tab\" option. Today, when everything works you need to hit \"Cancel\" but I don't have the close tab option - which is what the title suggests\nwhat are the steps to trigger this?\nUse Continue On Reopen on the Desktop from"}
{"_id":"doc-en-vscode-a9960e0d1c57f296152b62b7dada15776bcbaa06838e768a526078142a61d280","title":"","text":"This has been fixed for notebook diffs but is still an issue for text diffs: <img width=\"115\" alt=\"image\" src=\"https://user-\""}
{"_id":"doc-en-vscode-11c6cc1ca0367eb9b01c59f6c621e6990d290a02f5428d3304a32e3357607957","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --in my project, there are several extensions in our recommended , but vscode sneaks in recommendations that are not recommended by workspace, but by vscode itself: ! in this screenshot, the WSL extension is recommended by vscode, and the YAML extension is recommended by the workspace. you have to hover over the star icon to figure out which recommendations are actually relevant for this workspace. this makes it confusing for new developers who just want to install the extensions required for the workspace, and not random additional ones that vscode suggests. as a workaround, i can add these vscode recommendations to the array, but that's annoying because i would have to update that list whenever new recommendations show up.\nVS Code shows workspace recommendations on the top followed by other recommendations. If you are not seeing in this order, can you please share the steps to reproduce?\ni don't know if i could, since it seems to be some sort of recommendation algorithm that shows different results based on all sorts of unknown factors. (i feel like it'd be comparable to coming up with steps to reproduce an ad i saw in a social media feed.) but i'll give it a go\nit seems that the docker and wsl extensions (ones made by microsoft) always appear above workspace recommendations. steps to reproduce: to the workspace recommended extensions a to trigger the non-workspace recommendation to the extensions tab and look at the recommended extensions list ! ! even if these were showing up in the correct order (workspace recommendations first), it's still extremely confusing and misleading to the user to use the exact same icon for two completely different types of recommendations. i get that microsoft wants to get as many people using their extensions as possible, but this seems comparable to an advertisement that is not clearly labelled as one.\nI found out that there is a bug in showing them in the right order. Will fix that. I am not sure how much it makes sense to show different icons for different kinds of recommendations. It might be super noisy or confusing. I believe at present we have a right balance to show an icon for recommendation and explanation on hover."}
{"_id":"doc-en-vscode-ce9667507f942bfcedfefa74083a36a188cc15009835e07561ac32375b66ef0e","title":"","text":"<!-- Do you have a question? Please ask it on --- VSCode Version: 1.10.1 OS Version: macOS Is it possible for VSCode to support the following kinds of auto complete, please see screenshots Word completion: <img width=\"357\" alt=\"screen shot 2017-03-07 at 10 23 19 am\" src=\"\"Member function completion: <img width=\"359\" alt=\"screen shot 2017-03-07 at 10 20 22 am\" src=\"\"All auto-completions come with fuzzy matching. Steps to Reproduce:\nAgree, actually fuzzy matching only works with camel Case, but in sublime and atom you can get completion by typing any middle letter of the target word. I still miss this feature in vscode\nI like, tho will be some work and maybe some adjustment for folks that are used to our current filtering and ranking\nThis got implemented with <img width=\"280\" alt=\"screen shot 2017-03-20 at 17 38 54\" src=\"\"\nAnd this is valid for all languages then?\nYes, we have changed the default matcher to be more fuzzy.\nVSCode: 11.0-insider I appear to be getting stuck in a recursive loop, and according to DevTools (I clicked Pause in the Debugger) it is happening in . I tested this with . I was able to reproduce this with a simple HTML file: After typing the opening and IntelliSense appears, trying to type an additional character will freeze the whole VSCode.\nThanks for the heads-up. You are right, there are situations in which it keeps recursing forever. I have implemented a different approach today and just pushed the last changes. Will be in tomorrows insiders."}
{"_id":"doc-en-vscode-253e7901c389217ced692f2f82edff243d354c1fc337660d5d867f63df1713d1","title":"","text":"The cache is preserved throughout URI serialization. We should invalidate it when the URI crosses platform boundaries. Related to"}
{"_id":"doc-en-vscode-d89fdf6c87463ac0d54fb65ae52b62c6ebda3bf76b9f158df676670a8320ae30","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: head OS Version: n/a This is a bug/inconsistency in the VSC codebase, which is currently obscured because of the specific way the code is written. This wouldn't really matter, but when using a build step that generates files for dependencies, the gets evaluated down to and compilation fails. And overall, it'd be nice for the types to align in a sensible way. Steps to Reproduce: , replace with Expected: compiles fine Actual: there are a number of errors like Property 'getFirstLineText' does not exist on type 'never'. The intersection 'this &amp; IResolvedTextFileEditorModel' was reduced to 'never' because property 'capabilities' has conflicting types in some (2339)TextFileEditorModelIWorkingCopyIResolvedTextFileEditorModelthis.isResolved() is known to be true) says capabilities should be a readonly 2. <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes\nSince this seems to concern only you with your specific setup, wanna do a PR to fix this?\nI would be more than happy to, but I'm not sure what approach you'd find (in)appropriate. could type in as a number instead of an enum (since checks against it are anyways done via bitwise-and) could change in to actually be 2 instead of 0 (though it seems like you also use 0 for this field in other places like , and somehow TSC doesn't care...) could add a enum value of 0 to , since it seems like that's how you use it. I'm leaning towards approach 3. What are your thoughts?\nYeah I think we can give 3. a try in a PR."}
{"_id":"doc-en-vscode-d192dd9695f2b382f5cd4ffa9551a589045be89675624e968a0697f0112e7743","title":"","text":"When a test fails, it might sometimes look like this: ! By pressing in both editors, I can get it to look like this, which is more helpful for long strings: ! It might be better to enable word wrapping by default for string comparisons.\nCode pointer:\nIs this issue up for grabs? I would love to implement this\nYep\nCould you assign me to this issue?\nNo, only people in the Microsoft org can be assigned issues. You can open a PR without being assigned though\nI agree this is a better default. Thanks Maybe a setting to make this customizable? Not sure how many people would care.\nHow do I update commonEditorOptions dynamically? Replace every usage of commonEditorOptions with ?"}
{"_id":"doc-en-vscode-9ddb593e5f6c527affc39e2c291a4c133b639574ba79c7af52978453be052bbe","title":"","text":"VSCode Version: Code - Insiders 1.8.0-insider (, 2016-12-09T12:05:39.439Z) OS Version: Linux x64 4.8.0-28-generic Extensions: Steps to Reproduce: an untitled file and add content: the cursor after the shift, press down and then end, the selection should be shift, press home, the selection should be but it is because the select went to the start of the wrong line.\nThis used to work till yesterday... something broke. I'm pressing <kbdend</kbdafter going 3 lines above, and the cursor goes to the end of the selectionStart line. !"}
{"_id":"doc-en-vscode-07188f734e1ae6dd18d33223860b0c99378251effe8ae2ef80ee3748a282b5b5","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- Version: 1.60.2 Commit: Date: 2021-09-22T11:59:27.195Z (1 wk ago) Electron: 13.1.8 Chrome: 91.0.4472.164 14.16.0 V8: 9.1.269.39-electron.0 OS: Darwin x64 20.6.0 Steps to Reproduce: a large file with a long line to be scrollable over the file (x and y axis). right. on minimap. Actual: The minimap slider does not have full width. Expected: The minimap slider is always full width.\nCould you please also add a screenshot, I am not sure what you are referring to.\nPlease check. https://user-\ncan I work on this issue. if yes code pointers to investigate.\nPR welcome.\nthanks\nwhat would be the expected behaviour of the minimap for a horizontal scroll?\nI think we could just remove that the slider moves horizontally and let it stay put.\nI think it should be:\nThese lines of code should work if you support scrolling horizontally for the minimap slider. At this time, we can only scroll the minimap slider vertically.\nOkay, I can do that but I think it would be much better if we allow horizontal scroll in minimap. what do you think about that?"}
{"_id":"doc-en-vscode-fcf08d44d43027610cdb33c777e61b1199d6a5b24bcfd1523798e43cba4aa247","title":"","text":"Testing I find it a little weird how there is a cloud badge on the list item and that represents 2 different items origin/main and origin/HEAD in the hover? <img width=\"1031\" alt=\"Screenshot 2024-08-27 at 10 31 23 AM\" src=\"\"Should these be different icons? Should it be a single item ? Should they be the same color?"}
{"_id":"doc-en-vscode-eb38c5540de33ba66f351eeb068e2ac5ad3de6f4a176358feb3564d2e10790bf","title":"","text":"Repro: folder test task, wait to finish another terminal test task again <img width=\"1344\" alt=\"Screen Shot 2022-08-07 at 11 40 14 am\" src=\"https://user-\"\nIn repo, ctrl+shift+b is focusing the terminal: !\nReproduces on webgl only\nGuessing it's related to canvas recreation:\nYep: !"}
{"_id":"doc-en-vscode-e6c5c027cae4f40104d939e55c9a325d74ce20107ca33d9ec05ffc95e5c11541","title":"","text":"Using regex expressions am unable to find empty lines. Tried both the end of line sequence. Copying the same text in sublime simple n finds all the blank lines.\nThat has been an issue for me for quite a while. Never really checked again if it's already working because I started to automatically open notepad++ again whenever I needed to replace newlines.\nThis problem could manifest if the lines are being split before searching as opposed to doing a full text search. Interesting side note; seems to work in find in files (Ctrl+Shift+F), just not in the find dialog (Ctrl+F).\nCouldn't even find empty lines with ^$ (in case they were being split without EOL).\nLooked at the code, it does indeed look at each line individually. I'm created a new issue for multi-line matching"}
{"_id":"doc-en-vscode-a657058fdf88cfb4d7f2c1cf6a422ebc6a79f0196ee96d0c12127ca90875eb8d","title":"","text":"Steps to Reproduce: The cursor should go back to the exact spot where it was before the jump. Instead, it goes one sport further back...\nCan you provide more details like VS Code version, OS type and version etc. Also, sample code of what you are trying to do in your extension and a screencast/gif/video ?\nLikely\nThere is no extension involved. This is vanilla behaviour. Using version 1.4 on Ubuntu 16.04. Related to , but not the same. This is likely to be very easy to fix (jumpToDefinition event should push the current cursor position to the history).\nI agree we should record the current cursor position on certain actions like \"Go to definition\".\nPushed a change to improve this situation for the following cases: goto definition will store the current cursor location as history entry so that when jumping back you end up in the same line number opening another editor will now also store the current cursor in the history so that going back puts you to the exact same line number too"}
{"_id":"doc-en-vscode-ea3747249c371d07a09ce6c834bad92a1a2e310d14c8d2343ee3f7a086697db8","title":"","text":"Version 1.74.0, when I open the developer tools, it becomes a floating window, which is very unfriendly for extension developers, please add at least one option to use the old layout. !\nWas caused by an upstream electron bug with WCO Believe there are plans to restore the docking feature in the near future now that the electron issue is resolved.\nI found that if you toggle OFF the setting, Developer Tools docking functionality is restored AND you also still get the WCO experience (perhaps that's a separate bug?). Suggest docking at the bottom instead of to the right if there's any visual conflict due to the aforementioned electron bug.\nIt work, thanks!.\nThe setting will be removed at one point, keeping the issue open to provide and dock options when WCO is enabled.\nI manged to get it to dock again with 's workaround but I noticed in the dev tool now it shows the HTML for the VS code app instead of the file I have open in the preview !\nRelated Stack Overflow question:\nThis makes working on Windows really awful for me, I think the upstream bug is fixed.\nTo give some context, with WCO chromium has a bug in orienting devtools window when docked to the position where the Window controls are, basically they overlap making it harder to use the close, minimize or maximize button. Chromium always pop out the devtools window in PWA for this scenario and Electron inherited this by default. The fix in does not address the aligning behavior but rather provides a workaround that calling would be respected even in the WCO scenario rather than forcing to always detach. Now, on our end to incorporate this fix we need to change all our callsite with a mode otherwise we will default to detach as seen in this logic Say we chose as default and user later reposition it to right or left then this new state will be persisted to json file under the user data directory, in non WCO cases this saved preference will be applied next time when calling or without a mode value from the application side, that is how user's preferred position is applied. But in the WCO case, this does not take effect so we need a logic on VSCode side to read the preference value and adjust the start mode with it. I didn't want to make this change yet in core hence this issue remains open, I will tackle it in debt week."}
{"_id":"doc-en-vscode-590a70ebf1c797304e32c1a726dc56ea850a74ddd368eed7b0970980c491da7d","title":"","text":"I see this as an issue when you dock the devtools to the right, but why is the ability to dock it at the bottom taken away? This seems to work just fine with WCO. But on Windows, a user is not able to reposition the devtools at all? This functionality seems to be truly gone, so a setting for Windows only would be a decent workaround and good improvement over what we have today. If we decide on a setting that forces the location, I can see to do such a change.\nI am good with a setting to force the bottom position only on Windows. I will see if the runtime behavior can be improved to avoid docking the devtools to the position where the controls are present as part of this issue.\nThe dock options are back, though setting it to dock on the right looks funny with the WCO in the way. !\nYes that is the original bug in chromium which is not addressed.\nIn the new version, if the developer tools are on the right side, the close button will be blocked, which is not as good as the old version of . !\nuse a dock position that is not overlapping with the window controls, by default the devtools will not dock to the position were the window controls are to avoid the bug.\nI know, but I always put the developer tools on the right, which is where I'm most productive.\nIs it possible to revert \"window.experimental.windowControlsOverlay.enabled\" to a VS Code new version? In order for the developer tools to be correctly positioned on the right side, I now have to go back to VS Code version 1.81 and disable updates. !\nis there an issue where we are tracking how the WCO is overlapping the devtools?\nno we are not tracking it in the runtime, unless you have manually docked it to the overlapping position devtools by default would be undocked. The overlapping position in the runtime will not resolved anytime soon."}
{"_id":"doc-en-vscode-772e21bc17314c9e4dc5f6369579f2601f4804bdaf0f8807780c0bc0c1f32d68","title":"","text":"<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --When we use to jump to definition, would open the definition in a new window, which is inconvenient for me. It would be better if we can choose where to open definition, like in . !\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.492 --- <!-- score: 0.467 --<!-- potentialduplicatescomment --\nNice suggestion. If you agree with this, I'll create a setting so one can choose if 'ctrl+click' always opens the peek view.\nNormally, ctrl+click runs and cmd+ctrl+click opens . With this setting enabled, they both run . I don't think that's expected. Possibly change to ?\nyou can reopen verification-found issues\nUhm, I'll take a look on that. Thanks for your feedback!\nPlease file a bug and don't reopen feature requests\nI created"}
{"_id":"doc-en-vscode-05cfc19a0bc1c4f9c5c5c40ad88a039069c9d859e3e208fe305d0cc3fbb57868","title":"","text":"VSCode Version:1.30.2 OS Version: Applies to Windows and OS X When the cursor flips on line numbers the SVG used has inconsistency's with the system defaults, e.g. tail is shorter Does this issue occur when all extensions are disabled?: Yes"}
{"_id":"doc-en-vscode-a8bd2d93f4ab733a9e9b3237656bb25eb5c539a561813013797a229518a50b40","title":"","text":"developer tools console :bug: see a bunch of errors like this for github authentication, microsoft authentication, typescript language features, remote repositories etc"}
{"_id":"doc-en-vscode-98377c7fd4e808b0441822c960b0d80e6a0150892d2d907095a9432802abff70","title":"","text":"Testing Ideally there would be commands to navigate between changes. And they might deserve an action in the diff title bar. Same as the editor diff experience. !\nWhy is for these commands? I would expect them to be available from the command palette, since in a regular text editor Show Previous Change and Show Next Change both appear."}
{"_id":"doc-en-vscode-b7989d2610640ff5add3cf2df3db1ace969ce712a3da220db5efe49c70f847b0","title":"","text":"Testing Notice that the scroll range is below the decorations, even though they're on screen <img width=\"717\" alt=\"image\" src=\"https://user-\"Meanwhile if I scroll to where they're shown, they aren't in viewport <img width=\"713\" alt=\"image\" src=\"https://user-\"This was on the following gif: <img width=\"401\" alt=\"image\" src=\"https://user-\""}
{"_id":"doc-en-vscode-d9028ec1b3feb7de8443148bd8cac590e407006a9985b9db8b6a3f6288a92e55","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.71.2 (user setup) OS: Windows_NT x64 10.0. CPU: i5- Run command and pick a invalid uri (e.g. path doesn't exist) when having \"large\" (like a few hundreds) amount of recent history entries, vscode will freeze for a few seconds !\nBy the way, having \"large\" amount of recent history entries will also slow down the start up time (both vscode start up and open file time) for a few seconds as well, just tested that with out any history with a few hundreds\nthanks for testing that, do you recall how you got into this state of having many entries? Just by opening folders and files that contribute to the history? If you could do me a favor: open the command palette run the command \"Log Storage Database Contents\" run command \"Toggle Developer Tools\" click on \"Console\" expand the first \"Storage...\" output that is printed there find the entry (see picture below) how many entries do you see? <img width=\"1005\" alt=\"image\" src=\"https://user-\"\nTotal of 671 entries, most of them are temp files opened from things like MobaXterm and !\nGot it, we do have a limit of 500 entries when presented, but I still have to understand the performance issue. You could actually help out by doing a performance trace yourself when this happens by starting performance tracing before invoking the time consuming action and stopping right after."}
{"_id":"doc-en-vscode-8000e9877ce84178fa1c0570dc33b426794527ee7d235d152d542ee4717e213f","title":"","text":"The steps are outlined here:\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\n/verified\nThanks!"}
{"_id":"doc-en-vscode-7899e48b318a8d14ef9b6f3b7f807aa69feba56720325967202d7fd48d0e14df","title":"","text":"When you hover over a type in a TypeScript file, the tooltip that appears has a very low max width, making it difficult to read long type signatures.\nI would include a screenshot but it's not easy to take a screenshot keeping the tooltip on screen!\n! Errors are quite unreadable if the width does not adapt to the font size.\n! it would be nice if the tip used the available space instead of wrapping at an arbitrary point in c# too\n!\n! Same problem using rust's language server.\nGeneral styling control would be very helpful. Also line-height. Another approach would be to have a bottom panel that displays the full error and the tooltip displays a shortened version.\n!\n! When using the auto import feature it is often impossible to differentiate whats what as the import path is cut off due to the limited width of the suggestions window.\nIs this something we can adjust with custom css? It makes snippet infoormation very difficult to read as well. !\nSame in Python. We have to jump through hoops trying to flow doc nicely. And when it appears OK on Windows, it breaks on Mac... ! And it is inconsistent. Ex, hover is OK ! But completion list decides to wrap. What gives? ! Looks like hover respects markdown while completion item doc is plain string and gets wrapped randomly.\n!\nSomeone say long type signatures? !\nHere is a workaround by hacking the file \"C:Program FilesMicrosoft VS Coderesourcesappoutvsworkbench\" Simply append the code above after the original content of file, and restart VSCode. (Title bar may alert a warning message \"Not Supported\" and hacked changes would be overridden by updating VSCode) !\nIt would be nice to have some control on the width. The above does work but vscode presents the message your installation may be corrupt on restarting.\nInstead of hacking the CSS directly as mentioned above, I recommend using the to just override the current CSS. That way when VSCode updates, you can just reload the custom css file that you have specified (personally, I have a project where I hold configurations for my shell, Spacemacs, etc. and this custom CSS file goes there). Until it's officially supported, hacking the CSS is likely the best route."}
{"_id":"doc-en-vscode-89ec49a1050f99de5ea2bce6f357b9c5f6b40464c1905986e619c3b0468b46d3","title":"","text":"The suggested widths above were a bit much for me, so I settled on the following values for now:\nWhat is preventing the devs from providing a configuration option for this? This issue has been open for almost 2 years, and it bugs me on a daily basis. Would love to see it fixed.\n(since you seem like the most relevant person to ping) is anything likely to be done about this?\nI would love to hear an explanation from devs too. What are the difficulties of implementing this? I imagine it's settings a dynamic max-width and adding some more lines to add it in Settings. Are there other design blockers?\nWill you take a look at this? In 3 days it will be a 2 year anniversary for this simple feature request.\nAssigned to and as I remember we are going to have some UX improvements for editor tooltips and so on.\nmaybe now is the time to take this on? current iteration tries to solve parameter hints ux and this is not that much different\n! I employed CSS and put together full instructions: [Custom CSS and JS Loader extension][1] macOS VS Code: VS Code Insiders: Linux: : [1]: Restart VSCode Ignore \"VSCode is corrupt\" error You can choose to suppress this notice forever Run command \"Reload Custom CSS and JS\"\njust as a disclaimer, if anyone modifies the CSS/JS of the app, it comes up at “corrupted” in diagnostic checks, and you’ll typically get pushback if you report issues and include diagnostic information. Or, at least, I received pushback when I had custom CSS/JS enabled, and wouldn’t receive help until I turned it off (even though it was harmless/not related to the issue I was having). While this is indeed an okay workaround, we should just keep making our voices heard so that this is natively supported.\n100% agree. This is just a workaround. It's hard enough to read TypeScript lint as it is! :)\nCurrenlty using VSCode 1.30.2, followed instructions step by step, but it's not working. I don't even get the VSCode corrupt error. I'm running Windows 10 (V.1803). Edit: It was my mistake. For anyone who's having trouble, make sure to run Enable Custom CSS and JS from the command palette. Simply restarting won't work."}
{"_id":"doc-en-vscode-86e49b0fa3280b5d01e2b35e43210f57e78ef3661ddbffe8a9f21c6a2c3691a2","title":"","text":"This issue seems to be fixed in 1.32.0-insider (1.32.0-insider) . I hope this is an intended behavior. I want to thank the developers.\nThis change is related to . I confirmed that the max width of is still small as mentioned above. <img width=\"867\" alt=\"2019-02-22 17 06 19\" src=\"https://user-\"\nUpdate: in , only the hovers for problems and the hovers with code blocks become to have larger width. Others have smaller width.\nHello there,your solution is great,but I want to ask how to get rid of the word at the title.As sort of OCD,sufferring from that. thx a lot!\nThis extension was created specifically for solving this:\nThis has been a long standing annoyance, with an issue that's like 2 and a half year ago. It shouldn't take much effort to have a config property that overrides the width/height of the tooltip, surely? Like, 10 mins of effort? <img width=\"525\" alt=\"Screen Shot 2019-07-18 at 6 41 15 PM\" src=\"https://user-\"\nif it's really 10 mins of effort, then I'm sure you can submit a PR for this? I definitely can't do this in anywhere close to 10 mins, so it would be much appreciated!\nSeriously, I want instead of this: ! At least something like this (with syntax highlighting it will be even better): !\nThere's already been a PR. I really don't understand this. I figure most VScode devs develop VScode in VScode, are they comfortable with the current state?\nfyi, it would be great if the hover could be resized\nHow is this issue 4 years old and such a basic thing still isn't implemented?\nIs there any word on whether we could see this soon? The suggestion widget has had resizing enabled since v1.51 in October, 2020. Is this tooltip widget considerably harder to add this feature for or are there additional considerations to be made? Whether it's a click to resize, or a manual adjustment in settings, could we have an update on the status of this addition? Opening new editor tabs just to properly read code annotations for some more descriptive components on my large monitors does feel like an unnecessary extra step."}
{"_id":"doc-en-vscode-9a07415976f5afd5b6556b1b2063233628b3135a06d6425ab1d9e7498cfb23cd","title":"","text":"might be very related and close to what you did here:\nHi, Instead of configuring the max width as the title says, could the tooltip hover be resizable ? (Also check for more info on what was done with the suggest widget)\nIt's sad to see that it's still not a thing.. I would really like to adjust the height of hover description tooltips.. The big brother Visual Studio does a much better job already. Let VS Code catch up a little..\nanother possible view for this issue\nIt is 2022/11/11. Christmas is close and I made sure Santa knows what I want!\n\"Excuse me, it's 2023\"\nThis one will be my favorite. Copilot suggestions and TS errors just can't be seen in this tooltip. It's time to fix it.\nIf this is still open and no one is looking at it, I'll take a crack at it. Feel free to assign it to me if that's alright.\nHey all, thank you for your patience. A PR has been made which is currently under work but nearing its completion. You may find it here: Once this PR is merged, you will be able to resize content hovers within Insiders.\nWould you add \"Reset Hover Widget Size\" command in Command Palette? Like we have for suggest widget: \"Reset Suggest Widget Size\". It would be good if we could reset the size to the initial one.\nThank you very much. 1 Is it correct, that this would not stay across different sessions? 2 The default max width is very small, is there any consideration, of increasing that?\nWill this feature also allow us to define the window size of a hover provider in a extension above the current limitations?\nHi if is is decided that the hover widget size should be configurable, there will be a command to clear the persisted sizes. This feature is still under discussion. Hi depending on the implementation, it is possible that the persisted sizes are persisted or not across different sessions. This is still under discussion. Currently the implementation is such that the maximum natural width is the same as before. Hi the current implementation does not provide this functionality. This could be addressed in a separate issue as it is related to the usage of the hover widget by extensions. The current PR that is worked on is the following:\nThanks a lot."}
{"_id":"doc-en-vscode-3bf8dd03e36acfd2d438406838cdebcf802d1d799ace9e956196a2c6b61e6046","title":"","text":"And do you consider increasing the default size, that is unfitting to most real world scenarios? The windows is so small, that I feel the need to increase it even on my WXGA Laptop.\nHey all, thank you for the patience. A PR has been merged in order to make the content hover resizable. The change should be visible in tomorrow's VS Code insiders version and in the next stable release which will happen in the coming two weeks. There currently is no mechanism to save the content hover - we are discussing how best to approach it. You may track the issue on persisting the hover content here: we are currently not thinking about increasing the default maximum size. However it is possible that we implement the persisting mechanism as follows: when you increase the content hover size to a specific size, this size is persisted, and it is applied as maximum default size on the next hover appearances. This potential implementation is still under discussion by our team.\nJust want to say thank you you seem to have been tackling a lot of these high-level nitpicks that have been in the VSCode backlog for years, and for that I am very appreciative.\nI can't find a command to clear/reset hover widget size, is this has been ?\nHi currently the command does not exist because a persisting mechanism has not been put into place as of yet. It will appear when the latter is .\nOh okay, thanks a lot"}
{"_id":"doc-en-vscode-0cb9646b57142263ece28735a1b745cd369e60c44caaf290fee150aeddcba7f8","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- Version: 1.46.0-insider Commit: Date: 2020-06-01T10:32:27.369Z Electron: 7.3.0 Chrome: 78.0.3904.130 12.8.1 V8: 7.8.279.23-electron.0 OS: Darwin x64 19.5.0 Steps to Reproduce: a new untitled notebook (using VS Code Python extension) you add a cell or the like, then VSC will invoke the save method on the Problems Adding cells should not invoke save (this happens when user has auto save turned on) Errors thrown in the method are ignored. VSC will assume the notebook was saved and mark it as not dirty (the dirty indicator doesn't exist on the tab anymore) I don't think VSCode should call the method for untitled notebooks. FYI - We create untitled notebooks with a scehme . If this isn't the way to indicate an untitled file (scheme), we might need some API or the like in the notebooks implememtation. /cc\nGood catch, untitled file should not trigger save until users explicitly run Save, which will invoke SaveAs.\nWe should also improve the overall experience for creating untitled notebook documents, they can be created in two approaches: Users press cmd/ctrl+n to create an untitled file, switch to another view type (tracked in ) Users run a specific command to quickly create a notebook untitled file. command should take view type as arguments. cc we may also want to adopt this in Custom Editor world.\nWe now support opening a untitled file with specified editor view type through the existing new untitled file command (in master, and tomorrow's Insiders) When the command is executed, is invoked with an URI with scheme. is invoked only when users save so no auto save for untitled files as usual."}
{"_id":"doc-en-vscode-758d066a24bf4d74fd532995d74e281ebd2ae41be1f2ee28d160958ee9c8c476","title":"","text":"I find it strange that the entire area where the two action bars atop of notebooks are is focuseable: <img width=\"1007\" alt=\"image\" src=\"\"Other action bars across the workbench aren't focuseable, only the actions themselves within."}
{"_id":"doc-en-vscode-ebaa529fe9bcf3151e99c2309f8970f0b58ac785f7c7d4f6c57c4d403d8aa011","title":"","text":"debugging the loaded scripts view open debugging, notice exception in the console can not reproduce with Stable so it might be that my listener refactoring caused this If you pause on exception you will see that caused it.\ndid you have a chance to look into this? Or would you like me to investigate? I suspect this is breaking the loaded scripts view and is the root cause for\nno, please investigate since this and was working before, and I haven't touched this code for two weeks.\nend session events are not arriving in As a consequence the tree for the session is not properly removed.\nI see something differnetly, the end sessions are arriving however when the last session end arrives view starts refreshing the tree and while that is in progress the view gets disposed which throws an exception. Looking into a fix...\nI have created a PR which address this issue"}
{"_id":"doc-en-vscode-88108b8e7cf944d1c0e71835ba76403f2a64eb0c4e074d0ef469260796f709af","title":"","text":"VSCode Version: 1.17.2 OS Version: Win10 Steps to Reproduce: Terminal to bash <Ctrl+<Shift+<Pand type \"run active file\" command not work because path to the file is windows-style and <.is missing in front.\nI would like to fix this.\nLooks like the util was introduced in Windows"}
{"_id":"doc-en-vscode-e9d5c8864a2c9ddc9c22cff53701f6661f934f67cba2cd8839bdf2bf65d49602","title":"","text":"Repro Steps: install VS Code native menubar has no keyboard shortcuts Alternative Repro Steps: a keybinding the menubar does not reflect the change.\nMust fix.\nReviewed, tested, merged and cherry picked to . See\nsince you tested this can you please add a verified label"}
{"_id":"doc-en-vscode-c519faf8547708dbe339a490d02f8fcd7c27a066f96c038b59159e756e3fcebb","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.82.2 OS Version: MacOS 12.0.1 Steps to Reproduce: our extension (Evidence), create a new project using the VS Code command palette which pages open automatically text file is opened automatically on the left walkthrough is opened automatically on the right This was working until one of the more recent VS Code versions I believe. text file is opened in full walkthrough is opened in full, taking over the editor pane from the text file (i.e. you can only see the walkthrough in your editor) This is the function we use to open the walkthrough: Where maps to I was under the impression that the in this function will open the walkthrough \"to the side\", which was happening successfully before, but is no longer working.\nI had an older version of VS Code running on another computer and have confirmed it used to work in VS Code version\nshould this command be documented here\nI did update our docs last week to include this command. Waiting for the newer changes to be published."}
{"_id":"doc-en-vscode-29cb54e56f0ebcee7c5732001e0ba4f7a1a2bfff8d460e3c8ced274ae60f208a","title":"","text":"have an eager comments controller extension (see below) open a notebook create a new cell leak warnings are printed\nThis seems to be rooted at  It looks like a new is created without always disposing the old one. The looks highly suspicious. <img width=\"1586\" alt=\"Screenshot 2022-10-26 at 14 09 41\" src=\"https://user-\"\nSuggesting to tackle this for October because after a short while VS Code was crawling due to excessive listening\nthanks for good reproduce. The comment thread equality check is wrong, it's an oversight during refactoring iirc. I pushed a fix for this memory leak. Though with it fixed, I found there are issues with \"Editing\" state of comment thread for cells (focus issue), will look into this next week.\nThanks"}
{"_id":"doc-en-vscode-70e8740fdb0d04437b9d71f8e90e47c6ed22bea44eb4632f0c5fdcc1f7a77cd3","title":"","text":"a folder in WSL and show logs the error log !\nAre things broken in this case or just a bad stack trace. I'll take a look to see if I missed a module include somewhere\nJust a bad stacktrace from what I see, though I didn't intentionally go test things thoroughly"}
{"_id":"doc-en-vscode-4e6563fa45f54aa16806424a8a96b1da5421ee03a78e53e45e380c0b5a15e30a","title":"","text":"It's used in , this would be easiest after that's merged.\nI wish instead of explicitly checking whether proposed API is enabled in in the implementation, we could check it when checking distro for whether an extension is allowed a proposal instead.\nthe list of allowed extensions is not available in code - oss though on local or in CI\nThat's fair but do we really need to check in code - oss when developing? I would think is being tested out in the final build that's published.\nnot my area but being strict is good for proposals and it's good to align oss with the product in areas like this. Consider a fork of VS Code that does not have the check, extensions could use the proposals without warning which could lead to easy breakages."}
{"_id":"doc-en-vscode-7697a2845c8e189d279c8370f3b7cb31a3648776392bcbc42d5df2f6c6e6025a","title":"","text":"Testing Add a git pre-commit hook that sleeps for 120 seconds Make a change to a file, and commit the change Switch to another VS Code window Expected behaviour: Window should be active while the source control operation running Actual behaviour: Window becomes inactive even though the source control operation is still running\nCan this be adopted in SCM? Holding an activity lock is fairly easy, just call this method\nthese commands are being contributed by the git extension. Should be wired into the so that in case there is an on-going progress indicator the window is considered active?\nYea, I think that's reasonable, thanks for the suggestion. Will take care of it"}
{"_id":"doc-en-vscode-ec7c60cbbe2ae18852e7eb41c76215a30c0be400463b29445757859fb60601c8","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.69 () OS Version: macOS Ventura (13.0) developer beta 2, Firefox 102.0 Steps to Reproduce: an Insiders Codespace with any repository on Settings Sync until extensions are installed many walkthroughs being opened Here is a video demonstration: Something else that I took note of (also seen in the demo video) is that: Sync is turned off for every new codespace I create the extension setup, the page reloads every time in some stage of the process (this could be expected though) I have tested this both on Brave and Firefox with various different repositories. Original issue:\nThe respective component owner has to opt into syncing the state.\nThis user is stating it only happens in codespaces and not on or I'm not sure if settings sync is treated differently here but we do sync walkthrough state\nYes indeed, In addition, this does not happen for me in 1.68.1 (stable).\nCreated a PR fixing this Can it be cherry-picked to the release branch before tomorrows release :pray: ? This also affects workbench theme as it flashes from default theme to the theme from settings sync which is really annoying, in 1.68 workbench starts with the theme from sync server directly cc\nIt seems the newly created codespace is not getting initialised with synced data because the workspace says it is not new."}
{"_id":"doc-en-vscode-2c274158c1bf5ba81ff2f683f0e258f902eab97489d0a36c5836c90fbcd20320","title":"","text":"Issue Type: <bBug</bnew VSCode window new Terminal terminal: terminal: in Cmd and mouse over the two listing name. Only the /tmp/Foo.h will be clickable. won't be because of the . VS Code version: Code 1.43.0 (, 2020-03-09T19:34:44.548Z) OS version: Darwin x64 19.3.0 <!-- generated by issue reporter --\n/duplicate"}
{"_id":"doc-en-vscode-33089335b232e03dff4be3f0764089149035a63ce0a589e13ad11fc6597e95b1","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Not sure --the only reliable way I know how to reproduce it is by creating a spelling error with Code Spell Checker. If anyone has any tips how I can try to repro this with all exts enabled, I'm happy to take that on and post a new GIF. <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --Version: 1.59.0 (user setup) Commit: Date: 2021-08-04T23:13:12.822Z Electron: 13.1.7 Chrome: 91.0.4472.124 14.16.0 V8: 9.1.269.36-electron.0 OS: WindowsNT x64 10.0. Steps to Reproduce: exactly one information status to be reported in the status bar and observe the tooltip tooltip will read instead of !\nIt shows up in plural form for all 3(errors, warnings and infos) even if only 1 of them is present. Source\nIf you want to implement different cases, this would involve having separate translation strings for the singular and plural, as well as a third case for the 0 value. But is it really necessary? One could easily avoid the grammatical discussion if you put like this: , , .\nSeems very reasonable to me\nThis looks better and is understandable\nCan I take on this issue( If it is not assigned to someone else )? It looks fairly simple to me. But as I'm quite new to this so I might need some guidance.\nI have made the changes to correct the typos. Should I make a PR or it is already completed?"}
{"_id":"doc-en-vscode-3a62f9c9a4270e6e5abcbd5e68d29e7a352e260cbad50438b54a63d72d3a41f2","title":"","text":"This looks like a bug in VS Code, I'm seeing a lot of weird behaviors Basically sometimes the 2nd progress is not visible (but when i scroll or add a new cell it magically appears) The output is not displayed at the end (but when i scroll or add a new cell it magically appears) Originally posted by in\nVerifying the corresponding bug on jupyter and I found there's still a problem. VS code shows this: ! Jupyter shows this: !\nThe difference being there seems to be extra lines in the VS code output.\nSee"}
{"_id":"doc-en-vscode-1d6b0caa3612d45213309d85e8b2d0763cd21e83d4c1ee436bc7d6ba6790fe91","title":"","text":"Open VS Code insiders Open the following ipynb file in vscode <details<summaryipynb</summary</detailsOpen the same notebook sideways (on the right) in a text editor Now set focus to the text editor (on the right) and hit the save key (CTRL+S) Watch how the error output in the notebook (on the left) gets duplicated !\nThe reproduce steps are very impressive"}
{"_id":"doc-en-vscode-177dcaaacbb35d31f983e15f144a8f72e9bd0f12e20de289e14f805fa13cec31","title":"","text":"Type: <bBug</bWhen first opening vscode or reloading the window several context keys are set to incorrect values. These values may eventually be corrected when focusing another editor and returning to the original one. The keys I have noticed with this issue are: It can be a little tricky to catch some of them misbehaving. It depends on which editors you have open and how they are arranged. ! Steps to repro: Open 2 editors Focus the first editor Pin it Record context keys Reload window Record context keys Diff the two sets of context keys and notice they are not the same VS Code version: Code - Insiders 1.86.0-insider (, 2023-12-18T05:36:59.983Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </detailsExtensions: none<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nBen, maybe you could check for the ones in editor area since you recently fixed some issues there. Maybe its a regression?\nThe following context keys were not being updated on reload: Updating them now on reload\nThis seems to apply to a few additional context keys outside of . Just a heads up in case closing the issue was accidental."}
{"_id":"doc-en-vscode-b80736ebc52d2412b7980d255febb9e8805c328ca6f7248894b6b29ac08d02d6","title":"","text":"I found the easiest to reproduce is with this integration test: This correctly opens a group at the ninth column but the scrollbar is still showing as if it was not scrolled even though it is. !\nVery likely an issue in et al.\nJust submitted a PR.\nI see the scrollbar on the right, but it also looks like there is a second shorter scrollbar behind it <img width=\"1042\" alt=\"image\" src=\"https://user-\""}
{"_id":"doc-en-vscode-c92a778e67fea31d57c36db9115fb2e908543df3fc826c74b0323213aa08f28e","title":"","text":"Logging errors from the file service I notice that the terminal seems to stat the terminal prompt multiple times: ! Notice the at the end of the file path.\nWhen you hover a line in the terminal links are resolved, one of the link providers gets confirmation on whether it's a file or directory and gives a high confidence link (underline shows when you hover without cmd) and on macOS we treat as a valid path character. This is how it's always worked. but we could cache this stat result with a TTL (10s?) so it doesn't spam so many stat calls and will be able to return the link faster on successive hovers."}
{"_id":"doc-en-vscode-1651017ffe87aa784ae4ebfcef780fe7ea35b591764df08d92554f69fc6e3429","title":"","text":"Type: <bBug</bThe keyboard focus doesn't move to the \"SOURCE CONTROL GRAPH\" view, it only opens it, but without actual focus (can't move using the keyboard arrows between commits or open their context menu via keybaord) VS Code version: Code 1.93.1 (, 2024-09-11T17:20:05.685Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (8)</summaryExtensionVersion ------ vscode-markdownlint0.56.0 prettier-vscode11.0.0 vscode-pull-request-github0.96.0 rainbow-csv3.12.0 vscode-docker1.29.2 mssql1.24.0 remote-containers0.384.0 sqltools0.28.3 </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --"}
{"_id":"doc-en-vscode-1dae6f941cf30adca389d26a271b467dd2f1a68ad816c7fbbc181020328a461f","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --I saw and created this issue anyway because I think that issue is a) too old b) has too many other things going on. <!-- Describe the feature you'd like. --For me, personally, I'd like a simple way to enable/disable automatic extension updates per extension. I really care about updates for many of them and want to read the changelog before I update but for some, like Copilot, who get several updates a day, I really just want to always be on the latest. Copying section 7 from the above: Users could opt-out of Auto-Update for specific extensions, with toggle button next to Disable/Uninstall in Extensions Sidebar (under gear icon menu) and next to those buttons in Extensions Detail Page. This could allow users to stick with previous versions before new high-risk permissions became required This could allow avoiding updating to problematic versions of extensions until issues are resolved. This could allow users to manually review/approve updates based on reviews and changelog for higher (security or reliability) risk extensions This could be controlled per extension without disabling globally as may be desired by default for most extensions. Can be set per extension to \"Default\" vs Auto-Update vs. Disable Automates, like with Firefox, with Default behavior controlled through global setting. Could provide Undo Update button or choose from version history (like with Chrome/Firefox extension stores) on extension details page, to enable rollback to previous version after an update causes issues, instead of just disabling until if/when ever fixed.\nHave you tried to select specific version for extension? This should effectively force version and prevent updates.\nNo, would that still show if there was an update available?\nIt should show. Also September 22 update had few updates to including . Check it out\nSupported auto updating options per extension/publisher <img width=\"550\" alt=\"image\" src=\"\"This is enabled when auto update mode is disabled or set to (new config)"}
{"_id":"doc-en-vscode-11cd8e5dab524da3a34278b974837bc9ac63bb2b1320a5060c00289b4e7e223f","title":"","text":"VSCode Version: 1.6.0 OS Version: OSX 10.11.6 *.ini files comments-line starts with the SEMI-COLON. but VSCode has insert (or remove) the HASHMARK.\nPull request pushed, thanks a lot !"}
{"_id":"doc-en-vscode-7ced7a13217b90c1f733d9052c6bd3da322aff29c880d8bea39fe00ae9cf094c","title":"","text":"VSCode Version: v1.2.1 OS Version: Windows 10 Steps to Reproduce: file type to INI \"A=b;xxxx\" in file content Expected: \"b;xxxx\" colored as white Actual: \"b\" in white but \";xxxx\" in green \";xxxx\" should only be regard as comments when there is no other contents in the line before \";\"\nThis issue was moved to demyte/language-cshtml"}
{"_id":"doc-en-vscode-a810a7d6ab2e02b9c694c05b5f6e768c0db1e50e96f813f0a4579bd2a2a6d1c8","title":"","text":"VS Code Version: insiders/from sources Task terminals don't respect setting so decorations cannot be disabled.\nthanks for creating this issue. I remember noticing this then got distracted by some other issue\nwe're separating the task decorations from terminal decorations (shell integration), so will have a setting\nVerified that works"}
{"_id":"doc-en-vscode-7561593a0a1ccd1334587676b95dcc6803f7431622802f331cf2018b6cc1e760","title":"","text":"fyi\nVerification: opt into edit sessions using , then ensure the edit sessions view does not appear in the sidebar until you run the Show Edit Sessions action in the command palette."}
{"_id":"doc-en-vscode-3faa052f93240fc8e82cad95c03980c3dab72a579a8fa39fd7e0ce5c86571bbc","title":"","text":"Issue Type: <bBug</bWrite http://192.168.1.7:3000/ into a text file. Command-click it to try to open the link. You will get an alert asking whether you want to open the external website. Click “Configure Trusted Domains”. <img width=\"1025\" alt=\"link and dialog asking to confirm opening external website\" src=\"https://user-\"Look at the second option in the menu. It says “Trust 1.7:3000 and all its subdomains”. But 1.7:3000 is not a valid address. <img width=\"1024\" alt=\"link and menu with buggy entry “Trust 1.7:3000 and all its subdomains”\" src=\"https://user-\"I can only guess that VS Code thinks that “7” is a TLD and “1.7” is a full domain name. What I actually want to do is whitelist opening all links to IP addresses like . In other words, white-list the subnet . At least, I think that makes sense. But I am not a network engineer, so if it makes more sense to recommend to the user whitelisting some other subnet such as , or whitelisting only the address but on any port, that would be fine too. VS Code version: Code 1.40.1 (, 2019-11-13T16:47:44.719Z) OS version: Darwin x64 18.7.0 <!-- generated by issue reporter --"}
{"_id":"doc-en-vscode-6824c34f24939b6a04008daa346765745374586f725a840a80bbeb406ef8d33a","title":"","text":"Type: <bBug</bPinning a terminal editor does not set some of the context variables that are normally set when pining a text editor. Here are the ones I've noticed: activeEditorIsDirty activeEditorIsFirstInGroup activeEditorIsLastInGroup activeEditorIsNotPreview activeEditorIsPinned In the case of activeEditorIsPinned this makes it impossible to detect if the current terminal editor is pinned. Since there is no command to toggle the pin on an editor this means it's not possible to create a single keybinding that can intelligently pin and unpin the current terminal editor. In the case of activeEditorIsLastInGroup it means it's not possible to create a keybinding to move a terminal editor out of the \"pinned section\" of editors in the current group. When editors are pinned they are moved to the left side of the current editor group. When moving an editor to the previous position in the current group if it happens to cross into the pinned section it will become pinned. But the inverse is not possible: you cannot move a pinned editor out of the pinned section unless there happens to be something else unpinned to the right of it. So I have a keybinding to unpin the current editor if it's the last editor in the current group. Since terminal editors don't set the required context they break this binding. Steps to Repro: Open a terminal in the editor area Pin the terminal editor Inspect context keys Notice activeEditorIsPinned is not set VS Code version: Code - Insiders 1.76.0-insider (, 2023-02-10T16:22:19.445Z) OS version: WindowsNT x64 10.0. Modes: Sandboxed: Yes <details<summarySystem Info</summary </detailsExtensions: none<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nThe type of editor should not play any role in this. I somewhat doubt this is a bug in our code, so setting label that we need confirmation. Open for help, investigation, suggested fixes, PR.\nThis one I cannot confirm or reproduce. See this gif where I open a terminal, pin it and then inspect the context keys: ! What are some exact steps?\nLooks like it depends where you click when inspecting context keys."}
{"_id":"doc-en-vscode-db121aa69e1c340312ae715f0da98c8be42328e829ec45d9750fd74745cccf5c","title":"","text":"I'm clicking inside the terminal, where my cursor is, because I'm attempting to make keybindings that work when using the terminal. You're clicking the editor tab. So it appears the context is set for the editor, but when you focus the terminal within it that context is removed. Interestingly I still see some of the context so not all of it is removed.\nQuite likely that the terminal itself sets its own context keys.\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\n/verified, but while confirming the fix I noticed more issues with these context keys"}
{"_id":"doc-en-vscode-59089056b165d9c8242d2160114f4e771709902bce4dc5f20249c35da043d968","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --Version: 1.85.0-insider OS: Linux (codespaces) Steps to Reproduce: in the terminal prompt is contributed via env collection by Python extension. Normal terminal: ! Hovering says that collection is applied. Hidden terminal: ! Although it says that no environment collection is contributed upon hovering, collection is still applied as can be seen in the terminal prompt.\nit seems like this is actually a bug where the initially hidden terminal doesn't show the applied environments but should? Making a terminal hidden was just meant to be about visibility, not environment?\nI'm fine if we change hidden terminals to also apply environment collection, as long as we get support for Python extension decided not to send activate commands to hidden terminals: , as hidden terminals are considered private terminals belonging to a specific extension. One could probably argue similarly for not applying environment collection to such terminals.\nBased on offline discussion we're continue to go with NOT applying environment collection to hidden terminals, due to it affecting remote scenarios.\nWe investigated that env collection from shell integration in unix is still being applied on hidden terminals, i.e is still being applied when it should not. Any further updates?\nCode to disable for terminals: Where vars are set:\nI cannot reproduce this on Windows or macOS. Steps: Code - OSS a terminal and verify the environment does not include terminal-sample from vscode-extension-samples via symlink a terminal and verify the environment does and select the new terminal and verify the environment does not include Maybe you had the variable set in the parent environment (main/server?) which is why it showed up?"}
{"_id":"doc-en-vscode-116c710961a069e936b38e5b4951ab3bc93891930babd85e23beb57996743e58","title":"","text":"Please try it on Linux (codespaces), stated in the original report, this issue doesn't exist on Windows. And I don't have a macOS to verify.\nThanks, missed the remote bit\nSeems like this behavior has now changed with: So we should either reopen this one or , to accomplish\nyou can accomplish by using\nI see, suppose if is , but no is passed explicitly, I assume system environment variables and shell initialization scripts are still applied? What does window's environment mean here and will it be applied if is , but no is passed.\nWe don't have any environment to provide, all we want to do is skip applying env collection while the rest of the environment stays as-is, can you guide us on how to exactly use to achieve that? Also, we would expect shell integration to be active.\nbasing the environment on should do the trick\nGotcha, will try that. I assume shell initialization and integration scripts are still run?\nlooking at the code, I think shell integration does activate even when strictEnv is set\nWhat are some verification steps for this issue? For example, how do I get Python deactivate to show up?\nsteps at , remote only\nActually we reverted the change made in this issue with , so even hidden terminals have environment collection applied by default now.\n:+1: So the steps are: a terminal and verify the environment does not include terminal-sample from vscode-extension-samples via symlink a terminal and verify the environment does and select the new terminal and verify the environment does include\nHow do I do step 3? Am I opening the terminal-sample workspace after cloning and creating a symlink to it?\nyou can build the extension and do it in the insiders extension development host"}
{"_id":"doc-en-vscode-b87c95d5285f17e629520d933860f1c075fa59783fed8c173cba2dcd37cc577c","title":"","text":"Testing comments view open and scrolled to the top of the file enter in the node :bug: I get scrolled to the first comment\nThis also happens if I expand the node after collapsing it, which I found a bit surprising, as I didn't hit enter on a specific comment, just the file node\nThere are a couple problems here: We should only expand/collapse the file node when the twisty is clicked/entered (we should have both an action and an expand/collapse with the file is clicked/entered) We should just open the file and not scroll to a comment when the file node is clicked/entered.\nActually, since you're opening the file from the Comments view, it makes sense to reveal the first comment. I'm just going to fix the expand/collapse + action issue.\nI just verified this by making sure that clicking on the node doesn't expand/collapse the node (only the twistie can). Please lmk if that's sufficient!\nThat is sufficient!"}
{"_id":"doc-en-vscode-414b82276d31d1c832a99671457be8b6e7992187649aad19fdd74001da2421b7","title":"","text":"Performance Issue When using VS Code to debug a java program with many threads (~500) the UI becomes generally unresponsive and sometimes freezes for about 10-20 seconds when hitting/continuing from a breakpoint. The problem seems to be that VS Code tries to refresh the \"Call Stack\" very often (and each refresh takes 100-200ms with that many threads) VS Code version: Code 1.20.1 (, 2018-02-13T15:23:28.607Z) OS version: Darwin x64 17.4.0 <details<summarySystem Info</summary </details<details<summaryWorkspace Info</summary</details<details<summaryExtensions (10)</summaryExtensionVersion ------ vscode-eslint1.4.6 tslint1.0.28 prettier-vscode1.1.3 python2018.1.0 debugger-for-chrome4.1.0 java0.19.0 lwc-vscode0.11.2 perforce3.1.0 rewrap1.7.1 vscode-java-debug0.6.0 (1 theme extensions excluded) </detailsReproduces only with extensions VS Code example stack trace when freezing: <img width=\"349\" alt=\"stack-trace\" src=\"https://user-\"\nfixes this\nonce the insiders build is out on Friday you can try this out with your PR and let us know if all is good now - thanks\nWill do, thx!\ntried with the 2018-03-09T05:17:20.534Z insiders build and it worked great\ngreat, adding verified label."}
{"_id":"doc-en-vscode-87f9546f36761c32bc5d67f300e443e301c6f8756ac209e1d4612942d469452b","title":"","text":"Type: <bBug</bCC [I cannot easily find my authored issue so I am creating a new] Ctrl+LeftArrow and Ctrl+RightArrow does not work in the Output View (Ctrl+Shift+U) area. This happens on Windows. This means screen reader users cannot navigate word by word within the output view. Ctrl+Shift+U to focus in the Output View Choose any output (e.g., terminal) if you can navigate word by word via Ctrl+LeftArrow or Ctrl+RightArrow VS Code version: Code - Insiders 1.92.0-insider (, 2024-07-12T05:08:08.550Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (125)</summaryExtensionVersion ------ vscode-openapi4.27.0 zotenote1.0.1 android-dev-ext1.4.0 aiprm-lang0.0.2 Bookmarks13.5.0 openscad1.2.2 spellright3.0.134 zoterolatex0.4.1 mermaid-markdown-syntax-highlighting1.6.0 external-pdf1.2.0 doxdocgen1.4.0 vscode-markdownlint0.53.0 vscode-eslint3.0.10 vscode-quick-select0.2.9 vscode-deno3.38.0 gitlens14.6.1 EditorConfig0.16.4 prettier-vscode10.1.0 figma-vscode-extension0.3.5 vscode-firefox-debug2.9.10 shell-format7.2.5 vscode-google-translate1.4.13 codespaces1.17.1 copilot1.213.0 copilot-chat0.18. remotehub0.62.0 vscode-github-actions0.26.2 vscode-pull-request-github0.93."}
{"_id":"doc-en-vscode-ae252e6a66b4d6fa4b87e15d761683bbfcc40124ef05ca0a7b3c8ed9aa4d7511","title":"","text":"cloudcode2.14.0 overleaf-workshop0.13.0 cslpreview0.2.2 path-autocomplete1.25.0 latex-workshop10.1.0 lilypond-syntax0.1.1 scheme0.2.0 better-cpp-syntax1.17.2 commitlint2.6.0 language-julia1.90.1 google-search0.0.1 vscode-lua-format1.3.8 lilypond-formatter0.2.3 lilypond-pdf-preview0.2.8 lilypond-snippets0.1.1 vslilypond1.7.3 language-matlab1.2.4 zotero0.1.11 git-graph1.30.0 azure-dev0.8.3 vscode-azureappservice0.25.2 vscode-azurecontainerapps0.6.1 vscode-azurefunctions1.15.1 vscode-azureresourcegroups0.8.3 vscode-azurestaticwebapps0.12.2 vscode-azurestorage0.16.0 vscode-azurevirtualmachines0.6.5 vscode-cosmosdb0.22.0 vscode-docker1.29.1 vscode-edge-devtools2.1.5 black-formatter2024.3. debugpy2024.9. flake82023.13. isort2023.13. python2024.11. vscode-pylance2024.7.1 jupyter2024.6. jupyter-keymap1.1.2 jupyter-renderers1.0.18 vscode-jupyter-cell-tags0.1.8 vscode-jupyter-slideshow0.1.5 remote-containers0.377.0 remote-ssh0.113. remote-ssh-edit0.86.0 remote-wsl0.81.8 vscode-remote-extensionpack0.25.0 azure-account0.12.0 azure-repos0.38.0 cmake-tools1.18.43 cpptools1.21.0 cpptools-extension-pack1.3.0 js-debug-nightly2024.7.517 powershell2024.3.2 remote-explorer0.5. remote-repositories0.40.0 remote-server1.6."}
{"_id":"doc-en-vscode-60d4b51d141b35fdab9885aee7e94b691aa9c911335227875446358b251a9b66","title":"","text":"vscode-github-issue-notebooks0.0.130 vscode-node-azure-pack1.2.0 vscode-selfhost-test-provider0.3.25 vscode-serial-monitor0.12.0 vscode-speech0.10.0 vscode-speech-language-pack-ko-kr0.4.0 vsliveshare1.0.5932 windows-ai-studio0.3.3 autodocstring0.6.1 pandocciter0.10.4 publisher1.1.3 shiny1.0.0 shinyuieditor0.5.0 quarto1.113.0 r-debugger0.5.4 java1.32.0 vscode-xml0.27.1 vscode-yaml1.14.0 r2.8.4 multi-command1.6.0 AudioQ0.0.2 vscode-deepl1.1.1 abc-music0.4.0 lua3.9.3 latex-utilities0.4.14 cmake0.0.17 vscode-terminal-here0.2.4 windows-terminal0.7.0 errorlens3.16.0 intellicode-api-usage-examples0.2.8 vscodeintellicode1.2.30 vscode-conventional-commits1.25.0 vscode-arduino0.6. vscode-java-debug0.58.0 vscode-java-dependency0.23.7 vscode-java-pack0.27.0 vscode-java-test0.41.1 vscode-maven0.44.0 zotero-cite0.8.6 markdown-all-in-one3.6.2 grammarly0.25.0 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\ndo you know why this isn't currently supported?\nIt seems to work for me in Mac, are you seeing it otherwise?\nIt does work for me, although sometimes with the screen reader enabled, it's reading the word then a whole chunk of text. I'll try on windows.\nIt’s not working in Windows.\nCan reproduce on windows. I see that this is happening because there's a part of the when clause for and of for windows\nRemoving that from the clause fixes it. Looks like that condition was to fix this can you try removing that condition from the clause and seeing if you experience any consequences of doing so?"}
{"_id":"doc-en-vscode-2dcf4b807bb6d42d2876abf632786c248171f40ac25ee58bbaaf478ea0ec4ed6","title":"","text":"This looks and reads a bit weird to me: <img width=\"262\" alt=\"image\" src=\"https://user-\"This does not really do anything when I am in Column Layout which I think is even the default.\nCan context menu entries be hidden when a context key condition fails?\nI think or the clause of the command, I always forget: knows for sure\nprecondition just disables that entry, it doesn't hide it\ntry with:\nIt is the clause that defines if an item appears in the menu\nIn addition to hiding this when it doesn't apply: can we find a better solution for when it applies. Can we have for column mode Show Base for mixed mode (radio-logic for these two) Show Base At Top Show Base In Middle\nFixed <img width=\"167\" alt=\"image\" src=\"https://user-\"\nI still find this weird: <img width=\"198\" alt=\"image\" src=\"https://user-\"Throwing in some ideas: Column Layout I would think just a toggle for the base is fine so that it becomes a single entry. Mixed Layout A radio group of 3 entries: \"Hide Base\", \"Show Base Top\", \"Show Base Center\". Alternatively if we worry about too many entries, we could also do a submenu for \"Base\" specifically with these options.\nJo's suggestion in is also fine, but I think we need that third entry to allow for hiding when in mixed layout.\nVerification steps: Open a merge conflict and verify these menu entries: ! !"}
{"_id":"doc-en-vscode-64a4cb95ef1bf8a8117e1716b48cd838df1850a09087b2d9a74b0e396a99579c","title":"","text":"Upstream:\nIt's not true that it takes very long. It does not get visible at all. I have to downsize the terminal pane to a certain size to make it visible again. In my case it's about 5-10 lines. Seems to me that the canvas is not resized properly. There are no problems if I disable GPU rendering (code --disable-gpu). Also without any noticeable performance issues. Would be nice to have a flag in the preferences to disable GPU rendering as a quick workaround. One of the reasons for using code (instead of Atom or Sublime) was the integrated terminal which seems to be broken since the new canvas rendering was introduced. My System Config: OS X 10.11.6, GPU: NVidia 8800, 4GB RAM\nthat will probably get fixed as part of this as well, it's probably something to do with your GPU/Chromium not liking the texture sizes above a certain size."}
{"_id":"doc-en-vscode-ae7fd0746dcee3d13b58f35a8969bb76979302370ab37218d39f365be68eba63","title":"","text":"enable sticky scroll open a diff editor in a TS file toggle inline diff observe that the original line numbers are garbled We could disable sticky scroll in the original editor when the diff editor is rendered to be inline here --"}
{"_id":"doc-en-vscode-a2861c57dca3691d1b883ccf5b9e5371f5f4d7c61d7165c2c9696c802b46a603","title":"","text":"Type my long commit message Type and suggest widget for issues pops up aggressively Press esc to get rid of it All the commit text is gone I think this is new?"}
{"_id":"doc-en-vscode-df5c411510b3a4e641e11d57a5a867915378359f229560104c015b4599ea9de3","title":"","text":"open explorer select a file press and to copy paste a file in place :bug: nothing happens, in the past I would get a copy of the file\nHmmm. I cannot repro this and am able to copy and paste. Do you have any custom keybindings? What is your setting?\nIt's the default, this is from the keybindings log\nHmmm I wonder what has happening with your keybindings. Here's mine triggering the explorer specific ones with the proper when clauses. Is your explorer in a non-default position (i.e. not the activity bar)\nThis gets more and more interesting... The keybinding is there and not (re)configured but it looks like something is wrong with context keys. The seems AWOL. Maybe that's because I moved the explorer around very much last endgame? <img width=\"2819\" alt=\"Screenshot 2023-03-01 at 16 55 16\" src=\"https://user-\"\nThink I found the bug. the folders view to the right sidebar the remaining explorer view container in the activity bar Can't copy and past in the explorer anymore\nI know this is very early into the next release cycle, but could this be released in a patch to stable VSCode? I have been bitten by this in stable VSCode after the February 2023 update, by dragging Explorer over to the secondary sidebar. Or is there a workaround to restore functioning of these keys, other than resetting the user profile altogether? I've tried dragging the Explorer menu back, and also right-clicking and selecting \"Reset Location\". Neither seems to \"restore\" the functionality of these keys. This affects any command using in , such as deleting a file with the Delete key (hopefully people find this issue instead of creating a new one for every such broken key). The workaround I've managed to apply is to manually remove from every expression possible, which may have side effects, but it works in a pinch. Copy all the keybindings below (expand for details, grab bindings between the comments) into your to fix this in a pinch. <details</summary</details\nDoes the command in the command palette not fix it for you? This will reset all your views to their default positions.\nMy keybindings workaround bypassed the issue on my home computer yesterday."}
{"_id":"doc-en-vscode-764ad67c787ceb7e02550b5255a3f24b45cba49901474e178433105cbe0eeb6e","title":"","text":"Now I'm at work (both are Windows machines, VSCode up to date), and I went to try your approach. However, I have a lot of menus in my secondary sidebar, so I didn't want to on my main profile, so I did the following: Ensure my workaround from before is commented out Export the profile as Import the profile Switch to that profile Verify that the key doesn't work First, try the less invasive, right-click on Explorer button in secondary sidebar and click key, and others, works again, even when putting the Explorer back in the secondary sidebar Curiously, switching back to the profile, the key, and others, works again, even when putting the Explorer back in the secondary sidebar Never had to go the nuclear option and So, some funny business related to one of the steps above, probably the settings profile shuffle, restored basic functionality of those keys, without having to manually tweak the expressions as before, and without having to go nuclear and . It's possible to work around this with some UI shuffling, but your mileage may vary. My initial request to backport the patch to stable VSCode is not so serious since there are a few workarounds documented here. Also, thanks for the quick work in the PR! I'm looking forward for this behavior to stabilize in future versions, since secondary sidebar Explorer view (without the compromise of moving the whole primary sidebar over) is great!"}
{"_id":"doc-en-vscode-fdeb1e13a74c1c33c2e3a968de841e8174cf98a649f592423c0307b95a6e574d","title":"","text":"mac select an entry in the explorer press to trigger rename the file is being opened, no rename log\ntested on windows and this doesn't work for me - not sure if it's necessarily platform specific\nSteps to verify explorer that keybindings still work for the explorer\nMac Pro (Intel 2019) user here. I also have the same problem. Before pressing can rename files but now it just navigates to the file editor window.\nThis will be fixed in the next release which comes out sometime this week. In the meantime -should fix it for you.\nThank you for prompt reply. Yes, it works fine now!!"}
{"_id":"doc-en-vscode-05d77fa4ad32af0d89070acf9d8c37483b02b643b70d1ed854c6836f69e4aff0","title":"","text":"! Which causes the output bg to look slightly off: ! This should be , probably happened after the color token name changes during endgame testing."}
{"_id":"doc-en-vscode-b2e2ca600f5c1bcbedd64167a6d32865ee8c8bf836a05d9599609bbc2253511d","title":"","text":"While helping Marketplace improve extension search I noticed that for most queries there are no more than 5 good results. And we currently show an infinite list which depending on Monitor size would show 10 - 20 extensions. Also we sometimes struggle with the small size of extension items in the list - not enough space for all the decorations we would like to render. Based on the two above, I suggest we explore with increasing the vertical size of the extension items in the list. Ideally this would allow to present the information in a cleaner way when users search for extensions. I am not sure how this would effect the list of installed and recommended extensions. Ideally we would also increase the size of those to be consistent, but for those cases I am not convinced if it would be a win. what do you think? for ux thoughts Example from my screen !\nFrom a visual standpoint it makes sense to explore giving these some breathing room. As far as solving for quality results, I don't know if making them bigger will necessarily solve that problem alone. It could be that we have a sort of \"Top Results\" section up top with the remaining results in another section. Just one idea off the top of my head.\nI agree its worth exploring it give more space for an extension in the viewlet. But this might impact if the size of the screen is smaller.\nWe closed this issue because we in the foreseeable future. If you disagree and feel that this issue is crucial: we are happy to listen and to reconsider. If you wonder what we are up to, please see our and . Thanks for your understanding, and happy coding!\nI see this is merged and should be latest VS Code Insiders but I do not see any changes in my VS Code Insiders. am I missing something here?\nI'm seeing the latest after updating: <img width=\"320\" alt=\"CleanShot 2023-01-03 at 08 48 48 src=\"https://user-\"<img width=\"328\" alt=\"CleanShot 2023-01-03 at 08 49 32 src=\"https://user-\"\nI just got the update. Looks ok to me - I will need some time to adjust. Let's see what insiders users and the rest of the team says :)"}
{"_id":"doc-en-vscode-e2f7a90519b7d6e2bfdc3d2639de594e8e26360c6703e99603acb8f26f19bde6","title":"","text":"Testing Why is the setting name linked once and once it just shows the ID? Both should be linked, or maybe you can rephrase it to just use the setting name once like <img width=\"548\" alt=\"image\" src=\"\"\nThat's a good point yeah"}
{"_id":"doc-en-vscode-096f74b3f825ff3b19b3a3f34021d9d287635aa5a720322146aa4f5b87a19f28","title":"","text":"There is a injected text that is left over on line 52: <img width=\"883\" alt=\"image\" src=\"\"Saw this stack-trace, I'm not sure if it's related or not:\nI am trying to reproduce this issue but am currently not able to. Do you know what is the file you were triggering inline completions on, and what inline completions you saw ?"}
{"_id":"doc-en-vscode-fc7c4b7f2ab4e900389a975d730f5f424a08ac77ac6ffb5e4837241e624e63ea","title":"","text":"https://user-\nRelated ,\nhappening bc the first line's marker isn't disposed of (and thus the decoration isn't either) !\nso maybe we should have some API on xterm like so that can be handled specially?\noh here's the issue that solved but looks like that actually wasn't solved anyway\nJust to confirm, to verify this issue, are you typing commands followed by the clear command in between? Or are you clearing the terminal another way?\nTyping and clearing\nPretty sure this is the command, not run in the terminal"}
{"_id":"doc-en-vscode-abd342faf9e4f92af823560c3d79671493dc2423cc037bfde8412e4a94f74bae","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.50.1 OS Version: 10.15.7 Steps to Reproduce: system input source to shuangpin, before which add shuangpin as an input source. ! caps lock, with caps lock unlock, you will input Chinese character, which is an expected behavior. something in the integrated terminal, I can only input upper case letters. But when I input something in editor or some other application, I can input lower case letters. <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes. Actually I think it might be the problem of might decide the letters by cap lock. This is not as expected with shuangpin mode.\nThis looks like an upstream issue in\nThanks for replying. Opened an issue for"}
{"_id":"doc-en-vscode-8192e3345186279377469531cd16c8a81ccb73a0a344a1e6540ed39061f95f99","title":"","text":"Currently when using , ArrayBuffers just turn into an empty object. We should support these for efficient data transfer between extensions/extension hosts. We recently support for them on a protocol level, so support here should not be terribly hard...\nI see two possible approaches to support this: this logic specifically for commands (or even just for a specific command). This is what we do for webviews for example: our implementation aware of array buffers instead of only supporting VSBuffers. Take a look at how is used in our rpc code as a starting point: The webview serialization logic I linked to above would probably be a good starting point for this"}
{"_id":"doc-en-vscode-341cae1cabb9224d8125e32ecee9ac4fdcd89b36cf86e9c08da25b38641c2a03","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.88.0-insider OS Version: MacOs Steps to Reproduce: In the python and python debugger extension we use the QuickPick, we add an option to accept the value in the input. It was working in stable, but in insiders it doesn't work. Steps: a picker some text in the input box, make sure that it match with one option at first. enter, to select the text in the input box. last value that appeared in the list is the one that is selected. You can check the implementation . I was debugging the code a little bit, and found that in stable when you press enter, the function is called. But in insider when you press enter instead of call that function, the function is called.\nVerification: Install Python extension Open a python workspace with no Open the debug pane and click the \"create a file\" Choose Python Debugger Choose Django Type ONLY 'b' and notice that \"Browse Files\" becomes the active item Type then, 'z' and notice that the list is empty hit ENTER and notice that the gets created with \"bz\" in the\nThe fix fully resolves as well!\nResolves my issue as well."}
{"_id":"doc-en-vscode-1067dc2a0e17c198264c533a77ec813b5eb6d4eff5b74f9f48a4f7f458a46f9b","title":"","text":"re focus SCM input show context menu the is a entry \"Command Palette...\" which is awkward <img width=\"703\" alt=\"image\" src=\"https://user-\"\nRelated to this, on my system that context menu also gives me entries that have been contributed by a couple of my extensions.\nSame issue on macOS as well\nThis happens across all our inputs which use the Editor. IMO we should not include all those contributions and should just have , & in that menu. What's the simplest way towards this\nThe \"Command Palette...\" should potentially have a better context key for , not :\nThanks! But this is actually only half of the story. As mentions, extension contributed commands show up: ! Should we file issues to these extensions, suggesting to change their clauses?\nIt might be better/simpler to have another menu id, something like or even something SCM specific\nTo reproduce, you can use:"}
{"_id":"doc-en-vscode-f8c3fa9747eea7f3527c1b23a4389401809dc55f0527145467471bc37776cc49","title":"","text":"Testing The description of 'git.branchRandomName.dictionary' could be improved to say that each value represents a segment."}
{"_id":"doc-en-vscode-035deb8204300c837255739eb3e117c6e6a12adc650ff1919c983b639b6971da","title":"","text":"Testing Open the settings UI for the settings and add new values. Observe that you can any input (not just , ...) and there is no warning when a value is mistyped. This can be easily improved by adding an enum to the items: This will limit the items to the given values and show as a combo box in the settings UI: !"}
{"_id":"doc-en-vscode-f4de1b85a6c216613df65d3a4a7de1a23bec0a5b06852cc981c4bf2ad8119da5","title":"","text":"Type: <bBug</bWhen pressing to collapse all nodes in a list view it leaves selection in such a state that you cannot expand nodes with until you do something to change the selection. ! I believe what's happening is is leaving the originally selected node selected. So after executing the command you end up with 2 selections, your original selection plus the last root node in the list. The original selection is the primary selection, and if it has no children does not have anything to expand and is a no-op. As a user it looks like you have the root node selected, which has children to expand, but pressing does nothing. Pressing then expands the list to reveal your original selection and you see that you unexpectedly have multiple selections. The current behavior of appears to be: Leave the originally selected node selected Add a second selection to the last root node in the list The expected behavior here is: Move selection to the root node of your currently selected node Ideally this also retains multiple selection: each selection is moved to the root node of that selection. VS Code version: Code - Insiders 1.76.0-insider (, 2023-02-10T16:22:19.445Z) OS version: WindowsNT x64 10.0. Modes: Sandboxed: Yes <details<summarySystem Info</summary </detailsExtensions: none<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nLooking at this again in the File Explorer it looks like the selection highlight is left on the where the original selection was despite the fact that the node is now in a different position. However, the \"real\" selection is still on the node. This means it looks like one thing is highlighted and pressing opens a different thing.\nJust submitted a PR for this. Could you please review it?\nWhat is the behavior supposed to be? Here's what I see: keyboard to focus a tree node that has a parent to collapse all moves to the whole tree and arrow down takes me to the first tree element. Right arrow does nothing.\nThe current behavior sort-of fixes the issue, insofar as is dodges it by doing something less desirable - moves selection to the whole tree. Imo, selection should remain on the root node(s) of whichever tree branch selection is currently in."}
{"_id":"doc-en-vscode-bc4480146fea2f116c085f77b7196a42114eb622f276351c1b6e69b2cb9cd6e1","title":"","text":"I think this is what the previous behavior was trying to do, it just had an issue with leaving selection in a bad state. A couple examples ([x] denotes current selection): Start state End state, after Stat state End state\nI agree, feel free to send an issue for it. Your observations are the fixed state. Here's the entire story: Marking it as verified."}
{"_id":"doc-en-vscode-cd9d95ffdca96d053accd205883b3327246bd6b385e39a29d26835dce6c2f630","title":"","text":"Testing focus on trusted URI list of items tab to navigate to the X to remove an item. Press to trigger removal nicely gets removed from the list, however focus is lost Focus should be preserved in the list on the neighbour item (example: checkout how Explorer does that when deleting a file). I thought this would be tackled via\nThe trust editor does a lot of full page re-rendering which I should improve (remove) so I doubt this is an issue with the table widget itself."}
{"_id":"doc-en-vscode-2017250c8d2801b826208f2b56b0d9bb2b5de971138a39acf3e706fcb16d12d0","title":"","text":"Those who are used to having the collapse button on the far right might accidentally hit the search tree toggle. <img width=\"502\" alt=\"image (2)\" src=\"https://user-\"Ideally, move this icon to the other side to prevent confusion. cc\nTo clarify: I think any new position will have an impact to someones muscle memory, that is probably hard to avoid. For me personally the collapse action was heavily used in search results. maybe for guidance, I also feel like we have quite exhausted the number of actions for the viewlet, maybe about time to think about introducing a \"...\" and move some there?\nThat's true, it is getting a little crowded. What else would be in the ? Might seem odd if we don't have plans for anything else in there?\nFor reference, the button is here for source control (outside of the ) so I wonder if people will expect it to be an actual button with an icon in search? <img width=\"241\" alt=\"image\" src=\"https://user-\"\nYeah I like the idea of keeping this in sync with other views, though we probably missed out on keeping some of the buttons in sync, like the refresh button...\nShould we remove the \"Open Search Editor\" action? it is duplicate, there is a link action also from the results: !\ncc\nEh, I think you might want to open a search editor before doing a search in the sidebar too.\n! I did some thinking and I think it should go here because: shouldn't go in the far left or right because people probably have muscle memory as refresh/collapse as side-most buttons. shouldn't go on either side of because the icon for that and are too similar, may cause misclicks. So, by process of elimination, it ends up where I placed it in the photo. Any opinions?\nI created to move it to the place that I mentioned, but I'm open to other options in the future. Just seems that the new position should be better than what we have now, so I wanted to merge it into insiders at least.\nThanks!"}
{"_id":"doc-en-vscode-018c91068ac8836c4fe02a141b94b0f2e04b980388314de069a77932ac054609","title":"","text":"Here with the long path a \"...\" would help: <img width=\"673\" alt=\"image\" src=\"https://user-\""}
{"_id":"doc-en-vscode-67df2cff86dda9c8ac06aa344e6f9d37fb7432f3f447f477d903fb763a1ce8fd","title":"","text":"cc: Type: <bBug</bSteps to reproduce: at least 2 files in VSCode. orca. ctrl+tab to switch to the next open file. In my environment orca announces something similar to the following: Is this the expected behavior? VS Code version: Code - Insiders 1.78.0-insider (, 2023-04-06T05:21:17.374Z) OS version: Linux x64 6.2.8-200.fc37.x8664 Modes: Sandboxed: Yes <details<summarySystem Info</summary </details<details<summaryExtensions (42)</summaryExtensionVersion ------ doxdocgen1.4.0 vscode-markdownlint0.49.0 vscode-eslint2.4.0 EditorConfig0.16.4 copilot1.79. rest-client0.25.1 better-cpp-syntax1.17.2 docomment1.0.0 azure-pipelines1.208.0 vscode-docker1.24.0 csharp1.25.4 vscode-dotnet-runtime1.6.0 vscode-kubernetes-tools1.3.11 remote-containers0.289.0 remote-ssh0.101. remote-ssh-edit0.84.0 azure-account0.11.3 cmake-tools1.13.45 cpptools1.15.1 cpptools-extension-pack1.3.0 makefile-tools0.6.0 powershell2023.3.3 remote-explorer0.5. vsliveshare1.0.5834 oracledevtools21.5.0 fabric8-analytics0.3.6 java1.17. vscode-xml0.25. vscode-yaml1.12.2 sonarlint-vscode3.16.0 cmake0.0.17 intellicode-api-usage-examples0.2.7 vscodeintellicode1.2.30 vscode-boot-dev-pack0.2.1 vscode-spring-boot1.45.0 vscode-java-debug0.49. vscode-java-dependency0.21. vscode-java-pack0.25. vscode-java-test0.38. vscode-maven0.41."}
{"_id":"doc-en-vscode-45102dd8e4f2b8ea82dbaf33605a1e59303e5cd7434521de78eaef20b57ce81f","title":"","text":"vscode-spring-boot-dashboard0.12.0 vscode-spring-initializr0.11.2 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nThis is happening because of the fix for I think it is behaving as designed, though will leave that with\nThat's a bit weird. Since experience does not visually render a placeholder. So I would not expect it gets read out.\nQuestion is, should it say nothing, or is the aria-label for not correct and it should be ? I'm leaning \"nothing\" because otherwise they might have to hold down until the actual quick pick option highlighted is read out.\nI have a fix out with the \"nothing\" idea... but would love to know what you folks think before I merge.\nIn version 1.77, when I press ctrl+tab, orca informs me that the focus has moved to a list box and the name of the next file is read. Imagining the following scenario: two files, a.js and b.js. start editing the a.js file. press ctlr+tab. My screen reader reads the following information: List Box, b.js. In my view this behavior would be sufficient.\nIt is fixed, thanks."}
{"_id":"doc-en-vscode-6244b57512688af8024357ecce4bc3221765b26f9594b9050fc637e1fe060a74","title":"","text":"Type: <bBug</btypescript server keeps crashing after the latest update VSCODE ! I can reproduce this with all the extensions disabled. LOG: ! VS Code version: Code - Insiders 1.76.0-insider (, 2023-02-03T05:23:17.202Z) OS version: WindowsNT x64 10.0. Modes: Sandboxed: Yes <details<summarySystem Info</summary </details<details<summaryExtensions (11)</summaryExtensionVersion ------ ng-template15.1.0 atlascode3.0.1 vscode-markdownlint0.49.0 vscode-eslint2.2.6 githistory0.6.19 gitlens13.2.0 prettier-vscode9.10.4 vscode-yaml1.11.0 sonarlint-vscode3.14.0 code-spell-checker2.16.0 vscode-stylelint1.2.3 </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nPlease try collecting the TS Server log from the crashed instance: VS Code and reproduce the problem the section of the output panel the very top, find the first line that prints to path to the semantic error log file. It should look something like: That file contains the typescript logs. Look through that log file for the first error or stack trace you see. If you can share the log, I can also take a look to see if anything stands out\n, for some reason the log file location is shown as [Object object] (as pointed out by the pink arrow), I checked in the folder pointed out by the red arrow and the folder is empty. ! I have checked this for multiple folder I could see in the log prompt but all of them are empty.\na Microsoft member created same/duplicate issue , in case this could help with the debugging\nConfirming that I also see when using verbose setting and all the log files are empty for me, which is also noted and shown on microsoft/vscode.\nFor me the directory under the object object is also empty, is that the case with you?\nYes, same.\nshould be fixed in the next insider release: microsoft/vscode\nthe object object issue is resolved, however, the file pointed by the log still does not exist. can you give it a try if you can see the logs?"}
{"_id":"doc-en-vscode-f6de2f627805791259d818e05dcb6d8d319381bd899585991d6de2b636fffc51","title":"","text":"i did some commenting in the file and found that commenting out setting stops the repeated crash. Note: I am not sure why I have /enabled this setting in the first place.\nOh wow, exactly the same for me. I also do not remember ever setting , but it was in my and removing it fixes the crash for me too!\nAlso confirming that with latest update is fixed but log directories are still empty. I've closed my bug as a duplicate of this one since it's pretty clear these are the same issue now. Are you able to repro using is Also do you know if this would ever have been set on user's behalf, which could impact the severity here if lots of people would have this in their settings?\nThanks and I can confirm that enabling cause the server to crash on windows. Oddly I don't see it on Mac. The server is likely crashing before logging can start too, which is why the log files don't exist I'm not sure how this setting got enabled for you. It's only meant for detailed performance investigations and should never be turned on automatically\nYay, thank you and"}
{"_id":"doc-en-vscode-5f08f1280b67656f829ff0f4f2b4782bac7456cd3c82948509c44f762e7fddd4","title":"","text":"The stage is now a bottleneck to having the platform specific stages run: ! Each job compiles each target serially. We could further parallelize all targets across different jobs. !"}
{"_id":"doc-en-vscode-65c6304c1c364a81d88c72c1cd692da5a5f331474bda0bfc95121017d6be209b","title":"","text":"Verifying Version: 1.55.0-insider Commit: Date: 2021-03-25T16:24:13.227Z clone and open in vscode add a to F5 to run the extension Open an untitled file In the editor's tab, choose from the context menu, pick ! a different notebook is opened, no console log statement printed !\nAre you able to open markdown notebooks without issue on non untitled files?\nYes, that works. I also tried to open a GitHub issue notebook on an untitled file and while a notebook seems to open which has a 'Github Issues' cell, the function in the GitHub issue notebook source seems not be called. So I'm a bit confused\nI did some investigation with this and I can't reproduce it out of sources. When I select it through the reopen with dialog the notebook service receives the correct viewtype, although it always fails to open with the error . I'm not sure why it is failing as the inputs get created correctly and everything seems to be passing through the right way. Let me know if you want to investigate together."}
{"_id":"doc-en-vscode-967587d3fb068d4ade306f7fda37cd2ca969ea23b3a30a1ff858a6337e19b5de","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --In the source control, after changes are staged and a commit message is provided, it's no more possible to see easily the current branch name to commit to: ! The validation button displays a pop-up message, but this does not really add value: ! I propose to display the current branch name in this pop-up: ! This would dispel any doubts about the branch on which one commits. By the way, the same is possible for the \"Publish Branch\" button. Is it the right idea?\nI'll submit a PR for this.\nWhoaww, great, thanks. I had a look on it, mostly for curiosity, but I am very far from vsc dev currently... I will review with pleasure.\nIt's ready for review.\nIt looks so easy! Thanks. Do you think the 'Publish Branch' popup should be relooked too? The branch name appears in the commit textfield, but depending of the sidebar size it's not always visible. !\n~That makes sense. I'll update the PR.~ I first thought this was stated in the issue description too, but it wasn't. So, we should hear from on this.\nIt is, at the end:\nI guess I have to compile the whole vscode if I want to live test?...\nYes. If it's your first time, I suggest using GitHub Codespaces or dev containers, instead of compiling/running it on your plain OS environment.\nI've updated the PR to include the branch name on the \"Publish Branch\" button, too.\nOk perfect! I first thought the screenshots were from my issue, just as an illustration... But it's yours in live, so I guess it's ok. I will investigate for a dev container, it may be useful for next PR... Thx for that."}
{"_id":"doc-en-vscode-6e50036c1b7c3c69f9b8cc523528c63ad96d1c337ca89c2daf7dc338755ed2f5","title":"","text":"Verification steps: Commit action button Open a folder/workspace that contains a git repository using VS Code Insiders Make a change to a file, and save the file Switch to the \"Source Control\" view and hover over the \"Commit\" action button Confirm that the tooltip of the action read Publish Branch action button Open a folder/workspace that contains a git repository using VS Code Insiders Create a new branch Switch to the \"Source Control\" view and hover over the \"Publish Branch\" action button Confirm that the tooltip of the action read\nHi, Both verification steps are working. Thanks. Version: 1.75.0-insider Commit: Date: 2023-01-20T05:23:42.626Z Electron: 19.1.9 Chromium: 102.0.5005.194 16.14.2 V8: 10.2.154.23-electron.0 OS: Linux x64 6.0.0-0..6-amd64 Sandboxed: Yes\nDo we need to submit a PR to include this in the release notes?\nThanks for the nudge. Added"}
{"_id":"doc-en-vscode-f100421628d88415c457a4b9d0eafc69ea2fc603e7b1a9709f601423f2323051","title":"","text":"Testing I'm not sure why I have two entries for edit sessions and wonder if it came from another machine/sessions. It would be great to see some metadata about where the source is coming from, similar to the \"Synced Data\" view for Settings Sync. <img width=\"394\" alt=\"CleanShot 2022-07-26 at 10 58 31 src=\"https://user-\"<img width=\"373\" alt=\"CleanShot 2022-07-26 at 11 00 07 src=\"https://user-\""}
{"_id":"doc-en-vscode-1e9455ed832cc23583adbd96b5b8833f75c3fa380724ab47fd627ca20db5389d","title":"","text":"Following test throws: I found this while investigating At that point the editor component is quite broken because of a lot of errors being thrown. Notice how not even a tab shows for the editor: <img width=\"1058\" alt=\"image\" src=\"https://user-\"\nthat should already throw because at least a scheme is needed. It doesn't because was designed to be the runtime counter part of which never produces such output so some IPC deserialisation cost can be saved.\n, that works for me and will also throw when we get bad objects into such as the from\nYeah, my plan is to add an argument like we have in other URI-places. It will default to true and only IPC should be relaxed\nNot doing that because it actually already exists as . I believe the adoption/migration task is to use instead of whenever getting uri components from storage or esp from command arguments. Tho, there is one special case for which is that it automatically uses when the scheme is missing. We could make that configurable but a check for might make this obsolete\nClosing this after and the ask to use when creating uri from \"untrusted\" sources\nI will see to adopt , I forgot we had it"}
{"_id":"doc-en-vscode-de173daa49b82ee7cc6c5326d00f701cdf9c24abcdd7e877dae10ffc9b305470","title":"","text":"The colors we define in code should match our default theme, otherwise there is potential for ugly flickering as seen below: ! Notice how the title bar color changes, the active tab indicator appears late. It feels sluggish and like a UI bug. At one point I think we made \"Modern Dark\" our default, we I think we did not make a pass over our colors to align them with the colors defined in the theme. We need to ensure that the colors in code match the colors in our default themes, I would argue even for light and HC theme.\nI think this should be fixed in May, it looks really bad for first startups.\nIs there a reason to do it this way? I think I would have assumed the default dark theme JSON file to be empty (because it does not overwrite any colors) and our colors in code to be matching the default dark theme colors?\nWe can not change the defaults as all custom themes inherit these. I agree that it's not so nice to duplicate the colors but I couldn't yet come up with a simple solution.\nI understand the argument that we cannot easily change the default colors without breaking other themes. But the issue still reproduces even after your change: open a fresh Code-OSS and see how title bar color and tab active indicator only kick in late. Also adding since this covers UI components of his."}
{"_id":"doc-en-vscode-b8b582e58ea787b3f339b1d9f2f1b95c8f75f34494348c61c1de469bc139fcf2","title":"","text":"VS Code Version: code-oss-dev(1.64.0) OS Version: win10 Steps to Reproduce: vscode web file \"w\", and click readmore icon at warn suggest label, but show webkitURL details in right panel !"}
{"_id":"doc-en-vscode-29aa47cbcf905e07646d378f08567d1145f5a405bff836c87c6dd5a611c3ff17","title":"","text":"VSCode Version: 1.19.1 In Markdown files, the editor is incorrectly styling every other inner word surrounded by underscores as italicized. The preview correctly shows them as not italicized. !\nI think this is what Text Mate grammars' behavior. Tried in TextMate 2.0 (latest), same there. !\nI think is right on this. The editor has an error as it interprets the other word as if it was a separate word starting and finishing with `` (that would be italized). It is correctly displayed on Github editor :\nI created a PR to fix this"}
{"_id":"doc-en-vscode-0198ff3656b220db00e57cedfabff4bf91ab83b5f4aa81ea79c23a3bddd88c7a","title":"","text":"! Two problems: has no contrast and is being displayed over text mouse cursor over it should be Is this you?\nWe've always had this problem, the bottom part of the icon has no fill hence why the text shows through. The only alternative is to show this in the gutter when the first column of the row has text, but I'm not sure if that is technically possible.\nCould we add a white trace outline to the icon, making it popup in messy locations?\nUnfortunately strokes don't work on icons: ! We could try to add the same editor bg color with a slight opacity to account for this: !\nOh man, that looks good."}
{"_id":"doc-en-vscode-285c78bd65131bce3525b72e89e2c8af1f5aedf4af0ca34847ce4d1ead141c1c","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --Setting to stop at the bottom (or top) when navigating in the diff view through the changes with \"next change\" and reaching the last change. Like did in\nCan I provide a PR for this feature\nGo ahead.\nThanks a lot for implementing this. But what is the setting to enable this?\nLooks like this behavior got to the \"\" setting\nNo, does not affect this unfortunately."}
{"_id":"doc-en-vscode-fc8f9a9c3a0d17f96374c4d5ccd335a08a4e42120b9e1cb425636adf2d829fe7","title":"","text":"Testing Repro: /explain in inline chat to get a long response. using to focus the last response (does not work) to focus the last response using to scroll up in list view of chat messages. (does not work)\nSame for panel chat. I believe the keybinding simply isn't bound\nYou may have to press it twice, once to move the cursor to the top of the editor, and then another time to trigger the chat keybinding, does that work? It's working fine for me.\nOh, not in inline chat though\nI can't remember what terminal chat looked like in March, do you still want this disabled? I think we probably want it enabled in all chat widgets now:\nI fixed this so that cmd+up/down will work in the terminal and inline chat widgets. But in the terminal, once the list is focused, then the arrow keys still navigate history. It looks like the terminal widget is registering a bunch of its own commands for history and for moving focus around. You shouldn't need to do that anymore, since you can use the commands that the ChatWidget has."}
{"_id":"doc-en-vscode-8e055b5b546f6f27863da1e435cd392b3b5427608c302d5a406414b7392782de","title":"","text":"I think we used to indent inline chat to the indentation level of the line: ! Steps: have an indented line press enter to be on an empty line (still indented) press to bring up inline chat"}
{"_id":"doc-en-vscode-e9178b432b9bd65758fe9dfe02825eeba0ef2b797707996ecac1b94abc3830c6","title":"","text":"Currently sticky scroll is grouped with the breadcrumbs in the context menu which appears on right click on sticky scroll. Henceforth it should be grouped with it too in the menu. cc"}
{"_id":"doc-en-vscode-fb779126148365a7b5e056c8c19dc7f938463d374c7adde9f9cbc41349a44df8","title":"","text":"<!-- Do you have a question? Please ask it on --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.21.1 OS Version: MacOS High Sierra 10.13.3 (17D102) Steps to Reproduce: markdown preview in combination with local plantuml server code to use local server for rendering a markdown with embedded plantUML content error about unsafe content ! The current only allow to enable/disable unsafe (HTTP without the S) entirely. Would it make sense to add an option that allows unsafe content from localhost but not from remote sources?\nThis may be ok (at least for images): If someone wants to take a look at this, you would just need to update the content security policy to allow :\nLooking into it, thanks for the pointer."}
{"_id":"doc-en-vscode-249eba60bc05adeaa7e3878661b0ddcea198857cad7c1180c6153d30c1a28d20","title":"","text":"Issue Type: <bFeature Request</bTo see the changes i made to a source controlled checked-in file, a user clicks on the \"Open Changes\" icon on the top right. I would like to see a better looking icon that represents git and changes. The current one looks like a search button. VS Code version: Code 1.29.1 (, 2018-11-15T19:13:36.375Z) OS version: Windows_NT x64 10.0. <!-- generated by issue reporter --\nWhat do you think? Also note that changing this will affect other users and we'll soon get \"but I was accustomed to the old icon!\" messages.\nAt first glance, the icon does appear to look like some type of \"search\" button. Upon inspecting it, you can see that there are two files being compared (i.e. a diff). It's hard to see these details at a quick glance, so if there is a better icon we could use I'd be up to changing that. GitHub uses these icons for diffs/compare: <img width=\"79\" alt=\"image\" src=\"https://user-\"<img width=\"85\" alt=\"image\" src=\"https://user-\"<img width=\"499\" alt=\"image\" src=\"https://user-\"<img width=\"498\" alt=\"image\" src=\"https://user-\"I do like the last one the best (though it is an octicon).\nI like it. If you find me those assets, let's do it together!"}
{"_id":"doc-en-vscode-d07dcedef82f98df31f26da3b8e156879805bab7aa7036a7efcb4cc31e4f41ce","title":"","text":"This may be useful as a hint: Please, take a two minutes to look into source, it may work only for :\nI think adding a desktop file with that launches VS Code with is what's needed to implement this.\nnice relevant gist:\nVS Code has native support for opening it's own URL scheme via the hidden argument, I don't think there is a need for wrapper script shenanigans. I'm going to submit a PR.\nI've not checked too far into it (Just subscribed to this to keep up to date) but the wrapper script is probably there because in the standard would include the \"vscode://\" prefix.\nThat's what should happen, that's what handles. :smiley: You can test if you like. Alternatively you can test this (Which is what the PR adds): For OSS: For release version: Drop it into or and run on that directory."}
{"_id":"doc-en-vscode-9543d919900702968dfa07cff273b532dc56170c8480e9ecebb7a752e3345545","title":"","text":"There are some parts in notebook land that depend on . This service and its provider are deprecated and should be removed. As of today the following dependencies must be cleaned up first: , , The should be used instead, it will allow you to know if a agent per location exists or not"}
{"_id":"doc-en-vscode-c2e1ee53292b67fc236608e0fbd6678620e21191bf5f3b9d2a7b9d616efe06b5","title":"","text":"run you'll notice that in the main process before creating the renderer the update service takes its time (~5-10ms). while that's not much it seems unnecessary and \"block\" the renderer for no good reason <img width=\"1101\" alt=\"Screenshot 2020-01-31 at 14 45 07\" src=\"https://user-\"\nThis is only problematic in macOS where calls into the native which blocks.\nThanks"}
{"_id":"doc-en-vscode-7fdc992091e7c66e8e96fd5df156e752b6007779c6df7264f1b21e768bbbd4db","title":"","text":"a traditional . , listen for debugging the extension, open a file picker, then close it. session token's never gets triggered when the session is done (aka when the file picker is closed) :bug: This contradicts\nregression from"}
{"_id":"doc-en-vscode-f368bd977e9f6c95bc185d873e57b7e2aa17aa904237d1cdd78a48fae21c9d44","title":"","text":"Testing The type cannot be used with because that requires objects (no primitives). We should accommodate for that and change the API accordingly."}
{"_id":"doc-en-vscode-41da26ec4543153475c70fcffd10be0ce2cff7c7e7a6f474f5e8820544ed19f8","title":"","text":"Right now we can do pull but not push with Git's API . We are looking into allowing users to create pull request from the extension but without this API, we can only execute git command ourselves. Not sure about the cost of adding this API, the shape of in my mind can be like .\nFixed by"}
{"_id":"doc-en-vscode-f8343fd3d3d0e87f8d8566a558be3d7fc1579f6da4fe479bedaed80d2c403747","title":"","text":"open normal quick pick () type notice you want symbol search and select the cursor is random, within the word, and doesn't select the previous input anymore"}
{"_id":"doc-en-vscode-d4dbb1dda92f1d17e104e0bc840b3ce3971f58b1c8213724ced649b1205b0ddc","title":"","text":"To repro, cause the Ports view to rerender several times. This can be done by trying to set a label and escaping out about 8 times.\nis this a bad leak? Like does it leak just 1 listener per interaction or does it grow faster than that?\nIt grows faster than that, but it's probably only a medium bad leak. After 8 table rerenders there are 175 listeners. After ~5 more table rerenders there are 263 listeners. After ~5 more table rerenders there are 351 listeners. After ~5 more table rerenders there are 439 listeners. A table rerender can be triggered through port auto-forwarding. Once when the process is started and the port is forwarded, and once when the process exists and the port is removed. The number of listeners is depended on the number of ports that have been forwarded. My numbers above are with 3 ports.\nNo more leak, works as expected -verified"}
{"_id":"doc-en-vscode-98d5fa281e19d6f511ddce1049b52e1a6a3f7a7be75621528527bf90379b73cd","title":"","text":"Testing In the screenshot below, \"Chat: Insert at Cursor\" should be \"Chat: Insert into New File\" !"}
{"_id":"doc-en-vscode-9b17638c35cbe5af011f98e16fa831957fe6f69c282d2148e90a0379b3fcb574","title":"","text":"Testing There's a double close parenthesis here: <img width=\"305\" alt=\"image\" src=\"\"\nAlso a missing on the line"}
{"_id":"doc-en-vscode-18040fd7ada81fbe3cdc6d555537030b8338ab6d31ea587af64b667c5f3531be","title":"","text":"<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --I'm working on a Visual Studio Code extension that's supposed to make it easier to clone a git repository in a user configured directory. The basic idea is that the user configures the directory once (e.g.: in ) and then all the cloned repositories will be placed in this directory. The seems the ideal candidate for implementing something like this (especially now with the new ). However, the only problem is that the command is always asking the user for the destination directory. This feature request is about adding an option to the command which would allow other extensions to clone a repository in a specific directory without showing the dialog. One possible implementation could be as simple as adding another optional argument (for the destination directory) to the command ().\nFound a quirk:\nFixing the quirk is fine in the future. Thus pushing it out to the On Deck milestone since Joao is on vacation. please speak up if you disagree.\n+1 for this issue. I'm also working on a VS Code extension (specifically the C++ extension here in Redmond). We're working on an experience that involves operating on the files in a repository that the user has cloned via the command, but currently there is no way to know which parent directory was selected. Allowing the repository parent directory to be passed in would solve the problem entirely for us -- I'll be watching this issue closely. :slightlysmilingface:\nThis has been implemented () and you can already use it in Visual Studio Code version .\nThanks\nSince this is already implemented I am closign this issue. Thanks As for the quirk can handle as an independet issue\nshould this issue be labelled with again? I also wonder if the quirk has been filed as a separate issue."}
{"_id":"doc-en-vscode-5294d2c5868ca631af94ca4c5604afa40662fc7f81c97965e44ef775f52225b1","title":"","text":"Issue Type: <bBug</bCreating an untitled document (from a file URI amended with the scheme) results in different states depending on whether the document is created with the command or the function. I tried to reduce the problem to a failing test case and failed, but maybe can still be useful for debugging. Create an untitled URI for a new file () Open an editor for that URI with / Type some text Save the file Notice that the save dialog starts with the we had providedCreate a new untitled URI for a new file Open an editor for that URI with Type some text Save the file The save dialog starts with the we had provided, just like with the other method. The save dialog starts with the text we typed instead of the we had provided. VS Code version: Code 1.63.0 (, 2021-12-07T05:16:45.441Z) OS version: Linux x64 5.15.6-200.fc35.x8664 Restricted Mode: No <details<summarySystem Info</summary </details<details<summaryExtensions (38)</summaryExtensionVersion ------ vscode-unfancy-file-icons3.15.0 icons-carbon0.2.4 chalice-icon-theme1.2.16 awk-ide-vscode0.6.2 vscode-fish1.0.24 init-script0.1.3 make-task-provider2.1.1 npm-intellisense1.4.0 path-intellisense2.6.0 vscode-eslint2.2.2 EditorConfig0.16.4 gtk-dark-titlebar1.2.3 vscode-pull-request-github0.34.0 go0.29.0 terraform2.17.0 nix-ide0.1.18 language-julia1.5.6 rust-analyzer0.2.842 git-graph1.30.0 fluent-icons0.0.15 direnv0.3.0 vscode-docker1.18.0 python2021.12. vscode-pylance2021.12.1 jupyter2021.11."}
{"_id":"doc-en-vscode-e6421528d8a171bfee38f71508cd727d3fda0e91918ffbf4c2bddbbd5b8792fa","title":"","text":"jupyter-keymap1.0.0 jupyter-renderers1.0.4 remote-containers0.209.5 remote-ssh0.68.0 remote-ssh-edit0.68.0 vsliveshare1.0.5196 vscode-commons0.0.6 vscode-yaml1.2.2 edit-with-shell1.3.0 lua2.5.3 even-better-toml0.14.2 shellcheck0.18.2 redo0.1.0 (6 theme extensions excluded) </details<!-- generated by issue reporter --\nLooks like when you call we treat the URI as the associated resource: But we do not do the same from\nI tried this with the following code: and both files saved to the correct paths... but I wasn't prompted at all by a file dialog. Is that expected?\nYes. Untitled with associated path don't ask for a file name, that is the whole idea of \"untitled with associated resource\" vs. \"untitled without associated resource\"."}
{"_id":"doc-en-vscode-61e223e807d6749e143f1d91549791ddca6b82c1379871b26f10477c4977456f","title":"","text":"VSCode Version: 1.33.1 OS Version: 10.14.4 When editing and closing a git commit message file very quickly, the contents are not provided to git. Currently I manually save before closing any git commit file but it seems clear this should work properly - autosave should happen, then Code should inform Git that the commit file is ready. Currently it appears the two events are happening in reverse order, causing a bug. Steps to Reproduce: VS Code as editor () a git commit the commit message in Code close the commit message in Code, before autosave occurs Result: git complains about an empty commit message Expected: Autosave should happen before Code releases control back to git, resulting in a properly saved commit message Does this issue occur when all extensions are disabled?: Yes\nGood catch. Extra points for using auto save in VSCode :+1:\nvia"}
{"_id":"doc-en-vscode-143ce1fe860f04bcc1a4a86d68a2965e44bda1017dc9e6f8490d2eef511255d3","title":"","text":"Type: <bBug</bvscode with wsl: or similar tool to monitor CPU usage when vscode is idle or minimized to the taskbar. Expected: CPU usage for vscode's remote server should stay around 0% at idle conditions. Actual: CPU usage stays between 2-10 % VS Code version: Code 1.91.1 (, 2024-07-09T22:06:49.809Z) OS version: WindowsNT x64 10.0. Modes: Remote OS version: Linux x64 6.6.36.3-microsoft-standard-WSL2 <details<summarySystem Info</summary </details<details<summaryExtensions (10)</summaryExtensionVersion ------ remote-wsl0.88.2 material-icon-theme5.6.0 vim1.27.3 markdown-checkbox0.4.0 gitlens15.2.0 prettier-vscode10.4.0 hexeditor1.10.0 advanced-new-file1.2.2 vscode-fileutils3.10.3 even-better-toml0.19.2 </details<!-- generated by issue reporter --\nI had earlier opened the issue at where I provided more in-depth issue description. I'm just going to put the same description here. I recently noticed high CPU usage and faster battery drain (intel cpu) when working with any project under WSL. To find out what's taking so much cpu time (even when I'm doing nothing, apart from staring at the screen), I used to list processes in the order of cpu usage: ! Looked like vscode's remote server running inside wsl is taking CPU. So, I went a little further to profile the running instance of the remote server and I found: ! ! from is being called at a regular interval with a small delay (~1000ms). : ! I was able to trace back to : ! And I finally found where it is being called repeatedly: ! Now, why is it a bug? Because I've disabled all related settings in my config, yet the tunnel monitoring service is running in the background. I expect the service to not be running when or related configs are disabled. The service is currently always running.\nWow, great job, thanks for tracking down the problem!\nI'm having trouble understanding this: Is WSL going to end up being a ?"}
{"_id":"doc-en-vscode-a9340e14d9a49b451b30b32ace4caf69904e26c2c5f83cd5bf29e5d90449086e","title":"","text":"I don't think we actually want to show the ports view in WSL even if the Tunnels extension isn't installed.\nI agree. For non-managed remote authorities created just via the in the workbench, how did we determine whether or not to show the ports view? We should apply the same logic here...\nThe ports view is controlled by a context key getting set. It gets set in two ways: an extension that wants the ports view. This is what dev containers does. a tunnel factory getting . Since the tunnel factory is always now, we're always enabling it. I should be able to add a check if the tunnel factory is coming from an extension and only do 2 if that's the case.\nLooks like we also show the ports view when there's an extension with an actiation event: Is this perhaps no longer needed with the work you did for tunnels I don't remember exactly why I it.\nI've moved the check to happen when there's been a port forwarded.\nLooks like this was actually a feature: You can set to disable this. I'm also going to try to improve the description of and .\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nTo verify: the descriptions of and . Verify that you can understand what the settings do and how they're related. (These settings may be a bit confusing. We should consider deprecating and only have ) . to or if it isn't already. the WSL extension, connect to a WSL remote. the log level to Trace for the Remote Extension Host. the output for the Remote Extension Host. Verify that you see a steady stream of log lines that say \"candidate\" coming in. This means that port discovery is on. to and to . Reload. sure that the log level of the Remote Extension Host is still set to Trace the output for the Remote Extension Host. Verify that you DO NOT see a steady stream of log lines that say \"candidate\" coming in."}
{"_id":"doc-en-vscode-c645bd3d9cf7f188fccc560c3ae3daf3015754cd50e1377dc9c3da9798b6eb59","title":"","text":"In the WSL workspace, I don't see any log lines that say \"candidate\" even after starting a Python HTTP server.\nI can't repro this. Do you have an old version of WSL installed?\nI verified that on my machine the work.\nI haven't checked with Insider, but in stable I can confirm 'candidate' logs are not visible after Step 6 of the verification steps. Thanks. I also want to point out that it's not clear from the description that there's a process monitoring service running in the background, and that it can be disabled using . Should the default option be instead of ?\nThanks for trying it out If we set it to , then we might end up not showing any ports in the Ports view. In insiders, I updated the text of the setting: Sets the source from which ports are automatically forwarded when is true. When is false, will be used to find information about ports that have already been forwarded. On Windows and macOS remotes, the process and hybrid options have no effect and output will be used. It still doesn't explicitly say there will be process monitoring, but it should be clearer."}
{"_id":"doc-en-vscode-654ff5450d36d5dd01bdb52de717a4ee49ca0607d8ae25c0ac9af885884077f8","title":"","text":"Type: <bBug</bWindows 10, screen reader is NVDA, most recent insider build, as of December 13, 2022. The first time I start vscode for the day, after a short time (about 3 minutes), my screen reader stops speaking. If I press nvdaKey+tab (read current focus), it says \"setup\", but nothing else and I can't do anything. If I cancel out, I get back to vscode, but now every minute or so, my keyboard locks up, again with no speech. The only way now to get things going again is to switch away from vscode via alt+tab, then back again. I cannot see the screen, so cannot give any more info, but I'll bet whatever that setup window that came up first was, it's coming up again... Sorry I can't be more specific... VS Code version: Code - Insiders 1.75.0-insider (, 2022-12-13T05:22:12.417Z) OS version: WindowsNT x64 10.0. Modes: Sandboxed: Yes <details<summarySystem Info</summary </detailsExtensions: none<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nI cannot reproduce this. I am wondering if you are not in mode. Can you check the value of in your settings and make sure it's set to ?\nI can verify this issue, also with NVDA on Windows 10. It occurs both with the latest insiders and with the production 1.74.0. And yes accessibility support is enabled. The symptom is usually that the screen reader simply stops reading anything after a period of time, and it can be temporarily fixed by minimizing the window or switching to another then going back to vscode. I've also had the symptom where it will keep reading the same line after consecutive down or up arrow presses even though the cursor is clearly moving to the next or previous line. cc:\nSymptoms point to running with experiment as the cause. Can you try starting with and see if the issue persists ?\nDisabling that feature seems to have no effect. The problem persists. I'm also noticing that I can get NVDA to start reading again if I switch context (e.g. stops reading in file editor, switch to terminal then back, reading usually resumes)."}
{"_id":"doc-en-vscode-e1ecf8b1ca988eb8949411ec21219937df6c878870c7038a6345f970cc7edcd7","title":"","text":"I also have this problem in stable version 1.74. In many occasions I can't read the current line. In addition to losing focus, when returning from another window, the file browser and line navigation, when it works, becomes much slower. I have to restart NVDA every time that happens and it fixes it, but it is annoying and interrupts a lot of work. I would appreciate a quick solution please\nRunning new build with potential fix for testing, will update here once the builds are available.\nCan you test if the issue persists with the following insiders build, please make sure turn off updates temporarily in your settings with\nThese are terrible bugs i've encountered on production 1.74. The most annoying one for me was NVDA were trying to read whole document from top to bottom when ever I press up or down arrow. With the given insider 1.75 link above, it looks fine for now. It also feels a bit faster than the usual vscode, out of topic.\nThanks for filling this issue. I tried to reproduce with latest Insiders, NVDA on my Win10 VM and I could not repro, here are my steps. on NVDA, VS Code insiders the repo in VS Code Insiders this file in VS Code Insiders up / down arrow for 7-8 minutes inside the file. Notice how NVDA reads everything even after all that time can you tell me what am I missing in this steps? Also can you try with the build provided . That would really help us to understand this better since we cannot yet reproduce the problem on our side.\nI have not tested with insider, but in stable, if you press ctrl + shift + e to go to the file explorer, leave it still for a while, you can no longer do anything and it is necessary to change window. When I go back to the VS Code window, sometimes I can't read the lines, it reads the whole file from the beginning, and besides that, NVDA takes a long time to move through some controls. For example, when expanding a folder in the file explorer. All this is solved if I restart NVDA. But it is something I have to be doing very frequently to be able to work well. As I was saying, this happens to me in the latest stable version. I will try the above solution."}
{"_id":"doc-en-vscode-030fd6c9be16d5b106ac915ecb347c3db48fb236a5734a82a29adb31e5b8ff78","title":"","text":"Thanks\nThe above solution solves the performance problems I have been having for a long time when starting VS Code. It seems that the line reading problem has been solved as well. However, when I leave the focus on the file browser for a minute or so, the problem persists and nothing else can be done. If I press ctrl + shift + e and switch contexts, the focus returns, but now I have lower performance with NVDA, which causes me to have to restart the screen reader anyway. Thanks\nAfter some time, I have been able to verify that it is still not resolved, but it does not happen in all cases. The performance issues in VS Code at startup seems to be the only thing that was fixed, apparently. I hope you can keep and merge the changes from the above proposed solution because now I don't have to restart the screen reader after starting VS Code, which is much nicer. However, the other two problems still persist. Standing still for about a minute in the file explorer results in not being able to focus on anything. The line problem I can now only reproduce with an extension, but it uses the same Monaco editor as VS Code. Install eamodio.gitlens extension. Now, focus the file explorer and look for a file with the arrows. Press shift f10 and scroll down to open file history. Then, choose any commit and press enter, then f6 to go to the editor that opened. There the lines do not read well. For example, given this file from a Ruby on Rails application: Pressing up arrow or down arrow always reads this, except in the last line that reads blank: Thanks\nI just installed the build posted above: VSCodeUserSetup-x64-1.75.0- and now, everything works as expected. The screen reader no longer locks up on me every minute or so! This is great... Hopefully, this will give the team enough information to track down this bug! I will leave updateMode set to \"none\" until I hear that this bug has been squashed. Thanx for the fix! -- Rich\nI investigated further and the problem was indeed solved. While changing \"\" to none VS Code downloaded the update. I uninstalled and reinstalled and everything is fine. Thank you very much!"}
{"_id":"doc-en-vscode-2c318bf04b7e2c5805976b9726be5fb432352219d3712bf55ec03d2d892d6cda","title":"","text":"I hope you can release this solution quickly, as I can't use this specific code-insiders version with WSL. Try to install but I get 404 error.\nI agree that your test build seems to have solved the issue and navigation feels crisper and more responsive. Please update insiders and also a patch release to 1.74 since this does not work with WSL or devContainers.\nthanks for confirming the fix, it will be available in Insiders today. Can you also respond to comment in about any missing steps which will help us on our end to verify the fix. Thanks! We just had a patch release yesterday , you should be able to use the normal insiders with the fix on both WSL and devcontainers until we make a release. Also, we will need the steps to verify if we want to make a patch release to stable. I will update here again once the fix is available with Insiders.\nInsiders has been released with the fix, you can get insiders from\nthe steps are in this , for the line reading and freezing problem Thanks for the solution!\nFOr me, the problem would occur when I was simply navigating the interface. It was most apparent, and seemed to be triggered more quickly, when navigating a menu like view (alt+v, then up/down arrows). Question: I'm seeing behavior that I've never seen in other menus -- if I open view with alt+v, navigate via downArrow, escape out of the menu, then come back in with alt+v, it almost never focuses on the first menu item, but somewhere in the middle. Is this intentional? Thanx for all the work on this! -- Rich\nI cannot say I noticed a repeatable pattern, but it was much more likely to occur when navigating around the UI in areas like the explorer, settings, extensions, menus, etc. than just staying within a single file. So it doesn't surprise me that steps did not reproduce the issue.\nClosing since we have pushed a fix for this. Adding a verified label based on users confirming the issue is indeed fixed (since we can not reproduce ourselves). If some users can it would also be helpful to verify with the stable build that is about to come out in the next 24 hours."}
{"_id":"doc-en-vscode-a4183cf481113d91d2327aa7ff0b8b4b3b40bf4ccd7111ae9b170d83d45129f6","title":"","text":"Here's the links to the build depending on your platform darwin: darwin-arm64: darwin-universal: linux-arm64: linux-armhf: linux-deb-arm64: linux-deb-armhf: linux-deb-x64: linux-rpm-arm64: linux-rpm-armhf: linux-rpm-x64: linux-snap-x64: linux-x64: win32: win32-archive: win32-arm64: win32-arm64-archive: win32-arm64-user: win32-user: win32-x64: win32-x64-archive: win32-x64-user:\nYes, it's also fixed in 1.74.2."}
{"_id":"doc-en-vscode-488a7b9e1e6554ff4193ec522ea6581a61874d8549dc5a0f8f5b2a137f349671","title":"","text":"<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.29-insider OS Version: Windows 10 Version 1809 (OS Build .1000) Steps to Reproduce: Find in files and search for something that will return results over different result lines Badge and Action button (Dismiss) are on the left (aligned to the end of result text) I'm expecting that Dismiss action will have fixed horizontal position (on the right side) that will allow me to easily dismiss several results entries. With current position at the end of result text I have to constantly move mouse to hit Dismiss button. In stable version both count badge (for file) and Dismiss action were aligned at the right side of the viewlet which was convenient. At this moment it is hard to filter out result entries because texts have different length and button shifts for each line. I've found which says that button should always stay on the right side, however this isn't case in v1.28.2 independently from viewlet width (both badge and actions are on the right side). <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <details<summaryScreenshot</summary<div! </div</details\nI guess this only applies to the search panel?\nYes, I observe issue in Search panel. UPD. Checked - same behavior when Search: Location is set to sidebar.\nYou should see the actions on the right when search is narrow, and on the left when it's somewhat wide. The reason is that when the actions are on the right and the text is aligned to the left, it can be hard to associate the two, so it seems more user-friendly to put the actions closer to the actual result. Maybe the threshold could be tweaked, but it doesn't feel good to click an action on the opposite side of the monitor from the actual text. Does this match what you see?\nYes, that match what I see, but makes it harder to use Dismiss action with mouse. For badge with results count - it is good position, since it is visible without hover. On the other hands actions are only visible on hover, when background can clearly indicate what line will be dismissed."}
{"_id":"doc-en-vscode-6755a10e4d65811c4072c0e36bb2679691a1ee7308e9fabcc077cc217cce3861","title":"","text":"Previously (September and numerous builds before) I was able to put mouse over X and just click (even without looking) several times to remove unnecessary lines. Now it takes more time/attention to do it with mouse, and if next line after dismissed has longer text - it is easy to open file instead of dismissing it. Open editors view (Close editor button), Source control (action buttons for files), Extensions (Manage button) - have static position of actions that are not changed with sidebar width. My cases are 23\" monitor with 1920x1080 and Surface Pro 3. Never had problems when action were on the right side. Users with 4k 30+\" will have to snipe for X too.\nYes it was broken for a few months, but this was the behavior from before it was broken I see your point but I will leave this open to think about it\nthanks!\nPutting the dismiss button to the right would also make it less likley that someone presses it by accident which I just did. The middle of a search result might be just above/below the end of another so that clicking just a bit above/below where you intended could lead you to remove the other search result.\nIssue 1) The dismiss buttons are too close to the actual text of the search result making it very easy to accidentally dismiss something. Issue 2) The dismiss buttons are invisible until hover/focus. So if you had your focus on another panel or another window and then click on a search result to focus it, it's too easy to click on an invisible button and accidentally dismiss a search result because of issue 1.\nTrying to decide whether we should just always right align... but it seems weird to me to have the X really far away from the row content in some cases. Anyway, if someone wants to add a setting to configure this, I would take that PR."}
{"_id":"doc-en-vscode-9635e7375b605e601d4931f9d0febec9ee8b32660449477c7796a975f3f8b2d0","title":"","text":"Related: ! A better user experience for blocking/deferring subsequent modal dialogs needs to be implemented."}
{"_id":"doc-en-vscode-e29e83f8d259e838ca1ff10cbd9337bd2113b18cab9198e86646e6398a1d2545","title":"","text":"Open notebook with a header, code cell, and symbols in the code cell: # 1: symbols appear as a child of the header md cell, and the code cell is not present in the outline. # 2: changing the filter settings to remove code cells or symbols does not recompute the outline. Fixes: For # 1: an outline provider was being created with the incorrect target (sticky scroll was created with a QuickPick target), and this was being retrieved for the outline pane incorrectly. For # 2: one of the locations an outline change event is fired was shifted and subsequently removed, meaning that the outline did not recompute to reflect the filtered-out cells."}
{"_id":"doc-en-vscode-19350a4955672e97c53887f5ff72423265704fbd2a35131cf3205a00ba0a6312","title":"","text":"Hi ! I am running into an issue issue when using vscode 1.66 : when running pymc models there is a progress bar output which was working fine before the update and now the progress bar simply prints down :( See pic below ! This is how it prints in Jupyter notebooks (anaconda) and it was looking so in vscode 1.65.2 ! Here is the info about my session\nUpdate It actually works when I am not looking ie, I am not in that specific cell of the notebook ¯()/¯ !\nHere is a reproducible example: Code: Output: !\nI'm having a similar issue using enlighten progressbar. This likely started after a recent vscode update. A new progressbar is printed everytime it updates/redraws. I'm guessing this is the same issue. (This works/prints fine when run using jupyter-notebook on a web browser) !\nThank you for working so fast on this one! Looking forward for the next update/release !\nI couldn't verify this fix, using : !\nthanks, it's not being picked to the release branch yet.\nWorks for me too. !"}
{"_id":"doc-en-vscode-c64c8a4ec70353dea184392268b5343033508f1fdcb8c71270ce97951e94c4c7","title":"","text":"I wanted to customize my keybindings depending on the OutputPanel state, but it looks there's no clause to set in the stanza for it.\nThere is a context , please try that\nOk that works when if focused (but needs to be documented), what about when it's visible?\nWe accept PRs that update our docs. Checkout the vscode-docs repo. No this is not currently supported, we accept PRs that add this. Code pointers Similar to the viewlet service here The panelPart would add a context key\nworks but how does a user learn the panel ID?\ngood point. However we already had that issue with the viewlets, since we already have a \"activeViewlet\" conext key. We should document this properly"}
{"_id":"doc-en-vscode-cf5f00cd5ecc3be8df86cd69605e08ba6692f2d7913278f514d161bb34746dbb","title":"","text":"Including all who have commented : New SelectBox implementation: Standard native select allows navigation to an item using the first character. This can be done both on a unexpanded or expanded select. Should this behavior be implemented in the new custom SelectBox ?\nyeah good catch I think it should. If the control has focus at least. Please do not assign the \"candidate\" label to issues. It is reserved for issues we plan to fix in a stable recovery release.\nthough assigning milestones is usualy a good idea, if you plan to fix this until the end of january just assign to December / January milestone.\nNoted on the labeling issues (verified, candidate, milestones), sorry don't know all the workflows :-( Sorry to pepper you with a few more questions: I saw you wrote a draft test plan, would you like me to add some more specificity on keyboard support so people know exactly what to look for? If I have small tweaks/cleanup should I create an issue(s) or is using notes in a pull request sufficient? Essentially when is it appropriate Not to do an issue for work in progress? I will fix all issues posted for the December/January milestone assuming they are known in time.\nyeah feel free to make tweaks to the test plan item: I also published some release notes on it here: The advantage of having an issue is that it will get verified in the endgame if it is a bug. If the thing you are changing cannot be verified or is not a bug, doing just a PR is fine.\nYou chose selectBoxCustom for the filename, do you prefer that I re- factor the delegate class and symbols to use custom instead of list?\nno I think its fine\nFixed with:"}
{"_id":"doc-en-vscode-cf6327fbc6a4c4757ca80b13648cf0b7f6859b282a153dd6eb1e33e54b67be0a","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --With the above settings in effect and a terminal open in the bottom panel: If the terminal has focus, the cursor blinks If the terminal doesn't have focus, then the cursor remains visible without blinking Unlike the block-shaped cursor, which becomes hollow when the terminal loses focus, the line-shaped cursor doesn't change appearance when the terminal loses focus; it merely stops blinking. This makes it hard to tell at a glance whether the terminal has focus. So that it's easier to tell whether the terminal has focus, I would like an option to customize how the line-shaped cursor appears in the terminal when it doesn't have focus. Possible options (not exhaustive): Hide the cursor altogether Keep it visible, but make it (say) 50% opacity Show the same hollow block that's used with the block-shaped cursor\nI actually thought it showed the hollow block in this case, I think that makes the most sense\ntry try\nWe're going back on the decision made in and making unfocused show the rectangle regardless of cursor style. Discussion:"}
{"_id":"doc-en-vscode-d6ab6b31ef877c64e5a3e8daffe07d48476ae7f461d66455694970cc3f785f24","title":"","text":"Testing Using the simple file dialog this is a bit inconvenient: <img width=\"567\" alt=\"image\" src=\"https://user-\"\n(Also on Linux which I'm supposed to test.)"}
{"_id":"doc-en-vscode-7aa71f11ca5e5bb7cf55d695f3e56b2f24c2282284bdaed882fdc233f7236ebe","title":"","text":"Type: <bBug</bWhen editing a Markdown file, VS Code will apply LaTeX syntax highlighting to math in double-dollar signs (math blocks), e.g., will color and and the numbers. GitHub Markdown also supports a code block syntax (the triple-tick syntax; see ), so the above could be written in GitHub-flavored Markdown as <img width=\"200\" alt=\"image\" src=\"https://user-\"It'd be nice if it could apply the same highlighting as double-dollar signs. VS Code version: Code 1.75.1 (Universal) (, 2023-02-08T21:34:59.000Z) OS version: Darwin arm64 22.2.0 Modes: Sandboxed: Yes <details<summarySystem Info</summary </details<details<summaryExtensions (19)</summaryExtensionVersion ------ prettier-vscode9.10.4 linter-gfortran3.2.0 unicode-math-vscode0.3.5 isort2022.8.0 python2023.2.0 vscode-pylance2023.2.30 jupyter2023.1. jupyter-keymap1.0.0 jupyter-renderers1.0.14 vscode-jupyter-cell-tags0.1.6 vscode-jupyter-slideshow0.1.5 cmake-tools1.13.45 cpptools1.14.3 cpptools-extension-pack1.3.0 vsliveshare1.0.5831 rewrap1.16.3 cmake0.0.17 clang-format1.9.0 markdown-all-in-one3.5.0 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\n<!-- -96bd-47a8-a0e8- --This feature request is now a candidate for our backlog. The community has 60 days to the issue. If it receives 20 upvotes we will move it to our backlog. If not, we will close it. To learn more about how we handle feature requests, please see our . Happy Coding!\nYay, 23 upvotes!!\n<!-- -c9e0-7adb-d31b- --:slightlysmilingface: This feature request received a sufficient number of community upvotes and we moved it to our backlog."}
{"_id":"doc-en-vscode-6d9f419cd61641564d42b0fc4fbfd61c04c148eba298f48efb4d107e47cc9396","title":"","text":"To learn more about how we handle feature requests, please see our . Happy Coding!\nHaving math block and code-block specific code linting would be also great!"}
{"_id":"doc-en-vscode-6e7194868c75468a1f5fa19be7e4a69d9f42e837f9ca3d1375656ff4a8075bd5","title":"","text":"Testing Meaning that it would automatically highlight and reveal in the background when you type a valid number of a match."}
{"_id":"doc-en-vscode-4a163d3b491c79f2951736e1b44ee84257694a95d752f9462beeef266f524ad9","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --Version: 1.78.0 (user setup) Commit: Date: 2023-05-03T20:09:00.748Z Electron: 22.4.8 Chromium: 108.0.5359.215 16.17.1 V8: 10.8.168.25-electron.0 OS: Windows_NT x64 10.0. Sandboxed: No I've reported a bug related jupyter notebook before . After new release of vs code this below bug happens. Steps to Reproduce: https://user- a Jupyter notebook file that contains some images. to view images and see what happens.\nI'm not seeing this repro - can you try disabling all extensions and opening that same notebook to see if it still repros? Also check which renderer is being used for the image by selecting 'change presentation' in the output menu !\nI've provided info about you asked in the video. extensions are disabled was using 1. as seen in the image and video. tried other renderer as well to show difference. another issue is; VS Code is scrolling down own its own when notebook is opened first time and loading of output started. It can be observed below in end of video as well. I cannot hold the page and have to scroll back by myself. !\nHow many images are you displaying there? (I still don't see it with 500, though perf takes a big hit) Does it repro with a smaller number of images? Like just displaying the same image 10 times."}
{"_id":"doc-en-vscode-2d170d9d61e707087f11ec383edf471ae44c648dee1b6ff043a6188a61149a27","title":"","text":"Do you know if all the images are being displayed? I see some suspicious collapsed borders in between a few. Can you try and reloading to check if there's a setting causing this? ( to default to switch back)\nAlright, I have tried the way you told me and created new profile and didnt install anything at all. But it seems that doesnt change. I dont have a notebook that have few images. The notebook have 618 images as it could be seen in the video before images output says that 1/618, 2/618 .... 618/618. Those are the images is saved in storage after that they are being displayed. As far as i noticed those movements only occurs when notebook is being loaded. Because if load finishes that movement seem to stop as well. Another thing is there is something wrong about loading process as it could be seen in the video. I especially didnt use mouse's wheel not to confuse you, i only used scroll bar on the right. VS Code stops loading/showing images after some point untill i scroll down bottom and if i come back it suddenly appears. And i wanted to show how google colab loads in the browser.\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nare there verification steps for this?\nverification steps a notebook and put a small image in your local workspace a cell with the code: the cell and wait for completion, (finished for me in a few seconds) , close, and re-open the notebook the images might take a bit to all render, but they shouldn't bump up and down once they are visible\nSame bug (bumping up and down) still persists for me. Another thing is loading of images are stuck some point and it's fixed if i scroll down to bottom and come back. I showed and explained the bug which was in vscode.mp4 video and duration between 00:42-1:20."}
{"_id":"doc-en-vscode-e1b26aa0c690f8c544903d3af59b06d9b6187b66a4109ac8152875d5f93b0f08","title":"","text":"Version: 1.79.2 (user setup) Commit: Date: 2023-06-14T08:57:04.379Z Electron: 22.5.7 Chromium: 108.0.5359.215 16.17.1 V8: 10.8.168.25-electron.0 OS: WindowsNT x64 10.0.\nThis fix is only on VS Code Insiders right now. It looks like you're still using the stable version."}
{"_id":"doc-en-vscode-29f005761b2c01f7cdc70fd17ab721c2ad9bc68f5ab69eeb8601aae35dec9d3a","title":"","text":"Type: <bBug</bPressing Alt+] and Alt+[ in Accessible View includes redundant aria alert before new content. On Windows, NVDA and JAWS users hear the following info first whenever they press Alt+] and Alt+[ in Accessible View: I have tested with Accessible View against inline suggestion. Please exclude this message from the aria alert. VS Code version: Code - Insiders 1.82.0-insider (, 2023-08-17T05:33:29.141Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (89)</summaryExtensionVersion ------ android-dev-ext1.3.2 Bookmarks13.4.1 openscad1.2.1 spellright3.0.116 zoterolatex0.4.1 mermaid-markdown-syntax-highlighting1.5.2 doxdocgen1.4.0 vscode-markdownlint0.51.0 vscode-eslint2.4.2 vscode-quick-select0.2.9 vscode-deno3.20.0 gitlens14.2.1 EditorConfig0.16.4 prettier-vscode10.1.0 vscode-google-translate1.4.13 codespaces1.14.15 copilot1.104.343 copilot-chat0.7. remotehub0.60.0 vscode-github-actions0.26.1 vscode-pull-request-github0.70.0 easy-snippet0.6.3 path-autocomplete1.24.1 latex-workshop9.13.4 lilypond-syntax0.1.1 scheme0.2.0 better-cpp-syntax1.17.2 google-search0.0.1 vscode-lua-format1.3.8 lilypond-formatter0.2.3 lilypond-pdf-preview0.2.8 lilypond-snippets0.1.1 vslilypond1.7.3 zotero0.1.10 git-graph1.30.0 vscode-docker1.26.0 black-formatter2023.4.1 flake82023.6.0 isort2023.11. python2023.14.0 vscode-pylance2023.8.30 jupyter2023.7."}
{"_id":"doc-en-vscode-5d289aeca957328ba5cd80aa92b1c02c76371826bee63719b6c5712c47991ead","title":"","text":"jupyter-keymap1.1.2 jupyter-renderers1.0.17 vscode-jupyter-cell-tags0.1.8 vscode-jupyter-slideshow0.1.5 remote-containers0.305.0 remote-ssh0.105.1 remote-ssh-edit0.86.0 remote-wsl0.81.0 vscode-remote-extensionpack0.24.0 azure-repos0.36.0 cmake-tools1.15.31 cpptools1.17.3 cpptools-extension-pack1.3.0 js-debug-nightly2023.8.1117 powershell2023.6.0 remote-repositories0.38.1 vscode-github-issue-notebooks0.0.129 vscode-selfhost-test-provider0.3.16 vscode-serial-monitor0.10.0 vsliveshare1.0.5877 resourcemonitor1.0.7 autodocstring0.6.1 pandocciter0.10.3 shiny-python0.1.2 shinyuieditor0.4.3 quarto1.94.0 r-debugger0.5.4 java1.21.0 vscode-xml0.26.1 r2.8.1 multi-command1.6.0 vscode-deepl1.0.6 abc-music0.4.0 lua3.6.26 latex-utilities0.4.10 cmake0.0.17 errorlens3.13.0 intellicode-api-usage-examples0.2.7 vscodeintellicode1.2.30 vscode-arduino0.6.0 vscode-java-debug0.53.0 vscode-java-dependency0.23.1 vscode-java-pack0.25.13 vscode-java-test0.39.1 vscode-maven0.42.0 markdown-all-in-one3.5.1 grammarly0.22.1 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\ncan you pls try in insider's? I made some changes yesterday that should've fixed this.\nThis issue still exists in the latest insider that I have updated today.\nThat info, 'Inline Completions Accessible View' was formerly provided via the aria label."}
{"_id":"doc-en-vscode-1c25312822fff3904381d8dae47b03f91ee0213f165bba32bf24c6944c1f5754","title":"","text":"With this change, , it just says 'Accessible View', but when I use the keybindings to go to next/previous, I only hear the new suggestion via as expected.\nThat’s odd. I still hear that redundant label info in Accessible View for inline suggestion when pressing Alt+[ and Alt+]. I have tested with NVDA and JAWS on Windows 11.\nWhat happens for me is: inline completion accessible view \"Accessible View, edit text....\" 's some delay before the screen reader moves on from reading the initial text then I'm alerted with the new suggestion\nIf we wanted to prevent this delay, we'd need to hide the context view, wait 50 ms, reshow it. The issue with that is then it would do exactly what you describe and read the aria label + edit text before reading the new suggestion. That's what was happening in insider's yesterday and I realized was not ideal, so changed it.\ncan you reproduce this in today's Insiders?\nit is working for me as described. I also debug with NVDA, and confirmed that it is processing the focus events before the liveRegionChange event. However, if I use Alt + left or right brackets, it behaves as we expect. If you hide the accessible view for 50 ms it might work, simply because NVDA will not receive any focus event before the alert.\nI get slightly nervous about being so prescriptive when it goes to screen reader behavior. You just need a small change in the way the screen reader handle things for all these hacks to turn against us. There are a few that I feel comfortable recommending.\nThanks for confirming I've only tried macOS/ VoiceOver, so maybe the repro happens only when using NVDA. Here was the change where I had that delay\nI can bring that change back if it seems reasonable to you. It's just like the other one that you tested and said works well\nwe still need to setup self hosting for you .\nOh yes. self-hosting is fun."}
{"_id":"doc-en-vscode-957b8c2f4c57f6459a5aba7f0e88738a18eb68e0d13e40e450dff1a529b56fda","title":"","text":"Needs to be fixed to reproduce the issue Repro: log level of \"Pty Host\" specifically to trace a terminal to verify trace logs show up in pty host channel the restart pty host command a terminal to verify trace logs show up in pty host channel, they don't show up\nsome help would be appreciated here. When the process is restarted the logging just stops working completely. Here's where the utility process environment is created: Here's the logger initialization on the pty host proc: If you try repro you'll need this change\nI can reproduce it.\nIt seems pty process is creating the log file in a new folder instead in the logs home folder. <img width=\"386\" alt=\"image\" src=\"\"It is because a new folder is getting passed to the pty process from main process. Fixing that would fix this.\nWeird. Thanks, I'll look into that\nThis is causing following regression - only last 5 log sessions are being retained instead of 10. Hence moved it to current milestone and prepared a fix for it.\nso the logs do show up, but the terminal doesn't seem to get created correctly - I cannot type in it. EDIT: after restarting the pty host, I cannot type in any of the terminals.\nthe terminal created before the pty host restarted receives input but does not display it. the one created after it restarted does not receive it.\nI don't see this log\nI ran it again and now can type in the terminals, but there are no logs happening when I type or create a new terminal !\nAs long as the regression fixed in is good, this is low priority. Will check it out again in July.\nFlaky repro, but sometimes a window might not ack the restart. See the bg window correctly shows the restart and relaunches, but the fg doesn't: <img width=\"1008\" alt=\"Screenshot 2023-06-30 at 8 02 18 am\" src=\"\"Restarting again fixes it, I can't repro any problem with the log not working though.\nWeird, when this happens data events aren't getting fired for just one window: !"}
{"_id":"doc-en-vscode-83648e4ce3f8acb68b9d8d0c34a8576d99a79801fd191471c617751a3a031964","title":"","text":"On input in the broken terminal: <img width=\"747\" alt=\"Screenshot 2023-06-30 at 8 31 15 am\" src=\"\"\nThe 2 windows are connected to 2 different pty hosts: <img width=\"989\" alt=\"Screenshot 2023-06-30 at 8 38 10 am\" src=\"\"\nI'm closing this issue off again as the logging problem was fixed in the linked commit. I investigated the problem hit and created some more targeted issues to fix:"}
{"_id":"doc-en-vscode-30d3e92cbda311ce4420166dda5c84f13369957e8045ce2e39e955b0e315d744","title":"","text":"Issue Type: <bBug</bWhen : one mouse wheel scroll in extension view: moves 100px one mouse wheel scroll in settings editor: moves 100px When : one mouse wheel scroll in extension view: moves 100px, with slight animation one mouse wheel scroll in settings editor: moves 15~25px, without animation I'm using a Logitech G502 Lightspeed mouse, without any scrolling tweaking softwares. ! VS Code version: Code - Insiders 1.48.0-insider (, 2020-08-05T05:30:25.429Z) OS version: WindowsNT x64 10.0. <details<summarySystem Info</summary </detailsExtensions: none <!-- generated by issue reporter --\nI don't see this, any idea\nI believe this setting is ignored for macOS. Duplicate of\nActually not a duplicate. Interesting. I need to get my hands on a mouse to see if I can repro.\nCreated a PR that fixes this:"}
{"_id":"doc-en-vscode-6caa0a8d7e73b07f06d6268a756e3d9de5b424453f93b7d9660ea3a7e3122655","title":"","text":"Repro: in a terminal and select something ! We should hit ctrl+c beforehand () which should work properly on all shells\nWe should also do this when is clicked in the decoration's context menu\nI attempted this in but it was more complicated than I originally anticipated. Moving to backlog."}
{"_id":"doc-en-vscode-984ef73d9d262eaa9b569304dad6d804770b69f3403d7b1c2bb5f3fe22453b46","title":"","text":"Light themes have relatively low contrast for foreground and hover colors. ideally we do not need to create a new color. Either we can tweak the default list colors which are used for quick pick , SelectBox, Explorer et cetera. Light Default: ! Bad Hover contrast: ! I suggest we increase the contrast for the default listHoverBackground applicable to all users. For the foreground color we could make this slightly darker again for the default. I don't see why we would create new colors either, I think whatever is good for selectBox should be the same and good for the quick pick Tagging this \"under discussion\" hopefully that is correct.\nI finally figured out the listFocusForeground has no default, it's inherited from Monaco-shell. If we define it and listFocusHover in light-defaults we won't create any other colors. Screenshot does not do justice to foreground, you probably want to try it on your own machines. Let me know what you guys think. Below light+ with foreground, hover darkened: foreground === originally hover === # !\ncan you make a PR for me to try out your suggested color change?"}
{"_id":"doc-en-vscode-adbc9dcd7a8a106118ccfd069414ec071a555d997c509a0119b369b10877bc93","title":"","text":"When I use the command line code --install-extension {id} Cannot find the installed extension Vscode - userdata: $HOME /. Vscode/extensions/extensions. The json. This error only appears in version 1.90.0, but does not exist if you go back to version 1.89.0.\nThis only happened the first time, and it worked the second time I executed the command line to install the plugin. So did everyone else around\nAre you still seeing this issue? If so can you please provide steps to reproduce?\nMy plug-in is a private plug-in that updates through the code command. When I released the updated version of the plug-in, I used the code command to update it, and there would be an error the first time, and it would be successful the second time, which was stable. !\nTLDR: The issue happens when a version that was installed before is installed again I have pretty much the same scenario. A private extension that uses to update itself when a new version is available. This is the error that's logged: I do have an entry for my extension in the file though: I did some testing on a bare project to understand the failure mechanism. What I found out is that if a particular version of the extension is installed before (for example during my testing before I release the extension) then upgrading/downgrading to that version will fail. Using following steps I can reproduce this issue reliably: 1- Create a bare project: 2- Package the extension: 3- Increment the version number, and package again: 4- Install the first package: 5- Update to second package: So far everything is in order. The problem begins when we try to install any of those packages again: 6- Downgrade to first package: At this point the neither 0.0.1 nor 0.0.2 is shown as installed. So installing any of them will be successful. this explains why you're always getting the error on the first, but the second round is successful. 7- Install the first package again: 8- Update to second package again: Tested this with Windows and WSL Ubuntu. Happens only on WSL Ubuntu in my case.\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected."}
{"_id":"doc-en-vscode-7186fd3e8a3f9b4e672f3333f6266f2fb5ca8d5b9b79174b40b49028715e0528","title":"","text":"If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nLooks like this is working now in the latest VS Code Insiders. Tried with WSL: And using the latest stable (which doesn't have the fix yet) I can repro the errors that describes."}
{"_id":"doc-en-vscode-ffa05f93844bfd44faba15028f1d49c95a47eed4a95cbf51ab62c68a3716674f","title":"","text":"Testing microsoft/vscode-jupyter The setting description says that the 'interactive window session' is restored. But after a reload, it seems that the state has been forgotten e.g. Open the interactive window run run prints 9 reload the window the output is restored but running again will show Note that I had to retype print(x) as there was no way to rerun the previously shown cells What's restored seems to be the last output. Maybe the settings description can clarify that?"}
{"_id":"doc-en-vscode-3953879d94adbf615bab60c2ad96213ae530c8ecc6f9fd5452d420ecbe0876c6","title":"","text":"Testing microsoft/vscode-jupyter IMO is about restoring a editor with unsaved changes, but the interactiveWindow doesn't seem to have a dirty state. Maybe just call it 'interactiveWindow.restore`? It's actually not really a window ;-)\nGood point Martin, I think calling it makes better sense. is a bit ambiguous, we can revisit the name for it in the core when we stabilize the API."}
{"_id":"doc-en-vscode-f7f48a673ebd73a7022ee17c4e0d53aee6b22f2977c80628402bae592daf8d2f","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: No, issue is with extension contributed context menus. <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.89.0-insider OS Version: Mac OS X (ARM) Steps to Reproduce: a context menu in Stable and compare it to the same menu in Insiders. I verified the issue reproduces for multiple extensions including GitLens. Stable: <img width=\"424\" alt=\"image\" src=\"\"Insiders: <img width=\"420\" alt=\"image\" src=\"\"\nIf this is a confirmed regression, it would help to know which change caused it. We maintain a based CLI tool  that automatically downloads and runs previous VS Code insiders and asks for each build if the issue reproduces. It takes up to 8 steps to find exactly the build that caused the regression. Steps: install (if not yet done) follow the instructions until you found the offending build (its fine to leave the commit empty) report back the commit range in this issue Thanks!\nBisect points to My money would be on Reproduces by looking at the context menu of \"Commits\" when GitLens is installed: !"}
{"_id":"doc-en-vscode-880c74b31d2fe9905ad527e05bf2b6cd4ee75d12f52639e5440dc86f9715591d","title":"","text":"Notice how for the task terminal it jumps all the way to the top on the first ctrl+up, instead of going to the most recently run task: !\nWhen I run the build task and do cmd + up, the window crashes !\nThat's even worse\nthe call to get markers here returns undefined !\nI bet the markers are getting disposed of, so deleted from the map in but it looks like that wouldn't even happen since we don't set like we should. Could be why I see the renderer process crash?"}
{"_id":"doc-en-vscode-f230f6ada690183d71c6f9470cf6b799979957b28593e51c549d7abd416efd82","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --So far, I haven't been able to find any issues around this, but I also haven't found a ton of documentation on this feature either. Basically, I'd like to have finer grained control over how the terminal \"word link\" auto detection works. This is one of the only comments I was able to find about: Example: In the terminal, I have some output from a compiler that includes relative paths that can be clicked conveniently as an autodetected \"word link\": Specifically, clicking takes me to the line in question, which is great! However, at some point I tried configuring to include a forward slash (colon has related effects too), e.g. By doing so, the auto-detected file path is no longer clickable, since each path segment is considered as a separate word instead. Notably, the absolute path above is still clickable, but relative paths appear not to be (unless I remove from again). I guess there is probably more than one way to solve this, but a simple solution that comes to mind would be to have a separate config, maybe something like . If null it could fall back to the value, but at least this way it would be possible to configure different separators for double-click selection and link detection. Thanks in advance!\nThis case is meant to work with the regular non-word link detection as you can see here: <img width=\"490\" alt=\"Screenshot 2023-05-08 at 6 31 15 am\" src=\"https://user-\"This may not work is if the relative path is not relative to the current working directory or you have shell integration disabled (which is provides better cwd detection). In your example is relative to the directory the terminal was currently on? The word links are meant to catch the majority of cases where the verified file links fall through.\nEdit: never mind, I think I was testing with the wrong value of as I described in the original issue body :( I did however notice an interesting data point while integration was still disabled: the clickability of the file path seemed inconsistent. Here are a few screenshots showing what I mean (ctrl held in each case): Can't click in a compiler diagnostic, maybe due to the prefix on that line? !"}
{"_id":"doc-en-vscode-712d3f2913d3d223c3b5cb5e0211d42bae2e25e6d3c5b26cfee5397fd26a1239","title":"","text":"A command I wrote without line/column numbers works fine: ! A command I wrote with line/column numbers works fine: !\nIt's a little hard to tell if this is by design or not without knowing the exact cwd and full path of these folders. What we're going for is verify if the file is real if it's relative to the current working directory (which is citro3d-rs in your last 2, not sure for the first), otherwise fall back to word links. Since you changed wordSeparators, the fall back isn't as useful, but it looks like it's working properly.\nAh sorry I forgot to include those extra details. The cwd and path was the same in all 3 cases (i.e., the relative file path should work in each case, I think?): I guess the part that doesn't make sense to me is why the latter two cases work but the first one doesn't - since in each case the path is the same and I haven't changed directories at all. I suppose the workaround of removing and from is still an option but it would be nice if the linking was more consistent.\nYeah this is strange if it's all the same cwd. I Can't repro here I'm improving trace logging in , when that is released to insiders (the day after it's merged) you will be able to set the log level via the command palette (F1) and view the trace logs when hovering the terminal line by looking in the Output panel Window dropdown entry. That will give us some more info on why this is happening.\nThe logging changes should be in tomorrow's insiders\nthanks for that, here is the trace log output that I think is relevant for debugging. Please let me know if you need more info or another reproduction: Edit: the cwd again was and the relative path should resolve: <details<summarytrace output</summary</detailsOne odd thing I noticed - without changing any settings, I actually had one instance where it worked as expected and linked the relative path to . However I'm not sure how or why it worked. When I opened a new terminal the old behavior (not linking) came back. If I can reproduce that again I'll try to capture those trace logs and maybe it will highlight a difference in behavior that's helpful. Thanks again for looking at this!"}
{"_id":"doc-en-vscode-4252c37dbb92a739512ce5c6765ce345bca1ac6486e76d1b86017876094cf6b9","title":"","text":"So there's this in the logs: So it's parsed as a possible link, but not made a link candidate (which checked the file system before making the link). Somehow this part isn't adding the candidate: The only way I think that could happen is if doesn't return anything. I another log to verify the cwd and also improved fallback in case it is undefined. Can you do the same again on the next Insiders which should come out on Monday?\nThis issue has been closed automatically because it needs more information and has not had recent activity. See also our guidelines. Happy Coding!\nsorry for the delay, was a bit busy last week but I was able to check with a recent build (): The link is clickable now, even with and both in ! I'm guessing the fallback changes you did the trick. It seems the triage bot closed this as \"not planned\" but maybe you are able to update its status to \"fixed\" or similar? Thanks for your help looking into this issue!\ngreat news! I still don't totally understand how it was failing before but at least the issue is fixed"}
{"_id":"doc-en-vscode-509bb392bc913b82d12528213fee7d584fc4143e6cc7b5f34f5c3c9b979da33c","title":"","text":"Type: <bBug</bterminal in editor area. and hit enter. and hit enter alt+F2 to open accessible view ctrl+shift+O to open symbol list and select and hit enter Note: the focus doesn't move to VS Code version: Code - Insiders 1.83.0-insider (, 2023-09-25T10:41:26.295Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (91)</summaryExtensionVersion ------ android-dev-ext1.3.2 aiprm-lang0.0.2 Bookmarks13.4.1 openscad1.2.1 spellright3.0.118 zoterolatex0.4.1 mermaid-markdown-syntax-highlighting1.5.2 doxdocgen1.4.0 vscode-markdownlint0.52.0 vscode-eslint2.4.2 vscode-quick-select0.2.9 vscode-deno3.23.1 gitlens14.3.0 EditorConfig0.16.4 prettier-vscode10.1.0 vscode-google-translate1.4.13 codespaces1.15.3 copilot1.115.437 copilot-chat0.8. remotehub0.60.0 vscode-github-actions0.26.2 vscode-pull-request-github0.72.0 overleaf-workshop0.2.4 cslpreview0.2.2 easy-snippet0.6.3 path-autocomplete1.25.0 latex-workshop9.14.0 lilypond-syntax0.1.1 scheme0.2.0 better-cpp-syntax1.17.2 google-search0.0.1 vscode-lua-format1.3.8 lilypond-formatter0.2.3 lilypond-pdf-preview0.2.8 lilypond-snippets0.1.1 vslilypond1.7.3 zotero0.1.10 git-graph1.30.0 vscode-docker1.26.1 black-formatter2023.4.1 flake82023.6.0 isort2023.11. python2023.16.0 vscode-pylance2023.9.20 jupyter2023.8."}
{"_id":"doc-en-vscode-b657946cba64c525449c861168d025d9fc2163a65206f47832e8e5908681ec88","title":"","text":"jupyter-keymap1.1.2 jupyter-renderers1.0.17 vscode-jupyter-cell-tags0.1.8 vscode-jupyter-slideshow0.1.5 remote-containers0.312.0 remote-ssh0.106.4 remote-ssh-edit0.86.0 remote-wsl0.81.4 vscode-remote-extensionpack0.24.0 azure-repos0.36.0 cmake-tools1.15.31 cpptools1.17.5 cpptools-extension-pack1.3.0 js-debug-nightly2023.9.2117 powershell2023.6.0 remote-repositories0.38.1 vscode-github-issue-notebooks0.0.129 vscode-selfhost-test-provider0.3.18 vscode-serial-monitor0.10.0 vsliveshare1.0.5883 autodocstring0.6.1 pandocciter0.10.3 shiny-python0.1.4 shinyuieditor0.4.3 quarto1.100.0 r-debugger0.5.4 java1.22.1 vscode-xml0.26.1 r2.8.1 multi-command1.6.0 vscode-deepl1.0.6 abc-music0.4.0 lua3.7.0 latex-utilities0.4.10 cmake0.0.17 errorlens3.13.0 intellicode-api-usage-examples0.2.8 vscodeintellicode1.2.30 vscode-arduino0.6.0 vscode-java-debug0.54.0 vscode-java-dependency0.23.1 vscode-java-pack0.25.14 vscode-java-test0.40.0 vscode-maven0.42.0 markdown-all-in-one3.5.1 grammarly0.22.1 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nThanks for the report! Reproduces in panel terminals also\nFor step 5, after I select and hit enter, the cursor moves to the beginning of the line that contains , rather than after the prompt. Is that intended behaviour?"}
{"_id":"doc-en-vscode-00c5968faa9db89bd9c39cfedb1247c819e6bbb54b6a30ab7163237869410b04","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --When using the Git API extension Repository interface and the corresponding are not exposed. Having access to these functions would be very helpful, since using with returns early and doesn't provide useful errors.\nVerified by code review"}
{"_id":"doc-en-vscode-464cc3ca4d54ac01634733d95c572a66710ca6eacbb0b840dc2de7283312b6d2","title":"","text":"use generate cell in notebooks make a request while the request is active press stop a bunch of errors are printed\nIt seems not wise at all to maintain my own *chatcontroller. I'd love to look into how do I ensure I reuse the most of Inline Chat Controller later.\nIs the generate cell feature just with using ctrl+i from within a cell body? Not sure how I can trigger it."}
{"_id":"doc-en-vscode-c650496e0f299105626d025cb7dc764b4942c60db977ad97462cf63b75bb51ee","title":"","text":"It would be nice to have a entry in the configuration section of markdown-math extension. A place where users could put their custom math macros.\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\n/verified"}
{"_id":"doc-en-vscode-5086a239b16010ce421616da6537816297c7146b1d093b38bb5f0452ab297f35","title":"","text":"Friendly ping! Looks like this issue requires some further steps to be verified. Please provide us with the steps necessary to verify this issue.\nVerification steps: Open a folder/workspace that contains a git repository Use the \"Git: Checkout to...\" command to \"Checkout detached...\", and select a branch/tag Switch to the \"Source Control\" view and confirm that the \"Commit\" action button is shown instead of \"Publish Branch\""}
{"_id":"doc-en-vscode-63c24a01328ccadc46f2dc75701aa9553a8d5c124e73d98d8cbcb6c04ba1a500","title":"","text":"Testing When turning off , the decorations disappear. When turning back on, they do not reappear automatically. !"}
{"_id":"doc-en-vscode-087b549859019fce58ec48ace7186be4580d29c1fa4bf93cbbf14dba9e5ce142","title":"","text":"Testing After enabling remote tunnel access, a link like this is printed: The browser ends up redirecting to It would be nicer to genrate the vanity format"}
{"_id":"doc-en-vscode-d53ec58b37de46549635067af05b79be104ce7b9882dbe4a8716c20fbd5636c6","title":"","text":"Start legacy protocol debug session Open something via \"Open loaded scripts\" scroll down ctrl+tab to switch to another doc ctrl+tab back to the newly opened script It's scrolled up to the top. Is it being closed/opened each time we switch away?\nThis problem is independent from the \"Open Loaded Scripts\" feature. To reproduce: debug a simple node hello world program and stop on entry in the CALL STACK select an internal module press Control-Tab twice (to switch between two editors) Observe: you loose the scroll position in the internal module\nMarch to investigate\nWorkbench is not preserving scroll location of in memory resources. Same is true for diff view. Forwarding to since he probably already has an issue for this\nI do but I cannot find it right now, PR welcome.\ngive me some code pointers for the PR and I can do it in debt week\nwe do it for and via the and methods that are actually defined in and we could probably push this down from and and just use it from textEditor for any input that has a resource\nOk, makes sense. Assigning to me and April to look into it\nRelated:\nWith landed, we keep the UI state when you switch editors, though we clear it when you close a tab. The rationale is that the lifecycle of these editors that have virtual resources is not known, as such we assume the view state can be ditched once the editor closes.\nThat sounds perfect!"}
{"_id":"doc-en-vscode-8d7ce9627ab453856cdca4b967a241cd2275bbf8a8b7331bae3c5991a593f565","title":"","text":"Open Select the top cell and page down a few times Focus hits a cell that is longer than the viewport, which is maybe important, then the next page down moves to a higher cell\nWhen a list item is larger than the viewport, and it's also the only element in the viewport, pressing page down will reveal to the previous element, this is because we did a here (the catch is, the last page element is also the first page element)"}
{"_id":"doc-en-vscode-21822ecd1940874630a7448c75e3e3c7f67b77b05184c0c1f43b8245b739fffb","title":"","text":"only seen in OSS so far outline view shows 'No symbols found in document '*.ipynb' or a very limited amount of the outline logs show an exception:"}
{"_id":"doc-en-vscode-5687a053025a5e3da1068af91458b3f08cf4f69b34400d040fc4afc5ed758e57","title":"","text":"Testing I'm wondering if there's a way to add a schema so that users can get descriptions on the various fields they can use while creating a new profile. I also realize that the keys represent profile names, so that might make it harder to validate/provide descriptions for them. I guess the description for the keys would go under in that case, though there already is a long description for that field.\nthis would be nice, looked into it a bit but we're not sure how to set it up for such an object. We essentially want to enforce something like\nwould you be the right person to ask about schemas for settings?\nYou can use the property to specify the shape of all properties regardless of the key. An example is the setting: If there are constraints on the key, you can use :"}
{"_id":"doc-en-vscode-08bca2b1bebddfd153b09befe0c68b78b4ac86e5a35dad1b897147a51f97335a","title":"","text":"Testing Type in the settings editor. It would be nice if there was autocomplete for a list of languages. This is a finite list which we should be able to suggest fyi !\nIt's easier to discover language overrides in the editor because there is autocomplete there: !\nThanks to for pointing out the language service to me the other day. PR ready above.\n/verified? The autocomplete works but no matter what I write there I get the same set of settings? I suppose all of these apply to all languages. ! !\nYep, the idea is for it to show all language-configurable settings. Then, the user can read and write settings for that language. I plan on making it so that the tab says which language you're on, for extra confirmation."}
{"_id":"doc-en-vscode-2ab317419f324247b244a6db827dcc00a17f7a21c991b39d5365c9b013523653","title":"","text":"Type: <bBug</bterminal git bash Enter and you will get delay VS Code version: Code - Insiders 1.90.0-insider (, 2024-05-06T08:07:13.626Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (42)</summaryExtensionVersion ------ rust-bundle1.0.0 glance0.0.8 project-manager12.8.0 redmodscript-lang0.2.0 redmodtweak-lang0.1.0 lit-html1.11.1 glsl-literal1.0.6 vscode-markdownlint0.55.0 vscode-eslint2.4.4 composer-php-vscode1.45. intelli-php-vscode0.12. phptools-vscode1.45. profiler-php-vscode1.45. rust-syntax0.6.1 gitlens14.9.1 EditorConfig0.16.4 prettier-vscode10.4.0 auto-close-tag0.5.15 auto-rename-tag0.1.10 copilot1.186.856 copilot-chat0.16. todo-tree0.0.226 path-autocomplete1.25.0 redscript-syntax-highlighting0.0.3 auto-comment-blocks1.0.1 dotenv1.0.1 debugpy2024.6.0 python2024.6.0 vscode-pylance2024.4.1 remote-wsl0.88.2 hexeditor1.10.0 vscode-extension-auto-import1.4.3 indent-rainbow8.3.1 vscode-yaml1.14.0 rust-analyzer0.3.1950 crates0.6.6 shader1.1.5 vscode-styled-components1.7.8 vscode-stylelint1.4.0 vscode-mdx1.8.4 highlight-matching-tag0.11.0 vscode-import-cost3.3.0 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\ndoes this reproduce if you disable shell integration, ?\nIf I uncheck Shell Integration, it works well after terminal restart.\nI have more details in with what's going wrong."}
{"_id":"doc-en-vscode-43d111a71d7942b43433088c6b01a9437d2f74e8c7edf9899a6dd78720839e36","title":"","text":"The issue was introduced by this PR where becomes extremely slow\nI had the same issue and setting to does the trick.\nWhen I just hit the enter key on a bash prompt it takes about 4 seconds for the prompt to return. I ran process monitor and noticed many many (maybe over a hundred) instances of being spawned to run the shellIntegration- script.\nAny idea where this ticket is tracked? Every reference to the issue is closed marking duplicate.\nClosing an unfixed issue that everyone have ? I don't get the point\nI closed this accidentally, have reopened.\nworked for me !\nHey guys I'm having the same issue. Running a single is taking over 15 seconds to complete the command. After disabling , bash looks way better\nThe problem seems to be related to The issue happens also if I start (outside VsCode) my bash thus (same as VsCode does for its terminal): Each time the prompt must be shown (eg, after simply pressing the Enter key) a delay of about two seconds happens. Most of the delay is caused, it seems, by the function In particular, commenting out the line 178 seems to alleviate good part (not all) of the delay. Edit: More specifically, the culprit seems to be If I change the line 109 from to (effectively forcing \"the switch to the faster function\"), the problem goes away. (BTW: the faster function seems to have a bug of its own, it produces an extra prompt!)\nProcess monitor screenshot shows it re-running this over and over 89 times when I press enter at the shell prompt \"C:Program FilesGitbin..usrbin\" --init-file \"c:UsersarthuAppDataLocalProgramsMicrosoft VS Coderesourcesapp/out/vs/workbench/contrib/terminal/browser/media/shellIntegration-\" !\nDisabling is another workaround that improves the delay for me.\nHello Guys, This setting worked for me.\nThis helped me, thanks!\nwill move you to the old backend stack which significantly reduces the functionality/capabilities of the terminal. shouldn't do anything. is the right workaround for now, expect a fix soon though\nwhat caused so many bash invocations when shell integration is enabled?\nwe started reporting the bash prompt in order to better position shell integration markers and understand the prompt."}
{"_id":"doc-en-vscode-4a1eb52168fe83d0b9ff7bf301ac1508bf8da413cc844b51175f04b8c21653c5","title":"","text":"Apparently printf is just extremely slow on git bash\nI had this terminal slowdown problem, in git bash, and found setting fixed the problem - the terminal was back to being fast. However, after the last, or recent, insiders update, even with that setting set to , the terminal is slow again - and it's showing the icons, etc. like it is ignoring this setting: ! !\nyou probably just didn't restart the terminal. This is now fixed and will come in the next Insiders and likely 1.90.1\nActually, I had the shell integration disabled for the last several insiders updates, so it has been off for a while. Terminal windows have been created and closed many time since. For whatever reason, when launching VSCode and the terminal session was restored, this last time, the integration feature was ignoring the flag. I terminated the terminal and started another, and the integration went away again. So basically the integration flag was ignored when the terminal session was restored. I just did a quick test of quitting VSCode, with a terminal window open, then relaunching, and the integration was still disabled. Unsure what happened. Guessing, also, you type-o-ed the version? I am already running 1.91.0 insider !\nwhat happened was the terminal was restored which re-creates the process after vscode starts with the original environment, args, etc. No the next insiders build on Monday and v1.90.1 (stable)\nwhen will be the next minor update, not the insider build so this terminal slowness is resolved?\nFWIW, I can also observe the prompt's slowness on Linux (bash 5.1.16(1)-release on Ubuntu 22.04.4 LTS) on the order of a half-second or so (still annoying but bearable). looks like a temporary workaround for Git Bash only; will we fix the underlying cause of the slowness?\nI also observe this to be slow over linux using an ssh remote. It's much slower if I have a starship prompt enabled (4s) and around 1s if I have the default debian prompt.\nI have this issue (Win 10, git bash). The suggested workaround of disabling shell integration then restarting terminals worked for me.\nAlso having this problem. Disable terminal integration worked for me too. (Win 11, Git bash)."}
{"_id":"doc-en-vscode-4b9a2b6b12e56c0e2a6494f90cb7da4128fccc870c6c6e3ad804738cabe13d6a","title":"","text":"I re-enabled terminal integration with the latest Insiders - ! and it the problem seems to have been fixed. For me at least.\nYes, it's been fixed in insider's already.\nPerfect solution\nWhen is the release date for the VSCode stable one?\nno exact date, but probably later this week\n, Turning off shell integration only sorta helped me out. It does return a result slightly faster, but it does not return the prompt back for 4 seconds or so, but only with python poetry workspaces/folders opened. This slowness goes away if I don't have a workspace/folder open, a non-poetry python one, a javascript one, etc. There's now a '(packagename-pythonversion)' at the top of my prompt when I open a python poetry folder and that seems to be taking forever to generate after each command. Definitely wasn't happening before my last update. This seems like a separate issue? Or is it related somehow? Happy to open a separate Issue if needed.\nI don't know if this is the same as your issue but I've found that, if I already opened a terminal in a workspace before changing the setting, I had to kill the terminal for that workspace and open a new one so it would be initiated with the updated setting.\nYes. With the feature disabled, issue does not happen . But the feature will not work also . I can live without that feature but I like it. It makes my IDE look smarter. It's strange. I remember the feature was there earlier without any issue.\nWas the fix included in 1.90.1? The shell integration is still too slow even after I updated to 1.90.1. It still takes over a second for all those bash commands to run every time press the enter key at the terminal. I have to keep it off to keep my Windows git bash terminal usable. It seems like it takes around a second and a half to get a new prompt with that setting on. Is this expected?"}
{"_id":"doc-en-vscode-e1b22097ce4bcaab816a28149b2c3f579ff3ed6bcd34ab7f845aa714de211340","title":"","text":"I just updated to this version: Version: 1.90.1 (user setup) Commit: Date: 2024-06-11T21:01:24.262Z Electron: 29.4.0 ElectronBuildId: Chromium: 122.0.6261.156 20.9.0 V8: 12.2.281.27-electron.0 OS: Windows_NT x64 10.0.\nit should be fixed yes, can you run and paste in the results?\n$ for i in {1..10}; do time _vscreport_prompt; done real 0m0.236s user 0m0.045s sys 0m0.092s real 0m0.194s user 0m0.045s sys 0m0.076s real 0m0.210s user 0m0.030s sys 0m0.122s real 0m0.217s user 0m0.015s sys 0m0.152s real 0m0.216s user 0m0.061s sys 0m0.138s real 0m0.245s user 0m0.045s sys 0m0.123s real 0m0.202s user 0m0.015s sys 0m0.091s real 0m0.212s user 0m0.061s sys 0m0.091s real 0m0.230s user 0m0.000s sys 0m0.153s real 0m0.212s user 0m0.045s sys 0m0.108s\nIt seems like it's taking a long time to parse the prompt. My git bash prompt is user MINGW64 /c/path/to/dir (master), where the my /c/path/to/dir is a 29 character path. But sometimes it's a lot longer. Does a longer path make it slower?"}
{"_id":"doc-en-vscode-e5a0c83dfdae8c2ca67dcff8c4cb0f6436818a2bf9cf06d265d284624f7d05d1","title":"","text":"When I cd to a 135 character path, and try to time it it just prints 633;D;0: $ time vscreportprompt 633;D;0 $ vscreportprompt 633;D;0\nthose numbers aren't too bad, but regardless I think we need to rethink how we're doing this prompt reporting feature. I'm going to stick it behind a setting in so it's just disabled completely by default.\nThis is still unreasonably slow, IMO. If I'm running a shell with the extensions the shortest possible interval between two commands seems to be over 700ms on my machine: ! Running without the shell integration, it's closer to 250ms, and that may only be because that's how fast I can press the ⇧⏎⇧⏎ keys: ! Despite the improvement over 1.90.0, I still feel like I have to disable the shell integration. It's 2024. Having to wait a perceptible amount of time just for the terminal prompt to re-appear feels like a huge regression to 1980s computing. Is my floppy drive trying to seek to a specific sector?\nv1.91 which I think is being released today includes the changes I mentioned in , disabling the prompt reporting stuff completely (for the stable build): !\nI'm running 1.91.0 (insiders) and I still see a significant delay between prompts (on windows, using git-bash terminal). Turning the shell integration off speeds it up quite a bit.\nit's only disabled in the stable build. Insiders keeps the slower code as it provided reliability improvements and the plan is to optimize it soon.\nWhile v1.91 does seem to improve the delay, there's still a noticeable delay for me so I'll be sticking with the option disabled."}
{"_id":"doc-en-vscode-b8a3a9e93895dc03f4706ed203f321dd8154fff3c2853695655fe7a8c5f2fe2c","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Hello, we have been recently testing out our extension for accessibility bugs, and we believe we may have found one directly linked to vscode. The extension we are using is Bridge to Kubernetes. Does this issue occur when all extensions are disabled?: Yes VS Code Version: Visual Studio Code Version 1.73.1. (user setup) OS Version: Windows 11 Version 21H2 (OS Build .376) Steps to Reproduce: on NVDA/JAWS. Visual studio code application and setup Bridge to Kubernetes extension. cluster config file and load 'stats-apo' project. 'Ctrl+Shift+P' open command palette in View menu. 'Bridge to Kubernetes: Configure' and press enter. using down arrow key to select 'stats-api' in connect to kubernetes (1/4) stage and press enter. '3001' in connect to kubernetes service: stats-api (2/4) stage and press enter. whether NVDA/JAWS is announcing the given placeholder text at edit field or not.\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.73.1. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!\nHi The bug still exists with the new vscode version. I edited the original post. Thanks!\nI think you just worked on improving the accessibility of the quick pick - thoughts on this?\nA11yMAS; #A11yTCS; #A11ySev2; #DesktopApp; #Win32; #Visual Studio Code Client; #BM-VisualStudioCodeClient-Win32-Aug2022; #WCAG1.3.1; #Element:Placeholdertext; #AILimited; #JAWS; #NVDA;\nhave you looked into the feasibility of supporting this?\nSorry I'm not sure I totally follow here."}
{"_id":"doc-en-vscode-d3a4812e2f7d5c546593d3a8638fd801d3979e969ca48fb8818a1da480eb8ee6","title":"","text":"Just to make sure, can you attach a gif? /gifPlease\nThanks for reporting this issue! Unfortunately, it's hard for us to understand what issue you're seeing. Please help us out by providing a screen recording showing exactly what isn't working as expected. While we can work with most standard formats, files are preferred as they are displayed inline on GitHub. You may find helpful as a browser-based gif recording tool. If the issue depends on keyboard input, you can help us by enabling screencast mode for the recording ( in the command palette). Lastly, please attach this file via the GitHub web interface as emailed responses will strip files out from the issue. Happy coding!\nfrom the above description, I believe what expects is for the placeholder text to be read before the selected item is. From my testing though, that is working already (I tried and\nThanks for the comments and I will try and get a gif for you folks in the upcoming days!\nsadly gif does not record audio. I will record a video and send it to you internally.\nvideo uploads on GitHub issues work now so you can always attach here :)\nI have attached a video below! Basically, the place holder text should be read with NVDA. The accessibility folks have flagged this as a bug. Thanks! https://user-\ndo you have Accessibility mode turned on? Just checking. You should have gotten a notification about it.\nyou can turn that on with\nLet me test that today thanks!\nwhere do I find ?\nand search for . The alternative way would be to run the command via the command palette\nThanks for this it was set to auto. I will turn it to on and try again\nIt is still not pronouncing the text in the edit field. Would you classify this as a bug or is this just a part of VsCode?\nI can reproduce the issue - I'd expect the placeholder text to be read before the selected item is, but that does not happen.\nExample: In a multi root workspace, run . I'd expect it to read to me: \"Select current working directory for the new terminal\"\nLogging shows that the quick input's aria label is correctly set. After that though, when an item is selected by default, its aria label takes precedence\nPerhaps when is true, we could not select an item by default so that the placeholder text will have focus every time?"}
{"_id":"doc-en-vscode-4bec6a57c2bbbb7036ff4301de1083fd77ca215254cf77e4e9df7617eda95410","title":"","text":"thought that was the behavior today? Just opened the Command Palette... focus is in the input: <img width=\"830\" alt=\"image\" src=\"https://user-\"\nI mean when you run a command that opens a quick pick such as !\ncommenting out this code fixes it\nso I think we need to check there if and not focus\nLooks like this use to be reported in\n! odd I don't repro with that command you have... wonder if we have a conflicting setting. Also VoiceOver did read out the placeholder in this case.\nfor me, the placeholder is not read in most cases (but it is with the changes in the PR)\nmy repro steps were when in a multi root repo, run that command. the first quick pick placeholder isn't read\nHi and I see this as moved to the backlog. Do we have a rough date in which this could possibly be looked at again? I plan on telling the testers that you two are looking at the bug. Thanks!\nI did it in \"holiday mode\" as a lot of us are disappearing for the holidays. I'll leave it on January\nThanks for the update that totally makes sense. I will let the testers know. Happy holidays!\ndo you see a way to do this? it's not clear to me where the item is getting focused\nis open to testing the PR once it is available\nWe'll try to fix this in February with the following read by the screen reader via : placeholder of input box (note this doesn’t logically have to be the placeholder, it could be an aria label set somewhere) number of items in the list the label of the highlighted item in the list (aka the first item in the list)\nI just discussed with and he says \"I need an ARIA expert because I can’t get this to work nicely. is what we use today on the quick pick input box. That points to the focused item in the list. This is needed because true focus always stays in the input box. The problem is that if an element has AND the is ignored. I’ve tried a few other solutions and haven’t had luck, unfortunately. Looking for creative solutions\" might you have suggestions here?"}
{"_id":"doc-en-vscode-e606b808d6aee95e1e998c1027219ddaa9657203acee614959f7de0a91942883","title":"","text":"Just to provide an update... After chewing on the problem yesterday, my best idea is to try to mirror this design pattern as much as possible: Applying it to the quick pick means making the following changes: in a11y mode, we do not focus any of the items in the list. Doing this, we can easily ensure that the placeholder is read out because aria-activedecentant won’t be set yet in a11y mode, When the user types something in the input box, aria-activedecendent is removed and we remove the \"active\" state of the elements (thus focus goes back to the input) When pressing down the first element is highlighted. aria-activedecendent is set on the inputbox pointing to that item. The item gets read out. Nothing new here. While this pattern is fine, it's not really what I want because the QuickPick isn't really a combo box (though I'll admit it's close). I’m sad because I would rather not have specific a11y mode behavior here… but this is the best I think we can offer with my ARIA skills. Tl;dr: I need a solution for reading the placeholder of the quickpick input box first and then read out the first highlighted item in the list (the aria-activedecendent)\nHi team. I’m reading the thread and I’m probably missing something but I’m concerned that there is perhaps a misunderstanding about how the quick input experience works in general and the particular issue of the Bridge to Kubernetes extension. Perhaps I’m missing the problem, but the quick input list works well for me. Here is an example: the command palette by pressing CTRL +Shift +P a. NVDA announces the label and placeholder text for the editable field as follows: “Type the name of a command to run. combo box has auto complete editable Press 'Enter' to confirm your input or 'Escape' to cancel”. type git, for example, and select the Git clone command with the arrow keys and Enter. a. Upon selecting the command, focus returns to the editable section and NVDA reads the prompt as follows: “Provide repository URL or pick a repository source. combo box has auto complete editable Press 'Enter' to confirm your input or 'Escape' to cancel Provide repository URL or pick a repository source. blank”."}
{"_id":"doc-en-vscode-86ac8d8381d508aae87e6fa83c5e8e9b22d261b9532cd5e5708c0556c58e4af0","title":"","text":"use the arrow keys to select the option to clone from remote sources and press enter on it. a. Upon pressing Enter, focus returns again to the editable field and the prompt is read again: “Repository name (type to search) combo box has auto complete editable Press 'Enter' to confirm your input or 'Escape' to cancel Repository name (type to search) blank”. I wouldn’t expect this to behave any different. The only instance where I have observed a different behavior is with the Bridge to Kubernetes extension where, upon selecting the connect command, for example, screen reader focus seems to jump directly to the list instead of the editable portion which is the standard behavior, maybe because aria-activedescendant is set at a time when it shouldn't? why you would like the placeholder text and the first highlighted item to be read together? I understand the placeholder text as the label for the editable field, therefore it is relevant only when the editable field is focused. By focused in this context I mean that it has screen reader focus, not keyboard focus. When we manipulate aria-activedescendant we are communicating the screen reader to focus on a different control, therefore the label of the control with focus shouldn't be read. The label of the list portion is \"Quick Input list\", and that is actually the hint that we are now exploring the list of options. I fully agree that we should stay away from an A11Y exclusive mode as much as possible.\nbecause this is the behavior of a sighted users today. They are able to see the placeholder in the input box and the first item in the list is already focused (using ). This allows a user to be shown the quick pick and be able to hit to accept the first item in the list. Putting the \"quick\" in \"quick pick\". Today, users of a11y mode via the setting have a different experience, which is what you describe... no item is initially focused and a user in a11y mode must hit down arrow to get to the first item in the list."}
{"_id":"doc-en-vscode-50c4e3db7d0606953f1d0418bf508c98800b4bd20e54c6beeff0e26be50ab94b","title":"","text":"My wish is that we could have the same experience for users of a11y mode and not... but my research above, albeit not expert level, has given me the perspective that it's just not possible... and that it would be easier to fix the bug you call out which can be simplified to: \"When we chain quick picks we need to remove on any item if one exists so that the 2nd quick pick reads the placeholder\"\nThanks I think I understand now. Since this is really not a combobox, I would then suggest a change to the way this works from a keyboard and screen reader perspective. Please take this as a general idea which could be significantly refined or completely abandoned. and Shift +tab can be used to move focus between the edit field and the list box. Tab works now, but shift +tab doesn't return focus to the text field. Visual keyboard users can press Enter in the text field to trigger the first element in the list which is selected. Blind screen reader users can tab to the list to hear / read the element selected and press enter from there. text in the text field changes selection visually and updates an aria-live region with the selected item in the list. In a way it is better than manipulating aria-activedescendant because the text is still editable. note: It should be a visually hidden live-region which will only contain the visually highlighted item at a given time. while focus is in the list filters the list as it does now but doesn't update the aria-live region because screen readers are tracking ffocus in the list itself therefore there is no need to announce anything. focus is in the text field, pressing down arrow changes selection in the list and moves focus to the list itself. quick pick changes resulting from selecting an option, we reset focus to the edit field so that the placeholder is read. Screen reader users can then tab to the list to hear the selected item / navigate, or start typing in the edit field. remove the combobox role from the text field because we agree it is not a combobox. note: As a blind screen reader user myself, I don't know how the experience looks / feels like from a visual perspective."}
{"_id":"doc-en-vscode-64161c51c4ee0a811f5fa902a45e7cca75a3a0e131b845f58b7c052863581bb1","title":"","text":"For example, I don't know if sighted keyboard users can edit the text as they type, or if there still an insertion point visible in the text field as ones navigates in the list with the keyboard as it is now.\nI think we are on to something here... Thanks for your input. It sounds like, from what I can tell, you are suggesting we drop aria-activedescendant usage from the input box and go with your alternative. I'm intrigued by this idea. I have a couple of follow ups mostly because my ARIA skills are quite low: you: Could you explain the difference between these two approaches? How are they represented differently via a screen reader? Just to confirm, yes, the inputbox that currently has is still editable and should be editable. We want users to be able to continue typing to further filter the list... just like a combobox. said: Today, when you type a character, the active item in the list becomes the first item in the list. I would imagine we would still announce via if that item isn't the one the screen reader user is on? clarify what you said here: Yes. Visual users are able to see a cursor in the input box that they can move left to right as they choose and edit the text in the box while an item in the list is active (again, similar to the combobox example I linked above). This works today because true focus is entirely kept by the input box. \"fake focus\" or what we call \"the active list item\" never actually gets browser focus, it just appears \"focused\" or \"active\" or \"highlighted\" to the user compared to the items in the list.\nHi ! quoting you: Aria-activedescendant overrides DOM focus. It means that, as soon as we add the property, screen reader focus goes to the list and screen reader no longer track the cursor in the text field. If we use an aria-live region however, changes in selection can be announced while focus is maintained in the input field for the user to edit. In this way if an element is selected in the list by default, we can announce that element and the placeholder text will be read by the screen reader due to the focus event. We may not be able to fully control the order in which both events are read, but that's a lesser problem in my opinion."}
{"_id":"doc-en-vscode-6855730d5c321f93b0723eeb1cfb084d7ba9a803c27f3e896047cbce36d616b9","title":"","text":"Does it make sense? Yes, that is correct. actually, your visual explanation has helped me understand the entire experience better. I initially thought that there was no cursor in the text field at all while the user was typing.\nthanks a lot for your comments! I really want to sit down and play with aria-live some more based on your idea. Considering that this issue is pressing, I've made a small change to get the placeholder to be read out.\nhappy to test that change when it becomes available in an insiders build, or remotely with you in case you want earlier testing. for my idea to work, it would require some changes in keyboard behavior as well, and careful testing so that we don't break current user expectations around a widget which is critical for some customers. feel free to reach out if you need additional input.\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nI did some testing using the latest insiders build and NVDA Results with screen reader mode on: Placeholder text is announced when the widget receives focus. When typing in the text field, the number of results is announced by NVDA but not the highlighted option. This can be a productivity killer. I believe the announcement of the highlighted option is the most important piece of information while typing. the number of results is irrelevant at this point. It becomes important when the user starts to navigate with the arrow keys, and at that time the screen reader takes care of it. Screen reader mode off: When the widget receives focus, the number of results is communicated by NVDA followed by the announcement of the quick input list and the highlighted element, but the placeholder text is not announced, likely because the \"combobox\" portion doesn't have focus. when typing, NVDA announces changes in the highlighted option in Braille and speech as expected."}
{"_id":"doc-en-vscode-79f80db776b4e2699bf8db832deb2b76c78292c7e6e19c70ba9565f695b77f2f","title":"","text":"Version: 1.77.0-insider (user setup) Commit: Date: 2023-03-23T10:21:45.873Z Electron: 19.1.11 Chromium: 102.0.5005.196 16.14.2 V8: 10.2.154.26-electron.0 OS: Windows_NT x64 10.0. Sandboxed: Yes\nClosed Verified the issue in latest environment: Version: 1.78.0-insider (user setup) Issue is fixed. Hence closing the bug. please find attachment for reference. https://user-"}
{"_id":"doc-en-vscode-f2d3ca3871a08e35903ce53e5825917bfbc30187c59ba865f3a59aef1c0407b1","title":"","text":"Type: <bBug</bthe settings like below: : new terminal via ctrl+` from terminal input. Pressing Shift+Tab moves to the terminal buffer. Shift+Tab should be passed to terminal as an input key. The focus needs to remain in the terminal input field. We need when cycling back through shell auto-suggestion or ipython completion. VS Code version: Code - Insiders 1.81.0-insider (, 2023-07-19T05:34:51.441Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (90)</summaryExtensionVersion ------ android-dev-ext1.3.2 Bookmarks13.4.1 openscad1.1.1 spellright3.0.116 zoterolatex0.4.1 mermaid-markdown-syntax-highlighting1.5.2 doxdocgen1.4.0 dscodegpt2.1.14 vscode-markdownlint0.51.0 vscode-eslint2.4.2 vscode-quick-select0.2.9 vscode-deno3.19.1 gitlens14.1.1 EditorConfig0.16.4 prettier-vscode9.19.0 vscode-google-translate1.4.13 codespaces1.14.12 copilot1.97.271 copilot-chat0.5. remotehub0.60.0 vscode-github-actions0.25.8 vscode-pull-request-github0.68.1 easy-snippet0.6.3 path-autocomplete1.24.1 latex-workshop9.13.1 lilypond-syntax0.1.1 scheme0.2.0 better-cpp-syntax1.17.2 google-search0.0.1 vscode-lua-format1.3.8 lilypond-formatter0.2.3 lilypond-pdf-preview0.2.8 lilypond-snippets0.1.1 vslilypond1.7.3 zotero0.1.10 git-graph1.30.0 vscode-docker1.26.0 black-formatter2023.4.1 flake82023.6.0 isort2023.11."}
{"_id":"doc-en-vscode-f3c1f920695a09b31dbccafcef0488480f92121681807bd0d2a63e1321f8268d","title":"","text":"python2023.12.0 vscode-pylance2023.7.30 jupyter2023.6. jupyter-keymap1.1.2 jupyter-renderers1.0.17 vscode-jupyter-cell-tags0.1.8 vscode-jupyter-slideshow0.1.5 remote-containers0.301.0 remote-ssh0.102.0 remote-ssh-edit0.86.0 remote-wsl0.80.2 vscode-remote-extensionpack0.24.0 azure-repos0.36.0 cmake-tools1.14.34 cpptools1.16.3 cpptools-extension-pack1.3.0 js-debug-nightly2023.7.1717 remote-repositories0.38.1 vscode-github-issue-notebooks0.0.129 vscode-selfhost-test-provider0.3.15 vscode-serial-monitor0.10.0 vsliveshare1.0.5873 resourcemonitor1.0.7 autodocstring0.6.1 pandocciter0.10.2 shiny-python0.1.2 shinyuieditor0.4.3 quarto1.90.0 r-debugger0.5.4 java1.20.0 vscode-xml0.26.1 r2.8.1 multi-command1.6.0 vscode-deepl1.0.6 abc-music0.4.0 lua3.6.23 latex-utilities0.4.10 chatgpt1.1.2 cmake0.0.17 errorlens3.12.0 intellicode-api-usage-examples0.2.7 vscodeintellicode1.2.30 vscode-arduino0.6.0 vscode-java-debug0.52.0 vscode-java-dependency0.23.0 vscode-java-pack0.25.12 vscode-java-test0.39.0 vscode-maven0.41.0 markdown-all-in-one3.5.1 grammarly0.22.1 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nrelated this also reproduces in stable VS code\nWe're going to remove the keybinding altogether as it should enter the buffer when toggle tab moves focus is enabled and not (as designed) without that\nChanging the keybinding could throw users off,"}
{"_id":"doc-en-vscode-e94b06a2c80cc268a195be06b55c388617fc5f1b1c93fb3debedc1684a2ce063","title":"","text":"though is wondering if we can treat this like a widget with results/ input and allow moving between the two with keybindings consistent with\nShift+tab is still going to the workbench even when I remove that keybinding and have tab moves focus mode off\nWe now allow using or to go to the terminal accessible buffer instead. I think it makes sense to remove as a keybinding next iteration if we agree that makes sense and The only downside of that would be users might already have in their muscle memory.\nIf we cannot resolve the synchronization issue with autocomplete and shift +Tab, I am in favor of removing the key-binding. there are worse things than retraining mussel memory.\nYes, and I just discussed, and we agree it can be removed this iteration.\nWe will also remove the tab index\nI cannot reproduce the original issue in stable (either 1.81 or 1.78). What are the steps to verify?\nSorry, to verify, focus a terminal with screen reader mode enabled. Use . the accessible buffer should not open.\nThis does input and no longer shows the a11y buffer, but the event also bubbles beyond and moves the focus to the panel title incorrectly: !\nGuessing we need to do something with tabindex that must've been removed in order to get this to work\nI think it would be either shift+tab needs to be specially handled in or we should handle it and cancel the event in the custom key event handler.\nnot sure how this was working before. I dug through commits and tried reverting this one, which didn't help."}
{"_id":"doc-en-vscode-d8fef08f4ba68c673aaea45feef1e134a856948d40ee3bd973035637547dca48","title":"","text":"I love VS Code, But please, don't make that option default : mirrorCursorOnMatchingTag it's a nightmare Thank's\ncan you please share more detail about what behavior makes it so bad for you that you had to change the setting?\n/needsMoreInfo\nThanks for creating this issue! We figured it's missing some basic information or in some other way doesn't follow our guidelines. Please take the time to review these and update the issue. Happy Coding!\nHey I agree with with this one here. I first came across this new feature today after updating VSCode, and after some searching, saw the setting & feature mentioned over at IMHO - This should be disabled because it feels unnatural to select an HTML tag after it's initially created and want my IDE to automatically create multiple cursors on it's closing tag by default. I'm more likely to go to an opening tag and add attributes (or like when I'm working on an Angular/React project, add directives, inputs and outputs to my components) than want to have multi-cursors to rename the component/tags (which if needed, can be done with the Ctrl+D key binding anyway). Took a few minutes of head scratching to figure this all out, and seems like I'm not alone on this one. I'd say it's one of those things that could be enabled if needed, but disabled by default. :)\nI might have a narrow use case here but I use to write Angular code. These multi-cursors on HTML tags completely ruined my ability to edit the file. I tried copying the current line with and it copied both the line where my cursor was and the line with the matching HTML tag. I didn't notice this until I tried pasting with and it pasted both the line I copied and the line with the matching tag a couple lines above that. I noticed it changed more of the file than I intended but it took me multiple operations before I realized multi-cursors were the culprit. I tried using to open a line below my current line and it two lines because there were two cursors. Then I tried using to append to the end of a line (after an HTML tag) and it appended only on the first cursor (the opening tag), which meant I couldn't append to the end of the line. Thank you for adding the setting because this was infuriating to me."}
{"_id":"doc-en-vscode-a25dda6ddfdc33656af14930d85e8673aa2650b8dfdb3b00af5b058f6e69cfbc","title":"","text":"I couldn't understand how my vim plugin was so horribly broken. I could probably complain to the vsvim project that they should disable this setting when installing the extension but I figured I might as well throw my use case into this issue first.\nInstead of mirror cursor on matching tag (which conflicts with vsvim for me), have some feature which only changes the matching tag name. I don't want to add to the closing div tag (as it just happened to me). Also when using vsvim, and you press to open a new line, it opens a new line in both locations.\nWhile I like the idea of being able to edit start/end tags at the same time, the way mirrorCursorOnMatchingTag works currently has some flaws that are really easy to trip over. I noticed it has special handling for typing a space in the start tag which is nice, but there are some other cases that can cause problems. For example, pasting - currently if I try pasting an attribute into a bare start tag, my end tag gets the attributes too, which I don't want and have been frustrated by a number of times. There's also an issue with getting double comments when toggling comments for a tag that starts and ends on the same line. A third issue is that it doesn't seem to work when making a selection with the mouse (either by double clicking the start tag or by dragging a selection in the start tag), which is my primary way to rename tags. Suffice to say, my initial impression is that this feature is buggy and doesn't completely accomplish what it was meant to do, and so I don't think it should be a default option until it can work more smoothly and not cause disruptions to my regular editing.\nI just wanted to change the closing tag of a specific element. Since this feature enabled itself after the last update unawares to me, I could not change the tag because the feature also changed the opening tag that element, and also other identical strings in the file. I had to google myself silly just to turn it off. Then, when I tried to turn it back on to test it, it didn't work at all. I don't see any productivity benefit with this feature, at least not in the current unreliable state. The use cases where I'd want to turn if off far outweigh the productivity benefit that I might glean from having it on by default."}
{"_id":"doc-en-vscode-ca5232838ad1be9b44589625e77d5c2b9ec83bb3f5bd699a304deb73d825d14f","title":"","text":"It's irresponsible to introduce and enable a new feature by default. I wasn't pleasantly surprised at all.\nAs mentioned in some comments above, I don't mind the feature but the fact that there are two cursors simply because the cursor is inside and next to the tag name is a bit weird in my opinion. I mostly work using Angular and it's a bit frustrating when code is pasted in the opening tag and gets pasted in the closing tag too. Why not make the feature that the double cursor appears when any text of the tag name is selected (i.e. highlighted)? If I explicitly want to change the tag name I would then simply double click tag name name type therefore changing the opening and closing tag. ! If for some reason I select part of the opening tag name, the closing tag name gets selected too (and vice-versa). !\nAs we were in the comments, it is wonderful that VS Code allows a plugin mechanism so that I can have extensions that make all sorts of nifty things happen when I edit, such as adding matching tags, inserting tags from abbreviations, or whatever. That's what extensions are for. But the VS Code made two big mistakes here. First they turned on a a feature automatically that should be an extension. It fundamentally changes how the editor operates. I make some changes here, and I don't expect, without some further action on my part, that it will at the same time make changes over there. I see flashing cursors everywhere while I'm trying to edit it, and things I did not touch suddenly get modified. An editor is not for surprises, especially when it comes to modifying content. It was bad enough turning this thing on by default, but the other horrendous mistake was pushing out this half-baked, ill tested feature on us. It has huge bugs, which have been reported already by many—see for example . If I have an element that has attributes, when I select and delete the attributes this new buggy feature selects and deletes content outside the closing tag, completely corrupting content; removing essential, unrelated content; breaking the document; and losing the information. VS Code is becoming my go-to editor, and it is becoming beloved by a huge community. Don't squander this support by starting forcing half-baked and ill-tested gadgets on us."}
{"_id":"doc-en-vscode-326f90f603553d01828e513f88e5f7d8ebc6897e73568daeaaaf103b4f192f14","title":"","text":"There are thousands of gadgets we can turn on with extensions, and it's awesome that such a facility is available for us. At the least, the very least, this should have been tested and re-tested for six or nine months, made available in an extension or an option, and then if there was an outpouring of requests perhaps a dialog that pops up after an update that says, \"hey, you should really turn this on\". If all goes well, and still no problems, you could turn it on for new installs. But instead you threw something together half-baked and forced it on existing installations! Are you starting to see how it this interesting idea was executed all wrong? Look, if you have all these extra coding cycles and want to implement something cool, why not go which is one of the huge remaining bugs that mar an otherwise fabulous editor. Congratulations on the editor in general, and good luck learning from this lesson and making it even better.\nSince you can use Rename Symbol (F2 by default) on HTML tags, I think having automatic double cursors is unnecessary. In my experience, renaming HTML tags does not happen often enough to warrant the automatic double cursors - since I am already used to renaming symbols with F2, it's more intuitive to do it that way.\nyou the needsMoreInfo tag to this issue. Is there sufficient info now or is there something else you're still waiting on? ? Anything else you need?\nWe can consider disabling it by default, but that means this feature won't be discoverable at all. Some issues are hard to fix, others are impossible. We are working on an improvement. You can find more details in . This feature does what it says: Creating another cursor when you are on an open/close tag. Some actions, when two cursors are present, perform differently than one cursor. Like, if you press Enter when two cursors are there, it produces two newlines. We can't bend multi-cursor to work differently, just so when you press Enter it only creates one newline. I'm not sure how you can use VSVim in VS Code (isn't it only for Visual Studio)? During my testing it had some issues with VSCodeVim, because of the way VSCodeVim is implemented. But most of the time it works fine."}
{"_id":"doc-en-vscode-84cabc441e2dc0cee0e8da2c0b2e315f69d03f06a80f567fd04b7bdb3e4c959e","title":"","text":"In short: we'll use this iteration to find workaround for some of the issues. If there's no easy workaround, we might take a more general approach in , and meanwhile disable this feature by default. If you have issues with concrete reproducible cases, please open new issues.\nSo you're saying then that this will be Microsoft's policy moving forward, that is, to make new and unpopular features enabled by default in order to make them 'discoverable', so that users then have to sacrifice their time and productivity with vetting processes like this? It would seem that it would be in the user's best interest to let users discover new features in the release notes and let those who wish to test the feature do so on their own time, rather than forcing new features on the general userbase who actually depend on reliable editors to meet deadlines and earn their daily bread. Thanks for informing us of this policy now so that users can let it inform their choice of editor in the future.\nI can't dictate Microsoft policy. Please don't make extreme straw man statement to criticize it. We'll consider disabling it by default. That's why the issue is open and marked with .\nYep, sorry, I meant , totally messed up there. :manfacepalming: I think the issue here is that by enabling the feature by default, I had no idea what the feature \"said\" or that it even existed. All I knew was that multiple cursors kept popping up when my cursor was on an HTML tag. Yes, precisely. That's the problem. Some actions performed differently and I didn't know why. Nobody is asking for the multi-cursor feature to behave differently, we're saying that only the users who want_ multi-cursors should deal with those nuances. This GitHub issue is full of comments from people who were confused/frustrated by the change in default behavior, which is arguing for it to be disabled by default. I'm curious, what is the argument for enabling the feature by default? Is that discussion documented somewhere? I'd like to see why the feature was considered so useful as to warrant being enabled by default because as far as I can tell, the only benefit is in the case where you need to rename an HTML tag. And I agree with comment: Is there another use case that I'm missing?"}
{"_id":"doc-en-vscode-4be3d87f29d3ffcb88f5832e8fe734a4241cdcb0f14ae5789abc9158a8c72216","title":"","text":"With all due respect, it should be obvious that it's the decision to enable it by default that most people in this thread are objecting to. That's either a one-off decision that was made for this particular feature or it's a policy decision that will be made for other new features moving forward. I apologize if my comment came off wrong - Code is a great editor. What I'm saying is that I hope this is just a one-off, because if it were to become the default method of launching new features, it might affect users' willingness to use Code or to upgrade for fear of encountering unwanted new features that result in loss of productivity.\nas noted, I run into this multiple times a day working in Angular. I had thought it was an extension gone rouge, as it seemed like a feature that an extension would add. So I spent my morning disabling them one by one, until I opted to search and see if this was a new vscode \"feature\" . For what it's worth, I also disagree with the notion that new features should be enabled by default to encourage \"discovery\". That's what the update and changelogs are for. If I was a new VSCode user coming in post-update and I ran into this not knowing it was a \"feature\", it would turn me off from the editor completely. It's not what I would consider typical behavior for what an editor should do, but that's just my two cents.\nI specifically looked to find this issue/topic as I've been hit by it multiple times during a day of coding now and it did introduce bugs in my code. The worst case is when I reorganize HTML code and sometimes have to delete or edit the starting tag. Then the closing tag can get deleted too, but nothing is of course re- if I start typing in the new/changed starting tag because it has now no ending tag counterpart. Well, maybe it's immediately after my new starting tag, in the completely wrong place. This can introduce hard to immediately discover bugs in my HTML because of sometimes deep nesting in long blocks of code and removed or messed up closing tags (I've particularly discovered things like suddenly missing end brackets that I've only noticed early on thanks to syntax highlighting?)."}
{"_id":"doc-en-vscode-5c868a1e492a30163aa8edd091467ad273272aab755cb216ba5b7f40327b60c6","title":"","text":"It's especially frustrating because sometimes things happen and other times it doesn't seem to, and I have a hard time mentally getting around when and why it's doing things, so I end up having to baby sit the editor. I think that for me, it would be easier to have more predictable behavior back again. I'm very used to renaming both tags, no problem there. But sure, at least now I know what the setting is. I'm just offering this as feedback on \"how does this impact me negatively\".\nThis feature is a poor substitute for real code refactoring, and should never have been enabled by default (IMO it should never have been to the main application at all). If you were worried that users would not discover the feature, you could have a hint about it in the changelog. Or better yet; you could have made it as an extension, and advertised it in the extensions panel. Multiple cursors are really handy in certain scenarios, but it is vital that the user is fully aware and in control of where they are. One really essential problem with this feature is detecting when the user is actually modifying the tag name, and not just adjacent code. Another is that there are other ways of modifying the tag name than just adding or removing characters. I do not believe that these issues can be solved without proper code awareness, and if the editor has that then this feature is redundant anyway. Given this code example... ...I have three examples of how this feature breaks the code while I am editing: want to remove the extra whitespace preceding the class-attribute, so I place the cursor at the end of the tag name and hit <kbddel</kbd. Code is now broken because the end-tag has no closing angle bracket (). inadvertently typed in the attribute \"component\" without a space after the tag name \"custom-angular\", and I fix it quickly by placing the cursor between the two and hit <kbdspace</kbd. Code is now broken because the end-tag does not match the start-tag. want to remove the prefix \"custom-\" and I do so by placing the cursor at the end of the prefix and hitting <kbd<- Backspace</kbdrepeatedly."}
{"_id":"doc-en-vscode-c60ad86652faa3d29111bb1ea64dc4a77cd08f65de28952be2f345f973581397","title":"","text":"Now if I go back too far, accidentally removing the opening angle bracket (<), and then re-type that character to fix my error, then the code is broken because the end-tag has been converted to a new start-tag. Now all of these are simple examples, and if you are aware of the feature and/or see the errors as they occur, it is super easy to fix the code again. Either manually or by using the undo-feature. But if the code is longer and more complex and the end-tag is out-of-view when the errors occur, it may be quite difficult both to discover the errors and to fix them.\nHere is a use case where the default \"on\" of this new \"feature\" is incorrect. tl;dr - A proficient vim user can make a horrible mess very quickly with mirror cursor. I rarely use VSCode but at work I need to write up some documentation from time to time and I prefer to do so in html. So I fired up VSCode for the first time in a while and started typing up a doc. I use vim bindings and don't need to look at my editor too much especially when working in the way I was today. So I wasn't paying attention to my editor. About halfway through I opened my work in a browser to review how my old CSS settings were treating the work. It was a garbled mess of half paragraphs and repetition. Using '/', '?' and other regular vim navigators causes multi-cursor insertions EVERYWHERE. Change the default to \"off\" and test these features against different editing options like Vim and Emacs in the future.\nThis kind of falls into the same category as any setting that can help users reduce keystrokes (similar to ). When you've built up some muscle memory or expectation for how things behave, and that suddenly changes, it can take time to get used to. Personally I have many of these features disabled because of being used to unconsciously doing many of them manually, and the time investment in changing that and fixing mistakes that occur in-between hasn't seemed worth it. It can be especially frustrating when you are in the middle of something and a feature like this is introduced. Even more so if finding the name of the feature in order to disable it winds up taking more than a trivial amount of time."}
{"_id":"doc-en-vscode-b4b9e30fc6ca1b2e410c6b1db14df202fc5e98ad36120a00375a599ed0b97d01","title":"","text":"The worst part is for those who are uninformed, which is everybody who didn't see the patch notes, the name of this feature would remain elusive until you either find the correct patch notes on stumble upon this thread. How can one disable a feature that has no name? Personally I thought it was a bug with multi-select / multi-cursor that got enabled on accident. Still do actually.\nI'm also having issues when moving lines with alt-arrow . I want to move the starting tag higher to include block of code above, but ending tag also gets moved.\nI also got hit with this today. I mangaged to edit the opening element but it did not auto select the closing element. When I then tried to edit the closing element it auto-selected the opening element parent to the first one. See picture for example. I also ran into the same issue as and by then my blood was boiling and I literaly screamed from frustration. Features like this should not be activated by default when it clearly messes with peoples personal coding behaviour/styles. !\nHad four lines of elements. Selected the opening tags for all four as distinct selections, and then edited to add to each. Didn't realize it was corrupting content in the close tags. Uck. Too dangerous to use, let alone enable by default.\nA use case that this feature completely corrupts is reordering HTML tags using the + key binding. It took me ages to realise that when I was refactoring some html, reordering things, why my html was all of a sudden being corrupted because the corresponding tag was being moved as well, and more often than not, incorrectly. This should definitely not be on by default. Thanks for reading.\nCreated a draft PR, seemed like a pretty trivial change, just need to hear if I have to do any other tests/etc for this\nAll the complains above explains well, and I just want to add my current brand new face to face experience, which I encountered numerous times and today I finally, \"enough is enough\"... ! Anyway, thanks for the hard works, to those who work with this feature... but please make it off as default... Thanks...........\nI'll turn this off by default and work on , which will be an opt-in feature.\nGreat, it is nice to hear that it will be turned off by default"}
{"_id":"doc-en-vscode-f5188758fd86b319997cfccdc2e73e4bc44b316249fc9527b5a772731215afc6","title":"","text":"Testing Steps: Open locally cloned vscode repo with any branch Welcome page Connect to... Continue Working in Nothing happens after this for me... but I do see the commands get run:"}
{"_id":"doc-en-vscode-d35ef6d51a884f792314365b8ebe7aea12b7902a2189908258df6d3ba5b10d1a","title":"","text":"Issue Type: <bBug</bGo to menu affichage apparence afficher la barre latérale secondaire L'item indique ceci \" afficher la bar&re latérale secondaire\" Un petit And & se balade Have a nice day VS Code version: Code 1.67.1 (, 2022-05-06T12:37:03.389Z) OS version: WindowsNT x64 10.0. Restricted Mode: No <details<summarySystem Info</summary </details<details<summaryExtensions (3)</summaryExtensionVersion ------ copilot1.20.5902 vscode-pull-request-github0.42.0 vscode-language-pack-fr1.67.3 </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nthe translation lacks an extra ampersand to turn this into a mnemonic <img width=\"501\" alt=\"image\" src=\"https://user-\"\ncan you ensure there's a comment that says \"don't touch the ampersand\"? Additionally, you can add another comment in the form of: to ensure that they do not get \"translated\". See this example:\nI have the same comment in 275 places across the code base And this is the string\nBest I can do is ask for the string to be updated manually to ensure is not translated... but that won't prevent this from happening in the future due to human error. better is using a locked comment like the one above... could also have the be a runtime string instead... but really we need language-aware mnemonics where the position of the changes based on the language but I'm not sure if there's a process for that. do you know?\nso I add? what kind of enforcement does this do? does it just force that a mnemonic exists? its not really required that a translation contains a mnemonic even if the original does\nI believe it forces that it exists, though or are experts here.\nI know a little bit more... this does basically \"fail the string\" until it meets the criteria of \"there should be a somewhere in the string and a translator looks at this. did you get the chance to add this to the comments?"}
{"_id":"doc-en-vscode-74ab1f6c30761cc29032ba5e7f59ecc673dc1ffe8c2a4d2b2fb6385fbbcfd873","title":"","text":"a new file with the following content and commit it. the file to like as follows, save it, and create a stash with it the file to like as follows, and save it the \"Git: View Stash...\" command and pick the created stash diff in the stash should look as follows: <img width=\"921\" alt=\"image\" src=\"\""}
{"_id":"doc-en-vscode-d3af1b62f85eaca9345ae3c4bf6853c5bd5ecc60cf640da9d63a16a6f80c31a2","title":"","text":"Reported originally by where GitLens seemed to be breaking JSON auto-complete behavior -- causing an extra being at the end. ! I tracked it down to this contribution causing the issue I was able to reproduce this with a clean vscode user-dir/extensions-dir and a simple extension with that contribution\nThat JSON extension has configurationDefault for the setting: That configurationDefault is lost when the gitlens extension also makes a configurationDefault The bug is that all default overrides for the same language need to be merged, not set."}
{"_id":"doc-en-vscode-81f6864751bde98f530cef250ec7f59f92e08eef4e5ca9edafa73ba495e646ed","title":"","text":"We don't want to officially have the hex editor be recommended for non-textual files, but what about a setting like that can be used to open a binary file in a given editor instead of showing the \"binary file\" message? ! I would love it if I could have all non-text files automatically opened in the hex editor.\nTo verify ensure the setting works as expected. Setting it to a custom editor should automatically have that editor be opened when a file is detected as binary."}
{"_id":"doc-en-vscode-ec26a30ada59fc03a00eacb12a6b5b8b06d337800dfed7f5eabb42d5a608ab1e","title":"","text":"I do not see a duplicate of my issue: ! So I'm in the clear. <!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.32.0 OS Version: Windows 10 1809 Stable Steps to Reproduce: Visual Studio Code, click on the \"Search\" sidebar menu item to bring up the huge Search menu. ! for a common word used quite a lot. Like \"a\", \"let\", \"var\", \"i\", and so forth. the Search displays the total results count, dismiss a few of them. total results count should decrement by 1 per dismissal, or indicate both the original total results count and the remaining non-dismissed total results count side by side when dismissing the search results. For example: ! Does this issue occur when all extensions are disabled?: Yes\nFor me, it does decrement. You don't see that? You are clicking the 'x' icon on a result row, right?\nIt doesn't. !\nOk, so I was about to get it to decrement, but that's not done in real-time. ! Both of the GIFs are done in 1.32.1.\nThat's right, I thought you were talking about the total shown at the top (\"x results in y files\") in your original post.\nThis also happens when a new match is or deleted in an existing file match. !"}
{"_id":"doc-en-vscode-83190ad175224f2c309c22acf01fd0e4ad26492bc8b675e24d5fb1bce44e022e","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --When using Copilot, users would like to be able to only partially accept suggestion instead of the whole suggestion at once. The users demand either word by word or line by line completion. Desired behaviour: Press a predefined key and accept the whole suggestion. Use a predefined combination of 2 keys to accept the top line of the suggestion. (eg: For a suggestion of 3 lines, the user would need to press the combination 3 times to accept the whole suggestion). The prompt would not show the line already accepted.\nthis sounds to me like a UX problem. Is there a mock sketch for a potential solution here? Is there some other editor which already supports this? We can also discuss it in one of our UX weekly meetings.\nI don’t have a mock sketch. No other editors are supporting this yet but it is a requested features. I wanted to get the discussion going (doing the same for the other editors) to understand the constraints around it as well.\nTechnically, I think this would be rather easy to implement. I'd love to hear some more UX discussion before we implement this thought. What key-binding do you have in mind?\ndo you have some UX designers on your team that would like to help here? Once we have some good ux proposal we can bring it in our UX weekly sync on Wednesday to start the discussion.\nNo I dont have a UX designer in my team. I ll see how i can come up with a suggestion and get back to you\nThis is similar to Solution for me was 'CMD + -' to accept a word at a time. Something similar here would be great.\nFish shell allows word-by-word completion using the <kbdAlt</kbd+ <kbd&;</kbdkey-combo, which—at least on macOS—is used system-wide for navigating text word-by-word. I notice that I keep trying to automatically use this key-combo, so big +1 from me. ( Did not realise this is the vscode repo.) \nawesome idea. since you said this is low hanging maybe we assign to July for the initial investigation? What do you think?\nThis would be great."}
{"_id":"doc-en-vscode-4a4e4526621b2203bcc7f97f241432bb1cc928c82666e8f85373905125a3bcc5","title":"","text":"I am loving copilot so far, but this is my biggest UX gripe. Copilot's first suggestions in a new piece of code tend to only be accurate for the first few words. It tends to get much better only after more context is . Not a blocker, but it'd make it much easier to get started.\nWe did an initial investigation and this is not low hanging as I have initially commented above. Also, this milestone we have been mostly focusing on improving the merge editor. So this one will have to continue in September.\nHad a sync with the VS folk and here's some very early thinking: When partial completion is accepted the ghost text should not change - we should use the cache. VS did multiple user studies and users always get shocked when the ghost text changes beneath them Undo is the user gesture to undo the last partial acceptance (this should work out of the box for us). Simple keybindings like word next navigation, and navigate to end of line should be used for partial word acceptance, and whole line acceptance. So same keybindnigs as and commands. Open questions (probably not for first version): How to make this keybindings discoverable Should we support the mouse? For example alt + mouse click to accept until the mouse click\nI've also previously advocated for a way to accept only a single line out of the multi-line completion. I would strongly support doing this at a line granularity, rather than e.g. token granularity, since I ~never run into the case where I want to accept 'more than 0 tokens but less than a line' of a multi-line completion -- it's ~always either one line or maybe a couple, or the whole completion. Repurposing regular cursor movement commands for partial acceptance when a ghost text completion is open sounds error-prone. I see this kind of key-stealing behavior in some autocomplete systems (especially for ), but it's always optional, since it's really easy to interfere with muscle memory and frustrate people. This would apply to text navigation commands for sure, and it would be even worse in the middle-of-the-line completion case (which would also be wonky if we want an 'accept a single token' shortcut).\nHi! A use case for per-word completion: comment/natural language completion. I want this almost every time that copilot gives me a comment suggestion."}
{"_id":"doc-en-vscode-5cc35fe616f35799445419c7f4663f47fc2144742bad076d087009a964cf36fd","title":"","text":"To make sure I understand the shape of the discussion, I think three issues are separable here: 1) whether to offer partial completion; 2) at what granularities (here the use case and example-based analysis is most relevant, and feed back into question 1); 3) if doing it, what default keybindings at these levels to provide so as not to hijack muscle memory (but possibly still utilize it in a parallel modifier key universe)? Is that right, and if so at what stage does the team currently see the conversation? For 3), in Lisp-y language tools, Paredit's keybindings might be a decent template too: in some sense what we are doing here is structurally selecting part of what is an (falling back on a as above) structure before \"acceptance\" (which is already well-understood as a result of by default keybinding users). That would be a more elaborate form of the partial acceptance, though, requiring a language-sensitive parse, so definitely not low hanging fruit.\nIt's a marketing problem, they market the code you accept as efficiency. If they would also add the 97% of the code you have to remove carefully again it wouldn't be that a nice metric. I believe we need to adapt the plugins on our own or write a plugin watchdog, something that monitors completions, immediately removes them and offers a word-by-word insert of the code.\nA use case from He wants to accept the first couple of words here (but not the entire completion): ! ... ~~ A use case for line based acceptance: !\nFor jetbrains IDE's you can use my plugin () You press CTRL+ALT+D and it will complete the sentence word by word. So in your example you press it 3 times (you'll be after \"to\", then you type \"have a per\" and you can continue with the shortcut 3 more times on the new completion text. Or just \"TAB\" if all is good after your intervention, it doesn't block you from using the default behavior of copilot Though the plugin only works on IntelliJ IDE's (you need to install 'LivePlugin' and 'Stewardess')\nI love this suggestion. I encounter the scenario of wanting to only accept a suggestion partially a lot.\nwould you prefer word-by-word partial acceptance or line-by-line?"}
{"_id":"doc-en-vscode-08b37a100592e0dcae1c4e39d5646bf6068749b3dbad2a29f8c304442e5071ba","title":"","text":"It would be awesome to be able to do both, but if only one of the two is possible, then I would prefer word-by-word.\nThere is no reason why only one would be doable. A line based completion is a couple of lines of code. Word based completion is half a page if you want to do the word separation smart, or an abstract if it's dumb.\nFirst iteration has been implemented in"}
{"_id":"doc-en-vscode-83acfc0a346267028e4a2eeb23c7e62361cbfe1cc7f2ff1d2bf2e54860cab699","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --This is essentially a more general issue of the closed Markdown styling support are missing in the following, missing bold and italic: Solarized Light Red Abyss Solarized Dark Tomrrow Night Blue See screenshot below for an example. For High Contrast, section heading () are not in bold. I am unclear if it's a bug or intentional. Screenshot for Solarized Dark !\nHi, I tried bold and italic with these themes: theme-tomorrow-night-blue theme-solarized-dark theme-solarized-light theme-red theme-kimbie-dark theme-abyss hcblackdefaults(High Contrast)"}
{"_id":"doc-en-vscode-81930a35fb4935dad6e532d5c5345800b62eddb81a67f455813f2106ae7124a4","title":"","text":"Issue Type: <bBug</bRelated to and Steps to reproduce: and or corresponding JS settings. Create new TS or JS file and paste next code: and ! TS Version 4.4.0-dev. VS Code version: Code - Insiders 1.59.0-insider (, 2021-07-28T05:13:47.114Z) OS version: WindowsNT x64 10.0. Restricted Mode: No <details<summarySystem Info</summary </detailsExtensions: none<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\n/assign\n/ Can you please take a look? I believe that folding generates two new requests for inlay hints that VS Code then displays incorrectly. Here's the TS server trace from folding with only : The position of the hints seem correct to me. I don't know why VS Code requests the hints twice here though (or why it needs to re-request any hints on folding in this case since no new code became visible)\nThese are the visible ranges: ! And vscode sends two requests, one for each range. For some reason, each reuqest returns one inlay hint, so in total, there are 2 inlay hints: ! do you have an idea why this is? To me, it looks like a bug in the provider, there should not be a result for the second request.\nEmmmmmm... I think I know the reason. In the tsserver. For each request, we will get the hint in the area that overlay with the visible range (not strictly included). Honestly, I don't know about this behavior. We may could de-duplicate these items?\nCan't you just filter out all inlay hints whose position is not in the requested range?\nWell, do you prefer to do it in vscode or ts side?\nHmm, maybe actually in vscode.\nVerified fixed with TypeScript Nightly in"}
{"_id":"doc-en-vscode-5435587d620b2bf8b020b0c06cba6465a1f600a12d4a8ed9db4676fd055343e1","title":"","text":"Refs: Steps: enter high contrast theme hover over menu items they jump up and down while doing that"}
{"_id":"doc-en-vscode-57ad71b3d37ec6bbd7ccbd3338d59df1267a1c1967fe402ccedc149dfcd5559b","title":"","text":"Steps to Reproduce: the Activity Bar Position as Hidden Command: View: Focus Activity Bar Then: ! Activity Bar still takes up screen space, but doesn't display the icons Version: 1.84.0 Commit: Date: 2023-11-01T11:29:04.398Z Browser: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Code/1.84.0 Chrome/114.0.5735.289 Electron/25.9.2 Safari/537.36\nProbably depends on\nProbably\nNot related to\n!"}
{"_id":"doc-en-vscode-927c8fa9c3616c37c5aed9b2fe2e4f6af6cef1c91a02a7159bfe8760e9a02718","title":"","text":"Verification set to enable native title bar in settings open a floating window ensure you see no custom title below the native title"}
{"_id":"doc-en-vscode-99b358f63b6a1fde67dfb7412c0ffd4aad13ea997c9e4ce3334a98df04ce29b0","title":"","text":"Type: <bBug</bHi, My name is Ebrima Mbye. I have been using this wonderful IDE for some while now. However, recently I have noticed that when I run a program, the output in the terminal window does also include the memory location of the file or folder I am working on. Sample output: \"93c9--e1dddca87b23PS C:UsersUserDesktopNewEbrimaProgram\" This is not a super cotly problem, but I just wanted to bring your attention to it. Thank you for your valuable time, User, Ebrima Mbye, Java programmer. VS Code version: Code 1.88.0 (, 2024-04-03T13:26:18.741Z) OS version: WindowsNT x64 10.0."}
{"_id":"doc-en-vscode-0dc06b30d1cd7cc1a03835b03a3540f545dc4226a510c697025b146d099e2025","title":"","text":"Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (44)</summaryExtensionVersion ------ codesnap1.3.4 vscode-javascript-snippet-pack0.1.6 vscode-django1.15.0 githistory0.6.20 python-environment-manager1.2.4 python-extension-pack1.7.0 code-runner0.12.2 vsc-python-indent1.18.0 kodezi2.2.2 debugpy2024.4.0 python2024.4.0 vscode-pylance2024.4.1 jupyter2024.3.0 jupyter-keymap1.1.2 jupyter-renderers1.0.17 vscode-jupyter-cell-tags0.1.9 vscode-jupyter-slideshow0.1.6 cmake-tools1.17.17 cpptools1.19.9 cpptools-extension-pack1.3.0 vsliveshare1.0.5918 sqltools0.28.1 sqltools-driver-mysql0.6.0 autodocstring0.6.1 vscode-python-typehint1.5.1 fabric8-analytics0.9.4 java1.29.0 LiveServer5.7.9 pythonpack0.3.0 javafx-support0.0.1 tabnine-vscode3.86.0 open-in-browser2.0.0 bootstrap4-vscode6.1.0 cmake0.0.17 intellicode-api-usage-examples0.2.8 vscodeintellicode1.3.1 vscode-java-debug0.57.0 vscode-java-dependency0.23.6 vscode-java-pack0.26.0 vscode-java-test0.41.0 vscode-maven0.44.0 jinja0.0.8 JavaScriptSnippets1.8.0 grammarly0.24.0 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.88.1. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!\nHow may I upgrde?\nI meet this problem too."}
{"_id":"doc-en-vscode-7e2efaaad99c00b1da62d357afb2499865c53ffef146d7fdcc032c0a4f2473a6","title":"","text":"Today when I use python debugger to debug python files, something wrong happen in vscode 'Python Debug Console'. Output reads: ncher' '7541' '--' 'D:' ;-402f-491b-bdb5-ee7fdd8b6e0cncher' '7541' '--' 'D:' ;-402f-491b-bdb5-ee7fdd8b6e0cPS D:I have neven seen strange characters before. This string wants to overwrite what I want to print in python file and still appear with empty python file. ! My VSCode version is 1.88.1. Hope Greatest Microsoft Engineers will solve this problem. Thank you very much~ PS: To exclude influence of other entensions,I only use this four extensions. !\nI've been having a very similar issue for the last week or so. I'm running Windows Powershell (5.1) in an integrated terminal in VS Code on Windows 10. If I run a command that wraps over to the second line of the terminal, an unfamiliar GUID along with some other characters (see red boxes in screenshot below) appears interspliced with the text of the command I just ran. Makes a little difficult to use the terminal. PS I don't encounter the same issue when running the same command windows powershell 5.1 directly (ie not as an integrated terminal in VS Code) ! VS Code version info: Version: 1.88.1 (user setup) Commit: Date: 2024-04-10T17:41:02.734Z Electron: 28.2.8 ElectronBuildId: Chromium: 120.0.6099.291 18.18.2 V8: 12.0.267.19-electron.0 OS: Windows_NT x64 10.0. Powershell version info: ! Windows verison info: ! Installed extensions (both disabled -- enabling or disabling has no effect on the issue) !\nThanks for the issue, what happens if you comment out your file?\nsame thing happens with user commented out. !\nI believe that's just your current working directory, which reflects the folder you're in. This is working as designed. You can change the behavior by tweaking your setting."}
{"_id":"doc-en-vscode-228800659aeabe62fded7c5c815968b78c19e32a6cf0619796389585780d0960","title":"","text":"Unfortunately, it seems unchanged when set the value of as ${fileDirname}. And when clicking Run, it looks like to execute a to stop something runnning. Truthly, I would accept this strange long string in my sonsole but it wants to overwrite what has been output. So, if want to print some message in console without owerwritten, I need to add this code above the whole codes.\nWhat if you set it to something else? Like workspace folder?\nI've tried setting [] to other value with different directory and drive letters. But it only changes the initial working directory in Teriminal <pwsh, not in Integrated Terminal <Python Debug Console. Also I manually changed <Python Debug Consoleworking dir, it looks like the same problem. So it has no positive influence on this problem.\nWho closed this? Why close it when the problem is not yet solved? On Tue, Apr 16, 2024, 1:53 AM Megan Rogge\ndoes this reproduce if you disable all installed extensions?\nupdating my \"\" value in the user settings json did not resolve the issue. As I mentioned before, all installed extensions are already disabled and the issue still occurs.\nI think this is the shell integration set command line nonce being passed echoed instead of being passed in with the sequence. I see 2 versions of Windows mentioned and both are Windows 10 so this makes sense here. I'm thinking we'll just not verify the command line via the nonce for conpty under Windows 10.\nAre there verification steps for this?\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nRunning any command in Windows PowerShell (v5) is the verification steps\nI'm unable to repro the issue in Stable on Windows 11.\nI can't even reproduce the original issue on my Windows 10 VM at home with VS Code 1.88.1. has the issue been fixed for you in the latest VS Code Insiders? You can get a copy at\nThanks for your reply."}
{"_id":"doc-en-vscode-0e20dc5ba47f6ee28ab01080eff34394ab82fc2b4ece2cda7a1e872fe0cf3a8e","title":"","text":"I am getting the same issue with VS Code Insiders (see atached video file). Version: 1.89.0-insider (user setup) Commit: Date: 2024-04-25T21:21:03.843Z Electron: 28.2.8 ElectronBuildId: Chromium: 120.0.6099.291 18.18.2 V8: 12.0.267.19-electron.0 OS: Windows_NT x64 10.0.\nthis was a regression in 1.89, so it was fine in 1.88\ncan you try again but make sure you recreate the terminal?\nThanks for your continued support on this. I can still reproduce on the latest version of insiders. Version: 1.89.0-insider (user setup) Commit: Date: 2024-04-26T00:56:34.034Z Electron: 28.2.8 ElectronBuildId: Chromium: 120.0.6099.291 18.18.2 V8: 12.0.267.19-electron.0 OS: Windows_NT x64 10.0.\nSincerely sorry for replying so late on this issue. I donwloaded and installed the latest VS Code Insiders. Unfortunately, this problem still happen. My PC Info is listed below. Commit: Date: 2024-04-26T00:56:34.034Z (1 day ago) Electron: 28.2.8 ElectronBuildld: Chromium: 120.0.6099.291 18.18.2 W8: 12.0.267.19-electron.0 OS: Windows NT x64 10.0. I found something interesting that while you changed the TERMINAL widget size just like maximized the VS Code window or dragged the left side of terminal list, the strange string will disappear. It seems to mean when you resized a terminal window to control how many characters in a line to show on screen, the string which was already output will be changed or even erased. I captured two new videos of this probloem. The first is the origin problem reproduced in the latest Insiders and the second is changing the window can erase this strange string. Hope these useful for solving this problem.\nReopening issue based on feedback above."}
{"_id":"doc-en-vscode-976b29ce5a63f30fc1a67dbbea3eb7ad57061723a68e4f9f187572d88f06061c","title":"","text":"i found issue the didnt occur on 1.87 but then could reproduce after upgrading to 1.89. Version: 1.87.1 (user setup) Commit: Date: 2024-03-06T00:21:36.258Z Electron: 27.3.2 ElectronBuildId: Chromium: 118.0.5993.159 18.17.1 V8: 11.8.172.18-electron.0 OS: WindowsNT x64 10.0. Version: 1.89.0 (user setup) Commit: Date: 2024-05-01T02:09:22.859Z Electron: 28.2.8 ElectronBuildId: Chromium: 120.0.6099.291 18.18.2 V8: 12.0.267.19-electron.0 OS: WindowsNT x64 10.0.\nit isn't only when paths are long. in insider i'm seeing this with ohmyposh/poshgit in powershell terminal windows, even if i'm at the root of a drive where the path is only a few characters long (though probably related to whatever posh is doing?) ! i'm on win11, insiders 1.90.0\nanyway in 1.89, as a workaround i replaced ~AppDataLocalProgramsMicrosoft VS CoderesourcesappoutvsworkbenchcontribterminalbrowsermediashellIntegration.ps1 with the shellIntegration.ps1 from commit and it resolves the issue.\nLooks like the only remaining issue with Windows PowerShell (), the fix for that was only just merged yesterday:\n- did you have some reliable steps to repro this?\nin Windows PowerShell () run a lot command like a few times and ensure a GUID does not show up in the output."}
{"_id":"doc-en-vscode-46bd5baad834beb477f219a49c8dd3b3512817ce20dec9c3ea997264c1fb6f3f","title":"","text":"And we could use in browser.\nSorry, just confirm, Would this will be released in July (the next) version?"}
{"_id":"doc-en-vscode-851ab01659032da43a35c1644cda58bd6298e2ce59479c8dc5c8bdf1ad4678f6","title":"","text":"Currently we enable the persistent terminal/reconnect feature if a terminal is launched by an extension, unlike the env var relaunch where we disable outright because it can cause issues with the extension as it loses access to the terminal upon relaunch. Relaunching the PowerShell extension right now for example is a bit of a weird experience though because it creates the \"PowerShell Integrated Console\", and that reconnects upon reload but is no longer managed by the extension. Before reload: ! After reload it is restored, but no longer owned by the extension: ! Assigning to March for exploration. Thoughts\nI think just allowing opt-out/opt-in on this behavior would be more than enough for PowerShell's case.\nHas there been any progress on this? Now that persistent terminals are the default, this gets quite annoying and bothers users when extension-owned terminals come back but aren't actually connected.\nNo progress, we'll try get it in the next version\nproposal:\nIs this still ongoing and targeted for February?\nYes it is"}
{"_id":"doc-en-vscode-7c83c36523a0597d3d39ddbada578e546e9260f8402984a2a9c40aef2b9b563e","title":"","text":"Issue Type: <bBug</bmicrosoft/vscode-extension-samples the helloworld-sample folder , build, verift that its contributed \"Hello World\" command activates the extension and shows a notification. the vscode version references in in two places to use 1.43 minimum. code to activate function in , for example: , run, use \"Hello World\" command to activate. breakpoint on return line above. a folder. Explorer button or context menu to create a new file. Confirm that the event fires and your breakpoint is hit. from breakpoint. Explorer context menu to copy/paste or use drag/drop with modifier to drip as a copy. :bug: The event callback does not run, despite the API doc stating: Issue also exists in 1.64.2 VS Code version: Code - Insiders 1.65.0-insider (, 2022-02-21T05:15:17.930Z) OS version: Windows_NT x64 10.0. Restricted Mode: No <!-- generated by issue reporter --\ncc\nthe extension host's event is only emitting from 's event, but the explorer is using the 's , which doesn't seem to talk to the working copy service. What do you think the fix here should be? (cc\nthe bulk edit service is using the for its operations. If you look at , there is a bunch of usages for that service. The working copy file service is a wrapper around the file service that ensures the relevant events are emitted and can be participated on. I suspect that for the copy/paste operation or drag/drop we are possibly not using the bulk edit service but the file service directly? I suggest to make a pass over all explorer code and look for raw usages of to replace with bulk edit service. I think that would even enable undo/redo properly.\nMy bad, the real bug is that this block: doesn't handle events. cc\nPragmatic fix would probably to treat the same as ."}
{"_id":"doc-en-vscode-3c13be5aa4b18bc70596ab4d5d9f31a23fc3cfa638c03001a6187e836dae20d4","title":"","text":"a notebook that has several cells, and put this code in the second to last cell the cell so that it has output, and shrink the window so that the single cell+output takes up the whole scrollable region the cell again with shift+enter :bug: the notebook ends up scrolling up, instead of showing the full output and next cell. !\nThe current behavior of the list view is to use the second visible element as an anchor which can cause some strange behavior. In this scenario: next cell is focused and revealed, making that cell the anchor as second visible, since the executing cell takes up the rest of the visible region. output is removed, moving more elements into view from above is and the initial anchor is no longer the second element, so it is scrolled out of view I'll try to play around with some options, but my initial thought would be to use the executing cell as the anchor. The output could expand and push the next cell out of view, but typing in that cell will bring it back into view. cc\nI wanted to see the behavior locally. I really can't seem to repro this, my notebook stays pretty stable as the cell shrinks and grows. What am I missing?\nI think I found it. I believe you have disabled, am I correct? Can you reproduce this when flipping that setting back to the default value of ?\nIt repro's with or without - you just need to scroll down enough so that the resized cell is only partially visible. The strange behavior is due to the notebook cell list setting different anchors depending on whether the element top is above the render top.\nAs I looked into the remaining open issue and PR , I was trying to grasp what the issue really is, what has been actually fixed and what's left to fix. Then, I pulled latest and ran out of sources and this is even worse than my last recording. Everything jumps around even when the cell is fully in view: Did just break this even more? I suggest reverting , pausing, simplifying the problem, narrowing in on a solution and consider fixing it in debt week so as to not cause regressions during this endgame week. cc\nthanks for the testing, the two recordings you posted above were using different execution commands so I want to ensure we refer to the same behavior: in , it's re-executing the active cell (ctrl+enter), which should always stay visually stable."}
{"_id":"doc-en-vscode-015eecda31e56bfbfa641ea32296a9b124227e64b4b5a5145f8896145bd311cc","title":"","text":"I think it still works this way out from source in the latest , from what I saw, it's running shift+enter: run cell and move to next This command moves focus to the next cell, so the cell is now the focused cell Then the output of the previous cell (which is no longer focused) starts to grow The change introduced via 's last PR was to ensure that the focused cell stay visually stable, once they are revealed into view can you confirm if this is what you are seeing?\nI captured a recording for what probably was comparing, right side is Stable and left side Insiders In both setup, the next cell is fully visible In Stable, shift+enter moves the next cell up, and then push it down, but the new focused cell is still visible in the viewport at the end In Insiders, shift+enter moves the next cell up, and when the previous starts to grow, it does not move the focused cell downwards Even though the new behavior is more self explanatory (from our perspective), but unluckily is not the same as Jupyter Lab. It means we are changing from one custom behavior (our previous heuristics) to another (always keep focused cell stable)\nThanks for clarifying this It was not clear to me that (1) shift+enter and clicking the run cell action did different things, nor that (2) this bug was only about one of these actions.\nfixed by"}
{"_id":"doc-en-vscode-3f5aaa32e262cfe2a7cd1c2d523a91139fcb096fe0efd516a0a2e74e29e3bcd4","title":"","text":"Currently the scroll top position update is not context-aware, meaning when we have cell height changes, we don't know if it's because output height change, or users re-run a cell and outputs are updated/replaced, or users shift+enter to run cell and navigate to next cell. Users might have different expectations for each scenario but currently our heuristics for updating the scroll top didn't take the context into account. Let's use this issue to finalize the desired behavior of scroll top update. The polished/new design should cover [ ] Update element height API respects context (execution, execute+create, output update) [ ] Reveal API should handle out of sync view/content height Click the run button or press Ctrl+Enter to run focused cell If there is no enough room/space after the cell (e.g., the focused cell is the last cell) Users have disabled [ ] There is no output before execution, the view should scroll up to reveal the output fully [ ] Some users might want the bottom of the output to be fully revealed, when the output is larger than the viewport [ ] There is output before execution, the cell position should be persistent Users have enabled [ ] Reveal cell output If there is enough room after the cell [ ] Reveal cell output [ ] If the cell input is larger than the viewport, execute cell should reveal the output properly [ ] Collapse focused cell output should always keep the cell in viewport [ ] Type/Pasting content into focused cell should keep the cell in viewport [x] [ ] Notebook cell list reveal api might reveal a position on a cell whose view height is not synced with real editor height yet. [ ] Execute non-focused cell, should it keep the focused cell position persistent or keep the top of the target cell persistent? [ ] Non focused cell output height change should not change scroll top on scroll [ ] Shift+enter executes and moves focus to next cell, then new focused cell should always be in viewport [x] Scroll top position of inactive notebook editor should not change if the first partially visible cell's height change\nI noticed an issue recently, is this related? Paste a large amount of text into a code cell The notebook scrolls down so far that the focused code cell is out of the viewport I can open a separate issue if you want\nI guess what you were seeing is , I'll update the description to include this.\nCouple cases discovered in the latest notebook sync."}
{"_id":"doc-en-vscode-176de1d1c28db599ad70e969a36afd097686a305245780d8259bc419968a652e","title":"","text":"both of these are likely tied to outputLineLimit being large @ 100 when the shift-enter behavior is set to \"firstline\" the events are as follows: first line of next cell is focused and shown at the bottom of the viewport output is cleared and the focused cell is pulled upwards against the currently executing cell output streams and pushes the focused cell out of the viewport. my ideal thoughts: anchor the focused cell (with setting), maintaining the first line visible. output will stream in, pushing the executing cell upwards out of the viewport shift-enter set to \"fullCell\" and next cell has a large output focus shifts to next cell, and the viewport exposes the bottom of the output for that cell. there is no cell editor in the viewport my thoughts: \"fullCell\" behavior should show the top of the next cell editor at the top of the viewport"}
{"_id":"doc-en-vscode-f211d13ae58be00ece35a4c501dfc89270b686552b07d5613d5b6f21fa1ab213","title":"","text":"In there is , which seems to be called every time the scroll event of the editor fires. That method seems to clear the dom node and reconstruct it unconditionally. When smooth scrolling is turned on, the scroll event fires many times per second and the call seems to take some time: ! I think if you would only rerender the sticky scroll widget when necessary (and generally try to minimize dom API accesses), we could improve the performance here easily. I think could have an equals method and would just do an early return if the new state equals the old state.\nThanks for this remark. It is a very good optimization point."}
{"_id":"doc-en-vscode-74142053608a60f46e73e35b7e928971750a52ded9b26dbafde3e62f3f48926d","title":"","text":"Title.\nOpens on cmd+click. The external link to documentation in the first line works. !"}
{"_id":"doc-en-vscode-694a1dfee5034689c62f02eab19a3e46af102bc6d071d588ce9d779052c5dbb8","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --When the user starts an extension defined terminal through the \"Split Terminal\" submenu of \"New Terminal\", the new terminal is always created in a new terminal tab, instead of a split. When the user opens a terminal through this menu, VS Code should override the field of the new terminal profile to make it split, as the user requested.\nis this not what you're seeing or am I doing something wrong? !\nOnly when the custom terminal profile is created with . I have a terminal profile that is created with , that looks like this: This terminal profile is created like this:"}
{"_id":"doc-en-vscode-112225f9509e6b92ea0c082ed8fbc3168f54171527daca6a11b5b0ec4442331f","title":"","text":"For newcomers it would be nice if had a comment at the top stating something like \"This is a generated file, don't change it manually\"\nOpened a PR for this as I ran into this making my first contribution today"}
{"_id":"doc-en-vscode-82c9fe1f7e7ea6bf2e808dc8585580a8e7dc82c3b291c88b9b8682c152248aec","title":"","text":"I heavily use and love inlay hints but find their current styling too prominent for \"hints\" and think it harms readability and other editor features that use highlighting of words such as \"highlight occurrences.\" Try finding highlighted occurrence of : Current: <img width=\"1779\" alt=\"Screenshot 2023-06-29 at 11 51 00\" src=\"\"With less prominent hints: <img width=\"1779\" alt=\"Screenshot 2023-06-29 at 11 47 31\" src=\"\"I think in the latter screenshot, inlay hints are less prominent but still as useful, as they are literally \"hints\" and don't need to be so prominent. If not as default, having a setting to tweak the background & color of hints would be great.\nI think I've read a similar issue recently (should have the inlay-hint label).\nCan be configured via and leaving it up to to tweak this further.\nUpdatde the default light/dark themes in to make these more subtle. Still want to keep the backgrounds since these could otherwise be confused with the ghost text in Copilot, comments in some themes, etc. <img width=\"422\" alt=\"CleanShot 2023-06-29 at 11 23 35 src=\"\"<img width=\"430\" alt=\"CleanShot 2023-06-29 at 11 22 43 src=\"\""}
{"_id":"doc-en-vscode-f4d9428152502abab5893b47c2fc7287cc34eddd63985eb0b6d0540a8f01733e","title":"","text":"Steps to Repro: Search for an extension pack in extensions view (Say PHP Extension Pack) Open PHP Extension Pack in the editor Go to Extensions Pack tab Click on any extension Fails to open Close to open the editor Does not close Not sure how it happened but I had the extension pack open (as extension editor) and after a while (being on twitter) I wanted to close the editor and I saw error notifications. These are the minified stacktraces from the log the build was\nNot able to repro it. Also followed the code and I do not see a path that has no extension in the input. Do you remember any steps you followed during this? Is it possible to guess what could be happening in editor land from the above stack trace?\nare you using editors with preview mode, e.g. do they appear italic? Hard to say what the execution flow here is to be honest. Looks like a normal editor opening but failing to get the resource for the editor, which is used to associate the editor in the editor tabs model.\nBut this is happening while is trying to close the currently opened extension editor.\nI don't use tabs, so I am not sure what the model then is... Yes, closing it was. I had the extension editor open, also installed the extension, then waited, and then tried to close the editor. So maybe some cleanup logic hit me or installing the extension made a different...\nThe preview feature works with tabs enabled or disabled. The fact that an editor is opened can be explained if you closed an editor and the next one below that editor was the extension editor. Closing an editor always means we open the previous editor unless it is the last editor. I suggest you try to figure out how it is possible that is even though it is not typed as such.\nOk\nWhat are the verification steps for this?\nSteps to repro are in the description"}
{"_id":"doc-en-vscode-cfc23204e11bb0d898e11ff7cc0c424a0707ab1d8fd3f132f17f93cf8df53d71","title":"","text":"<!-- Do you have a question? Please ask it on --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.21 OS Version: Mac 10.12.16 Steps to Reproduce: editor mode should be in vertical mode on centered layout with one tab, you should see it is centered split editor mode to horizontal mode centered layout will no longer work until you switch back to vertical mode <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.458 --- <!-- score: 0.451 --- <!-- score: 0.451 --<!-- potentialduplicatescomment --\nThanks for figuring this out. I spent a while trying to figure out what the problem was. Thank to you, I changed splitting to vertical mode and now centered layout works. I can affirm that this big is happening to me too. I'm on a mac as well.\nI would expect centered mode to work if I have horizontal layout and only one editor. I tried removing the check to simply also enable it for horizontal layout but somehow the layout gets broken in that case. I plan to fix this in the following week but just to let you know that a PR is more than welcome :)\nI just let you know that money is more than welcome :smile: Just kidding... I run on fun xD you shall have your PR :) it has the number . The problem was, that the available size was calculated by using which holds the height and not the width of the editor when in horizontal mode. You have to play around with it a little... it seemed to work but I did no deep testing or anything As bonus I fixed two typos (one of whom I made myself :smile:)\nthanks a lot for your PR, I just merged it in I tried it out and it works nicely, also thanks for fixing typos. For money feel free to apply for our internshipts if you are a student or to our open positions :)\n:smile: I'll finish university in 4 month... so... I might very well do that xD"}
{"_id":"doc-en-vscode-2a861cada512b807f1333a996ec372c3beca4c221a26dcf369c23e4ada497ee9","title":"","text":"Used the stable built on v1.42 for sanity test and I see following error ! I suspect this is because we removed the search location setting and trying to remove its value as migration\nInstead of removing the setting, please deprecate it.\nBTW I see this error in every window until I manually remove the setting.\nTo verify: to and manually set to . the window ensure that: no error notification appears search is in the panel upon reload the setting is removed from your"}
{"_id":"doc-en-vscode-314cedd0246c4ff62f171882cbab235926798567eb3890b0dc14998fc50afc8e","title":"","text":"Create an ipynb file (with one Python cell) Save this Verify the entry exists in time line Change the language of the cell to Save this Verify we have an entry Click that item Bug On the left we need On the right there's no such entry in metadata This is incorrect. The ipynb has the above metadata. What seems to be happening is When the language changes, We serialize the ipynb and JSon contains the metadata However the model is not updated with corresponding changes to the ipynb file I.e. model change triggered =ipynb serialize to JSON =Json contains new metadtaa not in model =Missing step is to update the model again If we implement the last step, we could always end up in an infinite loop, Perhaps we should first update the model with metadata when ever we detect chagnes in the model before the save event. E.g. as part of pre-Save (onDidChangeNotebookDocument &/or onWillSaveNotebookDocument) This way, model will always be accurate and upto date when ipynb is serialized.\nEasier way to explain the bug Create an ipynb file (with one Python cell) Save this Change the language of the cell to Save this The cell metadata will not contain the new metadata that exists in the ipynb JSON file. (model is not in sync with ipynb json)\nSuggested Solution Perform model updates when we're about to save (or when we detect changes to model) When saving, ensure we save the metadata as is, without generating any new content I.e. JSON.stringify the metadata.\n/cc\nConsidering that a serialize operation should not have side effects (ideally), enforcing a \"save participant\" from ipynb extension makes very good sense to me. We can leverage which support additional async operations, like how we handle attachment clearnup\nFor such critical metadata change, we might want to handle both auto save and manual save as we want the json and model to be always in sync.\nAnother simple repro step Create a notebook, and save this Add a new cell The latest entry in timeline will be However there's a difference, one has and the other (current model state) does not have any Again, model does not contain metadta (as this is a new cell) and ipynb JSON contains metadata when serialized."}
{"_id":"doc-en-vscode-24f9e49b8618766031794f870b7860c05b9d56289687149474d11467fda5dcb9","title":"","text":"<details<summaryInternal Notes</summaryHere's the work flow Add a cell A change event is triggered, undo redo stack is updated Last undo redo is User hits save button Working copy starts to handle the save operation Save participants start We add a save participant and update the model via a work space edit Now a new change event is triggered This results in a whole new undo redo stack The latest (last) undo redo item is Save participants done Working copy keeps track of the last as the entry for Problems Time line does not include , time line shows only I guess this is correct, However from a users perspective, the labels are never going to be meaningful as we'd always say . </details"}
{"_id":"doc-en-vscode-8b81ec25d40859710f2ae54dae00a7e55e04802744ecf6a8352c901a14b7f90d","title":"","text":"Keeping this open as the main tracking issue as I keep an eye on error telemetry. Been quiet and haven't seen any entries in the errors site since early Feb with +\nClosing this as fixed by the above linked PR's. Error telem has been quiet with no entries for ~4 weeks."}
{"_id":"doc-en-vscode-d45dd1eef10e83ea02abcb27168df02ef2b1395add81d446110e363859d8e975","title":"","text":"! Version: 1.85.0-insider (user setup) Commit: Date: 2023-12-01T19:00:37.339Z Electron: 25.9.6 ElectronBuildId: Chromium: 114.0.5735.289 18.15.0 V8: 11.4.183.29-electron.0 OS: WindowsNT x64 10.0.\nI think that is expected, see This should behave the same for when multiple editor groups are opened in the same window.\nThe issue is that highlighted words remain until a new word is focused. In the main window, the highlights disappear when the word loses focus.\nnot for me in this case: ! Do you have a case where highlights remove to reproduce?\nIn the main window, If you click on a word and then on a blank line, the highlight disappears: !\nmy assumption here is that the other editor is over a word that then is highlighted in the floating editor. There is probably a heuristic to decide which highlight to apply. Can you double check my theory?\nit doesn't matter if it's on the same word or not but I found out this only happens if the same editor is opened on the main window or in other windows (in this case the first floating editor highlights disappear after losing focus). Highlights disappear after losing focus in: Main window Single floating editor (Not opened in main window) Only the first floating editor when there are multiple (Not opened in main window)\nI can reproduce but this is not a floating window issue. It reproduces even when you just have 2 editor groups opened next to each other: !\ntl;dr: Setting to \"off\" fixes it. I'm seeing what I think is the same bug, after upgrading to 1.85.1 (first upgrade in a number of months, so I've jumped a few versions). While writing this I think there are two distinct problems. The first is when a single file is open: -click a word. It highlights that word, and in a different grey highlights the same word in rest of document. esc: nothing happens. I now need to click a blank line to get the highlights to go away. (I'm sure esc used to get rid of it before??)"}
{"_id":"doc-en-vscode-3737a2344150ac65726ab3a65f4626a0bb01fbdee872dfd53767639d24918bd7","title":"","text":"The second is much more annoying, but only seems to happen when I have same file open twice, after using Split Editor Right. -click a word. It highlights that word, and (in a different shade of grey) highlights the same word in rest of window. In the other window it highlights some (but not all!) of the same word. doesn't get rid of it. Clicking a blank line does not get rid of it. All the grey highlights are there forever. the other window: still the highlights are there. I can now double-click a different word and see it highlight throughout document, click blank line gets rid of that one, but still my first set of highlights survive! only way to get rid of them is to close the editor and open it again. OR, reduce to just one window open, double-click it again, and then click a blank line. Playing around, I've noticed it does not matter which window I do the double-click in, it is the window on the right where the unwanted highlights survive. Regarding mentioned in the 1.84 release notes, searching for that in settings doesn't find anything. But I do have as mentioned in It is currently set to \"singleFile\" If I set it to \"off\", everything works as it used to. Double-click still highlights within the open file (only), and esc now cancels it again :-) BTW, testing the 3rd option, \"multiFile\", it appears to behave exactly as \"singleFile\" in every respect I described above. Anyway, switching to \"off\" gives me exactly the behaviour I want, so I'm happy.\nThanks for the detailed writeup! The editor highlights have been reworked to start bringing in support for semantic highlighting between multiple documents. This was primarily tested between multiple different documents and there are some bugs arising that have to do with split view of the same file. Will be addressing in the near future.\nFriendly ping! Looks like this issue requires some further steps to be verified. Please provide us with the steps necessary to verify this issue.\nSteps for verification: Open a typescript file in split view, with content as follows: In the left panel, click the symbol. In the right panel, click the same symbol. In the right panel, click anywhere else that isn't a symbol."}
{"_id":"doc-en-vscode-2f0f1fd97986542ee40306e204fc8e664a137b630278a0a9a80d0c017872f879","title":"","text":"choose any whitespace, etc in Stable: The highlights should disappear from the left panel, but persist in the right, even as you click around the right panel. Fixed on insiders: You should not be able to end up in a state with persistent highlights when your cursor is not on a symbol."}
{"_id":"doc-en-vscode-e85d9200f2694d60a0db8e96b30ee56825af53617d7afe775195ae079090391b","title":"","text":"Steps to Reproduce: an empty window the trust editor, and trust the window the window Expected: Window is trusted Actual: Window is not trusted\nThis is also repros when you have the setting enabled: <img width=\"1024\" alt=\"CleanShot 2021-06-08 at 08 17 39 src=\"https://user-\"\nPushed a fix for this issue into , in case we will want to take this as a candidate for the June release."}
{"_id":"doc-en-vscode-e5e8d1d5611c20aaf9ff710c8f9ccfe5adbdfe8c4b92bc3ebdd107e27f1b303d","title":"","text":"Version: 1.94.0-insider Commit: Date: 2024-09-20T05:03:57.400Z Electron: 30.5.0 ElectronBuildId: Chromium: 124.0.6367.243 20.16.0 V8: 12.4.254.20-electron.0 OS: Darwin arm64 22.6.0 Steps to Reproduce: Set Google IME for Japanese. The cursor is not visible during typing. On almost all the platforms and all the applications, the cursor is usually visible during composition.\nHi yes thank you for opening this issue. This is caused by some code that is related to the current edit context implementation which hides the blinking cursor because there is already a blinking cursor in the visible text area shown during composition. A PR has been made to fix this."}
{"_id":"doc-en-vscode-cff0ed39afdf61b157bc7020baa9a829f6f10423fb6d494dc8ca2f89a747e6ff","title":"","text":"Testing the GitHub code spaces extension remote explorer and try clicking on any of the help and feedback items bug Nothing happens In the developer tools I see:\nI cannot repro this, but I do see that after installing the Codespaces extension, if I click on any of the help and feedback items, no Codespaces-related entries are displayed\nI can't repro the buttons being busted, and I can't repro no Codespaces-related entries either.\nwhat version of Codespaces to you have installed? And, can you check that it's activated?\nthe codespaces extension isn't * activated anymore, so in a local window after installing the codespaces extension it shouldn't be activated.\nI would still expect it to be activated after clicking on the Remote explorer help and feedback items though as we run a command (ex. github.codespaces.getStartedLink in the original issue).\nThis is a bug in the codespaces extension, it's because the extension doesn't declare in its or . This used to be fine because of * activation but after we removed that the extension no longer activates timely. I'll open a PR to fix.\nWorkaround is to switch to the GitHub Codespaces view in the remote explorer or execute any of its commands, which will cause the extension to activate.\nHm I'm still seeing this even after reloading. Codespaces version: v1.14.0. Currently not running\nEven with my PR to the Codespaces extension, we need an additional fix in remote help land. Today there is a raced timeout of 500ms and we just return after that. Since the Codespaces extension now activates on demand (and usually takes more than 500ms to activate), this means that the empty string gets cached as its url value. Since the extension will eventually activate and the command will return a value, I'd propose that we update with the resolved value at that point. To the user this will manifest as the picker populating with more values as extensions finish activating (which IMO is fine as that is how our other menus behave as when clauses are evaluated for extension-contributed commands).\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected."}
{"_id":"doc-en-vscode-cb9e629e464eb85c86e66ecdc685a35b460778817775c693f8c50cfe74feca2e","title":"","text":"If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nThis should be verifiable in the latest Insiders and with the latest GitHub Codespaces release from earlier today.\nAre the repro steps just latest codespaces extension remote explorer to click on help/feedback items Or is there more to it?\nyeah that's it!"}
{"_id":"doc-en-vscode-42d827d5f09e9a6ab3d21d947a089428bcc257b8aac955f651e0b6ee8b3dd002","title":"","text":"Issue Type: <bBug</bRunning tasks with dependencies in a multi-root workspace executes the wrong dependency tasks, if there are multiple tasks defined under the same label but in different root folders. VSCode will just pick the first occurance in the order of the root folders defined in the workspace settings. To reproduce the issue: a simple multi-root workspace, e.g. each root folder, add a with two tasks. One task depends on the other. Example tasks of : and have the same tasks configured except that they echo their respective folder name instead. task of either or . VS Code version: Code 1.65.1 (, 2022-03-08T02:06:27.846Z) OS version: WindowsNT x64 10.0. Restricted Mode: No <details<summarySystem Info</summary </detailsExtensions: none<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nExperiencing same issue on: VS Code version: 1.65.1 OS version: macOS 12.2.1 BuildVersion:21D62\nTo verify: or create a multi root workspace with 2 roots, FolderOne and FolderTwo (or names of your choice). each of your roots add a .vscode folder. each of your .vscode folders add a file with the following contents: \"PrimaryTask\" in FolderOne (command \"Tasks: Run Task\" and choose the task with the label \"PrimaryTask\" that says \"FolderOne\" next to it). Verify that both PrimaryTask and DependsTask echo FolderOne's path. \"PrimaryTask\" in Folder Two (command \"Tasks: Run Task\" and choose the task with the label \"PrimaryTask\" that says \"FolderTwo\" next to it). Verify that both PrimaryTask and DependsTask echo FolderTwo's path."}
{"_id":"doc-en-vscode-88511c54bec15c99478a5fdddd1537ca036b4e60f0976f9008965f0db025ad87","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.64.0-insider () In the Catch2 C++ test framework it is possible to have multiple test results for the same test in one run. This happens for example for the following code: In this scenario, the section \"S\" is run twice, and the first time it fails, the second time it succeeds. The C++ TestMate extension will see the two test runs, and call for the first one, and for the second, with the same for both calls. This results in the confusing display that the section is green, even though it shows a failure inside: <img width=\"653\" alt=\"image\" src=\"https://user-\" does it make sense to change and (et al) so that when called multiple times, it sets the item's state to the max of the old and the new state? This seems to be analogous to what happens with the test messages (they are appended, and doesn't clear them). Or would you say it's the extension's responsibility to not set the state more than once for a given TestItem? (Cc Or maybe it is even Catch2's responsibility to accumulate the results for any section that is run more than once?\nIMHO, Catch2 reports it and definitely not the responsibility of vscode. It leaves it to me.. (Though I don't know how I can solve this properly)\nWhy definitely? It doesn't seem so unreasonable to me. As I said, it already does something similar with the test messages.\nAs I think of it, vscode part is just the UI, the controlling is in my hand. More precisely and . the test provider is the ."}
{"_id":"doc-en-vscode-b3dcb2865e31b4e48d6225e8caed1860893b68867ea601ac5238c96bf5ed6d5e","title":"","text":"I'm not so sure in this case. VS Code is also responsible for setting T's state correctly, you don't do that at all (as far as I'm aware). And note that it gets that one right, so it has some logic already for computing the resulting state of all child runs if there are multiple for the same item.\nI would be very interested in your opinion on this.\nYea, I think it would be reasonable to no-op if the extension tries to change the test state to one with a lower priority. For instance, going from passed-skipped or failed-passed.\ncan you add some steps?\nthe test sample: this block F5.Verify that, with these changes, test that should pass remain passed in the UI, and tests that failed remain failed."}
{"_id":"doc-en-vscode-39ad0fd202d6419a00716fe182e023335d84f0bb9c05176f6953c0bb40d1574b","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --Description: As of 1.44, I am not able to open multiple files from \"Go to file...\" menu (ctrl + p shortcut) when middle mouse clicked on them. I used it when opening similiar/dependent files (ts, html and css for component, etc.) As of 1.44 middle mouse click only opens last file clicked. Steps to Reproduce: dependented files (.html, .ts and .css) with same name. for that name in \"Go to file...\" (ctrl+p shortcut). mouse click on multiple files. <!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.44 OS Version: Windows_NT x64 10.0.\nI found that holding alt + middle mouse button does open multiple files in 1.44. If it is not regression please fill free to close it.\nWorkaround: configure in settings."}
{"_id":"doc-en-vscode-8e2347e6b4e880b13466128570a1637f9375286dadd85e033d935773cb2a047b","title":"","text":"Testing open execute a cell that shows issues select copy output no output is copied I'd except that this \"just works\", at least for its text/markdown output representation"}
{"_id":"doc-en-vscode-f16d9876d169a3ef5780822c398ced7a1bd7a5ca62da6f4778b94f9bfc07d9a8","title":"","text":"It would be nice to invoke current line codelens actions without moving hands of from keyboard to mouse. It could be done via context ctrl+. where codelens actions are or from ctrl+p, or even new shortcut key which opens context menu which contains only codelens actions. They are sometimes available if extension implements them that way that they are to menus/context etc but this probably could be handled more generic way in vscode itself?\nHi! I just opened a PR with the ability to trigger codelenses with the keyboard. Hope you find this helpful as I do :D I appreciate any suggestions and comments from people coming to this issue"}
{"_id":"doc-en-vscode-9de999d6267f4a631fb7892d3744d6714b77c4fe90130fa105315a16e13e85c7","title":"","text":"I've verified that I can see the terminal hint in insiders. Does this also need to be tested in stable?\nYes this needs a stable build as it checks quality. You could verify out of sources by setting quality to stable as well"}
{"_id":"doc-en-vscode-497c942bd8489f00b4f42fb8e0dbbc5a152a814c26c964bb95f3541738e81d2d","title":"","text":"Can this support right click to toggle the setting to hidden? !\nright click shows our terminal context menu, perhaps we should just put it in there rather than adding a new one? !\nthis isn't a feature the others offer, thus not a bug IMO"}
{"_id":"doc-en-vscode-41925bd92939a7ce8239b4d736089bea64e55b4329d6edcd63c7b80cbbb4cef6","title":"","text":"run voice over open terminal start accessibility help via while it start reading select :bug: the reading continues all way to the end\nI'm not sure we can / should prevent this from happening. Screen reader users have a way to abort the reading of certain contents. or might have insight here\nThis is very strange. The behavior is as if Voiceover wasn't detecting the change of context. In other words, upon pressing Escape, Voiceover is not detecting that focus is moving from one control to a different one. This is often related to focus events that are not happening. I have a question. Is the terminal accessible help somehow being removed from the DOM or hidden from accessibility without removing focus first? When debugging with NVDA, the first thing that happens after pressing Escape is an exception, as if the screen reader had lost access to something that should be there (DEBUGWARNING - NVDAObjects.IAccessible.IAccessible.getIAccessibleRole (15:05:00.655) - MainThread (5320): accRole failed: (-, 'Unspecified error', (None, None, None, 0, None)) ). Voiceover not stopping upon closing the accessibility help not only happens in the terminal. It also happens in the editor, the copilot chat and anywhere we call the accessibility help. Another thing I have observed is that, if I press f1 from the editor, I hear the name terminal accessibility help even though the terminal is not in focus and the content is related to the editor. This happens in both Windows and macOs. Finally, I have also observed that the command Open accessibility help doesn't show up in the command palette on the mac side for me, although pressing Option +F1 opens it.\nThanks for your findings You hear terminal accessibility help because I accidentally had that as the aria label. Fixed here I will comment on the others in a bit\nThis wasn't showing up there 2 days ago - are you on the latest insider's?\naccRole failed: (-, 'Unspecified error', (None, None, None, 0, None)) Yep, we were closing the context view before calling , which is the function that focuses the other element. Will fix\nHm, that didn't fix it. What fixes it is after pressing escape, moving my cursor in the editor."}
{"_id":"doc-en-vscode-cef8b50dad7b0e194f0132a20a36c4b2e7ae4e9a7f2324c1c50c331700cf8d15","title":"","text":"is correct though\nIt says you are currently on a application, inside of web content. When the editor is focused. !\nRight. From the VO perspective, VS Code is a website.\nHovers use the context view service as well - they handle this correctly though when escape is used. that is done here. we probably need something similar\nI clearly had not updated when I thought I had done so. I'm in the latest now and the command is showing up.\nGreat, thanks for the debugging. This issue should be fixed shortly.\nThat didn't work either, going to keep trying.\nOk so hiding the context view sets to and the element's display to . but for some reason, the screen reader continues to focus on that\nFigured something out - when I use , then it works. So this is a timing issue\nThis seemed to work correctly for me... though my ESC button is a part of the touchbar so I have to do extra steps when VoiceOver is on to actually trigger ESC."}
{"_id":"doc-en-vscode-c5586c9b19498f880eec2c158a652e0492cf2ea1e7782858d5a44ac02f0cf964","title":"","text":"Do we do this for other widgets? It seemed off to me at first sight. <img width=\"405\" alt=\"CleanShot 2022-10-04 at 09 58 26 src=\"https://user-\""}
{"_id":"doc-en-vscode-c46eda2c4ef7c6626ec01f599dcdea61c6a8aec8bbfc238e1d2fb0bf15baaf99","title":"","text":"When rendering Markdown preview, Unicode newline characters U+2028 LINE SEPARATOR and U+2029 PARAGRAPH SEPARATOR are replaced with an empty string at the source preprocessing stage. This behavior was introduced in to but it breaks the document in multiple ways: double-click word selection does not work as expected, screen readers may announce the text incorrectly, line wrapping and hyphenation does not work correctly, markdown-it plugins that rely on or may work incorrectly, even linkify fails: and so on. Since then it was fixed at the markdown-it side, and is no longer needed: It would be better to revert this change, and keep the Markdown source as is before feeding it to the parser. Steps to reproduce: a Markdown document like this: preview: Expected result: LS and PS are displayed as is, links end at Unicode newlines. Actual result: no Unicode newline characters displayed, links also consume the immediately following word. <!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: OS Version: WindowsNT x64 10.0.\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!"}
{"_id":"doc-en-vscode-88350a42f9f541f0cf2750e47dc3294e1fa74ecb2ea736f26f6876886e44e4fd","title":"","text":"The call was removed in\nIt's weird, seems like in stable we should be leaking terminal processes then but I see them shut down immediately after closing the window in stable..."}
{"_id":"doc-en-vscode-e1193c8e2e5c49ef23b421bf53f7e2131b1c60a985c4fb180c98c450e67b2146","title":"","text":"Type: <bBug</bI have noticed two problems with my vscode-insiders installation which I assume may form one issue. a notebook in nbformat 4.5 is opened, new cells do not receive cell ids, even though this is required by the format spec and is a documented feature of vscode: a new notebook is created, the format is always 4.2. I am not sure whether this is expected. Steps to reproduce: a notebook with jupyterlab or any other way (jupytext etc.), and verify that it has nbformat = 4.5 the notebook in vscode, and generate a new cell, then save. the JSON of the saved notebook - the new cell does not have a cell id. Some additional points which may be of interest The issue persists when all extensions except jupyter/python are disabled. I have recent versions of nbformat etc. installed, but I would assume this is bundled with vscode anyway. The issue also exists on the current pre-release version of the jupyter extension. I haven't found any configuration option which would enable or disable support of notebook formats 4.2 Maybe I have missed a configuration option which I should have set? VS Code version: Code - Insiders 1.84.0-insider (, 2023-10-23T05:33:33.712Z) OS version: Linux x64 6.2.0-35-generic Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (18)</summaryExtensionVersion ------ vscode-neovim0.9.1 vscode-eslint2.4.2 githistory0.6.20 fuzzy-search0.0.3 black-formatter2023.5. python2023.19. vscode-pylance2023.7.40 jupyter2023.10. jupyter-keymap1.1.2 jupyter-renderers1.0.17 vscode-jupyter-cell-tags0.1.8 remote-ssh0.107. remote-ssh-edit0.86.0 remote-explorer0.5. remote-server1.5."}
{"_id":"doc-en-vscode-7696ae3abb6fbc431dbdf8c317f744092a101e09192906d91784b498d83704e4","title":"","text":"even-better-toml0.19.2 my-vscode-utils0.0.2 pyexec0.0.1 </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nThank you for filing this issue and sorry you are running into this. I can replicate this at my end as well. Will look into this."}
{"_id":"doc-en-vscode-cf365a15576edae35f742d52eca205a3671d61ddbca495c64b93d31d9a6c2fd4","title":"","text":"see gif open GH notebook with a few cell that reference a variable run peek references and select entries notice how notebook cells are being highlighted, like \"Milestone Work\" get selected !\nI can repro in Insiders but not master, I'll check again tomorrow\nThis happens because of this listener: Tho, not yet sure why the selection keeps changing... Maybe some inner list event (from the references list) bubbles up to the outer list\nThat listener is fine, it's the nested lists\nHey do you have any thoughts on handling this? We have a list inside another list. The inner list handles the touch event properly. The outer list then handles the event and picks up the wrong index because it finds the data attribute inside the inner list in We can Make sure the event is fired on the outer list with the correct element, by fixing to check whether it's in the right list Not fire the event on the outer list at all, when we see that it originates in a different list Add some property to the list event that says it originated in a different list so the consumer can decide In this particular case, I think any of these are fine\nThinking more, I don't think that the event should ever be handled by two lists at all. Meaning we either mark the browserEvent as handled, like with preventDefault/isDefaultPrevented, or we make the list check every event to be sure that it originated in this list.\nThis sounds good to me. This also sounds good to me.\nrelated when I do peek references, I see markdown cells being selected? ! Notice I'm looking for under-discussion but \"Configure\" is being highlighted as one of the references\nFrom Good catch, I didn't notice that\nThanks for the quick handling!"}
{"_id":"doc-en-vscode-6554f56bf523bc4b47327aaa1734df4847c0ee2a784df9a5955d07132b19c8ef","title":"","text":"Hi, I would like to request that upon restart vscode, the terminals can be restored. By restored, i mean the layout, the custom names, icon and color for the terminals are retained. The content DOES NOT need to be retain. Otherwise efforts to customize all the opened terminals would gone to waste when vscode is restarted. Alternatively, vscode can provide new workspace configurations, where developer can manually enter a list of terminals' custom names, Vscode, upon starting, will automatically create new terminals based on the configured list. Even better, each time, deveoper manually open a new terminal , vscode will auto update the workspace configuration file to add a new terminal to the configured list. This applies the same to renaming of terminal Hope that this can be achieved without relying on plugin. *Do note that this might not be a duplicate of other look similar kind of feature request. Cos for this request, content does not need to be preserved. And the terminal's custom names should be preserved (sadly, persistentSessionReviveProcess, does not maintain the terminal custom names)\nexcept it doesn't save custom terminal names.\nNot only it doesn't save custom terminal names, it also doesn't save icon and color.\nThis is meant to be supported with including content as well, there are currently some issues with it causing it to be flaky though. They might have to wait until v1.63 where we will focus on testing. Related:\nfixed !\nIt's persisting the icon, color, name, and layout on window reload. Doesn't seem to persist icon, color, or name on restart.\nYes. My request is focusing on restart, not window reload"}
{"_id":"doc-en-vscode-93b24be3e8729d700ddc89e65b4b30e7e5abb40bd5ccd464bc36e2105b480ce2","title":"","text":"Issue Type: <bBug</bWhen pressing F8 in editor without error markers in current file, tab will switch to another opened file with error. Is it possible to prevent tab switching when F8 is pressed and there is no errors in current file? VS Code version: Code 1.22.0 (, 2018-04-05T17:25:43.982Z) OS version: Darwin x64 16.7.0 <!-- generated by issue reporter --\nIt's caused by a change in behavior in 1.22 (). I think it's a change to the worse: it's very easy to get used to hitting F8 to get a quick feedback for whether the current file has any errors, and this change breaks this workflow in a very noisy way. It also punishes those who work on 'dirty' projects with lots of errors and prefer to concentrate on one file at a time. I think it would be great to have a config switch for \"F8 checks all files/current file\", or at least a pair of keyboard bindings for the old behavior. I'm actually surprised that it wasn't implemented with this change.\nUnderstood. The goal was to keep the number of setting low, but I understand the different use-cases and muscle memory. Until then, a workaround, albeit weak, is to hit Shift+F8 which takes to back to the file.\nYeah, I agree that \"convention over configuration\" has its benefits, and it seems that the suggestion that caused this change was quite popular, but IMO this warrants at least a pair of shortcut bindings. Speaking about conventions, I've just checked that for example similar shortcuts in PyCharm (and I assume all other JetBrains IDEs) only jump through errors in the current file. One upside of the old behavior was that when the current file had no errors, there were no visual changes after pressing F8, and the cursor stayed where it was. Now in this case Shift+F8 after F8 won't even bring you back to that file, it would jump to another file with errors. \"Go Back\" or \"Open Next Recently Used Editor\" works though. It was also very useful that F8 would wrap around and bring you to the first error in the current file if you are working near the end of it and did some changes that e.g. made some imports or vars unused."}
{"_id":"doc-en-vscode-34424d5ba7ccf9cd53358b7508c8d44ff7b4582d261dafe4980edacfbe331859","title":"","text":"I think I was relying on this behavior 80% of times when I pressed F8. And now it's become quite difficult to predict whether F8 or Shift-F8 would keep you in the current file or move to another one.\nWe have the two old commands back (in addition to the new commands). That means there is now \"Go to Next Problem\" and \"Go to Next Problem In Files\" as well as the other direction. We have not changed the keybindings, so F8/Shift-F8 walks across the project and you would need to reassign the them to get old, old behaviour back.\nWow, I didn't expect this minor issue to get resolved so quickly! Big thanks to the team, it's really a marvel to witness the development of this amazing project.\nThanks to\nI verified the new command works as expected, however is the name \"in Files\" really clear what it does. Why not \"across Files\"?\nWe chose \"in Files\" to match \"Find in Files\". \"across Files\" makes sense too though and might be clearer for users who haven't used \"Find in Files\" before."}
{"_id":"doc-en-vscode-8b952e685f65b7d3271dca65ad33105d1b3323acee36d65cc4ce3f170febf313","title":"","text":"mentioned that the diff editor experience is a little confusing because we don't explicitly state that the current editor is the modified one vs the original editor. AFAICT, there's also no way to jump between the original and modified ones. Is this correct? & I'd appreciate your thoughts here. Edit: action items: [ ] add info in help dialog about switch sides command [ ] get accessibility signals to work in original editor\nThere is the Diff Editor: Switch Sides command that moves focus to the other side. Maybe the wording could be improved. So not sure what needs to be done here? Feels like similar to the other issue you created about missing clarity in announcements on original vs modified.\nYes, I was unaware of that command. We should include it in the diff editor help dialog and include a default keybinding for when in screen reader mode as that's definitely something users would want to do IMO.\npointed out this should maybe be a tab stop if possible\nI have the following keybinding, but there was no noticeable change (at least to from a screen reader user's standpoint) after switching into the other side. :\ncould you tell there was a focus change at all? Is there also no difference if you run the command vs use the keybinding?\nAudio cues do not work when I'm navigating in the original editor. !"}
{"_id":"doc-en-vscode-42a315430675af4ca2df6b788cc163ce4d7dc49f5acc09123ed7e41d6e444a05","title":"","text":"There's an unnecessary whitespace in \"i18n/vscode-language-pack-\". The whitespace after \"command:\" makes the link dead. !\nyou can add a comment to this string like so: to ensure that the command never has a problem like this. Notice that the example string has, in the Locked area, a at the start of it. This will make sure there is no whitespace between them."}
{"_id":"doc-en-vscode-88814189936f248f98419ea02b900a5fff888f7fbd6e7755da417c955f33c3c6","title":"","text":"Steps to Reproduce: text into PowerShell (with Terminal suggest enabled) see batched pasting behavior\nmaybe related to\nHaha, I think this is it triggering completions while the pasted text is being send then kicking in. Pastes should never trigger completions so let's track that here.\nWhen verifying make sure you're on macOS."}
{"_id":"doc-en-vscode-a49201987a4d4f4a5d019210111fa4090752b6d5fa51adeeb9b2de62a3308ea6","title":"","text":"todays insiders do any inline chat query the edits are done live and shows inside the chat diff editor <img width=\"946\" alt=\"Screenshot 2024-06-14 at 10 59 47\" src=\"\""}
{"_id":"doc-en-vscode-d46f20ea20fc07fa480fa40763a9d623b03bcb46d13fc3937129e5d7da328844","title":"","text":"Testing When I was on the section of the walkthrough I clicked on \"open remote repository\" and it had the dropdown. I noticed that I couldn't just click outside the dropdown (back on the welcome page) to make the dropdown disappear and instead had to use the \"esc\" button. From my experience with other dropdowns if you click away from the dropdown it disappears and think that behavior would make sense here.\ndoes the quick pick come from ? If yes, who should I direct this issue to?\nYes that's where the quickpick comes from. owns the desktop git extension.\ncould you please help with this? Thanks.\nClicking outside of the dropdown still doesn't escape. !\nIt turns out that the \"Remote Repositories\" extension does not leverage the picker from the git-base extension so the fix that I made is not relevant. To verify this, please download the latest pre-release version of the \"Remote Repositories\" extension."}
{"_id":"doc-en-vscode-1615a9448524643cd1d66b72b14f3fa895944edc668a9c1e48f84a42fe0a59c5","title":"","text":"On previous versions of VSCode (1.47) you received a message when attempting to save a file that was marked read only: ! As of 1.48, you receive a different message that lacks an overwrite option: ! When a file is simply marked as read only, retrying as admin is not an appropriate option and fails to resolve the issue. Please restore Overwrite functionality as it is very useful for us that use archaic source control systems that rely on making files read only. VSCode Version: 1.48.0 OS Version: Windows 10 (1903) Steps to Reproduce: a file with the read only flag set a change to the file the file Does this issue occur on the latest insiders build?: Yes Does this issue occur when all extensions are disabled?: Yes\nThanks, a regression we can probably fix for recovery.\nWas just about to post this as well. This changed about 3-4 weeks back in VSCode Insiders as well if that time frame helps.\nWorks great, I have verified the fix on windows with read-only files."}
{"_id":"doc-en-vscode-659e03d027d0a62a11d1c910df7e644970de8d3c7bcbf327c3e37a5cf006368c","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version:Version: 1.64.0-insider (user setup) Commit: Date: 2022-01-30T20:09:01.122Z Electron: 13.5.2 Chromium: 91.0.4472.164 14.16.0 V8: 9.1.269.39-electron.0 OS: Windows_NT x64 10.0. Steps to Reproduce: to install version 1.8.0 (pre-release) or 1.8.1 (pre-release) of the C/C++ extension using VS Code (possible) versus VS Code Insiders (not possible). FYI, we have 1.8.2 (non-pre release) and 1.8.3 (pre-release) released.\nBefore I was getting this dialog After the fix I can nicely install -verified !\nClosing as it appears to have been merged into release branch and verified"}
{"_id":"doc-en-vscode-e10c250da56e86ce0397bfef48d2b613c9b006bc22ac6b8678b912182b53726b","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.46.1 OS Version: Ubuntu 20.04 LTS x86_64 shell: bash When alt-clicking in terminal (e. g. to open a file or URL), it gets filled up with lots of or : ! Steps to Reproduce: integrated terminal (or, for a more practical example, in a temp dir) -click anywhere <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes\nHaving the same issue in WSL with Ubuntu 18.04 when alt+clicking any link.\nLooks like this is an issue where the alt+click feature (move cursor when clicking) is conflicting with links, only when is used.\nThis issue also occurs on MacOS, only with Opt+Click instead of Alt+Click. I thought at first that this was related to the Opt+Click on a detected link in the terminal (since that is an Alt+Click on Windows/Linux). But this truly does occur when Opt+Clicking anywhere in the terminal while a command is waiting for input. I can confirm that this occurs on my MacOS regardless of the value of setting.\nset in , now the way you click url is cmd + click, and it wont happen again\nnope, that only slightly changes the junk that is dumped into the Terminal, but it doesn't get rid of it... With (default): With :\nemmm, you say is right, i just change a way to open url in vscode terminal to avoid echo ^[[D but not fix it , this also occurs in , if you use option + click. by the way,my final setting as follows (if , vscode still let you click url with option + click ) !\nOh! That's a great observation, Alt+Click fills the terminal with junk even in the standard MacOS Terminal app and also in iTerm2. Now, I am aware that is originally reporting this for Ubuntu. I checked my Linux laptop (Fedora 31), and Alt+Click in the standard system terminal emulator does nothing. It only occurs in VSCode's terminal."}
{"_id":"doc-en-vscode-77184ecd9867d371cbffd3aad1da92d11849e4f4b3243f7163bb7de0ae844bc0","title":"","text":"But this still makes me wonder if all of this is system-level issues, or perhaps terminal \"emulator\" issues, not really a VSCode issue. Hard to know. But the Microsoft devs would know.\nAs for your workaround, thank you so much for that!!! I still don't see any change in VSCode's behavior when using , but changing definitely avoids the core issue. I hope you don't mind me describing in more detail... According to , \"The Go To Definition and Open Link gestures will also respect this setting and adapt such that they do not conflict. For example, when the setting is ctrlCmd, multiple cursors can be with Ctrl/Cmd+Click, and opening links or going to definition can be invoked with Alt+Click.\" You probably did what I did when originally installing VSCode... I'm using to using Ctrl+Click in Linux and Cmd+Click in MacOS to multi-select files in Finder, items in a list, etc. So I set to be . However, this means that attempting to click on a link in the terminal (for example, an error message referencing the offending line in your source code) requires the use of the other shortcut (Alt + Click). Since it seems that Linux and MacOS have various strange behaviors when Alt + Clicking in a terminal, you are suggesting we switch back to the default . Now we'll use Ctrl+Click / Cmd+Click to follow links in the Terminal, avoiding all that ^[[D^[[D^[[D^[[D nonsense that sometimes causes the terminal to hang. For multi-selecting in the editor window, we'll have to get used to using Alt + Click for that, but in my opinion, that is better than having terminal hanging issues unexpectedly.\nSo to clarify what I originally posted to Alt+Click will still cause junk to fill the screen, regardless of the value of \"editor.multiCursorModifier\" setting. However, the above workaround means we don't need the Alt+Click anymore.\nit's a feature that's sending keys to move the terminal cursor to the position of the mouse cursor. It needs support in whatever program is running in the terminal though which is why you might see the printed, we cannot detect this support though. The advice right now is to avoid alt clicking if it doesn't work in the program."}
{"_id":"doc-en-vscode-d4af10f74cf7d344810300f2ae19ce758c89e1978e00c3da710004566338c015","title":"","text":"could you clarify... Is this a feature in Linux that's doing it, and vscode is not able to see that it's happening? Or do you mean this is a feature in vscode to move the cursor, and vscode is not able to tell if the terminal shell supports it?\nit's a feature in , you can see the PR here\nwhat an interesting feature, and I never even knew that was there. Thank you for the perspective. I looked through that PR and searched the settings in VSCode; unfortunately, there doesn't seem to be any way to disable that cursor-moving feature.\nYeah, that feature might be useful but the ability to disable it or, even better, change a modifier to another one (so that it doesn't conflict with the “open file in editor” one) :)\nI'm seeing this happen regardless of the setting, on Ubuntu 18.04\nJust to inform that is waiting for a community PR for this.\nUm, how is project related to VSCode?\nHey great question. The package is a direct dependency as you can see here: means that the behavior will first have to be patched in the underlying package and then it will likely be made available through a setting in VSCode in a future release.\nI opened a PR with which would make it possible to disable this feature:\nThe PR was merged. What is required in order to pass the new option to in VSCode?\nI meant no offense; perhaps my \"Um\" conveyed an unintended tone. I was simply asking, as a non-developer user who has never looked through the VSCode codebase, but is experiencing the OP's issue... how an xterm project would be blocked by an issue in VSCode.\nAh sorry, I assumed you were familiar with the package because of your previous posts (and although it’s no excuse) I’ve learned my lesson and will henceforth refrain from posting prior to my morning coffee - the tone was indeed unnecessary. The original message has been duly modified.\nSo, does anyone know how to disable it in vscode?\nThe problem with has been fixed. Could you kindly suggest what are our next steps here?\nSorry, I've been out for a while. We updated in vscode yesterday, so the next steps are to expose it as a setting in vscode. Searching for a similar setting should show how/where to do something like this."}
{"_id":"doc-en-vscode-7bd624b5d10c486609e621200abff6c153cd73c800288b4dc6913c577fcc794a","title":"","text":"if no one picks this up from the community it would be a good change for you in debt week.\nThank you so much! :tada:\nOnce it's released in insider's, to verify, set\nHi! Thanks for fixing this Just to understand, may I ask: Why will the future setting default to true? One year ago the problem didn't happen on a fresh install, so why not fix the regression Why is this issue closed but the bug still exists in the latest Insiders? Thank you\nAs far as I understand this is a new feature in . The special characters only get displayed in certain situations (probably when the running command prevents the shell from moving the cursor). Also the default setting for multicursor is <kbdCtrl</kbdso you would not use <kbdAlt</kbdnormally unless you want exactly this new feature.\nI just want to open a file from within the terminal out of the box ! I typically do when running in watch mode On Linux the shortcut is Ctrl+Alt+Click even if the screenshot says otherwise Can users do this without having to find this issue and the corresponding setting? Also, when the filling happens, often the terminal gets stuck, I cannot kill the process with anymore, should I open another issue about this?\nThank you for voicing your concerns - I have to agree that OFF should be the default behavior as Alt-Click is mapped to the file open command in VS Code, and ON is therefore most likely to be the undesired behavior.\n<kbdAlt</kbd+<kbdClick</kbdis not the default. The default for opening links and definitions is the one key not used by the multicursor modifier which is by default <kbdAlt</kbd. You can confirm this by looking in the docs or viewing the default value in the settings UI.\nJust tried in Portable mode on Linux with no settings nor extension installed. The problem still reproduces. fixes the problem but I'm not sure how new users would know about this\nYou may be misunderstanding what this option attempts to solve. It's unfortunate that the name was chosen for the setting, since the setting is not about vs. , but rather about enabling or disabling the pseudo-mouse-cursor-moving feature in xTerm entirely. I concur with go back a year and VSCode didn't have this annoying broken \"feature\" of the terminal."}
{"_id":"doc-en-vscode-502343640578038a0879070d3c2ca3daa15992cbeef92f3bb437e7d6c46426a7","title":"","text":"Now that we've recognized it's broken and updated the xTerm dependency to a version that allows us to get rid of the broken \"feature\", the default behavior from a VSCode perspective should be to have the feature disabled, since we really never wanted it in the first place. Also, as pointed out, when the typical user hits this strange behavior, are we just expecting them to \"VSCode terminal fills with ^[[D^[[D^[[D^[[D^[[D^[[D and locks up\" into Google and hope for the best? This \"feature\" does not work 100%. It should be considered beta/experimental and disabled by default.\nI can attempt a PR that Changes the setting name to perhaps (something that doesn't mention vs. ) Default value of But I'm not sure how the open-source community handles requests like this. Any advice on how to proceed? Or is this a closed/merged PR and conversation over?\nAfter re-reading this I realized this is capturing the bug (alt+click conflict) as well as the feature request (new setting), I thought they were separate issues. to fix let's: Override the value of to be false when . Document this override in 's description, that also makes the name correct as this feature only ever works on alt+click. I think that name is good as it aligns with . we're a little tight on time as we're closing in on the end of the iteration, can handle it\nThanks, and I'm really looking forward to being able to click links in the terminal again.\n/verified"}
{"_id":"doc-en-vscode-703995a52b98ccd355480f52d9b783cbe44aefbd7fc68cfcc6e527ef4201a516","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --- VS Code Version: 1.54.3 OS Version: Windows 10 20H2 Steps to Reproduce: a file with the same words with different casing a find replace with preserve casing switched on the preview does not work properly (and the replace also doesn't) preserve casing does not do anything ! I'm pretty sure this worked as expected before.. <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported direct to the extension publisher. The 'Help Report Issue' dialog can assist with this. --\nThe feature came via\nThis case is not handled. When the match is mixed-case and starts with a lowercase character, I think the replace pattern should have its first character lowercased. Would take a PR that adds this."}
{"_id":"doc-en-vscode-d473eafb800f1f3f75c3e8d7b35c229a10fe8a3b47f9a636e34c207fd21a9394","title":"","text":"Testing a notebook and add a markdown cell with content at the top the replace widget and find an existing string in the markdown cell focus somewhere else in the notebook, then press ctrl+f to focus the find widget markdown cell is bumped up out of the view !\nyou found good bugs!"}
{"_id":"doc-en-vscode-d230795fa46c02d67687a67ef054a2352f9a13430961ec403963c944aa48bb9b","title":"","text":"Originally posted on without success. <!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.69.2 OS Version: linux Steps to Reproduce: project has files like \"Bar`1.cs\" to represent generic types, as is done in the dotnet SDK build task contains \"problemMatcher\": { \"base\": \"$msCompile\" } When such a file has errors/warnings, I get this: is fine. But I cannot control-click on that filename to open it. The file's path is not underlined as normal - there is a clickable part before the bactick, and after. Someone from SO offered this guess: \" probably backticks are not part of the regex that matches the filepath, or backticks are not in the w regex set\"\nI can reproduce this, it is correctly validating the link, but it then recognizes it as a folder: ! I guess nothing happens if Test doesn't exist? As for the fallback \"word links\", you can support this type of link if you remove backtick from word separators:\nThis falls apart when converting the link to a URI in !\nAnd that happens because backticks aren't allowed in the local path regex:\nThanks"}
{"_id":"doc-en-vscode-8c0e91139d2c33890d8fab91f46ac2b428a34fa49e9d4e2ab92cc3b156e5cb4b","title":"","text":"VSCode Version: 1.19.0 OS Version: openSUSE Tumbleweed (Linux) Steps to Reproduce: \"version\" in no entry, but should As enhancement it would be nice if the show-all-command would have an entry to display the Visual Studio Code version (I know that I could pass the flag on the command-line). Such a feature exists e.g. in Emacs, where it is possible to enter and get the current Emacs version displayed.\nI'd like to take this up. Any pointers? far I though to extend the current to execute and show the result? Or is there a place where we store version globally? we open up the and show the result of there?\nNo need to execute , that information can be obtained by requiring . Output is good, but a is better.\nraised a PR . Currently in the action under . Any better section to add this in?"}
{"_id":"doc-en-vscode-9d7a282551717914d25f66f8027cd22c9f2edadc3f0a211353aaf2e791b48742","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.45.1 OS Version: macOS Mojave 10.14.6 Steps to Reproduce: Recently, the autocomplete auto import text has become cropped so short as to be completely useless. In many cases over the past couple of days, I've had the option to import 2-4 same-named classes/components. Normally, I can select the correct one based on the additional text Now the package name is cropped to the same text for all options: This makes the auto-import feature completely useless (one of the most important features in an IDE). See example: ! project with React Native + React Navigation in 'Header' and cmd+space to open autocomplete (It may be that additional conditions apply, for example what the other options presented in the list would be.) <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes\nPress <kbdCmd</kbd+<kbdSpace</kbdagain to see detailed info for suggestions. Changing width of suggestions box already tracked in and\nOr just remove this completely useless feature which was around January 2020.\nAgree with and others, that current implementation does not make too much sense, and adding this description to the same line as a code snippet makes snippet name and description basically useless.\nSince was closed (with reference to some obscure problem from 2017), this ticket is the closest I could come to the new issue. The suggestion-list window is just too narrow for the info that is being presented there. Introducing a new feature to put more information there made the situation worse. It's simple not enough room to read the fully qualified names many are using. Snippets for PS/XML and other languages. (Sample image included) Alternative fixes : Allow us to turn OFF the 'descriptions' in column 2 of the snippet list. Allow people to customize the width of the list-window. Roll back the feature that gave us 2 columns instead of just 1 column (with the name only). I sincerely need a workaround or fix. Very frustrating. !"}
{"_id":"doc-en-vscode-2d4204c18e20ac11ad0cb1b9ae6efd198c6cbcf0949e7e7c44ea9e5089b70142","title":"","text":"The autocomplete just got even more useless! Here are the options that I see (cycling using ctrl+space): <img width=\"735\" alt=\"Screenshot 2020-06-25 at 10 54 43\" src=\"https://user-\"<img width=\"761\" alt=\"Screenshot 2020-06-25 at 10 55 24\" src=\"https://user-\"Previously the detail screen showed details on the option, not details why this suggestion is being provided. What's the use of that to a developer? Have I unknowingly enabled some obscure autocomplete-debugging mode or something?\nYes. Press to toggle the suggest debug mode\ndoes nothing. I couldn't find anything similar in the command palette either nor by googling.\nAha! To en/disable the suggest debug mode (on European keyboard): the suggestion dialog (i.e. the same hotkey as for (un)commenting a line) (I figured it out by seeing that on a US keyboard is .) So if you accidentally have the suggestions open when trying to comment out a line, you'll end up with a useless autocomplete system and no idea how to make it work again. May I suggest that a functionality that is useful to 0.01% of users and causes a lot of confusion for 99.99% if accidentally used would not have a default hotkey? It could be clearly labeled as \"Toggle autocomplete suggest debug mode\" or similar in the command palette, which would be just as easy for suggestion debugging cases but users would not accidentally trigger it.\nJust a heads-up that for the suggest widget and its details parts are going to be resizable. See This issue will all about the specific TS case and how to improve the situation for auto-importing modules\nThere might be more to it and things I don't know esp. the doc for says it might not be human readable but it kinda looks nice already. <img width=\"584\" alt=\"Screenshot 2020-11-02 at 12 37 51\" src=\"https://user-\""}
{"_id":"doc-en-vscode-a8d92f996e6c09a8287ab34a77079e950e66d1585ed3705b51ceeef567a83a52","title":"","text":"similar to the 'open in data viewer' command from the jupyter variables view, we should allow extensions to contribute a view for a variables for specific language/types on the built in tree view.\nThat would essentially be number 6 in this screenshot from the Julia extension , right?\nSame idea, either with an icon like that or through the context menu on that item. In the jupyter extension's case, we are planning to allow contributions for data viewers from extensions that didn't necessarily provide the variables.\nCool. I think icon and context menu would be great ;) It might also make sense to think a bit ahead there: there are probably some kind of viewers that are relevant for a lot of languages (tabular data, plots) and others that are very language specific. From our end, we would love it if VS Code at some point gained a generic tabular viewer, and a plot viewer that we could utilize and get rid of our custom implementations as well :) So maybe some kind of API design where in the future one could hook into a as of yet not existing generic tabular/plot viewer.\nExample implementation here The menu contribution point name will likely change names when we expand the usage beyond notebooks, but any needed migration will be pretty straight forward.\nI created a sample extension and cannot see any variables showing up. Do I need to enable something in VS Code for this feature to work. I couldn't find any settings/experiments that had to be enabled for this.\nFound that I needed to enable a VS Code experiment (I was looking for Jupyter experiments). Not sure the API is useful in its current form. Currently when i create a viewer for DataFrames, lists, objects, anything I do not get the name of the variable. Hence I have no idea what I'm looking at, all I get is a value, and thats basically a stringified version of the value. Thus its not really possible to create any viewer, unless of course I'm missing something here.\nMoving to March for now. Feel free to move back to February if the issue has been verified or once steps are provided and the issue is verifiable in the latest Insiders. Edit: I confirmed using the sample extension that the viewer does not report the variable name."}
{"_id":"doc-en-vscode-a1d1f9f4837f5dd49bf004398c48df4ebf1f8d511a9b1a2cd0095c75735b7344","title":"","text":"verification: sure you have jupyter and python extensions enabled the notebook variable view the or create your own some notebook cells that create variables and open the debug sidebar to see the notebook variable view that the command contributed to the menu is provided information about the variable ( should be the expression that would access that variable)"}
{"_id":"doc-en-vscode-cdc44542ec38db777a9f2a68680e3e85de308b103b6e7a5a68995dce5f1f9342","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: Stable & Insiders OS Version: Any Steps to Reproduce: Jupyter extension the following cell against a Python kernel The expected output is fairly simple: ( can appear some where in the middle) However VS Codes output is incorrect We are missing the line breaks (looks like this PR seems to have broken that) (using blockinline-block` fixes this). We have duplicate entries (as you can see below, we have multiple times) i think this has to do with the fact that we're trying to re-use the previous output and append the new stuff into the previous output element Note: This can be repro'ed in stable as well, but if i were to remove the code that tries to append to the previous element, then it looks as expected. Not sure when this part of the code broken though. Didn't go through the change log to determine this, you might have a better idea.\nnote that this is from the test, so that could probably be modified to check for the fix\nIt looks like this was briefly solved (at least for me as I reported in ) but the problem is now reappearing.\nI agree, I can replicate the same bug again, hence reopening this issue. Looks like subsequent change regressed the fix.\nseems to be a worse regression now (only in insiders) and repros without needing to do anything with background threads (unless I just have some bad setting): !\nI'm seeing something similar with the JavaScript/tslab notebook I use for teaching async and promises. Previously, the output was inconsistent across runs with sometimes missing or duplicated output."}
{"_id":"doc-en-vscode-ad73b091b70fcdc6a32dc644f422491b57fbda926436db4e715fe3ac37136a65","title":"","text":"Now all the expected output is present but with line breaks appearing only after some async code (e.g. ) is executed. <img width=\"661\" alt=\"image\" src=\"https://user-\"\nI reproduced output shown by above with missing new lines in code-insiders at ~14:00 UTC +1 (CET). In a code-insiders session that has been running last 24 hours. Restarting code-insiders approximately an hour later (15.00 UTC+1), the output is now shown with correct line breaks, see screenshot below. Version: 1.75.0-insider (user setup) Commit: Date: 2023-01-09T05:29:27.679Z Electron: 19.1.8 Chromium: 102.0.5005.167 16.14.2 V8: 10.2.154.15-electron.0 OS: Windows_NT x64 10.0. Sandboxed: Yes Output after code-insiders restart: !\nconfirmed that this no longer seems to reproduce on the latest.\nAdding label as confirmed above."}
{"_id":"doc-en-vscode-3154c6d6dd0cf14e6af909f82d5f3dd1f68764d14a28c74d38a0257357fd08b4","title":"","text":"From ! It took me a while to be able to use Copilot again. Here are the steps I took (also captured in the attached video): Reloaded VS Code Clicked \"Show Extension\" from the toast above, I'm brought to Copilot Chat extension in Extensions view Clicked \"Uninstall\" for Chat extension, nothing would happen Decided to instead search Copilot in Extensions view (thought maybe I was having issues by trying to uninstall Chat without touching Completions, though I would've expected a toast telling me that) After searching, the Copilot extensions automatically start to update Need to restart just Copilot (not chat), so I click to restart I try to uninstall Copilot Chat again, but nothing happens again Reload VS Code again Search for Copilot again in the Extensions view I can now uninstall Copilot (though I initiated the uninstall from the Copilot extension rather than Chat)\nVerification steps are in"}
{"_id":"doc-en-vscode-51dd697bc6c9580141c60be66c9c3a88705001a673f6e6f863537fd2f797c1f5","title":"","text":"PR introduced a max-height on comments, which goes against my team's use case when using the . Since our tours have animated GIFs that support our hands-on coding labs, the reduced max-height makes it impossible for a user to see entire images in the comment frame at once, which imposes a negative learner experience. (See: for comparison across versions). Proposed feature is to allow a user to set a max-height rather than imposing a max-height.\nI also think it should be settable through the API. My original on the PR has a comment exchange with about this at the place where I originally hardwired it.\nGenerally we try not to add API unless we see a clear case for having it. At the time suggested the max height I wasn't aware of a use case that would require a comment peek view to have a larger size. your case makes sense, and makes the case for having API for this. I would make this API rather than user configurable.\nYes, making the max-height configurable via the API makes perfect sense, since any extension can be configured with that functionality to set a max-height attribute that makes sense for the use case. Would the default behavior go back to expanding the height with the content?\nI'm not sure yet which would be default. I will discuss with our UX folks and see what they prefer.\nMy humble preference would be to adopt the old behavior as the default. Though the CodeTour extension is open source, it’s only seen one update in the past year. My team has a stale PR from 2021 on the CodeTour repo that was never commented on, so I’m afraid that even if we opened a PR with a fix for the max-height problem, it might not ever get accepted and merged. On the other hand, my team has been discussing forking CodeTour so that we can control the addition of new functionality and bug fixes.\nThere wasn't enough interest in adding API for this, so I the user setting .\nthanks! Is a boolean? If so, is the default (maintaining the as by default) or (no defined?\nIt is a boolean and the default is true, meaning the max height of a comment will be 20em.\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected."}
{"_id":"doc-en-vscode-063a7782725ddc73d9fd61dae09e1c0aa346c2ed02b3e54e1fd25e1f1eaa7536","title":"","text":"If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\n/verified Works great! Thanks!\nThanks for verifying!"}
{"_id":"doc-en-vscode-797ffb3fd4552c97d78bbe592276093ed1b7c78a9679614ecf2c3e4151580d61","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.72.2 OS Version: Windows 11 / MacOS Monterey Steps to Reproduce: . the icons using the command. that the arrows are still not hidden. The vscode-icons extension has a setting called , which, as the name suggests, hides the explorer arrows next to the project folders. This look something like this: ! With the introduction of the tree mode in the search panel, I would not have expected to see any explorer arrows there. Alas, the panel has the search arrows, regardless of the extension setting. ! I already opened an issue on the vscode-icons repository, which you can find .\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.72.2. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!\nStill happens on 1.72.2!"}
{"_id":"doc-en-vscode-521c260803cc312a2a7aaa150958ee72dee42fc063f7b5e831258ba15f370cc8","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No, this is an extension-related issue that arises across extensions with walkthroughs <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.91.1 OS Version: Windows 11 Extension walkthroughs are not launching upon install. Expected behavior: walkthrough launches automatically upon install. Tested on C# Dev Kit extension and Python extension. Seemed to work in 1.90.0 but does not work in 1.91.1. Steps to Reproduce [Windows Sandbox]: a Windows Sandbox Edge and download VS Code (stable) from the VS Code website the C# Dev Kit extension. Note: Upon install the walkthrough is supposed to launch automatically. This is not happening, but I can manually open the walkthrough using these steps. command palette. command 'Welcome: Open Walkthrough' 'Get Started with C# Dev Kit' ! steps 3-7 with the Python extension. Same behavior. Steps to Reproduce [VS Code stable and VS Code insider]: VS Code C# Dev Kit (if you have it installed) VS Code C# Dev Kit extension. Note: Upon install the walkthrough is supposed to launch automatically. This is not happening, but I can manually open the walkthrough using these steps. command palette. command 'Welcome: Open Walkthrough' 'Get Started with C# Dev Kit' steps 2-7 with the Python extension. Same behavior.\nclosedWith\nusing the steps above, I am not seeing the walkthrough open when I reinstall the extension\nWas just talking with about this- if you've had it installed in the past, it won't work, but for me it works in a window with a clean user-data-dir"}
{"_id":"doc-en-vscode-047be27a5e07a7e4b849b24d302f1d7fa11829e0b0de8ed4d39541fe525f13b1","title":"","text":"To prevent issues like this: and like this and many more\nthis is a good idea, I've been getting tagged in a lot of these recently. it's difficult to say what exactly the limit/requirements should be though. atm we require title and description (the rest of the info is fluff that is auto included). Maybe force description to be longer than 10 characters?\nWe can start with 10characters enforced on both title & body and see how issues keep coming. We could also maybe crowd-source ideas from the team"}
{"_id":"doc-en-vscode-fe974120fba056a0a8e0742f56a3bccc6c14f0afd128ac1b5574929dec28a076","title":"","text":"While testing . VSCode Version: Code - Insiders 1.15.0-insider (, 2017-08-01T05:07:59.267Z) OS Version: Darwin x64 16.7.0 Steps to Reproduce: 3 terminals from the command palette <kbdArrowDown</kbdto the last terminal in the switcher <kbdTab</kbdto rename the terminal -keyboard focus disappears from the input box (and seems to be nowhere). I need a second <kbdTab</kbdto move the keyboard focus to the rename action in the selected row. A single <kbdTab</kbdshould suffice.\nThis is also an issue with task configuration"}
{"_id":"doc-en-vscode-4b58c7d46ae03a652691c43206849bdb8d9be78a8d47c28b388731afb145df43","title":"","text":"Screen reader and low vision (or other accessibility) users often use our menu items to discover things. We should include a menu item to open the getting started walkthrough under .\nand WDYT?\nSeems reasonable to me. The title suggests that it could also be a walkthrough, or part of an existing walkthrough. The tradeoff there would be discoverability.\nI see an error in the console when opening accessible view from that welcome experience:\nCc\nRelated:"}
{"_id":"doc-en-vscode-6271f4841c6d7d4d6d8b5c978e2b187c82986824ec45541c3eab5ceae6cee0e5","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: OS Version: Steps to Reproduce: a ntoebook click on a cell, and notice the order of the cell actions same entries for regular editors is All places the order is We might want to consider changing this.\nis this a papercut or ux issue I'm concerned that changing this now could result in issues due to muscle memory (i.e. users being used to this order). Thus changing this might not be a good idea, then again, i don't think it is. Thoughts?\ndiscussed with team, Lets change this"}
{"_id":"doc-en-vscode-fc132de78ef3fa2ad509ceeb8fbc6116b1f1507afce25e87a207a128fc008151","title":"","text":"Testing Show extension display name instead of extension id. You can also make them links to open the extension editor. <img width=\"522\" alt=\"image\" src=\"https://user-\"\nwhat's the right way to do this? I had a look and could only see async methods to do this which seemed odd to me (and would require some refactoring to get the tooltip to work with async).\n1 I see this as well\nThere are multiple ways depending upon the source of this extension id. Since it is a running extension, I would prefer to use this api since you do not want to have async\nthanks will use that. You don't have an id -name map to avoid iterating over it multiple times though?\nThere exists one, but it is async"}
{"_id":"doc-en-vscode-a3abf23560c2ae6bb83c5fbfcde1a6b3c439f4eb0fd36d6f5cdb49c9c529d738","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: OS Version: Steps to Reproduce: type and enter that issue reporter is opened with correct extension selected. !"}
{"_id":"doc-en-vscode-738a226cd3236fa34e2a5853d1f2134dca01eebefdfb1aecea9c204c4f4bf4c5","title":"","text":"/gifPlease\nThanks for reporting this issue! Unfortunately, it's hard for us to understand what issue you're seeing. Please help us out by providing a screen recording showing exactly what isn't working as expected. While we can work with most standard formats, files are preferred as they are displayed inline on GitHub. You may find helpful as a browser-based gif recording tool. If the issue depends on keyboard input, you can help us by enabling screencast mode for the recording ( in the command palette). Lastly, please attach this file via the GitHub web interface as emailed responses will strip files out from the issue. Happy coding!\nThe terminal's usage of may have regressed?\nThe issue I'm seeing is no longer seems to flip the terminal link modifiers and column select modifier. Additionally the cursor isn't showing anymore when using column selection.\nActually I don't think it ever flipped the column select modifier\nWhen is set, alt+click still opens fine. After the crosshair will be working as intended, the link pointer cursor will override it correctly.\n! I can still repro if . Is there something else I should set? It works fine if I set .\nSince verification steps are needed and is OOF, this issue will be moved to the February 2024 milestone.\nin case you know the verifications steps.\nI verified that regardless of the value of , I can open links in the terminal with ctrlCmd and holding alt allows me to do column selection, which I believe is by design.\nIt turns out that the bug has nothing to do with the cross cursor. The real cause was extension host keep crashing after a few minutes after opening a big TypeScript monorepo."}
{"_id":"doc-en-vscode-ae15ef45aaa43cee51ae3159cb5bc1e3b8f7297c6317b01882fff4943915cc18","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.72.2 OS Version: macOS Monterey 12.5 Steps to Reproduce: custom HISTFILE (example custom HISTFILE per project) integrated terminal (zsh) in VS code will be /Users//.zsh_history instead of custom history file What i use: vscode I found that the changes made to break the ability to use a custom HISTFILE because the HISTFILE is always overwritten by these changes:\nTrying to set HISTFILE directly in settings, fails the same way HISTFILE is always set to\nHi I have opened a PR with a possible solution to this issue. Let me know, if it solves the problem.\nthanks a lot Sorry for the stupid question, but do you have any macos silicon build that I could try ? I never contributed to vscode and finding how to compile it from source may take some time\nUnfortunately, I don't have a Mac, however, I use Linux with Zsh so I could reproduce the bug and I could test the fix on it.\nI think this might be related to with McFly."}
{"_id":"doc-en-vscode-f0fb1d493521e0db7ffa5001814ba9bc344f8efa435abce702169d6dcc3ec5f3","title":"","text":"Type: <bBug</bTo reproduce: VS Code a folder and choose a git repo with changes a file with changes and notice the gutter is shown a folder to the workspace and choose a git repo the gutter is still visible close the file and reopen it Expected behavior: the gutter is shown Actual behavior: the gutter is not shown this may be related to . actually pretty much the same issue. However a reload (I assume that is ) doesn't help in this case. VS Code version: Code - Insiders 1.77.0-insider (, 2023-03-10T05:24:23.805Z) OS version: WindowsNT x64 10.0. Modes: Sandboxed: Yes <details<summarySystem Info</summary </detailsExtensions: none<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nThanks for the @ mention and for the repro steps!\nAlso having the same issue. Installing a previous version () helped to solve it (temporarily, until it's fixed).\nFolks who are seeing this issue: Could you please install the latest insiders build and confirm that you no longer see the issues there?\nYes. The issue resolved in the latest insider version.\nRelated Stack Overflow questions: I'd prefer using this one as a duplicate target on Stack Overflow since it's asked much cleaner.\nThanks for confirming that the issue is fixed in insiders!\nWow, so it was a bug, not an actual symlink-issue as I thought. So happy now, thanks! Context: I'll try the fix/update.\nThis issue seems to be fixed in version 1.76.2 for git repositories, however it is not yet fixed for Mercurial ( extension).\nI am still seeing this bug in\ncan you share some exact repro steps that we can use to reproduce the issue?\nI can't reproduce the issue on a new clean git repo, hmm\nIm having the same issues, reinstalling helped fix the current file having the issue but any file created after wont show the indicators. Disabling extensions wont work either and Im not sure on how to reproduce the steps Version: 1.76.2 (user setup) OS: Windows_NT x64 10.0."}
{"_id":"doc-en-vscode-841bba6c52980e6841b3d9ca7c1211718c085d6cae9e7df7acfd09667164e4da","title":"","text":"Testing If there's no error case, I now need to return something like:\nHi we have branched off from today, so I am pinging about the milestone here. Would you like to make this issue a candidate or move it out to the milestone?\nMoving to March for now. Feel free to move this issue back to February if it turns out to be a candidate.\nIt's already a , which allows 'undefined' but not void. Is it weird to say ?\n+1 on this, I have to explicitly today"}
{"_id":"doc-en-vscode-e5df8cbd8418b9909a4f3923ff955bd5084755308d455dd9da8e57635c87b134","title":"","text":"<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.39.0 OS Version: Linux RHEL 7 Steps to Reproduce: code with --no-proxy-server flag \"Option 'proxy-server' is unknown. Ignoring.\" error appears. I think the proxy is still disabled, the error message is simply wrong? If I run without the flag, I get prompted to log into the proxy, if I run with the flag, the proxy appears disabled (no log in prompting) but the error appears on the command line.\nI suggest to include that fix in the recovery builds to avoid any confusion due to the message printed by the cli. The cli says for all chromium options, but in fact, they are used. My understanding is that most of the chromium options are not for the end-user except and the proxy settings, so I these to the list of the valid (supported) args. Please speak up if I missed one of the options. PR is\nThis is not fixed in the latest stable release candidate. The reason is that the PR was actually not merged and this issue was closed erroneously\nThe discussion in yesterday's standup was to take this, so we've now merged the PR..."}
{"_id":"doc-en-vscode-12292fc211916cd321219bbc6f256940d22c15810d3e151e13fd7518dacf8217","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.67.0 OS Version:mac os monterey 12.3.1 Steps to Reproduce: to settings (command+,) for markdown drop <img width=\"719\" alt=\"Screen Shot 2022-05-20 at 11 55 58 AM\" src=\"https://user-\"\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.67.2. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!\nConfirmed on latest insiders Version: 1.68.0-insider Commit: Date: 2022-05-16T05:30:27.563Z Electron: 17.4.3 Chromium: 98.0.4758.141 16.13.0 V8: 9.8.177.13-electron.0 OS: Darwin arm64 21.4.0 <img width=\"636\" alt=\"Screen Shot 2022-05-20 at 1 11 00 PM\" src=\"https://user-\"\nFixed as of\n<img width=\"967\" alt=\"image\" src=\"https://user-\"\nHey I see with the same fixes, already merged.\nYeah but it accidentally got reverted via This just brings your changes back in"}
{"_id":"doc-en-vscode-91672e583f499a939a01f7ca17a7801ac68987c516b906bdf6f4c353abaa50eb","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --<!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: version 1.76 OS Version: Windows 22H2 Steps to Reproduce: a debugging session, and click on the \"Edit Condition...\" button (the pencil). Ensure that the cursor is in a window with a short width. ! appears. ! window. The text appears to be cut off. ! to look like this: !\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.76.0. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!\nProbably we need to relayout that breakpoint editor when the parent editor width changes\nI'll get it done\nMy phone is not working unless connected to wifi I need serious help as I am to be in GA at 7am I realize this phone is outdated a Motorola it is used but my Apple I cannot see the screen on it and I obviously can't carry my pro on the road cuz it is not activated can someone help me you can record everything l do not care I need help\nAll kids are grown? The Samsung used to belong to my son now belongs to coworker On Sun, Apr 9, 2023, 8:29 PM Amy Scarbrough"}
{"_id":"doc-en-vscode-75a40935178775d1158d99cb17c8b89d0bc7b9e2e7640a075d0e70ae0d6998d3","title":"","text":"Running Extensions context menu -in extensions have disabled menu for !\nI don't remember creating the \"Report Issue\" action, so I don't think I own it. Maybe you did this?\nYeah I did this. Code pointer The issue here is that the JavaScript Debugger does not seem to provide the repository URI in the Thus the action gets disabled here can you maybe provide the repository information of the JavaScript debugger so this button gets enabled here?\nThere are multiple built-in extensions with disabled menu item, like, , , ... (Not only js debugger).\nYeah none of them have a seperate repository. So you shuold just file in vscode repository However for js-debug we have a sepearte repository so it makes sense to have that action enabled.\nI have an repository in the js-debug I think the issue is actually the line above that exclude built-in extensions: I don't think this line is necessary. Deleting it will just have people who open issues on extensions in the vscode repo open them here, which is desirable.\nYeah, agree.\nReport Issue on opens this repository . Some other built-in extensions still have disabled menu item.\nThis action is available only for few builtin extensions (jsdebug, emmet) and this action reports issues in VS Code repository. IMO it shall be consistent and show this action for all builtin extensions. If repo is not found and extension is builtin we shall enable the action and create issue in VS Code repo. Let me know if you agree, if so shall I reopen this issue?\nWorks for me and let's reopen. I am opening this for community contributions and adding help wanted\nI think there are two ways to do it. attributes need to be to the of each built-in extension In the , judge the extension type. If it is a built-in extension, set the base url to I think 2 is better\nYeah let's go with 2. Sounds like a good approach to me\nI have submitted a PR, please help CR\nthanks to your PR this is now fixed.\nFixed for built-in extensions, though just a note that it's still disabled for the github.codespaces extension."}
{"_id":"doc-en-vscode-1c287b7a7b38039a8a9cc161cdd1bc2fb074a111a64ace8062d879c0f3e72392","title":"","text":"<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.34.0-insider OS Version: All Steps to Reproduce: This issue is difficult to reproduce as it requires a special version of an icon theme extension that supports theme reloading ().\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.592 --- <!-- score: 0.528 --- <!-- score: 0.482 --- <!-- score: 0.45 --<!-- potentialduplicatescomment --\nBeing a maintainer of the extension, I stumbled on this issue while testing the implementation of this feature on our extension. As I'm the author of the theme reloading feature, I pinpointed the cause of the bug. The reason for it is that on we don't call the event, resulting in the not getting notified. The solution is to add a call to on the function. Although much of the code execution of the is overkill when we reload the icon theme, it has no side effects. Possible actions are: and extract the calls to telemetry and event triggering to separate function and then use only the extracted function nothing and just add the call to and don't care about the redundant code execution since it does not have any side effects. I'm more than willing to provide a PR according to the chosen action, although I suspect that you guys prefer action 2. Awaiting your response.\nI like the suggestion of calling . PR is of course very welcome!"}
{"_id":"doc-en-vscode-c3810b1fbb7be95b63f5532526d29ee498d6534d8b893998c52a6744be6c6738","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. -- (merged as ) and other colors relating to the foreground of icons in the activity bar when it is configured to be at the top of the sidebar. I'd like colors for the activity bar background when it is at the top. Specifically, I'd like these new colors: to set the background of inactive icons and the region of the top activity bar not occupied by icons, corresponding to . to set the background of the active icon in the top activity bar, corresponding to . See also , .\n<!-- -96bd-47a8-a0e8- --This feature request is now a candidate for our backlog. The community has 60 days to the issue. If it receives 20 upvotes we will move it to our backlog. If not, we will close it. To learn more about how we handle feature requests, please see our . Happy Coding!\n! vscode 1.86 Windown 10\nWe two new color. One for the activity bar background when it is set to the top / bottom () and one for the border between the activity bar when set to top / bottom and the views ()\nAdded\nVerification steps: activity bar location to the top the following colors in of your file and see the color change in the activity bar at the top : Background of the activity bar : Border color between activity bar and the view container : Background color of active activity item\nThank you very much for your efforts! What needs to be is: Can you add the following color setting items: ! This color setting project has been a long time coming, but no one has ever made it happen.\nCan you create a separate issue for this? These are colors that we don't yet have for the default activity bar, right? In that case, please also include a very short description for each color in the issue.\nNo need at the moment, thank you very much!"}
{"_id":"doc-en-vscode-23255f2667477718cb2a87316316a3108c5a5e942d5eb0c48a9f26b7c755902f","title":"","text":"Due to how colors are blending together. Notice the color of the input and the selection of quick open and how the black background blends into the viewlet, so the input looks misaligned. <img width=\"1035\" alt=\"screen shot 2017-06-27 at 15 23 22\" src=\"https://user-\"\nHello, I'd like to help with this! UPDATE: I believe I've fixed the issue by changing the \"\" property in the monokai theme from \"\" to \"\". Before the change: ! After the change: ! Shall I submit a pull request? Thank you!\nYes, a PR would be great!"}
{"_id":"doc-en-vscode-4d171807d049adad47c7c3098c979c5853a85f2c581164f16b802d95b615517c","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.78.0 OS Version: Windows 10 (x64) Steps to Reproduce: \"Centered Layout\". the editor by independently moving its sashes (holding down Alt). and restart VS Code. The position used to be remembered in previous versions, but starting with version 1.76, the position gets reset on every start. Only the width is being remembered. It's still broken in the current insider build of VS Code 1.78.0.\nDo we already know, which commit causes this regression?\nI looked through the commits for 1.76.0 and I suspect it was .\nYeah, it is solved for me."}
{"_id":"doc-en-vscode-93428c09150e33551d9e3374851265fddbaf87abd8d110f1f8b0ae0ec694836e","title":"","text":"This throws the exception below. This happens when logging objects without prototypes, e.g . Instead of the property access should be used"}
{"_id":"doc-en-vscode-0f3f8a29174bdaf5e15861fe865fb7e12a8c98d18c088b20129758b1a75d76ba","title":"","text":"I am forming an openEditorAtIndex command using the following syntax: '' + index 1-9 work correctly (and are default keybindings already). Would it be possible to add support for any index? I currently need to use this command if <= 9 and if 9, I have to go to index 9 and then next count - 9 times to get to the desired pane with a risk of wrapping around\ntrying to understand the scenario here, I do not think it makes sense for VSCode to introduce commands for more indices because this number is pretty much unlimited. Can we have another command that just goes to the next index, from the current one and you could use that instead? E.g. does it work if you use ?\nLines 55 - 89 are what I use to be able to go to any tab My case here is what I am trying to avoid Ideally I could do and never think twice about it The problem with using nextEditorInGroup is that I have no way of knowing the max editor number index, so if I say go to tab 10 and there are only 8 tabs, the current logic will try to go to tab 9 with the absolute exposed command, then use a nextEditorInGroup command.\nthat switch-case statement looks quite horrible, an alternative would be for the workbench to expose a command where you just pass the index as argument into. We already have that for other commands, e.g. see the \"move editor command\" () Would that be better?\nThat would work for me too! It is not very urgent since I don't think people navigate to tabs they can't see very often but for completeness it would be nice! Thanks!\nGreat. I would accept a reasonable small PR on this feature request.\nI'll try to write a reasonable small PR for this :)\nlanded, thanks to\nSome steps so I do not have to read the whole discussion? What command with what arguments should do what now?\nhere you go: write an extension that executes and verify the tab at that index opens.\nWorks nicely. Though the counting starts from - might be a bit confusing.\nAh yeah good point, but I think that is fine."}
{"_id":"doc-en-vscode-b1fc0102325ba333dc837067c677cc3056dc70ff79ead6cfbcd71a446b184808","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.63.2 OS Version: macOS 12.1 (ARM64) I can't get more than two launch configurations to run at the same time. I can run the rest of the configurations afterwards and everything works (debugging, etc.). StopAll will also only close the two configs that were successfully launched. I haven't found any mention in the documentation that this is limited to two launch configurations. I've only seen that two are used in every explanation or tutorial about this feature. Steps to Reproduce: multiple launch configurations in multi-root workspace inside of project roots. launch configurations to compounds in .code-workspace file debugging two of the many compound configurations will run Examples:\nVS Code imposes no limit on number of launch configurations in compound config. I've verified this with this simple example project (which only relies on built-in VS Code features and not third-party extensions): unpack zip file in VS Code use \"Open Workspace from file\" to open workspace file inside folder switch to debug viewlet in debug config dropdown select and run \"4 Debug Session\" Observe: 4 debug sessions: <img width=\"1384\" alt=\"CleanShot 2022-01-26 at 11 45 23 src=\"https://user-\"Maybe there is a limit on dotnet/coreclr?\nThank you for taking a look at this After inspecting your configuration and some research in my own project I was able to isolate the culprit. The problem arises when you try to make use of inputs in the launch configuration."}
{"_id":"doc-en-vscode-eaf0a1bdde8ff0c769fa60ebfe7d797899fbff5a5dcd4bba4ae329ec41d31d18","title":"","text":"I created a demo for download, the process is the same as you described above.\nthanks for the additional info. I was able to create an even smaller repro without a multi-folder workspace: unpack zip file in VS Code open the \"compound\" folder switch to debug viewlet in debug config dropdown select and run \"Compound\" Observe: only one debug session starts\nIs it possible to implement the input that allows blocking another input request until the first input has completed? Latter quick pick also cancels former ones so it needs to be fixed as well.\nDo you own the configuration resolver Is my solution correct or should the service guarantee correct behavior when calling into it multiple times simultaneously?\nyour solution looks good. The configuration resolver service doesn't currently make any guarantees about what will happen if it's called into multiple time simultaneously. I would also be fine with the change being in the configuration resolver service though. We can change it's behavior so that it does make that guarantee.\nYes, making the fix in the configuration resolver service sounds better.\nThis fix doesn't fully cover our scenario (tested in latest insiders). This fixes case, and yet we also prompt for input or quick pick in , which is not covered in this PR. Is it appropriate to wait for as well?\nGood point. Yeah, I think that makes sense.\nI guess we'll need Queue's for and Would you like to make the changes, or should I work on this?\nI can take care of it\nThis gets a bit weird. Since one launch config can be calling into the variable resolver (the first fix) while the second is calling the extension config provider, we also need a queue that orchestrates those two things. Easiest fix is to start launch configs one at a time instead of kicking them off in parallel: and I'm not sure what other impacts that could have. Perf to some extent, probably. So I wonder whether we would ever consider queueing calls into the quick pick control so that invoking the quick pick while it is already visible would wait instead of cancelling the first one as it does currently. Example of what I'm talking about (but in my real situation it's an extension competing with code in core): This extension code only shows one input box, while we could have the second wait for the first to complete. This could possibly be a breaking change for some extensions though, so it could be enabled with a flag."}
{"_id":"doc-en-vscode-2ea0124283e5818b1640548f929c5ebc1ea618600c933557411d0590c49a1bbd","title":"","text":"As an easy workaround, I can put a queue around calling into the vscode variable resolver and calling an extension debug config provider. But, putting a queue around two unrelated pieces of code seems to say that I'm solving the problem at the wrong level.\nOr maybe make them the same level by moving the resolve queue to\nJust to be clear. The two callers of the are in two different places right (core and extension)?\nRight\nOk I've not heard of any feature request like this but feel free to open one... I wonder if maybe instead Core could have concept of \"important\" quick picks that can't be stomped over by extensions and instead the extension calls wait until Core is done. Then we don't need any more API\nIt can also go the other way. The problem is just that there are two async processes that have to run and show a quick input, and we need both of them to complete and not to get canceled by the other. Filed I'll go with a workaround for now, so I'm not blocked, just figured I'd start the discussion\nSorry for the late reply. I tested this today and it works for our scenario. Thanks for the quick fixes!"}
{"_id":"doc-en-vscode-96565f4ee8929f6ee448467c763ab036c3d16c790772d64fe5eccb9f89f7e873","title":"","text":"Issue Type: <bBug</bThe description for is copied from VS Code version: Code 1.64.2 (, 2022-02-09T22:02:28.252Z) OS version: Windows_NT x64 10.0. Restricted Mode: No Remote OS version: Linux x64 5.4.0-100-generic <details<summaryExtensions (15)</summaryExtensionVersion ------ jupyter-keymap1.0.0 remote-ssh0.74.0 remote-ssh-edit0.74.0 vscode-icons11.10.0 gitlens11.7.0 cmake-language-support-vscode0.0.4 git-graph1.30.0 vscode-dotnet-runtime1.5.0 python2022.0. vscode-pylance2022.2.3 jupyter2022.1. jupyter-keymap1.0.0 jupyter-renderers1.0.6 cmake-tools1.9.2 cpptools1.8.4 </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --"}
{"_id":"doc-en-vscode-89f95048f395dae9f4aa5b13feadbfb602ec986d185a48d59540e7458dbaa941","title":"","text":"Track picking up\nTo verify, just open a TS file and click on the language status to check that the bundled TS version is 5.1.6"}
{"_id":"doc-en-vscode-b94efc6e439941a39369dcff95c68ea4d272bc7c5b48aa3369e677b694f91c85","title":"","text":"in VS Code stable and VS Code insiders all cells in each quickly up and down through the notebook :bug: in VS Code insiders scrolling over notebook outputs is not smooth anymore; scrolling appears to stutter and halt over notebook outputs I took performance profiles in stable and insiders but devtools won't let me save them unfortunately. Nothing stood out though in terms of memory usage or a particular function call increasing in percentage across profiles. LMK if I can provide more info\nIs this with a trackpad? I might be seeing this. Could be a regression from\nI've the same issue with a trackpad on Windows.\nYeah, trackpad on Windows for me as well"}
{"_id":"doc-en-vscode-dcbb752dcef949bb324e6994ecb1b58fed66f30b65ddf33b8ccfe3b3b4a464f3","title":"","text":"Variable substitutions that are available for (see ) do not include currently selected text. I am looking for something similar to $(CurText) that is available in Visual Studio Arguments for External Tools (see ). Please add a variable ${curtext} which would be replaced with the currently selected text in the editor window. If no selection is available, it could use the current line at cursor position.\nNice idea, but I think it's too late for implement that in the current milestone...\nFixed by in\nVerifiers: what happens if all text in a 300 MB file is selected...\ngood point. Maybe we should put a limit on this. Or only to copy the howl line at most (even though the line could also be 300mb I am fine with that) thoughts?\nSelecting lots of text, at some point I get . If there's a predictable limit where this will fail, we should just cut the string off there. Unless only sending part of the text could be dangerous or unexpected for the user. In any case I think that sending multiple lines of text could be useful.\nthere is not a predictable limit since the variables are used in different scenarios. If somebody has an idea for a limit I am open to PRs"}
{"_id":"doc-en-vscode-ff83a646f55e9d93b65e72163fd4193e67f760834ba43523c5299c906c97d7b0","title":"","text":"Issue Type: <bBug</bPlotting a large plot (e.g seaborn pairplot) with many columns via jupyter interactive mode and then viewing in plots viewer is very slow to view/zoom/pan VS Code version: Code 1.45.1 (, 2020-05-14T08:27:35.169Z) OS version: WindowsNT x64 10.0. <details<summarySystem Info</summary </details<details<summaryExtensions (11)</summaryExtensionVersion ------ vscode-django0.20.0 vscode-markdownlint0.36.1 python-extension-pack1.6.0 gc-excelviewer2.1.35 r1.4.2 vsc-python-indent1.10.1 MagicPython1.1.0 anaconda-extension-pack1.0.1 python2020.6. vscode-yaml0.8.0 vscodeintellicode1.2.8 </details<!-- generated by issue reporter --\nThis sounds like a problem caused by the extension providing the Jupyter Notebooks. Please file the issue there.\nHi same problem here, are there any workarounds in place for this?\nI've moved to"}
{"_id":"doc-en-vscode-256d3e255c7365350e0bbbb9c2854a01a052ce7be4e7cdac653b674cdf81e659","title":"","text":"Following the discussion in Repro steps on Windows: use a monospace font, make sure ligatures are not used, etc. (i.e. default editor related settings) change the scale setting of the display the editor decorations / selection should render correctly (on top of the correct text) (the problem was that the font measurements done with a certain display scale were used across all display scales)."}
{"_id":"doc-en-vscode-b99b6babf90cc225822069dbe429bc14f5326d8a67c9fa9254d9513d7cf1fadc","title":"","text":"<!-- Do you have a question? Please ask it on --<!-- Use Help Report Issue to prefill these. --- VSCode Version: OS Version: Window 10 Steps to Reproduce: gif: !\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.562 --- <!-- score: 0.491 --<!-- potentialduplicatescomment --"}
{"_id":"doc-en-vscode-b3081962777acf4acb456b6ed337a494353189af5c6a561c25217f0df0631983","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.77.1 OS Version: All Steps to Reproduce: This is not a problem in the extension, the extension code is the same, but a commit changes the way webviews are loaded and it's breaking it. This commit: Has introduced a regression that makes my extension unusable. It introduces a problem on webviews in which workers are not available, and causes the following errors in the console: Uncaught (in promise) TypeError: Cannot read properties of undefined (reading 'active') at I have been tracking all changes since 1.76.2 release, building VSCode myself, and setting the build to the previous commit, my webview works flawlessly, if I checkout this commit, it's totally broken. I have not been able to send you are reproducible test case, but if interested, I can show you a video on how to use my extension, so you can check for yourself the issue. Thanks!\nTo add more information, check these two videos: Not working scenario, with the code at 1.77.1: ! Working scenario at the commit previous to the one I mention. Just remark that checking out the code at the commit I'm pointing, causes the not working scenario: !\nSorry you are running into this. Do you have an extension you can share that shows this issue? I can't simply revert the change because it was made to fix an hang with webviews opening. If I can repo this locally though, I can try to investigate\nThanks for checking this one!"}
{"_id":"doc-en-vscode-dd78ac4e0edf17acad34e3f957d59a73b018450997883514ee03935c10fc771d","title":"","text":"My extension is this one: Find attached a sample project: These are the steps: Install the extension Unzip in its own folder Open the folder with VSCode With selected, issue the command Toggle Design/Code This will open up the webview, and with the changes, you could see in the console the issues Find here a video with the steps on how to open the webview, and the behavior when its working: ! Thanks!\nNeedless to say that if I should perform any changes on the extension to make it work, it's not a problem at all. That is, if I can get an idea on what the changes are, have no problem on that. :-) Thanks!\nThanks for the repo steps I have a potential Please try the next VS Code insiders build that comes out and let me know if you still see the problem. With the fix, things seem to be loading correctly: <img width=\"1129\" alt=\"Screenshot 2023-04-12 at 4 06 24 PM\" src=\"https://user-\"\nChecked insiders, but not sure if the fix got through, as I see some checks failed: ! Also tried update from the repository, and it seems still is not merged. In any case, what I see in your shot is that the Tool Palette is missing content. The content is provided by loading an external file called \"\" and it seems the webview is not able to load it: ! Will keep checking, THANKS!\nignore my last message, I have merged your changes locally and works perfect! So once this is merged, consider the issue fixed, thanks a lot!! !\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\n/verified"}
{"_id":"doc-en-vscode-160cb85ad698695082c8536cb9049cf46a88bde3c511e14318413e9b94f7c622","title":"","text":"creates a tree view opens the tree view disposes the tree view while it's still open creates the tree view again The tree view will stay empty until it's is toggled Real world repo (also works as steps to verify): to a Codespace (to use a specific build of VS Code you can specify a commit in the URL with , ex ) the Remote Explorer the page. This will cause the Codespace to load with the remote explorer visible 'll see that the Codespace view in the remote explorer is empty\nupdate"}
{"_id":"doc-en-vscode-dfa21e7bd2e83c7ca4d29fe18c89b993c980c4c1b5fe6afaa82f00f30e5f92e7","title":"","text":"<img width=\"1792\" alt=\"image\" src=\"https://user-\"Hi, like visible on the screenshot, more than half of the terminal width in vertical mode is taken to display the path of the file where the log originated. I think this is not relevant to many of us since the path will point to the logger file At some point it gets really hard to read the logs Just to have an option to display path in cli would be enought for me. My screen resolution is 3072x1920 and my zoom level is -0.5 (at 0 it is even worst)\nThis has always been a bit off- I think only showing the file name would be sufficient\nTo be honest, who even needs the filename? When debugging I know what my messages mean and where they are generated. If I don't, a simple clickable elipsis will save space and perform the same function by bringing me directly to the line of code."}
{"_id":"doc-en-vscode-16f0e01d50bd28d771d38f6f79d66da50ced8e1e0bdd42d082d7db7a361ffa7e","title":"","text":"Tracks enabling the new drop into editor features () for code cells in notebooks Was disabled as we were incorrectly showing the drop into indicator while dragging cells within a notebook\nActually looks like may have the most experience in this area. I'll submit a PR shortly that may fix the issue but will need your help verifying it"}
{"_id":"doc-en-vscode-df9b44a01349b98d479da3f816a8784561a8cd42153d8efe9930a6bfe6b86574","title":"","text":"Type: <bBug</bSteps to reproduce: a fresh, vanilla VS Code Insiders installation without any extensions. and open a new folder. , open it, and save it with the following contents: Actual results: A warning is reported in Problems tab and via yellow squiggles at line 10, column 25 (= character of the line specifying the ): Expected results: No warning is reported. Additional information: The problem matcher has been copied directly from the (). VS Code should not display warnings for unmodified examples from the docs. This also occurs in the current non-insiders release 1.74.0. VS Code version: Code - Insiders 1.75.0-insider (, 2022-12-08T05:20:51.149Z) OS version: WindowsNT x64 10.0. Modes: Sandboxed: Yes <details<summarySystem Info</summary </detailsExtensions: none<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nI can repro this. probably caused by your changes to the schema - could you pls take a look?\nOf course. I'll submit the fix.\nThe PR is ready for review.\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\n/verified Thanks for the quick fix!"}
{"_id":"doc-en-vscode-83a544869f671138567d599d420167a82ffff4c41a679a0d96c1de3c9903ba84","title":"","text":"When TS is configured with both and set to , TypeScript generates sourcemap files ending in . These files are not treated as JSON by VS Code."}
{"_id":"doc-en-vscode-7b518f1656950412180ecd1a0386b88fc79bf567b4deb1f73b2b3ad2eabecb8f","title":"","text":"I see that settings sync is suspended because of invalid token and token is not getting refreshed / updated. Sync LogMicrosoft Account Log Did not work even after Reloading the window / opening new window. Restarting VS Code worked. any insights? Is there a cache on the extension?\nFrom the account log, the token gets refreshed before it expires. This should trigger an event which causes the new token to get set, but it seems like something about this is failing, since we then get into a loop of getting a 401 and trying to resolve the access token.\nEvery time we see a failure, we unset the token and send token failed request and we update the token by getting it from the account sessions. Looks like we are getting an invalid token from the account sessions. Can you please check if that can happen?\nFound the issue! The token was not being set in the shared process\nEasier steps to verify: in with an account to preferences sync out of the same account in to the account again Preferences sync should restart without any error."}
{"_id":"doc-en-vscode-8109276848a29a4a9788323235e93cacee32e209dd3cb7011ce7cb8ceffdddfc","title":"","text":"Text in TypeScript file: With the Light High Contrast theme, I don't see the unicode highlighting outline. !"}
{"_id":"doc-en-vscode-07884e2761a82f3f143f49e15648a8c5a136ee5c4757f0883a061bf98f536838","title":"","text":"Maybe this already exists, but it would be good if <kbdctrl</kbdrendered this in markdown: <img width=\"35\" alt=\"image\" src=\"https://user-\"We could then add nicely styled keybindings to various hovers: <img width=\"125\" alt=\"image\" src=\"https://user-\"cc"}
{"_id":"doc-en-vscode-c5953d52e7bf222f57104488698f997fea2ec63f06092414a97263704b505c3e","title":"","text":"image in Luna paint, save. on entry in SCM viewlet to see diff some more a drink to resize the window diff editor loads Haven't tried in other custom editors, don't know if this is an issue with custom editor openers or in Luna.\nSeems like a core issue, also happens with hex editor with:\nThis seems like an important regression. No idea what changed in custom editor land\nThis seems to have been caused by /cc The current calls causes the webviews to end up with height in this case. On the first call to never gets invoked. Previously it got invoked before the webview was even created Looking to see what that proper fix would be for this. Some potential ideas: I feel like should probably always be invoked at least once on a newly created editorI'm not sure why the passed in always has zero height but that seems wrong. If that element has , then I believe the code would work The issue seems to be that and both have zero height\nI get the following error with these steps, unfortunately... So I could not repro this. ! I am pretty sure the grid calls layout to its views whenver their sizes change. Is it possible that some layer between the grid and your webview is caching something and not calling on you? Namely the editor group layer. For example, I see a potential problem here: If the layout comes, and is , the dimension will be lost forever, since the grid will no longer send repeated layout calls when dimensions don't change. The instance should remember the last and pass it along to the active editor pane, as soon as it comes along. cc A very similar situation happened here:\nI hit the same exception when following hex editor alternative steps. This might be a new bug?\nMaybe is the same and has easy steps to reproduce.\nThanks the lack of milestone was hiding that issue from me! Will check it out.\nbtw the is storing the dimensions and applies them here already when making the control visible:\nThat exception you see I receive in the official build of insiders, but not when running out of sources. It implies that the resource URI is undefined which shouldn't be possible. If you have any ideas or would like to help me discover what is causing this let me know\nFor the hex editor, I had trouble reproducing this until I set ."}
{"_id":"doc-en-vscode-6314fb7cd2d48e185ecdb2786c114044bf9e3554382afa226f8074c0aaf215bc","title":"","text":"The data inspector (usually) seems to trigger a relayout for me My reliable repo: a TS file and open the diff\nThanks I can repro with those steps! I will look into this tomorrow if you don't figure it out today."}
{"_id":"doc-en-vscode-7f5c7c09ebd3688f57599a13e4f271c776e7a59992adec2e94fbd4cf170c5773","title":"","text":"Testing The setting says \" speech recognized\" which reads a bit weird. Maybe leave \"speech\" away or says \"Controls whether the editor listens for the 'Hey Code' phrase and what to do with it\"\nBonus points for mentioning that this happens all locally\nYeah, I think I will go with this:"}
{"_id":"doc-en-vscode-1310841f4b049c7e42d9bdd4148fcaaeda5206bce51808ca7b1fb30a43fd985f","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.68.1 OS Version: Windows 10 Steps to Reproduce: a Markdown file and open \"Open Preview to the Side\" add line: ! image in preview second line: ! second image shows up. Remove the line above image preview shows 2 images and does not update properly !\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.68.1. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!\nStill happens in 1.68.1. Updating bug\nDoes this reproduce in the with all ?\nYes still happens with all extensions disabled"}
{"_id":"doc-en-vscode-c455e598b0715aa0b7c9749d24c398f0a9ebeb3986a4656459469cdf9e20ca5f","title":"","text":"chat in sidebar something into input Session in Editor :bug: input is cleared\nThis still repros for me\nThe input is preserved but the input for the new chat session in the sidebar is not cleared"}
{"_id":"doc-en-vscode-f6ea98e300acda006ca70ad83befda5dfa5095db4d53a5f4b91f39ebc4631da4","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.67.2 OS Version: MacOS Monterey 12.4 (default zsh terminal) Steps to Reproduce: a folder containing an \"Umlaut\" (äöü) folder in VSCode and launch/use a terminal (for testing I used echo something in zsh) VSCode restarting, VSCode fails to restore previous session as \"the directory doesn't exist\" However, if there is no \"Umlaut\" included in the path somewhere, this problem won't occur. This issue should affect at least all german VSCode-Users (like me) who want to open a folder on their Mac located in OneDrive. The problem is that the filepath for such files looks like \"~/Library/CloudStorage/OneDrive-Persönlich/...\" and restoring a session isn't possible for any folder/file in OneDrive, which is a little bit inconvenient. There might also just be a problem with my terminals encoding (currently set to UTF-8 so should work I guess) that it can't open the working directory with how it spells (in this case) the ö in the filepath: <img width=\"390\" alt=\"Bildschirmfoto 2022-05-23 um 00 04 17\" src=\"https://user-\"(when I try to cd to that filepath it doesn't recognize the oxccx88 as ö and says the path doesn't exist) Thanks for help in advantage :) Greetings, Fussel\nRelated:\nIs this a windows issue only? I can't replicate this issue. I'm running Arch Linux."}
{"_id":"doc-en-vscode-c575e1ba6305fc6d151f49e470a307485cb4e79b1d36ef24011fd4acafd7c588","title":"","text":"the OP reported on macOS which shares most code with Linux\nI'm trying to replicate the issue because I want to see if I can contribute to fixing it, but I can't seem to replicate it. I have a folder named \"äöü\" and I open the integrated terminal and run \"echo hey\". I close VSCode and reopen it, but it restores the session like it's supposed to. I even tried running a task that runs \"echo hey\" and it does the same.\ndo you close all VS Code windows and reopen? Just closing one of many will keep the original process running and reconnect to it.\nI am trying to trace this bug. But it is kind of tricky because it is NOT reproducable on OSS dev build. It is reproducable on stable or latest insider production release though. On production release, after running a command in the terminal, somehow Unicode characters in the current path was saved incorrectly. (You can also see an incorrect terminal title updated.) On OSS dev build, Unicode characters in the path seems saving correctly and there is no issue restoring or splitting the terminal.\nthat's odd, could it be one of your settings?\nI have zero settings for insider and OSS dev version. I only have some settings for the stable version. My system environment is same as OP's.\nAttached a video to show the issue. https://user-\nI wonder why this isn't happening in OSS: ! I think you must have settings in Insiders there because the default value is just ? Unless that's the same format zsh outputs its title?\nNo. I don't have any settings for insider. ! The final tab title is affected by other settings such as or ! After some testing, I can confirm that isn't updated into title sometimes. It might be another bug there but it is off-topic.\nIt seems like event fired unexpectedly with wrong path.\nI also tried to reproduce the problem on Ubuntu but I wasn't successful (both the original problem and the split-terminal one). I even tried Persian and Arabic special letters but still had no chance.\nTo verifier: When testing make sure shell integration is off"}
{"_id":"doc-en-vscode-b042aa9ac46d949173a046c3b49c30766cf854a37c2ace6051015e74be0ce1f8","title":"","text":"a directory at path: /Users/ma//demoproject.(contains Chinese component) the directory from VSCode top menu: File - Open... a new terminal from top menu: Terminal - New Terminal. and terminal window automatically goes to path (demoproject). Command + Q shortcut to quit VSCode completely. VSCode from mac dock icon. and VSCode shows error: The terminal process failed to launch: Starting directory (cwd) \"/Users/ma/xe6x9cxacxe5x9cxb0/demoproject\" does not exist. the path in step1 does not contain Chinese(like /Users/ma/local/demoproject.), no error shows. the error only shows when the path contains Chinese. Extra VSCode information Version: 1.65.2 (Universal) Commit: Date: 2022-03-10T14:33:49.188Z (1 wk ago) Electron: 13.5.2 Chromium: 91.0.4472.164 14.16.0 V8: 9.1.269.39-electron.0 OS: Darwin x64 21.4.0\nTo verifier: When testing make sure shell integration is off"}
{"_id":"doc-en-vscode-84c24afd57f807815264d1325d2ca0ebbd58c76c0818e4790f97bbba89ebf2f6","title":"","text":"when split terminal in chinese path,error!\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.609 --- <!-- score: 0.558 --- <!-- score: 0.557 --- <!-- score: 0.499 --- <!-- score: 0.499 --<!-- potentialduplicatescomment --\nI think this is language's problem.English path is very well.\nIt looks like your report is missing some important information, See for more info on how to report actionable terminal issues.\nThanks for your reply! version: 1.39.2 submit: Data Time: 2019-10-15T15:33:00.827Z Electron: 4.2.10 Chrome: 69.0.3497.128 10.11.0 V8: 6.9.427.31-electron.0 OS: Darwin x64 19.0.0 !\nThanks, you can probably workaround this with this setting:\nI has try this settings by your suggestion. And it really not error.Thanks!But this way to create a terminal is not my default way.When I create a new terminal.It go to its father's file path , not it is before. If has a way to realize my need? !\nYeah that's what the setting does, you can't get what you're after until we fix this bug.\nOK!I understand.Thank you!\nI guess this is obvious, but the problem also occurs with European characters (at least french), and very likely with Arabic, Japanese, Indian, Korean characters as well. I also switched to in order to work around the problem.\nI had this issue with the project folder itself containing unicode (greek letters). To my surprise, the hack still worked, the new terminal now sucesfully launches in the project folder.\nAlso happens when the cwd is set via runInTerminal\nNote that in the latter case, it's not just the cwd that's affected - it's also all the command line arguments.\ngood point, this is probably an issue in debug land actually or related to encoding in the shell. Reopening it.\nTo verifier: When testing make sure shell integration is off"}
{"_id":"doc-en-vscode-71d451431d3eb54452d492870fe9b2c65b8ae4ef38771001491f522414bb1b5e","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --The built-in git extension exposes methods for most of the base git operations. Unfortunately not , which would be used to unstage changes. Could easily be to , i'll open a PR for this sometime today.\ncould you verify this by reviewing"}
{"_id":"doc-en-vscode-968a96313b8fbfd2d3fe68ff7548e1a0cd57ed9585e45e7ed08c0c26f6705830","title":"","text":"in vscode , the formula is rendered with katex, an KaTeX parse Error occurred ! in pycharm , it's rendered properly with mathjax ! latex code:\nThe equation renders fine if you wrap it in so I reckon it's related to how we handle bare math blocks. Relevant code:\nwow\nHi Can I work on this?\nsure\nCan you let me know where to get started? As per the last comment, it works when wrapping with \"$$\" so I figure I need to look into how vscode is rendering lateX in general? any other pointers?\nthe reason is probably katex can't process environments outside $$ , but i wonder if katex has a similar option 'processEnvironments' like MathJax provides\nThis is the same for .NET Interactive\nThis is indeed a pain - especially for someone still going through the learning process and not a LaTeX expert by any means."}
{"_id":"doc-en-vscode-b2a861202a6f18589a874da3cf637a31f9c664bff8dce9957f7f8725404e2157","title":"","text":"They used to finish in 3m24s before - […]-5f82-0dbc-&t=-371d-5daa-d7a2-\nI can confirm that tests started timing out after this commit - The first commit that is started happening is Checking what commits in between would have caused it. FYI\nAssigning to as one of his commits has caused this timeout.\nMac seems to be passing pretty reliably now but now windows fails consistently on two tests Been adding a bunch of logging to investigate this. Here's the relevant new error: The service worker should be intercepting this request so it should never be going to the network at all here. Taking a look to see if I can figure out what is going on\nActually it's not windows specific: its Firefox specific. The windows browser tests just happen to be the only ones using Firefox Can reproduce on Mac too"}
{"_id":"doc-en-vscode-578ceb4208d3157692dad292c1ec92be857cf6596e3377c44d7b6e77e305e39b","title":"","text":"For example: Context:\nAs we anticipate we might need shell scoping as well in the future, use something similar to instead. Updated proposal: One issue with this is that carries optional properties. Hence an alternative could be:\nThe required would need to be removed when we a second one anyway, we can just treat and the same as\nSorry what do you mean by the \"second one\"? Mostly I was trying to follow : But I guess we don't have to strictly follow it?\nby a second one I mean a second scope, we don't want the user to explicitly enumerate all the scopes to set it which would do.\nUpdated proposal:\nClosing in favor of as these issues were only open to finalize proposed API.\nMoving this out of June as it's a feature request.\nI was expecting this so it's easier to understand for an API user:\nThat's what I had initially had , you're the one who recommended that perhaps we should put it inside itself:\nAPI sync also seemed to be in agreement with that, but maybe we should check again.\nmy mistake, I forgot some of the conversations as this has gone through a few iterations\nBased on discussion in API sync, we should probably deprecate existing API and merge it with the scoped one (similar to )\nAssigning myself too so it's easier for me to track\nWhen viewing contributions, is it possible to open the preview of the markdown directly instead of the file? ! as opposed to ! I can open preview using: !\nsince this is an advanced feature I think the raw markdown is more flexible as the user can navigate and copy section of it better. So I don't think this is a necessary change personally\nFeedback before finalization: Add an interface implements Add to the Clarify how scoped interacts with the non-scoped collection Don't deprecate\nShould we also call out how it's applied in addition to (and after) the global collection? Looks good otherwise\nSure, I have modified the docstrings to append it I plan to have the implementation ready but only merge and adopt it in Python extension next week to avoid multiple iterations."}
{"_id":"doc-en-vscode-0853a1efdb7751a6bfb63928f9aacd1951e9fb8a82273401bbd689a939b85c6e","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.80.0 OS Version: macOS Steps to Reproduce: vscode itself assigned this css var to but monaco-editor doesn't have this, it only has so should be , too.\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.80.0. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!"}
{"_id":"doc-en-vscode-f5d9638b55fe5ef0c9e31866be2d86242ad0faeed82ccd9ed3be581dae6e0b49","title":"","text":"Issue Type: <bBug</bSemantic token colorization fails when a token type ID matching any object.prototype property name is provided. Some examples are: , and . is the most problematic name since it is likely to be used as a token type ID. Any string should be able to be used as a token type ID without causing the colorization to fail. Because tries to call on what was thought to be a but was actually a , which throws a TypeError and causes the colorization to fail. The easiest way is probably to use a instead of an object for . Another solution might be to instantiate using instead of an object literal in order to avoid the object.prototype. VS Code version: Code 1.47.2 (, 2020-07-15T18:22:15.161Z) OS version: Linux x64 4.19.0-9-amd64 <!-- generated by issue reporter --"}
{"_id":"doc-en-vscode-65c09052c7328c9c26e8cb616811503d5cebd0fcbda203620495d0dafd12f044","title":"","text":"Issue Type: <bBug</bIn the September release, we updated the suggest widget to respect the class. See The recent changes in the suggest widget for has caused a regression and brought the icons back when class is set on the widget. VS Code version: Code - Insiders 1.29.0-insider (, 2018-10-19T06:50:39.461Z) OS version: Darwin x64 15.6.0 <!-- generated by issue reporter --\nI still see icons in the Setting suggest widget: !\nI meant the settings search <img width=\"646\" alt=\"screen shot 2018-11-02 at 10 15 38 am\" src=\"https://user-\""}
{"_id":"doc-en-vscode-14643b144bf7604028845961d1ddf6a2c76c9df4eebe66ca0f41b1b04dea0f12","title":"","text":"<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.34.0 OS Version: Windows 10 Steps to Reproduce: terminal any command that makes any change of your current project directory after a while, like 'New Folder' action refreshing directory by the terminal command, 'New Folder' input will disappear and File Explorer remains inactive <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes !"}
{"_id":"doc-en-vscode-1ea7dc73767abbac5a08ee11969afdb733d8584d66422744306d049756ed00ac","title":"","text":"Steps to Reproduce: an untitled file with JS mode code with errors/warnings on one =crash ! Is the resource scheme file but the URI untitled? Looks like it\nLooks like is registered this as a file resource where are is not having any issues.\nThanks for the investigation. should fix this"}
{"_id":"doc-en-vscode-66e026b9e83d8c2d8dcfd9a87d885e11382f3b23f3d0f5d5fbc8deec2c84a3f4","title":"","text":"Just to confirm that if you are aware that this change will not show the keybinding for Search view anymore at following places: <img width=\"367\" alt=\"image\" src=\"https://user-\"<img width=\"161\" alt=\"image\" src=\"https://user-\"Not sure if it is intended. Originally posted by in"}
{"_id":"doc-en-vscode-3d36dace239f23e0fce6eb89643b1a76e40cc34b9b44a6344b00b319cecf3b2b","title":"","text":"This was also discussed in and apparently fixed in , but I'm still experiencing this. Steps to reproduce: to in settings. a file, and add the following contents: Ensure that there's an empty newline at the end. Save it. Then hit Undo. Nothing will happen. Hit Undo again, then it will work as expected. You can even hit Save multiple times, and each time will add one extra step to the Undo list. So hit Save 10 times, then you can hit Undo 10 times and nothing will happen. Could you take a look at this if possible?"}
{"_id":"doc-en-vscode-06c0d78c68701f4bde53b377883f2d48396a5676953d5c6266940a0ff9b183eb","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.51.1 OS Version: on both Fedora 32 and Linux Mint 20 Steps to Reproduce: clone issuesvscode1400markdownPreviewRootRelativeImg . doesn't display in ! <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes This issue had already occurred in past: In the latest insiders vs-code build - it's not fixed as well.\nI took a look at this and found that the issue is that the resource vscode-webview-resource uri is invalid, is missing before I was looking how to fix it but I can't find the place where the gets mapped to a can you give me some code pointers?\nhere is my quick and dirty fix for this issue:\nStart by looking in the markdown link normalizer I believe this is passed the raw image links and tries to transform them into url that VS Code can render in the preview\nThanks and after spending some more time looking at this I think I found a proper fix for it"}
{"_id":"doc-en-vscode-daaaeb0087441b88c70b66e892f4a44673b4f865156ceedc26ac175994f04d86","title":"","text":"Issue Type: <bBug</bOpen extension recommendations Click on one In the editor, click the ignore recommendation button Note that the extension is removed from the recommended extension list as expected Click on the undo button in the editor Note that the recommended extension is not placed back in the list. VS Code version: Code - Insiders 1.25.0-insider (, 2018-06-26T05:18:23.085Z) OS version: WindowsNT x64 10.0. <details<summarySystem Info</summary </details<details<summaryExtensions (15)</summaryExtensionVersion ------ html-snippets0.2.1 jshint0.10.19 vscode-eslint1.4.12 tslint1.0.33 vscode-great-icons2.1.35 vscode-gutter-preview0.12.2 vscode-azureappservice0.8.1 vscode-cosmosdb0.7.2 azure-account0.4.1 csharp1.15.2 vs-keybindings0.2.0 team1.136.0 debugger-for-chrome4.7.0 addDocComments0.0.8 cordova-tools1.4.0 </details<!-- generated by issue reporter --! Testing:\nNote: this only applies to non-workspace recommendations. Will look into\nClosed with"}
{"_id":"doc-en-vscode-1b8c42d4394fd18a3090754bf9b1f1b0a218c467cc1642e61519fcec78528f8c","title":"","text":"If the knows which languages a user tried in their first day, it could unlock more specific questions to further improve the language onboarding experience. How though. tracks the edited count for languages that run surveys; which are the languages we care most about right now. Only issue is that the storage key isn't exported, so the doesn't know about it. any general thoughts on the idea, and best practices on how to make the storage keys known to the ces survey code; so they can be read and attached as params to the survey URL?\nThe CES code already has access to the same . Can't you copy the maps the surveys to storage service keys over to CES and read them from there?\nSure, was mainly worried that this creates some copy/pasta. Will take that approach for an initial pull request draft and gather feedback.\nNo worries. We can also extract the common code to a new file, sibling to those two.\nare there verification steps for this?\nSorry that they are so complex and time-intensive. I can also verify that I see survey responses from insiders that include the language data. a fresh VS Code Insiders instance , ensure that you are in the treatment group a Java file and save it. VS Code open for 1 hour to trigger the new-user survey popup and click to open the survey that the URL it opened contains\nOk, see you in an hour!\nThe treatment service does not seem to like me, for some reason. Unmarking as verified for someone else to give it a shot. Note that you need to repeatedly open VS Code windows with new user-data-dir's until you get into the treatment.\nAlso attempted this but no survey after an hour. Confirmed I did get an initial\nwhen the experiment enables the survey you should also get a event? Did that work?\nTesting again\nWorked! [redacted]&javaLang=1"}
{"_id":"doc-en-vscode-ca4c8e284638e875a328a782f04eeaa29d63e04dda9b4a40ecce0a88b760b5dc","title":"","text":"In the recent Edit Sessions UX study we ran, we saw several participants searching in vain for the terminal item in the compact menu on , sometimes for minutes: ! Could we consider bringing it back? I'm guessing we intentionally suppressed it because terminals can't run in the web, but participants were simply confused that it wasn't there without connecting its absence to the fact that was a browser application. Additionally, RemoteHub contributes a welcome view for the terminal which is one of the primary ways we explain that users must continue working elsewhere to run or debug code, and the fact that the terminal is completely suppressed means one less opportunity for users to organically learn the limitations of through our welcome views. ! cc\nYes that is why it was removed. It doesn't do much but if it helps reduce confusion I guess it's worth it?\nNot against the idea of having the terminal menu in web but in its current form it's useless and doesn't allow me to actually show a terminal pane, it just deadends all over https://user-\nReverting in\nwhen I enable the menu again and fix up the terminal command, the run/customize task items give this on the scripts/code-web sample-folder: <img width=\"1229\" alt=\"image\" src=\"https://user-\"That does actually run, not sure what it's doing exactly though. Is that a custom execution task running in the web worker ext host? If so you might need a context key on those menu items to only show them when a custom execution task is registered?\nUnless I'm missing something, I believe only registered tasks will show there - if they're running when you select them, then I don't think there's an issue.\nI think there's a built-in fake task when running code-web is what I mean, when run in actual then running it might do nothing. In that case it should hide similar to how we're changing the behavior of New Terminal depending on whether extension contributed profiles exist"}
{"_id":"doc-en-vscode-8f3f44a93c9bc54ab8885dc41bf92d3fc693ba39858ea9dd6c3523a5d2a05a86","title":"","text":"I wonder if this could be a 18 issue? //cc\nYes , this is addressed in Node v20 and we have also backported the fix to Electron in However, we haven't backported the fix to but we can."}
{"_id":"doc-en-vscode-cdb0459802b2b8ab34de50e2780d4a55b77e583d33815f4174e3cd19443d135a","title":"","text":"VSCode Version: 1.7.1 OS Version: Win 10 Pro Add an option to set environment variables before starting the terminal. Something like: Or maybe you want to split variable and value something like: Obviously the same goes for Linux and Mac\nI'm curious what your use case is for wanting this in VS Code but not in Windows?\nHere are some use cases (first two from personal experience): Different paths to libs per workbench Test your application with different languages/translations. It would also have made a much better work-around for You just don't want to set set the right vars every time you open a terminal.\n! how should i fix this problem\nIf you are asking because I mentioned in my previous comment, I have no idea about that issue. That one is not my bug report. I just ran into that issue when I was searching around if the request I wanted to submit was not a dup.\nIts not hard to see in the Python context an app needing a specific PYTHON_PATH, and other languages needing class paths/library paths. My thoughts here is that the tasks may be the workaround - a task can have associated env vars - and a make/rake/grunt/gulp/fabfile or something could be used to set it all up for your project.\nI would love to see that too. For me it's about setting workspace env variables to be used with node gyp. Like OPENCV location\nHow does one configure this on the Insiders build?\nIt will be in the next build of Insiders, you will be able to use the new settings:\nI can confirm that it works great for me, i now use it in almost all my projects Thanks!\nUPDATE Again: Just found :man_facepalming: ! UPDATE: DOH! I just realized . God I love this editor! Amazing job. Is there any way to make this project specific? I am using this now for GOPATH but would love if it changed to the current working directory, which the integrated terminal seems to be aware of when it starts up. Maybe there is a variable I could set the envs to that contains the same value used by the starting terminal.\nAnything in is already specific to the root folder that is open in VS Code. This is called the Workspace settings.\nhow to specify the GOPATH env variable inside the integrated terminal to be the same as the one inside your ?"}
{"_id":"doc-en-vscode-825e3de69c7912b595dc936bea3584d7c786b3b58fb269b7d32d5ec2acf5f5ed","title":"","text":"I am trying this on 1.17.1 and does not seem to work:\nThanks but in my case the following is what I'm after, but doesn't seem to be working. The use of the variable seems to work fine because it stops the underlined errors in code related to gopath, but the env var on the terminal still is bork. I'm mostly good because I also setup an install task that works. Now if I could only get that to like the libraries do correctly so things would compile automatically on changes (which I realize many would not like). ! By the way, here is an (old) task that works. But, it is interesting that everyone has different variable names for the same thing (workspaceFolder v.s. workspaceRoot): !\nYou are using in the env var for the which doesnt get resolved. Should we add resolution of such variables when using the setting? For vs see\nResolving variables in the setting is tracked in\nAfter upgrading from 1.17 to 1.18, the setting doesn't work anymore (OS Version: OSX High Sierra) version 1.18: <img width=\"1064\" alt=\"screen shot 2017-11-15 at 10 12 25\" src=\"https://user-\"version 1.17: <img width=\"442\" alt=\"screen shot 2017-11-15 at 10 16 13\" src=\"https://user-\"\nJust tried on my Mac, it works as expected. One thing to keep in mind is that the env vars will not affect terminals that were created before the env vars were to the setting. Can you create a new terminal and check there?\nit was indeed on a new terminal after the env was . I've retried it on two high sierra machines recently.\nI tried on a co-worker's high sierra machine. Works there as well. Can you try on our Insiders? Can you try with extensions disabled?"}
{"_id":"doc-en-vscode-60b24c7e54ac832da6d521d4219395ac475a41f0dafd583d0bc078924097784e","title":"","text":"Type: <bBug</bThe issue shows up if I have a custom PS1 prompt that looks at the exit code of the previous command (using ). Here's a minimal configuration in the .bashrc file on my remote Linux host: Expected results: When I run in a terminal outside vscode or disable automatic script injection in the integrated terminal, I get this: Actual results: when I enable integrated terminal shell injection, I instead see this output running the same commands: That is, the exit code is now always 0. I tried manually injecting the scripts using the command from ; no change in behavior. Looking at the script that gets loaded (.../server/out/vs/workbench/contrib/terminal/browser/media/shellIntegration-), I see these lines: As an experiment, I tried moving the call later and opening a new terminal, which fixed the issue for me: So it looks like the step that clobbers the exit code is the step. Remote OS: RHEL 8.10, bash 4.4.20 VS Code version: Code 1.91.0 (, 2024-07-01T18:52:22.949Z) OS version: WindowsNT x64 10.0. Modes: Remote OS version: Linux x64 4.18.0-553.5.1.el810.x8664 Remote OS version: Linux x64 4.18.0-553.5.1.el810.x8664 Remote OS version: Linux x64 4.18.0-553.5.1.el810.x86_64 <details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nIn VS Code version 1.92.0, I still see this issue. In this version I now see that the method calls instead of just , but running this after the function still clobbers out the return code to zero. says this about running :"}
{"_id":"doc-en-vscode-a8231eb0fd409d89fa691f3987c81c0ffe7dc7796de8bfb8b1a0167e9ede270c","title":"","text":"<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.31.0 OS Version: Windows 10 Steps to Reproduce: a system with user home containing a space some gulp task configured you will have: <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No I think it was introduced by this commit: it is now using a global gulp which is installed in user home instead of local one. The command has never escaped space in path though\nThank you for catching this! The fix will be available in the next Insiders build.\nWhy are you looking in now, was that on purpose?\nAh that was already fixed"}
{"_id":"doc-en-vscode-74615f5a45cf30385b40b6f167520c558156af22d8574f1906e396586c055049","title":"","text":"Steps: multi root workspace with 2-3 folders right click into empty space below last folder select \"Remove Folder from Workspace\" nothing happens //cc\nI guess this should remove the last workspace folder.\nNot sure, if you invoke this command from F1 list, you get a picker asking which folder to remove. We could do the same here.\nSometimes the empty area takes the last workspace folder as context and sometimes not. So we should be consistent and check what other actions do on the empty area. Maybe the only exception is actually drag and drop, which on drop on empty area drops to the last workspace folder.\n! context menu of empty area should be something like this.\n! and also change this icons when there is no folder selected.\nworks fine, but \"Open Folder Settings\" does not bring up a picker, should it?"}
{"_id":"doc-en-vscode-832b620c17a18d07fe731981e65127ef462d966e430c9535c87bcb964758535b","title":"","text":"Somehow I missed the change in the and only now found it out. What I do not udnerstand is why we need all this code? I think this handler should be removed and the action should be called like it was done previously. Code pointer: It would be great if we could tackle this in December. Thanks\nThe reason for this code change was that creates a handler for command that passes no arguments to the action: So instead I used with a handler that would forward the argument. It looks like there are other debug commands registered in this way in , so I think moving the code for startDebug there makes sense. The handlers there are shorter and less generic, I'll see if rewriting the startDebug one that way causes any issues\nwhat you need is calling the start debug with a configuration, and I believe we already have a way for doing this. can you please provide a pointer to where you are already calling start debug with a configuration. are you calling start debug from the main side or where from exactly? I can also look into this. You can just let me know how to verify that I did not break your scenario.\nThe extension API's implementation passes a debug configuration to the debug service's startDebugging here:\nI'm trying to start debugging from the process explorer. Since it's a separate browser window, it doesn't have access to the debug workbench service, and it's also not an extension, so it can't leverage the extension API for starting debugging. Currently, the process explorer window sends an IPC message back to the issue service in the main process. This then sends an IPC message to the main renderer process to run the debug start command: The path is also used by the native menu to trigger commands from the main process.\nok got it. So the best way it seems is to what you have done going via the action. Can you try to clean up the handler. And if that does not work I can look into an alternative solution.\nSure, I'll work on cleaning up the handler today"}
{"_id":"doc-en-vscode-5be6721b894b1a469911c0825a683b37edf680e9c038b084bf78a336da082a7e","title":"","text":"On Windows: something with a rich hover in a tree view, like an extension or an item in the activity bar +tab off the window back to VS Code the hover is still there and stays around around until you hover something else with a rich hover, or click on it and hit escape"}
{"_id":"doc-en-vscode-3103de3e5d1f72f38b5152ad4f31731eb0ae50d68187d0c8f944cffe8f965774","title":"","text":"Type: <bBug</bThank you for all the improvements to walkthroughs in the latest release, however I noticed that the following issue started popping up after the changes to \"save\" your place in a walkthrough when changing tabs. When you have a walkthrough open: to a different tab in the same editor window back to the walkthrough that the media on the right side of the walkthrough is no longer present steps 1. and 2., note that the media has reappeared. If you repeat steps 1. and 2. once more, the media will dissapear again. This process can be repeated any number of times to \"toggle\" the media's visibility. Jumping between steps will also cause the media to reappear."}
{"_id":"doc-en-vscode-1b603920c27aa6d0dfd0bbd63380ca8c452eb01444483e0ef595a0730bb479a6","title":"","text":"Screen recording: VS Code version: Code 1.87.0 (Universal) (, 2024-02-27T23:42:56.944Z) OS version: Darwin arm64 22.6.0 Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (34)</summaryExtensionVersion ------ vscode-markdownlint0.54.0 vscode-eslint2.4.4 gitlens14.8.2 EditorConfig0.16.4 prettier-vscode10.1.0 go0.41.1 vscode-drawio1.6.6 debugpy2024.2.0 isort2023.10.1 python2024.2.1 vscode-pylance2024.2.3 jupyter2024.2.0 jupyter-keymap1.1.2 jupyter-renderers1.0.17 vscode-jupyter-cell-tags0.1.8 vscode-jupyter-slideshow0.1.5 remote-containers0.348.0 makefile-tools0.8.22 vsliveshare1.0.5918 ansible2.13.148 java1.28.1 vscode-yaml1.14.0 sonarlint-vscode4.3.0 code-spell-checker3.0.1 intellicode-api-usage-examples0.2.8 vscodeintellicode1.2.30 vscode-java-debug0.56.2 vscode-java-dependency0.23.5 vscode-java-pack0.25.15 vscode-java-test0.40.1 vscode-maven0.44.0 </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --"}
{"_id":"doc-en-vscode-93c5e4df7e774ff1c494b588c78cae0035430b1d1d8fa6f45e21830139afb07c","title":"","text":"We should filter out the terminal if this is undefined, otherwise bad bugs like could happen where we restore a terminal into the wrong window: The chances of this happening are pretty slim so not doing this in Feb.\nFalling back to 0 as it has been seems to work well, if we need to handle this on reconnect we will have a persistentProcessId of 0 which isn't valid anyway."}
{"_id":"doc-en-vscode-2edc0cb37ef6a48b076354894c97bd8edb74a7629c6824d3a46e00996d0ef335","title":"","text":"Copied from My workaround is export VSCODEGITASKPASSEXTRAARGS=\"--title=foobar\" but the proper bugfix should happen in VSCode's GIT_ASKPASS implementation.\nI experience the same problem, from the moment been released. Working through remote ssh the git processes stay open, without anything happening.\nStill facing this issue on WSL.\nWill be fixed in upcoming 1.62.2. Meanwhile a workaround is to and then manually\nI can confirm that this is fixed after updating to 1.62.2 ()."}
{"_id":"doc-en-vscode-9e042d9bea477506757630ac62f0ed8d098cdd9baa457d5ed622fc561ee30049","title":"","text":"Steps to Reproduce: to 1 window on a folder open the command line run = only the folder opens, but not an empty window I would expect here 2 windows to open, one is the folder and one the empty window since no argument was given to open any folder or file."}
{"_id":"doc-en-vscode-e4572014a0855abb10d0906466cbf39d298bc7966e83dc59b55e706bf27fdb56","title":"","text":"Testing a Markdown file with headers up to H3, enable Sticky Scroll and scroll up and down in the editor. Stick Scroll a header leve and press . Try to scroll up and down with the arrow keys. Unclear where the cursor focus is. Try typing, the editor doesn't have focus Focus should should probably go to the editor on the selected header. That is the behavior with breadcrumb navigation\nThis is the general behavior of Sticky Scroll and also reproduces if you select a level via the mouse\nVerified on OSS"}
{"_id":"doc-en-vscode-e32c78fc23424df54fd41f8832dade1301cffb7aca5b4f9e7cf8bf47f43b97fd","title":"","text":"Type: <bBug</bA corrupted file is stored in CachedExtensionVSIXs and never re-downloaded, preventing extension from being installed until the file is manually deleted. Repro: Attempt to install an extension while the Azure DevOps service is not available and returning errors. Result in this comment: The corrupted vsix contains the following HTML instead of zipped data ! Shared Output: VS Code version: Code 1.84.2 (, 2023-11-09T10:52:57.054Z) OS version: Darwin arm64 23.1.0 Modes: Connection to 'SSH: ox' could not be established Canceled <details<summarySystem Info</summary Connection to 'SSH: ox' could not be established Canceled </details<details<summaryExtensions (31)</summaryExtensionVersion ------ vscode-ron0.11.0 vscode-eslint2.4.2 prettier-vscode10.1.0 kotlin0.2.32 copilot1.139.0 copilot-chat0.10.2 copilot-labs0.17.1121 vscode-github-actions0.26.2 better-cpp-syntax1.17.2 vscode-peacock4.2.2 vscode-clangd0.1.25 membrane1.6.1 vscode-dotnet-runtime2.0.0 vscode-postgresql0.3.0 remote-containers0.321.0 remote-ssh0.107.1 remote-ssh-edit0.86.0 cpptools-extension-pack1.3.0 hexeditor1.9.12 makefile-tools0.7.0 remote-explorer0.4.1 proto0.0.4 vscode-yaml1.14.0 vscode-gitweblinks2.9.3 rust-analyzer0.3.1748 even-better-toml0.19.2 shellcheck0.35.0 es6-string-html2.12.1 vscode-lldb1.10.0 vim1.27.1 clang-format1.9.0 (2 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\ncc who was looking into this in\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected."}
{"_id":"doc-en-vscode-0680c980612616ed25b20c1561eed0512e5a359cecdeb33baaa71965cb61a456","title":"","text":"If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nbumping to see if this issue is still occurring and if the fix has been verified!\nIt's somewhat fixed. You can try again and the second time it will work. But the error message that appears doesn't indicate that. perhaps it should retry at least once if the zip is corrupted before showing this error? Note the bottom right popup asking to manually download where a retry could've sufficed. <img width=\"1304\" alt=\"Screenshot 2023-12-06 at 6 14 41 PM\" src=\"\"\nThanks for the feedback. This bug is to fix not to cache corrupted VSIXs and please validate that. Regarding trying again, I do not think we should be showing this as this might be random.\nVerifying with a code review"}
{"_id":"doc-en-vscode-ed0c60cd90e1dbe66fcb3102062c06ef31163396b4b77414dc785b4a183463fc","title":"","text":"String Text: Show the name of the file. When tabs are enabled and two files have the same name in one group the distinguinshing sections of each file's path is . When tabs are disabled, the path relative to workspace root is shown if the editor is active. Context src/vs/workbench/electron-browser/main.contribution:English (en) Resource vs_workbench Query: Typo in source string. distinguinshing-distinguishing. Should'nt it be \"the distinguinshing sections of each file's path are \" and not is? Reference: $/?issue=yes"}
{"_id":"doc-en-vscode-d47e2e36a7f4fa69689e3e4cf115cf1e13c7227d72a674d1d95fb81331d88ccf","title":"","text":"To reproduce. click smiley in the bottom right of editor window. The feedback form should appear enter several characters, press \"Enter\" (the new line should appear) continue entering characters while you can You should stop at \"1 characters left\" (declensions?) After that you will not be able to enter anything On the screenshot. There's a newline after \"1\". I reached the end of input and unable to enter more characters. But counter still says \"1 characters left\": <img width=\"468\" alt=\"\" src=\"\"Mac OS 10.10.5 Yosemite\nThis is interesting. I created this to reproduce the behavior. do you have any thoughts on this? Here are a couple ideas we've talked about before. 1 - Remove the text area completely and replace with just the tweet button. This makes sense because we're submitting to twitter anyways. 2 - Hack the text area element to get better behavior. This would involve looking for new line characters based on the batter described by OP. I'm not a fan of this as it seems error prone. If we go with option 1, I'll work with Brad and get some design love on the form. any thoughts? I created PR to fix the \"s\" on singular count issue.\nThanks for your amazing job and thanks for asking! :blush: If the \"Tweet\" button will not require having a Twitter account, then it's a solution, I guess. I feel a bit weird reporting those super-tiny issues that still require developers time. Is the related (then we should star it)? Probably even disabling line breaks (like turning into line input) could be a solution.\nYes that is the exact issue. Starred it! I thought about disabling line breaks, but I think it would be a good thing for the user to do, considering Twitter supports line breaks. The Tweet button wouldn't require a Twitter account, but once the Tweet button takes them to twitter they would need to login.\nWell, the need to login on Twitter is a requirement to have a Twitter account :blush: I wouldn't think it's a good UX :relaxed:\nRight. Unfortunately Twitter doesn't support anonymous tweets.\nAfter some internal discussion we decided to go with a dialog redesign. Will work with our UX designer to get some help.\nClosing as it will be fixed with ."}
{"_id":"doc-en-vscode-1b9440f4896c59291f09b59cb2a75a53b3216256bd3568fca66e1051a0d9f2fe","title":"","text":"Does this issue occur when all extensions are disabled?: Yes Search rearranges the items according to how well they match, so items are no longer in the original groups. Because of that we remove the separators. (See, e.g., the theme picker.) Initial state: <img width=\"622\" alt=\"image\" src=\"https://user-\"When typing: <img width=\"651\" alt=\"image\" src=\"https://user-\"Steps to Reproduce: items and separators in the quickPick (grouping). to search VS Code Version: Version: 1.77.3 (Universal) Commit: Date: 2023-04-12T09:19:37.325Z Electron: 19.1.11 Chromium: 102.0.5005.196 16.14.2 V8: 10.2.154.26-electron.0 OS: Darwin arm64 22.4.0 Sandboxed: Yes\nHi which extension is providing this quickpick?\nHello ! This is my debugger for .NET apps QuickPick implementation:\nAny updates?\nI can repro this. I'll get this fixed ASAP but as a workaround you could listen to the event that fires when the user updates the filter value and then clear out the separators from\nThank you,\nI have a fix for this. It won't make it for this release on Wednesday, but it'll be in the following release (late May/early June)\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nVerified in the task list. !\n/verified"}
{"_id":"doc-en-vscode-c175afe3688b70ff239d00dffeb868c9b3d2e0d5fd04168970b5caf2a95a314e","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --There is Title Case, Snake Case, Lowercase, Uppercase, but no Camel Case. It would be nice to transform: -\nIt would be a great feature. You could use that extension: until then. I also think that they should work on the to exclude articles and even custom words. instead of which is not accaptable in many academic organizations.\n<!-- -96bd-47a8-a0e8- --This feature request is now a candidate for our backlog. The community has 60 days to the issue. If it receives 20 upvotes we will move it to our backlog. If not, we will close it. To learn more about how we handle feature requests, please see our . Happy Coding!\n<!-- -a7ce-58b9-b1f9- --This feature request has not yet received the 20 community it takes to make to our backlog. 10 days to go. To learn more about how we handle feature requests, please see our . Happy Coding!\n<!-- -c9e0-7adb-d31b- --:slightlysmilingface: This feature request received a sufficient number of community upvotes and we moved it to our backlog. To learn more about how we handle feature requests, please see our . Happy Coding!\nUp for a PR anyone?\nI to do this. I have more details in the PR description about how different use cases should be handled. I don't really have a strong opinion on it. I wrote a personal extension long ago that has it's own set of rules, so I probably will keep using that over the one I implemented in the PR."}
{"_id":"doc-en-vscode-6b1e05d8a4fe46f6203806dd1138585b5cdba94f37cf7f053ac30eda7965fb58","title":"","text":"In OSS not insiders (so not a candidate) Steps to Reproduce: shows error This is because the gets run in a separate process than the Extension Host and so the vscode.l10n API isn't available. Proposed fix is to not localize the single error message being localized since it's not surfaced to the user."}
{"_id":"doc-en-vscode-f6b15949b0f92dc97b80ad96bdeae8b204e77da62e6bd7f5d6a29cf22baee360","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.85.1 OS Version: macOS Sonoma 14.2.1 Steps to Reproduce: a python file or other file with a debugging extension installed Debug view and see the welcome view showing Run and Debug Actions. a diff Editor for the same file debug view now shows an Expected: Debug view shows the same welcome state as when not in diff view as this language has debug extension installed. <img width=\"694\" alt=\"image\" src=\"\"Actual: !"}
{"_id":"doc-en-vscode-69052d4a276d38f78f8294d067419f9e59bd79390fea2806addc93bf4707fc24","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: N/A OS Version: N/A Steps to Reproduce: presence of in the example above means it should apply only to documents, not \"all\" as the text suggests"}
{"_id":"doc-en-vscode-22cf17ee6b4c5ac18457bd99ddecdcce0ceaf648238ad39dca66dca646244389","title":"","text":"Type: <bBug</bFuzzy search icon is visible in Go to Symbol view (Shift + CMD + O) which on press opens terminal recent search with fuzzy search. I think that button should be excluded from Go to Symbol view and be visible only in terminal recent search view. VS Code version: Code 1.73.1 (, 2022-11-09T02:22:48.959Z) OS version: Darwin arm64 22.1.0 <img width=\"607\" alt=\"image\" src=\"https://user-\"<img width=\"597\" alt=\"image\" src=\"https://user-\"<!-- generated by issue reporter --\nSeems like after the icon is not cleared and shows up in any command palette entry"}
{"_id":"doc-en-vscode-686d1c88f451b6e272db4e815ca7ba679a8166f2a00d058c8674307e936cb7ec","title":"","text":"STR: VS Code with locale set for language prompt and close it VS Code and repeat 1st step Language prompt shows again. has logic for storing an ignore key after a prompt was shown, but is using the wrong key (language instead of locale). The same variable mixup has caused prior breakage of the feature.\nForwarding to since he owns localization.\nI partly own localizations support in VS Code especially recommending. It seems you come up with a fix. Do you mind opening a PR?\nIgnore above comment, I see the PR now.\nI'm still seeing this in insiders I clicked the on the notification and then restarted. After a few seconds, the prompt came up again\nFound it. Only \"Don't show again\" triggers that logic, but just canceling doesn't. I wonder what the right behavior should be. Other prompts also depend on \"Don't show again\" (using the option, ) for a similar logic and otherwise pop up again on next launch. cc for guidance.\nClicking is never meant to be not to show again and user has to explicitly select action. I think this is general behavior for all prompts. I would recommend to discuss this UX change separately and mark this as verified otherwise.\nMarking as verified based on comments above"}
{"_id":"doc-en-vscode-1811e963fcbce8b326e250ac979b40b5aaad5c4bbbf96c72550a6be21e53e197","title":"","text":"A11yMAS; #A11yTCS; #A11ySev2; #DesktopApp; #Visual Studio Code Client; #BM-VisualStudioCode-Win32-Nov2021; #Win32; #Narrator; #NVDA; #ScreenReader; #WCAG4.2.5 #AINotInScope; #FTP; #STP; Application Name: Visual Studio Code Client Windows Version: Windows 11 1)Turn on Narrator/NVDA. 2)Open Visual studio code editor. 3)Open the command Pallet by pressing \"F1\" key and type \"Theme\" 4)Using Down arrow key select \"Preference color Theme\" and press enter key to select. 5)Using \"Up/Down\" arrow key select different available theme. 6)verify that screen reader is announcing the associated label for the 'recently used' and 'Other commands' or not. Screen reader does not announce associated label for the 'recently used' and 'Other commands'.. Screen reader should announce the associated label for the 'recently used' and 'Other commands'. Screen Reader users will not get the complete information about the controls. !\nThe issue here is that the group label \"recently used\" does not get read. I suggest to append the group name to the aria label of the first element in the quick pick that belongs to that group. I think it should be a low hanging fix."}
{"_id":"doc-en-vscode-a49d4a86721e1622dbae5acdc94a24fe8af1aa3c6f148ebe3ef266ee78130861","title":"","text":"VSCode Version: Code 1.12.2 (, 2017-05-10T13:20:36.315Z) OS Version: Windows_NT ia32 10.0. Steps to Reproduce: \"JSON\" file Expected outcome styled as (invalid) content token. Actual outcome styled as comment. Still properly marked as invalid, though. Remarks is OK. Same steps in JavaScript file is OK. ! Ad \"JSON\" in quotation marks: ."}
{"_id":"doc-en-vscode-696563d0313ebe57cd0323a2260fbbdf219950e44a3939558bf6d34a76c22c8b","title":"","text":"I am writing a debugger adapter. I use ProgressStartEvent/ProgressUpdateEvent/ProgressEndEvent to report some debug opertions' progress information. In previous versions, the progress information was shown in right-bottom corner, however, in 1.86.0, the progress information is shown in status bar instead. I think this is not a good idea, as user might not be able to notice the progress information. If there is a workaround if I still want to show progress information in right-bottom corner of VS code? Thanks! Bill Zhang\nI haven't yet found a 1.86 commit that specifically changes this behaviour. Have you perhaps been trying out this new 1.86 feature?\nIf this is a confirmed regression, it would help to know which change caused it. We maintain a based CLI tool  that automatically downloads and runs previous VS Code insiders and asks for each build if the issue reproduces. It takes up to 8 steps to find exactly the build that caused the regression. Steps: install (if not yet done) follow the instructions until you found the offending build (its fine to leave the commit empty) report back the commit range in this issue Thanks!\nThe progress information was shown in status bar, not a pop-up window as before versions, so I have no way to enable/disable notification. In addition, notification was shown as before in right-bottom corner.\n85 does not have this issue.\nIf you cannot find the actual commit causing this using , then you need to provide a minimal repro for us to try out."}
{"_id":"doc-en-vscode-2360d1e279f598b97c03ad04a2e3392b96202bbbebb02137a733a6aa07fa9955","title":"","text":"I have tried the vscode-bisect, but it failed: PS C:gitBIOSDBGV4srcextensionnpx --yes --reset [build] deleting cache directory C:UserszhiqzhanAppDataLocalTempvscode-bisect √ Commit of released insiders build that reproduces the issue (leave empty to pick the latest build) ... √ Commit of released insiders build that does not reproduce the issue (leave empty to pick the oldest build) ... [build] fetching all builds from [build] total 171 builds with roughly 7 steps [build] downloading build from [build] unzipping build to C:UserszhiqzhanAppDataLocalTempvscode-bisect.buildsVSCode-win32-x64-1.85.0-insider... [build] starting desktop build ... √ Is good or bad? » Good [build] downloading build from [build] unzipping build to C:UserszhiqzhanAppDataLocalTempvscode-bisect.buildsVSCode-win32-x64-1.86.0-insider... [build] starting desktop build ... √ Is good or bad? » Bad [build] downloading build from [error] Error: read ECONNRESET You can run vscode-bisect --verbose for more detailed output and vscode-bisect --reset for a fresh start without caches.\nAt first, it is quick in downloading vscode-insider, but later it takes a long time to download it. I am not sure why. !\nthanks a lot for trying it though, maybe some connection trouble. I can try it if you tell me how to reproduce.\nI create a simple debugger for testing. code: vsix: I am not sure whether you know how to start a debugger or not: 1) install test-debugger- 2) open a empty folder and then create a ! 3) select Debugger For Test ! 4) Start Testing debugger ! 5) You will see progress information in status bar !\nIf run in 1.85.2, it is as below: !\nthanks! Is it possible that you have snoozed notifications either globally or for the extension?"}
{"_id":"doc-en-vscode-6bc082b9da2d271ad170a306581f5f49c2d9fd5b32121bc7e0fd6d8f895f270c","title":"","text":"You effectively see the same notification but when it appears in the status bar, it means the notification is hiding because of the \"Do not disturb mode\". When you click the status bar entry you can see how it changes to the notification toast.\nYes, you are right. But even in below setting (BIOSDBG V4 is my extension): ! The notification is hidden unless you click the right-bottom icon. ! I think it makes me confused and it is not easy for user to notice this. Thanks!\nThere is indeed a change that happened as part of our latest release, but I think for the better: debug had always decided to show notifications for debug progress in the status bar via this directive: A change I did last milestone actually made to support it now: So I am deferring to debug component to further triage, but I think this is actually expected behaviour.\nI think we should revert to the old behavior for consistency as that's what users have historically seen"}
{"_id":"doc-en-vscode-5d2adad21175f2dd129fd4ab79d67342065dee049d6a25ac34c9039099e623c0","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.79.0-insider OS Version: macOS 13.3.1 Steps to Reproduce: VSCode in a computer with a relatively old version of Git (2.16.2 does it). a project under version control. control fails silently and only shows the Publish button. Trying to publish causes an error. This seems to be related to a change that makes use of the option in , which doesn't seem to be supported by old-ish Git.\nsorry about this. I will try to get this fixed this week.\nI am also experiencing the same issue. VS Code Version: 1.79.0-insider OS: Windows_NT x64 10.0. git version 2.17.1\nIt looks like the option was in git 2.18.0. Ideally you would want to keep your git installation up to date to keep up with the security fixes. Having said that I do understand that there are scenarios in which one cannot install the latest git release, so I have pushed a fix for this issue. The fix will be included in the next Insiders release.\nThanks a lot! I agree with keeping git up to date. Unfortunately I use VSCode with the remote extension in many servers where I don't have control of what version of Git is installed, so having at least some basic Git functionality there is much appreciated."}
{"_id":"doc-en-vscode-3d8c2038401a1dbbd4392de3f875a1c5aae4ce1d15eb702c81cb050a02d75ec1","title":"","text":"Hi, I encountered the same issue with my VSCode + WSL1 (I haven't upgraded to WSL2 for some reason) git version: 2.11.0 VSCode Version: 1.79.0\ncould you please share the complete contents of the ? What version of git are you running inside WSL? This flag should only be in case you have a version of git that support it.\nSame issue here\nAffecting me too. This seems like a major breaking change. Is there a bugfix release planned? Thanks.\ncould you please share more about your environment? What version of git are you running (git --version). Are you also using WSL, or any other remote environments? Anything that helps me reproduce the issue would be extremely valuable (ex: the contents of ). Thanks!\nThanks Ladislau Using git 2.17. I don't want to upgrade to insiders build Get Outlook for iOS\ncould you please run and share the exact output. My current hypothesis is that for some reason we are unable to correctly detect the version number and that is why we incorrectly add the flag.\nLet's continue the investigation in this issue Thanks!"}
{"_id":"doc-en-vscode-b6e4fee69a290bffdb841f012de480a211a127e700be60a5d289710122250459","title":"","text":"Light Theme the Extensions view the Extension item in the list -all letters turn black, the contrast is not good !\nLooks like these buttons are inheriting the codicon styling, and as a result, the color. I removed it from Light+ V2 to reset it back to null. Ideally this button wouldn't use codicon styling cc"}
{"_id":"doc-en-vscode-ab8f8a01304c1e789eadff48c6c398f5da96d90877b714987774e9df98990e4f","title":"","text":"Remove Manual Folding Range doesn't work Originally posted by in select a range in the editor invoke . Selection turns into a folding range put the cursor at the beginning of the collapsed line invoke . Should remove the manual folding range\nIt works when the cursor is at the second column or to the right of that, but not when it's right at the beginning of the line, which is exactly where it is placed after running the create manual folding range command."}
{"_id":"doc-en-vscode-2cc856f740782963019fedcda4ea658c2e9868c575554b8454755b828d96ec0c","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --- VS Code Version: 1.56.0-insider OS Version: MacOS Big Sur 11.2.3 Steps to Reproduce: Settings and and editor it manually. a rule: editors can't be edited anymore. <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported direct to the extension publisher. The 'Help Report Issue' dialog can assist with this. --"}
{"_id":"doc-en-vscode-12b106b7330b04624dfd9b354f7daf2971fc63935da2a4dde1b087cfd8662666","title":"","text":"Request completions, commit on Expected: Regular completion, no method body OR completion that is otherwise identical to both settings being . Actual\nI want to understand this better, since this kind of behavior is also present in other completions I think: how should interact with completions that can or cannot contain snippets? For instance, the class member completion can be a snippet, but it doesn't necessarily have to be, which is why it still shows up in the example above when is false, but the difference is that it's no longer a snippet (i.e. there's no tab stop in the body of the completed method). I guess this bug is saying that this completion should never be offered in its non-snippet form? Is this the case because not having the tab stop in the method's body degrades the user experience? Is it because of some other reason?\nUgh - okay, so there's plain completions completions with insert text completions with tab stops and for some reason, is the one that enables tab stops. Did I get that right?\nI think so, yes. Not sure if this this is relevant, but note that while is a setting that is user-configurable in vscode, while is not. Andrew B once told me that's because is supposed to indicate whether an editor supports insert text in completions or not.\nI believe it's controlled by setting\nand say that this should never have been user-visible in the first place. The flag shouldn't have been passed in as-is to TypeScript. If anything, the editor should use a flag like to blanket enable/disable other snippet features.\nI don't see the removed setting anymore so marking as verified."}
{"_id":"doc-en-vscode-4cf1796b9bbbd9da74776faab278548788a46d14cc766b3df8f0f2739c50f4b3","title":"","text":"I'm using to mount a remote Azure Blob storage container into my local file system with FUSE. Once I open up that mount point (say ) in VSCode, network traffic immediately goes up to around 100 kb/s. In the storage provider's API logs I can see many thousand operation requests per second to files located at any hierarchy level inside my storage container, but also files that do not exists (e.g. ). I was assuming this is related to VSCode attempting to watch these files for changes, but can't use inotify on the remote share, so falls back to polling the files with a crazy frequency. However, applying the following settings didn't have any effect: Same issue occurs with or AWS S3 + and I presume also for sshfs, davfs2_ and the like. When killing the sub-process, I can observe a short decline in network usage until the process almost immediately restarts.\nIn this case I wonder if we should simply not install the file watcher at all if a setting is configured to disable it. I see no other workaround other than maybe addressing this in our external file watcher where we pass on these exclude rules.\nVerification: run with and filter dev tools console to to see only file watcher related output configure to have an entry open a folder verify that in the output for the watcher you see no indication that the folder is being watched compare this to how the output looks like when you not have such a setting\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!"}
{"_id":"doc-en-vscode-5da332785beef63f199ffee9b60da3420b3b93542b1a9c623222fb5742312f4d","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --Steps to Reproduce: created this simple snippet: I expand the snippet, I get this: Expected this: It uses the first variable as a part of the placeholder of the second variable. The second occurrence of the second variable is not substituted properly. !\nConfirmed... Looks like we capture the value of 1 \"replication\". For now you can workaround by \"inlining\" the second occurrence of $1, e.g have two times"}
{"_id":"doc-en-vscode-fb452f54361640b1e6ffeee826e404ad575fc6a605439d54553fd95c7a6d2829","title":"","text":"Type: <bBug</bSimple Browser to visit a webpage that offers a download. the download. happens :bug: I plan to submit a PR. VS Code version: Code - Insiders 1.80.0-insider (, 2023-06-14T09:02:23.566Z) OS version: Windows_NT x64 10.0. Modes: <!-- generated by issue reporter --\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\n/verified"}
{"_id":"doc-en-vscode-c38fcf6e79785d4c9ccc37af2b82644e8c6d8b229d9c6a5b2babc18950fcfe4c","title":"","text":"Version: 1.94.0-insider Commit: Date: 2024-09-09T17:10:54.809Z Electron: 30.4.0 ElectronBuildId: Chromium: 124.0.6367.243 20.15.1 V8: 12.4.254.20-electron.0 OS: Darwin arm64 23.6.0 Steps to Reproduce:\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!"}
{"_id":"doc-en-vscode-676bcab0f5d8b491513ef2dcc30d3c4fe14d8beef18d39ca00a19cd4b53fd4e0","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.62.3 OS Version: Chrome Version 96.0.4664.55 (Official Build) (arm64) macOS 12.0.1 Steps to Reproduce: vscode web version on localhost (use any build similar to ). Make sure the test repository has a jpg file such as the . Use a chromium based browser (not safari). on a .jpg file in the side pane explorer. Note that it takes 1s for the image to load, even when hosted from localhost. Click back and forth on other files and note that text files display immediately but any custom editor or image viewer that uses a webview always pauses for a second before loading Chrome/Edge service worker console and look for the service- version; note that the version increments each time the same webview is opened (and closed). It takes approximately 1050ms to load every image as shown in the following screenshot ! The service-worker for any webview is instantiated inside  with a query string that includes the unique id of the host frame. This is not actually needed or used by the webview code in but causes chromium based browsers to reinstall and reactivate the service worker even though the underlying service-worker code has the same code, eTag and content hash. For some reason Safari does not have the same delay. If the code is changed to only include the query parameter actually used by the script then the end to end browser time is reduced to 50ms on localhost, a 95% reduction The slow load time is experienced when navigating ordinary files in the editor with no extensions installed."}
{"_id":"doc-en-vscode-2a17e5d3aec038f4da06f5169754b1def2ae30b178a7746f18598e66a2042781","title":"","text":"It is also experienced (and why I noticed it) when developing custom editor extensions that use the webview. This makes it appear that built in editors such as the monaco editor are almost instantaneous, but third party editors appear to be 20x slower to load. ! Please see accompanying pull request, as a one line code change."}
{"_id":"doc-en-vscode-462a85b9713f56a025064af6c0ee490e472b91c8914911b30ed74279122d7c06","title":"","text":"<!-- Please search existing issues to avoid creating duplicates. --- VS Code version: 1.62.3 (Universal) Jupyter Extension version (available under the Extensions sidebar): v2021.10. Python Extension version (available under the Extensions sidebar): v2021.11. OS (Windows Linux distro) and version: Mac Python and/or Anaconda version: 3.7.12 Type of virtual environment used (N/A ...): conda Jupyter server running: Remote I expect code cells to begin executing immediately and (for simple code snippets) to finish executing immediately (e.g. print('hello')) Once the jupyter notebook starts to contain a non-trivial amount of stuff, I start to observe the following behaviors: (1) code cells that contain simple tasks (e.g. ) start taking several seconds or more to complete, (2) it may take several seconds or even minutes for vscode to even visually show that it will start executing a code cell, and (3) in severe cases, the whole vscode editor will become unresponsive and I will need to force shut it down. Below is code for a jupyter notebook that contains a minimal example. The problems start to appear once you've cycled through the code cells 2 or 3 times. I think one of the main contributing factors to this issue are just notebooks that have a lot of \"stuff\" in them. To demonstrate that, the last code block prints 50 scatter plots as pngs, which the native jupyter lab server over the browser handles just fine... but vscode seems to have issues. Moreover, in this simple example I have observed instances where it starts to take a while to save the notebook and the editor may even become unresponsive. Other things to note: I am running this jupyter notebook on an AWS EC2 instance with ~90GB memory 48 CPUs (using a AWS linux ami: ). I have not yet observed these same issues when I have the kernel running on my local mac machine. <details<summaryJupyter example </summary<p</p</detailsHere is a screenshot where it takes almost 3 seconds for a simple print statement. I've seen worse in some of my code related to real projects, though. This is just what I was able to reproduce with a minimal example."}
{"_id":"doc-en-vscode-1bd943722f13ce6069b928fd218c97cba9233bd48143f6d103a2338438eaab62","title":"","text":"<img width=\"1593\" alt=\"Screen Shot 2021-11-22 at 4 48 27 PM\" src=\"https://user-\"\nThanks for filing this issue & I'm sorry you're running into this issue, please could you try the following: Check the performance withouth any data frames in the outputs You can run most of the same cells, but not display the outputs (e.g. don't run the cells with ) I'd like to check whether the perf is a result of large outputs. If that doesn't make any difference please could you: Install VS Code Insiders Go into your settings and search for and change the logging level for Jupyter extension to Reload VS Code Try the same thing & provide the logs from the output panel (use the command to get to the output panel)\nThanks . Not displaying large amounts of output does help. I experimented with the code a bit and here are my observations: (1) If I disable the cell with just the statement (but still outputting the figures), while it helps a little, I still see a significant slowdown. In particular, once I execute the code block that prints the figures (see below) and it is done, I try to execute the first code block again (containing the import statements), but there is a good 30-60 second delay until it will start to executing. (2) If I still create the figures but no longer display them (see below), this helps immensely with performance... there are basically no issues. (3) While still hiding the figures as in (2), but I go back to displaying and increasing the number of columns and rows of that are display to 200 (see below), then there is small decrease in performance. (4) If I do the same as (3), but take it even further by displaying 500 rows and columns of , then performance significantly degrades (see the screenshot). Once it was done displaying , it took ~19 seconds for a print statement, ~35 seconds to make the figures (normally takes a ~3-5 seconds), and it took over a minute to execute that last empty cell (the execution time was hidden once it was done). <img width=\"1585\" alt=\"Screen Shot 2021-11-23 at 10 14 06 AM\" src=\"https://user-\"\nThanks for the feedback."}
{"_id":"doc-en-vscode-4a1dd0ac65753085b89c4530834de527d29f43147225d5897d39b99d7efcca3d","title":"","text":"Please could you: Install VS Code Insiders Go into your settings and search for and change the logging level for Jupyter extension to Reload VS Code Run the command An editor will be opened with some text such as , please copy all of that & paste that into this issue Try simple print statements (without any data frames, and figures, lets keep it very simple) Please ensure the notebook is empty & does not contain any outputs only 1-2 cells with simple print statements. Provide the logs from the output panel (use the command to get to the output panel) Reload VS Code Repeat what you did earlier Provide the logs from the output panel (use the command to get to the output panel) Please note, I'll need two logs, one without any data frames, no plots, just print statements. The other is the one with dataframes and the like.\nHere is the text for : Even if vscode totally froze up while trying to run something in jupyter, I was still always able to ssh into the server no problem. Here is the Jupyter output from just the print statements: <details<summaryLog</summary</detailsHere is the Jupyter output where I print large dataframes and plots: <details<summaryLog</summary</detailsFew other things I've observed: (1) In the second log above, I see there are some lines like in it, but I was not always able to reproduce that output. (2) Here are some additional logs that I observed while it took 15 minutes to execute a cell with a simple print statement (but it was executed after cells with the dataframes and plots): <details<summaryLog from </summary</details<details<summaryLog from </summary</details\nThanks for the logs. Transferring to VS Code due to connectivity issues with remtoe SSH\nAlso having this issue with jupyter over ssh, in a notebook that has a decent number of plots.\nThis issue happened with me also (especially when i am insider a docker container inside the remote machine)\nI have the same issue, even when I freshly open a notebook with some outputs it is very slow and unresponsive. <img width=\"273\" alt=\"image\" src=\"https://user-\"I have a stable connection to the server: !\nSame problem, sometimes it works fine, sometimes it is unusable, stable and insider !\nAfter discussion in sync."}
{"_id":"doc-en-vscode-3bb3db59d90de7b3579405c9b5ed423b8ac75441333deb90b5e853bc209762f5","title":"","text":"Possibly add logging for every message that's only turned on if the user explicitly is told to (for debugging purposes).\nexperiencing similar issues when working on jupyter via ssh to an ubuntu server from an ubuntu pc\nALso had the issue where running a cell lags(or starts after some time) for jupyter notebooks when using remote SSH aws azure\nStill the same problem\nI also experience the same. I observed that: It happens with Jupyter notebooks, but not in the Interactive window. It does not happen if you connect to a Jupyter kernel that is not managed by VScode. It happens when your notebooks has many figures. I hope these may help diagnose the problem.\nSame here..\nAs an additional info to what enlightened us. Jupyter execution on remote ssh machine slows down (i.e., gets stuck) when notebook size is large (64 MB). VS code executing notebook cell normally for smaller sized (ie., 1MB ) notebook\nWould it be possible to share such a notebook? Is the size 64 MB with or without ouput in the notebook? If its a large file you could optionally email it to one of us (us = Microsoft maintainers of VS Code)\nsorry that is confidential work for the time being.\nExperiencing the same issue as well. Networking is not a problem since I connect to the notebook's machine on a local network\nI've also been having this issue lately although I think it only happens in certain circumstances. I can't share the notebook I'm working with at the moment as it's private (sorry!), but it does have many matplotlib figure outputs and dataframe outputs. The notebook itself is only 3.6 MB in size. I'm running the notebook on a Linux server with Python 3.9.7, specifically I'm connecting to the machine remotely with \"Connect to Host\" to edit and run code on the server remotely. Latest version of VS Code and all extensions, connecting from a machine running Windows 11. It seems connecting on the same network as the machine with my Linux laptop doesn't result in any issues, and I haven't noticed any problems with running the notebook from there. However, connecting from a different network when working from home causes the same notebook to be very laggy, including lagging for 30+ seconds on simple code cells containing e.g. just a print statement."}
{"_id":"doc-en-vscode-285e07bf16996fa9d3ad915ca9e552d07872402bee00df5938df0f3d2d855e3c","title":"","text":"I don't see how network speed/my connection to the server from home could be the issue (12ms ping, ~5 Mbps upload / ~50 Mbps download connections). Maybe there's something in the different OSes having different results, could also be unrelated though. I should try connecting with a Linux machine not on the same network and see if I can reproduce the results.\nSame problem here, just a correction it is not dependent on the figure number or commands difficulty, will take to execute 30 to 40s, and yes with a separate instance of Jupyter Lab server (not managed by VScode). Directly connecting to Jupyterlab over ssh port forwarding order of magnitude faster. Workflow: Windows to Linux remote SSH (cluster) Another instance in the cluster used to set up JupyterLab instance Change Jupyter server to remote of the previously created instance execution super slow (even delay between starting and finishing) ~ 30 second delay followed up by ~ 40 s execution Connecting directly from Windows tunnel to JupyterLab ~ few seconds execution Note: Both the Jupyter Lab instance and the initial remote ssh share access to the file system, just different hardware. Also after changing to remote, some shenanigans happen with the vscode, unable to open files, etc, had to restart everything to make it work again.\nI've had the same issue with VScode and Jupyter - notebooks becoming unresponsive after a few cells are used. The console shows continuous (every 20 seconds) reconnecting. with the level set to \"trace\". I can share a relevant file if it is helpful.\nI've had the same slow issue with VScode and Jupyter when connecting to a remote docker container via ssh, and I've also observed that keeps producing tremendous streams\nIf someone can share a notebook that triggers this issue (lots of outputs, graphs with lots of data, etc?) that would be helpful, thanks\nI noticed that if I switch the format from svg to png (by changing the setting ), the notebook would run smoother."}
{"_id":"doc-en-vscode-8704fe045554f06c40f8b6f5060f1e1c25ba6e5793b094e41b1109b7668f9982","title":"","text":"My network: Roundtrip latency: 5.520ms Up: 2898.6 Mbps Down: 3007.5 Mbps Authority: ssh-remote+** Roundtrip latency: 21.320ms Up: 25.8 Mbps Down: 385.0 Mbps A test notebook (1KB zip file), need , and to run:\nHaving the same exact issue and it renders the jupyter unresponsive: ! And it's still running. A temporary solution I found is to click on \"Clear outputs for all cells\" and more often than not it works smooth again. Only works if the cell stops executing though...\nI think notebook size may be the reason.\nI resolved my issue by closing the JUPYTER: VARIABLES in the terminal below Maybe in my case was an issue with the output variables inside the jupyter section ..\nI resolved it by using a jupyter server running on the remote machine which was not started by VSC. I'll use that until this bug is fixed unfortunately\nlife saviour! \"Clear outputs for all cells\" worked smooth as butter (:\nHappy to help. I saw in one of the other tickets this bug has a synergy with notebook size, and since clearing all outputs reduces notebook size, this might be a temporary fix until the size increases eventually and we need to do it again. I would love to see this fixed as this bug is the biggest hindrance to me day-to-day vscode experience.\nSame here, this issue was driving me crazy. That's the solution indeed, I had a lot of local memory usage in my code, not sure about what was the actual issue. EDIT: it seems I spoke too early. When a cell takes very long to compute, vscode seems to \"disconnect\" from the kernel, and any subsequenc cell evaluation waits forever without running.\nI pushed a change which should mostly fix this slowness with large outputs. I think there are more optimizations we could do. The issue comes when a save happens, and this blocks the remote extension host process for some period of time. Then a symptom of that might be that an execution appears to hang or take a long time. Would appreciate anyone trying this out in tomorrow's vscode insiders build."}
{"_id":"doc-en-vscode-789e8de54250f186e8143e03ed9e9609983dc64094f18c9c157d4508bf247cb9","title":"","text":"If you're seeing an issue related to the variables view, that wouldn't be fixed here, however I recently pushed a different fix that may help with that, and you could try it out in insiders + the jupyter pre-release extension.\nI have been experiencing these slowdowns (specifically on an AWS server) and I don't use the variables view. I definitely found that clearing cell outputs eliminates the hangs. I'm confused how this is caused by saving though, because I just found that the notebook I was working on wasn't being saved. I was doing some work tethered to my phone as I rode a bus into remote mountains. Certainly noticed that the bad connectivity meant even more waiting for things to execute than usual. Unfortunately, when I got back to stronger internet vscode was complaining it couldn't connect and was hanging when it tried to save. I figured it was fine since I saw 's comment and naively interpreted that as meaning I should feel comfortable that my notebook was getting saved each time it hung. Nope. When I finally had to reload window to get things back to working I found that all the code I wrote on the bus was no longer in the notebook (no worries, I had copied it out just in case, so no loss). I'm trying the 2023-01-13T07:49:05.588Z insiders build right now (is the \"tomorrow's\" build that should have the changes?). I think it is still too soon for me to be able to offer an opinion on how much of an improvement this fix might be. It is hard because the problem is erratic: sometimes cells execute immediately and sometimes it takes 20 seconds or longer.\nYeah, that version has the latest fix, you might have hit a different issue. And the issue might still have been kicked off by the notebook attempting to save. I'll leave this issue open to track namely, we still send notebook data between the processes more often than we technically have to, and I think that can be improved. Sorry about that, glad you had it saved, the intent is that the unsaved changes would have been persisted on the frontend.\nI can confirm that this change is a big improvement for my use case. If those additional optimizations ever get prioritized that would be amazing and appreciated, but I'm already grateful for this improvement. Thank you!"}
{"_id":"doc-en-vscode-5e7fc9f0e49cf55d8f60cfa388e462bc8d027aaddacbfddf73d01520561a1910","title":"","text":"On second thought, closing this issue, and moving other work to\nI have tried the insiders version on a podman container accessed on a machine over ssh (so I do vscode remote via ssh to the machine and then I use remote containers to access the container). It does not open a terminal and seems to be spinning forever with the following error in the logs: 2023-01-27 10:40:00.790 [info] [LocalProcess0][resolveAuthority(attached-container,27)][1013ms] waiting... 2023-01-27 10:40:00.818 [error] Shell server failed: Error: stream ended with:4 but wanted:9 at u (c:Usersvince.vscode-insidersextensionsms-vscode-distextension) at c:Usersvince.vscode-insidersextensionsms-vscode-distextension at s (c:Usersvince.vscode-insidersextensionsms-vscode-distextension) at Socket.<anonymous(c:Usersvince.vscode-insidersextensionsms-vscode-distextension) at (node:events:538:35) at endReadableNT (node:internal/streams/readable:1345:12) at process.processTicksAndRejections (node:internal/process/task_queues:83:21) May this be related to the recent changes or shall I seek for another bug report and eventually open a new one?\nNot related to this change- you can open an issue against the Remote Containers extension, I'm not sure how to interpet that error\nHello, I have installed latest Insider's VS Code, maybe it's a bit better now, but in general issue is still there. Very slooow Jupyter Notebooks on large content. I'm using jupyter/r-notebook docker image to run on AWS EC2 instance and have 2 options to acces it: from browser on 8888 port to web version of Jupyter notebooks. It works lightning fast and responsive. over SSH to EC2 instance and then \"Attach to running container\" So if I compare those 2 options there is a huge difference in UI when notebook has some data and charts. For me it feels like VS Code continuously sends data to server. Even dropdown for intellisence takes few seconds to load."}
{"_id":"doc-en-vscode-9ebb3f52b85d79210668c2df6b30611743e9c7ee27e2683ea00e4ddd03ba7398","title":"","text":"It's so sad that I'm unable to use VS Code Jupyter Notebooks for my work. But if I create a brand new empty notebook in VS Code - everything works like a charm!\nWe have experimental saving logic for Remote SSH, it would be great if you can give this a try and see if it improves the performance on large notebooks Install latest VS Code Insiders Remote SSH into your remote machine Add to your Remote/User settings Reload the window Then test the scenarios which used to slow or block the network Thanks in advance!\nThe improvements are noticeable, thank you for your great work! However, the latency is still there and appears to be significant from time to time. I love this ssh-notebook function, and I believe there must be more folks like me. Hope your team can prioritize this issue. Appreciate!\nHi, I just checked VS Code Insiders 1.81.0 version with \"Remote Save\" feature enabled in settings. I have generated in a loop 30 charts in first cell and after this I run code in a new cell. It still hangs for a while after generating charts for the first run. If I run second cell code again, it executes immediately as expected. So we still have an issue when a lot of data (or visual data) was generated recently. I checked the same code in web UI of Jupyter - and it works fast. Anton. From: Peng Lyu Friday, June 30, 2023 12:33 AM To: microsoft/vscode bo44arov Mention Re: [microsoft/vscode] Jupyter over remote ssh sometimes becomes slow and/or unresponsive (Issue ) have experimental saving logic for Remote SSH, it would be great if you can give this a try and see if it improves the performance on large notebooks Install latest VS Code Insiders Remote SSH into your remote machine Add \"notebook.experimental.remoteSave\": true to your Remote/User settings Reload the window Then test the scenarios which used to slow or block the network Thanks in advance! Reply to this email directly, view it on GitHub, or unsubscribe You are receiving this because you were mentioned.Message ID:"}
{"_id":"doc-en-vscode-3d016a5e7ddbd3c92fe39cb66166c2af0636ce1a21c77d0ee7149250f7f0731d","title":"","text":"Issue Type: <bBug</bKeybindings GUI (Sometimes optional) Scroll down/up over empty when columns shows when contexts from some other rows VS Code version: Code - Insiders 1.31.0-insider (, 2019-02-01T18:34:04.940Z) OS version: Windows_NT x64 10.0."}
{"_id":"doc-en-vscode-bef2b9e14d760e0c1c7bf08f377861be42921b5affd2a287518732931e83373f","title":"","text":"Chinese (Simplified) : <img width=\"321\" alt=\"image\" src=\"https://user-\"English : <img width=\"272\" alt=\"image\" src=\"https://user-\"\"Create Manual Folding Ranges from Selection\" is not in Command Palette when use Chinese (Simplified) VSCode Version: 1.71.0 (both macOS & Win10) Chinese (Simplified) Version: 1.71.\nThank you for your feedback. We are currently reviewing it (reference ).\nCould you please assist here? Thank you.\nThe problem is this label: should accept an  but it only accepts a string. deeper: and deeper:\nThe workaround is to use  to find this item."}
{"_id":"doc-en-vscode-4f26d5423de665764064d5fbc9da13cb1f5ce9a6d27a67eb50c2d1cdd573ba0e","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --This problem occurs on the vscode server. Here is the version information. Version: 1.91.1 Commit: Date: 2024-07-09T22:08:03.155Z Browser: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/126.0.0.0 Safari/537.36 Steps to Reproduce: I'm not sure if this issue has been fixed yet. I reproduced the problem on the vscode client, installing the plugin, and connecting to the remote. My system is Windows OS. ! I'm trying to reproduce this problem. On a windows server, the fsPath is spliced by . When verifying, The authority of the URI is localhost:8080. Therefore, the verification fails and an error is reported in the search box. I took a look at the code and changed fsPath to path in line 257 of the file. The query works fine. But fsPath is still concatenated with . I would like to ask, what is the difference between fsPath and path? When does fsPath use to concatenate. Looking forward to your answer. !\nI found the reproduction scenario of problem three. This is the IDE version information."}
{"_id":"doc-en-vscode-1c33ee08784437cad2654c56c77cc53edd42ac682601f20fea773852e0a65bce","title":"","text":"Version: 1.91.1 (user setup) Commit: Date: 2024-07-09T22:06:49.809Z Electron: 29.4.0 ElectronBuildId: Chromium: 122.0.6261.156 20.9.0 V8: 12.2.281.27-electron.0 OS: Windows_NT x64 10.0. Steps to Reproduce: the vscode and remote-ssh extension to a Remote Server any project a file that is not in the current workspace the search view and select Search only in open Editors button Keyword !\nCan you file separate issues for these three? Or at least a separate one for the first two? I'm not sure if I understand what you're saying here, can you rephrase? What OS is your remote? I'm assuming it's windows if the path uses path separators?\nOkay, I've submitted questions one and two separately. This means that you open a file in the editing area that is not in the working area. You can refer to the file opened using the Open File... menu in the gif image. My remote is linux.\nAlso, please confirm if this issue is also present on vscode server, I am seeing this issue on vscode server as well\nHmm, yes, I can reproduce this issue."}
{"_id":"doc-en-vscode-ab53ade72394fb3b6eb698e1a26e1795b2866c3b8a2a2a70edb9f81acc1d02b7","title":"","text":"Testing Issue Type: <bBug</bextension search box, type Bug start showing up instead ! VS Code version: Code - Insiders 1.26.0-insider (, 2018-07-30T05:11:25.700Z) OS version: Darwin x64 17.7.0 <details<summarySystem Info</summary </details<!-- generated by issue reporter --\nwhat should we do in this case? We aren't able to change how these match. We could disable fuzzy finding entirely"}
{"_id":"doc-en-vscode-8ad5afbabd1bf964bae7ee00be6958460382c249eb3b1f453a4645336b2f0d1b","title":"","text":"Have file named in explorer Copy/paste it Window hangs This is from We shouldn't attempt this when the number is = Number.MAXSAFEINTEGER. It loops forever here: This bug will happen whenever incrementFileName fails to modify the file name.\nI guess only interesting thing about this what we should do in this case as we need to create some new file name? Add without changing the original number?\nthat sounds reasonable"}
{"_id":"doc-en-vscode-c391d3f0b27331369f3de99115052de9c84b35d38da5c7853baf263be7a62c3b","title":"","text":"All of a sudden, I no longer have GHPRI installed. When I go into I see Settings Sync removed it 11 hours ago, but I'm 100% sure I never uninstalled it. I know I have Settings Sync enabled in both and are those scenarios doing unexpected things to my settings sync? !\nI am sure that 4 days ago I re- the extension. Today, I'm on another machine, and GHPRI is again gone.\nLets sync on this\nClosing as I couldn't repro.\nI was able to reproduce this issue. will sometimes remove and then readd the GitHub Repositories, Issues & PR, and remotehub extensions, except it doesn't readd the keypair to the config, so when I return to my local installation the extensions will sometimes disappear. Not sure why it doesn't always reinstall though. Sometimes won't readd the missing extensions in time, so it gets removed. When the extensions are without the keypair, the local machine can occasionally readd the missing config and keep the extension installed; other times it won't and the extension is removed. Local machine logs from yesterday: Extension config history: Remote removes extensions: Remote readds extensions, but not marked as installed: After I reinstall the extensions locally: logs from a repro today: Extension sync history: Removed extensions: Readd extensions:\nThanks for logs and info. Reopening for investigation.\nI am able to reproduce this with following steps: 1: Install GHPRI pre-release extension and sync .dev: Trigger Sync. Note that this is a builtin extension here. 2: Trigger Sync - Make sure GHPR extension is installed. 1: Uninstall GHPRI extension 2: Trigger Sync - Make sure GHPR extension is uninstalled. .dev: Trigger Sync. Make sure GHPRI extension exists in builtin extensions. 1: Trigger Sync. Nothing changes. 2: Trigger Sync. Nothing changes. This step and above step are important because, even though nothing has changed, these machines update their last sync state to have GHPRI extension has builtin extension. 1: Install GHPRI pre-release extension and sync 2: Trigger Sync GHPRI extension is not installed. Come back to Desktop 1: GHPRI extension is removed.\nImproved extensions merging to make sure the builtin extensions are handled properly."}
{"_id":"doc-en-vscode-e1ec9f681715a74da30dd29e044030a4074a3b4f301c3a3a8f6beab7ebe58890","title":"","text":"Each machine remembers what builtin extensions it syncs and make sure it won't remove them while syncing. Also some logs to get more information if the issue still occurs.\nIf I uninstall GHPRI from desktop and then Sync Now in , the extension appears to be 'uninstalled' on , is that expected? ! After reloading the extension is still installed though.\nI don't have a second desktop to verify with at the moment.\nYeah in your case it is expected because, GHPRI extension seems to be user installed and settings sync is removing that."}
{"_id":"doc-en-vscode-ed60c383789b1f157617e5c07f663fef924a0ad4d3d93b534f9782eef7efc92d","title":"","text":"Steps: extension that registers a document and a document range formatting provider open a test workspace that activates the extension and trigger a range format. You get:"}
{"_id":"doc-en-vscode-e927c498c62a6c3a8c2cc19e10fba65531ca4e4dc9ef3718733d295af08ac31c","title":"","text":"run open in your favorite browser add a breakpoint at notice how the breakpoint is hit every few seconds this ends up transmitting data from the server to the client\nA heartbeat fires every 5 seconds so we can detect when the pty host becomes unresponsive: Open to ideas on how to improve this but it's working as designed right now.\nIdea: determine if the pty host is responsive or not on the server process only communicate the status of the pty host from the server to the connected renderers when the responsiveness of the pty host changes.\nThat should work, we only care about the event when it's going from unresponsive-responsive anyway."}
{"_id":"doc-en-vscode-46c1b4bdd13015efbd6ec63322456a53ce84f9ff6613725b853609341da4d14f","title":"","text":"Seeing 2 test failures today for integration tests in Web ():\nfeel free to skip them if needed, they're known to be flaky. This was my top priority for debt week but it was completely consumed with the settings sync preview.\nyou mean flaky in general or only for web?\nFlaky in general, it seems especially on web though.\nWe can leave them running, I am not seeing them failing often, just found this one. It is not something to look at now, but eventually. Will revisit if I see this again more frequently.\nok seeing them fail again today, pushing to disable them when running in web via checking for . Your tests seem to have a questionable use of , I see one error being that is called multiple times. Why do you not rewrite to use async and promises?\nAnother one:\nI ended up disabling them all for web:\nMaybe this has now leaked from the tests into the code itself. I'm creating a new terminal within an extension using the API and give it a pseudo terminal. Sometimes the terminal is shown but no output is displayed and does not work. But I can also see no error or trace. This happens for me in vscode 1.44.2 using linux.\nIs this still an issue?"}
{"_id":"doc-en-vscode-57a9f64b254c68b9266597db911bac4b1697c673092e4bbf998b5ecf83ca22b5","title":"","text":"VSCode Version: Code 1.15.1 (, 2017-08-16T18:07:25.676Z) OS Version: WindowsNT x64 10.0. Extensions: ExtensionVersion ------ overtype0.2.0 python0.7.0 theme-dark-plus-contrast0.1.101 cpptools0.13.0 monitor-file1.0.1 (4 theme extensions excluded) Steps to Reproduce: there is such line in a *.bat file rem set SERVERVERSION=5.7 cursor on it, and uncomment with \"Ctrl+/\" expected: set SERVERVERSION=5.7 actual result: REM rem set SERVERVERSION=5.7 Looks like uncomment is unnecessarily case-sensitive. It should uncomment despite it is 'rem' or 'REM'. I don't care whether with next \"Ctrl+/\" it will add Upper-Case 'REM' instead of 'rem' - it is not problem. <!-- Launch with to check. --Reproduces without extensions: Yes/No"}
{"_id":"doc-en-vscode-75d46fb86d9f416d4dd0cb1b9d10aaad7849a8903abcac45602f44606f4a5928","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: Version: 1.74.3 (on ) OS Version: Linux, Gnome, Chrome Steps to Reproduce: an image in the media-preview extension click on the image and select copy to paste the image somewhere Nothing happens. Also note that Cut and Paste make no sense here as menu items. I assume that these are just the default context menu options from - perhaps these should be removed entirely from the preview extension and we could manually implement copy?\nHey I would like to work on this issue.\nassign this to me\nAre these the correct steps to reproduce the issue? I'm unable to verify it on my mac. Can you provide more information? I previewed the image and then copied and pasted it. <img width=\"950\" alt=\"image\" src=\"https://user-\"\nI would like to wokr on this issue.\nI would like to try solving this issue.\nYou right-click an image and select \"copy\", then you click a folder in VSCode's explorer and select \"paste\". The image should be pasted into that folder but it doesn't.\nI would like to work on this issuse. here, I tried to copy and paste (Ctrl+v) it. But it doesn't work. https://user-\nAfter a day of researching, I think I can explain why this issue happens."}
{"_id":"doc-en-vscode-8890bd7eb39ee7986f1dcc1b7f516ab867a4f71fdd63110feeb7c4e5123adbfa","title":"","text":"Basically in the core webview module (see ), three default context menu actions (cut, copy and paste) are to the menu, which are accessible throughout all webviews, and happens to be one of them, so after you right-click anywhere in a image previewer, you will be able to see these three actions. After you clicks one of these actions, the editor will post an message to the image previewer, telling it to cut, copy or paste. If there is a correspondent listener for this perpose, the image will be actually cut, copied, pasted, but unfortunately there is no such listeners defined. This problem has existed for quite a bit of time. The original modification that cause this problem is from here . It is origianlly meant to allow users to copy selected text within an extension's README page. There is also a closed issue that describe such issue in . Personally I think providing action cut, copy and paste for all webviews is a little bit hasty, since there are not universally required for each webview, and even some of them need, the subject to be copied may vary from different webviews, or even different places in a same webview. I can solve it by disabling the default actions first (there is a option to achieve this), then let image previewer to implement its own copy and cut functionality, but I think this is rather a quick and sloppy solution. There should be another better way to solve this issue.\nWhat about adding a handler :\nHere's what I tried VSC insiders without any extensions Open a folder with a .png file Open the png file in vscode Right click and select Paste into folder in vscode = does not work Paste into markdown cell in vscode = does not work What am i doing wrong?\nPasting image data into folders isn't supported by VS Code. Pasting into a notebook cells works fine for me though. To verify though you can also paste into an external image editor"}
{"_id":"doc-en-vscode-bf7b7502aa7ea902a186559542b99b64c850974bee6956d9877d1fd23ee32d04","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.75.1 OS Version: macOS 12.6 Steps to Reproduce: any CLI utility that displays a loader animation that uses braille, such as every step of the animation, you should see an array of dots that don't use up a full square space, but leave some blank space and look like a \"snake\" (this works fine in Windows) This is how it looks in other terminals such as iTerm2 or <img width=\"33\" alt=\"image\" src=\"https://user-\"This is how it looks in VS Code (i've tried changing fonts, it doesn't help) <img width=\"31\" alt=\"image\" src=\"https://user-\"This is the character: On Windows version of VS Code, this behaves correctly on macOS, instead of white space it shows circles with transparent filling\nany news?\nHello, it's been half a year and it's been tagged many things, I'd love to know if this is being taked into consideration since it's very bad UX\nInteresting that this works on windows but not mac.... This issue reproduces in the monaco editor\nit looks like for we are using the python3.9 spinners\nI found this - have you tried other fonts?\ni found out that it's Chromium's fault, the issue happens in Chrome for macOS and every other child browser, but doesn't happen in Firefox nor Safari, so I opened a bug report there\nLooking at it appears that whatever font is being fallen back to is adding the empty circles?"}
{"_id":"doc-en-vscode-bd2998762ae30057e9e05b46bd52afd027da04aa32b5685abb0dec11e9777d6b","title":"","text":"If that's the case the fix here would be one of: Try falling back to something like in the font family setting, eg. Support custom glyph rendering for those characters, I really would rather not supporting this unless we absolutely had to though.\nThis bug is still present in Version: 1.84.2/macOS 14. I tried using Symbola, SF Mono, Menlo and Monaco as the main fonts, without success.\nI double checked today and it looks like Chrome Version 127.0.6533.89 renders Braille characters without hollow circles on MacOS. <img width=\"952\" alt=\"Screenshot 2024-08-02 at 10 50 29\" src=\"\"(I have a macOS app that's rendering the top bar differently)\nSo... I spent a bit too long today investigating what was happening here and here's how I've resolved this issue for myself: Before / After <img width=\"163\" alt=\"Screenshot 2024-08-07 at 22 22 32\" src=\"\"<img width=\"163\" alt=\"Screenshot 2024-08-07 at 22 22 42\" src=\"\"In case the diagnostic steps are useful to someone, here's the portion of my search tree that resulted in this solution / bodge.... I looked through my macOS font book for which fonts were providing symbols for (U+2879) and ended up finding \"Apple Braille\" and \"Apple Symbols\". \"Apple Symbols\" had the wrong amount of whitespace when I set that as the fallback font and the \"Apple Braille\" looked like a drop-in replacement. Turns out, Apple Braille looked different in the OS's Font Book vs in VS Code. The reason for this ended up being that Apple Braille has multiple variants and... for some reason... the version of Chromium being used in VS Code uses either Outline6Dot or Outline8Dot variants by default (these have the hollow/outline circles) instead of the Regular variant which does not have hollow circles. I assume this has to do with the font-renderer-level behaviour. None the less, the of the specific font variant within the \"Apple Braille\" family that I want to use was \"AppleBraille\" and adding that to the stack makes it use the correct-for-me variant of the OS-provided Braille symbols!"}
{"_id":"doc-en-vscode-39dbe8efde7d998f46be470b91b82be0fccc7b36ab65cfca2b9b7e37f4f8903b","title":"","text":"Alternatively, you can also do what I did recently and change to a font that includes braille characters (like ).\nI considered creating an issue for this for the editor as well, but the visualize progress isn't as important there and it's more possible people may want the filled/empty circle rendering in the editor to more easily identify which circles are filled.\nThis failed the macOS build when merged:"}
{"_id":"doc-en-vscode-d02530010d441cc0fd8b0e0cdb73a124819c8beefd04f639592c0e020281004f","title":"","text":"We should also take out the \"Terminal\" menu when using the custom menu: ! //cc\npushed a fix for this only to the release branch. What we do for master is up for discussion. please verify once the build is out in 30 mins. Thanks\nThanks I didn't know about this code."}
{"_id":"doc-en-vscode-68d72fca28452235d46883ce2d01ce1746864d809f352dba50b92992f0da58f3","title":"","text":"VSCode Version: 1.8.0 OS Version: Windows Version 6.3.9600 (Windows 8.1 Enterprise) Steps to Reproduce: High Contrast on Windows VS Code, look at the glorious High Contrast theme a different theme, because you would rather work in that VS Code, your theme choice gets overridden. Essentially, I'd like to be able to disable the auto-detection of the High Contrast theme I'm using in Windows via some setting, because while I prefer Windows in High Contrast mode, I dislike looking at VS Code's High Contrast theme.\nI wholeheartedly agree. I'm okay with the default being auto-detection, for the accessibility scenario, but it should definitely be possible to disable this functionality via settings.\nAn option makes sense to me. If someone wants to provide a PR\nDone,\nCould you please re-open this issue? Unfortunately, my PR only affected the high-contrast swap that occurred when changing Windows theme while VSCode is open, not the initial theme that is set when VSCode starts, or opens a new folder. I have a new PR that completes the fix.\n[x] To verify (Windows): Normally VS Code switches to HC mode automatically when windows is running in HC mode. Verify this new setting can be used to prevent this automatic switch from happening (both on startup as well as when Code is running and you change HC mode in Windows)."}
{"_id":"doc-en-vscode-62013da4f4c85e308647e475a45a7a2388425848eb7b8553243dd1bbc4aa825a","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --- VS Code Version: 1.56.1 OS Version: Windows Steps to Reproduce: a (reasonably long) markdown file and scroll down a bit. the dynamic preview for the markdown file. It opens the preview at the top of the file, but should open it where your text editor (containing the target markdown) is scrolled to. This is not adjusted until the user scrolls or resizes the window. <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --"}
{"_id":"doc-en-vscode-e7264969dde688e46c8990494ff5f8d5271fb21cf57e4b61c25ed6ac877cf4ca","title":"","text":"One desired behavior of writing a custom is that when the user accepts a given suggestion, the provider is immediately triggered again for another suggestion. We need to track when users accept (or reject) suggestions so that we can measure improvements that we make. Without using the experimental API, the best way to track if a completion is accepted is to provide a value for the field in . When supplying a command, the retriggering behavior is disabled [1]. As a workaround, I tried to indirectly trigger an inline completion via the command , however, it does not work (not sure why, this does work with my unit tests but not when I try to call it from a command I supply via ). [1] I believe this is the code block that retriggers automatically, one can see that the retrigger only happens if the command is the default command:\nNote is related.\nNote that [1] points to the autocomplete source code, not the inline suggestion feature. I'm not aware of any disablements.\nOk, that's incorrect -- however, adding a 'command' to disables the automatic retrigger.\nPlace the cursor after and press enter. Then accept. Notice that is not suggested.\nThat monaco repro / editor is really cool\nThanks! The best thing is that you can use the \"Monaco Editor Playground\" launch config in the VS Code repo to debug the playground example :)\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!"}
{"_id":"doc-en-vscode-983a2412360bf360239b2635de19c65e7d279019a863783e1993b8211599ef0a","title":"","text":"Using the workspace TS version, 5.5.0-dev.: ! ! !\nClosing for verification"}
{"_id":"doc-en-vscode-79a1091d98d09cee43c610381fd148a5404088152af71aa38349f4d12cb885e2","title":"","text":"Diffing the files: original: modified: <img width=\"1127\" alt=\"image\" src=\"https://user-\"It would be better to line up things differently and give preference to having lines with smaller indent be boundary lines. fyi\nVery nice idea!! Very valuable example. I can implement this in the experimental diffing algorithm, and soon we can discuss to deprecate the smart diffing algorithm ().\nImplemented in the experimental diff algorithm (not smart). Let's try to deprecate the smart diff algorithm in favor of the experimental in the next milestones.\nClosing, as this is implemented in the experimental diff algorithm, which is now the default. To verify, try it out in the playground:"}
{"_id":"doc-en-vscode-0fe708a240a64c1e376ae00e8e3e678584f0e07c8d1275b42ffe37ff9574cf72","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.44.2 OS Version: macOS Steps to Reproduce: an extension providing in-editor comments via the comments API an editor containing comments the terminal at the bottom a command which causes the comments to be updated. For our comment provider, we automatically detect comments associated with your source control commit. So, for example, use the command line to checkout a different commit. comments will change, causing some of the previous comments to be hidden. When this happens, the editor is focused. This causes you to lose focus on the terminal. The end result UX is that you're typing commands in the terminal, and your focus gets stolen to an open editor for no visible reason, making it annoying if you wanted to keep running more commands. Notably, this depends on the way the comments API is used. I believe many comment providers do not respond to source control, and require clicking a button to activate/deactivate, so it's likely that's why this hasn't been reported before. For us, this is really annoying, and I could see this hitting other providers in the future. <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: No, because you need an extension which contributes via comments API\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.562 --- <!-- score: 0.513 --- <!-- score: 0.501 --- <!-- score: 0.493 --- <!-- score: 0.492 --<!-- potentialduplicatescomment --"}
{"_id":"doc-en-vscode-06b919a6de9ade426f217af83d34805e1c52b24498dc7d08c920969eb52cc4d2","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.86 OS Version: macOS 14.3 Steps to Reproduce: run profile programmatically with isDefault = false seems to restore the previously selected state \"isDefault\", or if there is none, makes the first profile \"Default\". behavior works for me. But unfortunately after the profile is created and before its state is changed manually in the UI, it reports isDefault = false. Expected: isDefault matches the UI state at all times. //\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.86.2. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!\nThis works for me, but note that the synchronization of the new profile is asynchronous after its creation. This is intentional because it's common for extensions to create a profile and then set its properties, like we don't want to cause that to override what the user previously set (especially since making isDefault a 'live' property was a later addition to the API). For me, adding a listener like triggers shortly after its creation\nFascinating! I can't reproduce this bug today.\nIt seems like there is some inconsistency here\nA simple snippet like this surfaces the issue for me: The console is saying And the GUI has run 1 as (Default)\nLooks like this is working now:"}
{"_id":"doc-en-vscode-e3719f72d237b4a4422394b52588f731696eecef6944c94a5ee81dd8f580e7ea","title":"","text":"Add the possibility to persist the content hover sizes\nTo test, please read the following. Currently the implementation of the persistance mechanism of the content hover works as follows: The content hover has a default maximum size which is defined as follows: const defaultMaxHeight = (().height / 4, 250); const defaultMaxWidth = (().width 0.66, 500); Whenever the content hover is resized, the last hover dimensions are stored. Next time the content hover is shown, the maximum dimensions are updated as follows: const currentMaxHeight = (().height / 4, 250, lastHeight); const currentMaxWidth = (().width 0.66, 500, lastWidth); The content hover is rendered naturally within these constraints. If the lastHeight and lastWidth are smaller than the default max dimensions, then the current max dimensions are unchanged. Therefore the next rendering of the content hover will be the same as the initial rendering. Test that the content hover works according to these constraints. There is some discussion going on about how to better implement a persistance mechanism. We could also persist the size for example on a per-token basis in the editor. Let me know what you think is best!\nDo you have an example of how it should work? Regardless of how I resize it, it is not remembered.\nYou may see an example of how it works here. The content hover is resized to a bigger size in the gif. The current max width is increased to a higher value and at the next rendering of the content hover, the widget takes on a bigger size. To see this behavior you should choose a content hover where the content does not fit in the visible widget, where there is a horizontal scrollbar. I opened an issue for persisting the content hover on a per-token basis here: as an alternative idea."}
{"_id":"doc-en-vscode-d6bb527da40d2fd5e284052b6646988682fd718fcbbf387a7f38c4c50e778abd","title":"","text":"deferring from as I spent more than an hour on it\nThis is a little different to how we do it in the terminal because the command palette entry: is registered separately to the command: also doesn't show up at all as a command anymore: Insiders: ! Stable: !\nCommenting out the schema fixes the problem for runTask and the commands below.\nhere needs workspace trust:\nremoved because the change was reverted so it doesn't work"}
{"_id":"doc-en-vscode-0990a780289dc27fe00628fb4790ae19e9a555b08d1b827fa9dd84adc6cb8851","title":"","text":"reported by the aria label for search results should read content first then location IE line/column info ! cc for historical context\nMakes good sense, and should be an easy fix.\nInteresting - this works as expected on macOS\nWhen replace is disabled, we have a label that looks like Otherwise, if replace is enabled, we have: Would you rather see something like for non-replace and for replace? Not too sure why this would be different on mac. What does the aria say on your machine?\nThanks for looking into this. I have not tried on Windows, but when shared his screen (on Windows), the announcement read the line/column info before the actual content. Do you know where that would be coming from?\nshould be fixed in\nI just tried on Windows and cannot reproduce the behavior that demoed - in which the line/col info was presented before the actual content. We will need to hear from him if it still happens and if so, how to reproduce.\nDid you just try on insiders? I just changed the functionality\nand Note: is redundant b/c that's the user input search term and users already know what they are looking for, and it takes longer time for screen reader users to hear the actual content.\nthanks for clarifying. Re The reason for was to distinguish between that and a . Do you think that we should keep the part of or get rid of both given they require the user's action to get into that state? Another option would be to append to the end of the message.\nThe reason that we state the match again is for regex users. If they have a fairly general regex, they might want to know what actually caused it to match. Although we could also argue that the line context is more commonly valued by search users, since many people don't use regex.\nThanks for that context I agree that we should optimize for the most common case here.\n-- I recommend taking out redundant prefix, including 'Replace'. This is similar to one of the recommended best practices creating alt text. We recommend avoiding any redundant words, such as \"Image of \" or \"Graphics of \" b/c users know what they are interacting with.\nUsing the example from before, we could probably do something like 'VSCode is cool' found in match 'VSCode' in line at column 15 when replace is not active. What do you think the replace case should look like?"}
{"_id":"doc-en-vscode-93c2b50f7d9be5851381062b32e59fe25685e64ab324480bb3678a28e3cd3c59","title":"","text":"The user will know if they're triggering a or , so the idea is to remove both of those terms. So, for search: for replace:\nSince the replace hasn't occurred at this point yet, wouldn't be past tense in the results. We can say, instead, but is that too wordy? Also, I feel that the match string is still useful here for those who search for regex, so perhaps that could go at the end? Or we could just read it if they have regex search enabled.\nfyi, 'm going to move this to that may iteration because I will be offline for the rest of the April iteration.\nI might fix it for you\nIt sounds like line number is still expected from the comments here, but only the column number is stated. e.g. 'matched context' at column 34 found (1) 1 of 1 Also, finding a match in the first column will state that it was found at column 2\nI see why you were confused but is text not a line number there. So this is working as designed.\nnot sure what you mean by this\ncreated since it seems like a different issue"}
{"_id":"doc-en-vscode-01ac1020cde12da61cd7df151c863212bd492d2508f0046b4249637b185eec90","title":"","text":"Currently in VS Code when users open a notebook document, they don't need to pick a kernel manually only in following two scenarios: The notebook was opened before and users run the notebook against a kernel already. That kernel will be picked automatically, no other kernel would replace it unless users pick one manually. There is only one kernel contributed to this notebook. When users run a cell, and there is still only one kernel, the kernel will be used. Before users run any cell, and there are more than one kernel, the kernel picker will not show any active kernel info (it shows \"Select kernel\" instead). Above two scenarios cover two use cases: open an existing notebook, or work on single controller notebook (e.g., GitHub issue notebook). However this doesn't cover two common workflow open their existing Jupyter notebook (created through other Jupyter frontend/service) in VS Code create new Jupyter notebooks when they have already specified their preferred kernel Especially that VS Code is still growing to be one of data scientists' Jupyter frontend/editor, almost every new users will run into workflow 1 and do manual kernel picking though often Jupyter has a good guess which kernel is preferred for current document, and also every existing would run into workflow 2. The concrete issues reported: [AZML/Open in VS Code] on AZML website, click open in VS Code, it opens a remote window and opens the notebook document in the editor. No kernel is not picked though users already specify one in AZML website. notebook A in File Explorer, pick a kernel. Create notebook B in File Explorer, the kernel info is back to \"Select Kernel\" again. untitled notebook in Editor Area, pick a kernel and run. Save as a new notebook document on disk, the new document has no selected kernel. . Create a notebook in JuliaLab, specify a kernel. Open the notebook then in VS Code, preferred kernel not selected. Proposals For AZML/Julia users, they are mostly working on single language notebook (other than polyglot), meaning the notebook serializer is the ipynb extension we ship out of the core, and kernels/controllers are contributed by a single notebook extension installed from the marketplace (Julia, or .NET, or Jupyter). If this is the case, we can pick the kernel whose affinity is the only highest without prompting users to pick one."}
{"_id":"doc-en-vscode-2ee58cf3fcc317103467b5e27328f89c529fb665e07939394ad88617ebd0c53a","title":"","text":"cc\nHad discussions with about this. In addition to the proposal mentioned above (optimizing the single extension contributing multiple kernels scenario), another idea came out of our discussions: Allow users to choose the default kernel. One example is users have both Jupyter and Julia extensions and both of them contribute notebook controllers for Julia execution. Users might have their preference for the default kernel, allowing them to specify which kernel should be the default can help avoid users picking the default kernel in every newly opened notebook.\nI thought about filing a new feature request on this. But it seemed better to keep the discussion contained here, at least for now. In particular it relates to the last comment about selecting a default kernel as well as the Gear icon for the notebook controller. Speaking personally, the majority of my issues with kernel selection would be alleviated by the ability to select a kernel to be my default, auto-selected kernel at both the global level with the ability to override it at the workspace / resource level, just like the rest of VS Code settings. All the rest of selection could work like it currently does. But I want the ownership to be able to explicitly specify kernel selection with some type of manual affordance that always trumps any type of heuristics or metadata or forcing a manual selection. At a surface level I like it as just a setting, but the issue here is that controller IDs are not something that's exposed to the User. The Gear icon might be the right place to surface this, and might have even been what the Gear icon did before (it does nothing now for me in insiders).\nAgree 100%. Even though we could save the controller id to settings when users use the gear icon, it's not great as users won't be able to read or tweak this specific setting as the controller ids are not always human readable. With that said, we could tackle this differently to archive the same thing, instead of asking controller id, we could ask users if they want to use the lastly used controller, use a controller suggested by extensions always, use a controller which matches the workspace stats the most (use Julia controller if the workspace stats are all around Julia files).\nYeah, that's the challenge here. Controller ids aren't nice to read and not guaranteed to be constant. A better concept is user state, e.g mementos, which also roam across machines."}
{"_id":"doc-en-vscode-a8a166f95cb4857d75ac1d64fdca0f4112c693a9f0e635f8097e89359ad6570a","title":"","text":"Something already exists but it isn't utilised properly and might need further tweaking. The knows and persists what kernel was used most recently for what . Maybe it's enough to make more brave kernel selections based on that information"}
{"_id":"doc-en-vscode-79e6d23efc88e535558f833e5906fae6a1273f419d6f969551118b9f82295007","title":"","text":"When I press Ctrl+Shift+P to open the command panel and type \"\", it shows \"No matching commands\". So I can't execute this command by Ctrl+Shift+P. I can only bind a shortcut key to this command and then execute it via the shortcut key. I think this is a bug that the command panel search is incomplete.\nPlease edit this issue title to become:\nindicates it may already be there and you just need to know what the command text is.\nOK, but according to my observation, there are other commands that cannot be found in the panel. For example, \"\".\nThe issue I raised is different from him. His issue is that the shortcut key does not work after binding. My issue is that I can't find the command in the command panel. In fact, some other commands cannot be found, indicating that the search of the command panel is incomplete.\nSome commands never appear in the Command Palette because they define their clause as\nI open the page where I set the keyboard shortcut and see that their when clause is - not false.\nSorry, I wasn't clear. This is what I am referring to.\nI read your link and I probably understand. Please set the when clause of \"\" to \"inDebugMode\" instead of false. This is more in line with our needs.\nLooks like that command is missing from here: is also missing.\n/confirm\nI will submit a PR\nThanks!"}
{"_id":"doc-en-vscode-8a0be80bd127cae7bcbd16f9df95ec4052115f8542e50920149321bc2b4fa98d","title":"","text":"In the following dialog that appears: <img width=\"696\" alt=\"image\" src=\"https://user-\"in and opening a protocol url, the 'Install VS Code Insiders' button should route to a new link for installing insiders specifically rather than just Install VS Code Insiders should not dismiss the dialog, so that the user can return and try again cc\nlikely a separate issue: Was not sure how this modal was arrived at, so I clicked the remote indicator and I only saw an option to download VS Code - not Insider's as I'd expect. !\nI then did the flow and saw the dialogue. Clicking on the button took me to the Insider's download correctly. !"}
{"_id":"doc-en-vscode-cf1bc07c825af4523ce1e23afaf1fa249b37fff7c10acf0766e17030cbcdd1ad","title":"","text":"Open some See that it's in the \"Open Recent...\" list Close vscode Delete that code-workspace Open vscode Use Open Recent to select that deleted workspace Get a message that doesn't exist anymore See Open Recent again - the workspace is still in that list If I follow the same steps with a folder instead of a workspace, then the folder is removed from the list as I'd expect.\nI cannot reproduce, tried with insiders on macOS. Here is what I did: open a workspace saved it as another workspace on my desktop closed vscode deleted that workspace opened vscode (opens empty) try to open that deleted workspace via file open recent dismiss the dialog try to find the entry again =it is gone Since this involves a menu update, could it be that you opened the menu too fast for the menu to update? I think the current delay is like 800ms?\nInteresting, I am using the \"File: Open Recent...\" command. When I use that, it isn't removed from the list, but when I use the File Open Recent menu, it is removed.\nI can reproduce and it seems to be a regression from the work to support URIs on the main side. The flow is: opening the workspace fails and we call remove from recently opened () since we pass over a URI we think this is a single folder workspace (which it is not) and fail to remove it moving to (cc\nThis was the change from adopting URIs for files."}
{"_id":"doc-en-vscode-845808ae7b3a012a8fa5821361e1ee9f40b6959a32443676ba2a0672be52fd02","title":"","text":"I wanted to copy this command to a diff't window and see how the selection didn't work so well https://user- Since we know where the prompt ends, could we do something about this? If not, should we have a copy command action?\nDouble click is all driven by the links system which works pretty well generally, adding prompt knowledge in there would be a bit of a mess and need some more APIs to override specific cell double-click selection handling since links live in but shell integration is all in vscode. The way I do this is to double click and drag it to the right a bit. I'd say this is out of scope but we can talk about it more if you want.\nYeah I thought abt the complications with the existing link/selection code and decided that would be too much. I wonder if we should have a copy command action though?\nSure we could do that, it would also be good to get some more order to this: ! Proposal: Rerun Command Copy Command - Copy Output Copy Output as HTML - How does this work?\nVerified. Looks like the proposal to me. <img width=\"277\" alt=\"image\" src=\"https://user-\""}
{"_id":"doc-en-vscode-3aae00abdc6af1af39af3fe357c669f1de4f44f9a93cd4d9760c286fd52b5945","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: OS Version: Steps to Reproduce: the extension in VS Code and run the \"Run Extension\" task from the \"Run and Debug\" panel a new untitled file and change the language to \"HTML\" the command or start typing \"example\" to reveal the \"example\" completion items that one of the completion items has a red block and one does not Expected Behaviour: Both completion items should display a red block. Suggested Fix: When the color extractor checks the property it should account for values: For example the piece of code above could be changed to the following:"}
{"_id":"doc-en-vscode-9bf64aa338b731d7a9477f6adc6de17cec973bec4fcc5549b86a472c7b2e5395","title":"","text":"I've seen this a lot over the past month and it seems like a major issue for notebooks, please consider investigating further for November ! ! ! ! While running integration tests today (23rd November): ! Version: 1.74.0-insider (user setup) Commit: Date: 2022-11-23T05:25:25.108Z Electron: 19.1.3 Chromium: 102.0.5005.167 16.14.2 V8: 10.2.154.15-electron.0 OS: Windows_NT x64 10.0. Sandboxed: Yes\nit's likely related to the position performance optimization this month.\nCan you try debugging. I can never reproduce this The best place to start debugging is: Let me know if you want to pair on this later this afternoon too\nWas finally able to repo this but not reliably. It seems to rely on a new being created, but this doesn't always happen. The most reliable repo for me: with a notebook and text editor in a single editor group the notebook editor down to create a new group below the text editor grab the notebook editor and drag it right to move the bottom editor to the right I think the root cause is that the notebook widget is passed the wrong position. I'm going to work around this by making the notebook editor ignore the position it is given, however this will also undo the performance fixes we made this iteration for notebook layout I don't fully understand the grid layout code but it seems that is being set incorrectly in cases where a new BranchNode/SplitView is created. Here's what I see happening after step 3 of the repo above: setter is called calls which creates a new the constructor, a new Split View is created is called on it. At this point, of the split view is: the time layout is called on the notebook editor widget, is computed as zero. This results in the wrong layout. Looking at the code, we can see that is passed in as the absolute offset in step 4: <img width=\"1104\" alt=\"Screenshot 2022-11-30 at 12 48 54 PM\" src=\"https://user-\"If I then slightly resize the window, here's what happens: call on the split View for the new group now is: we layout the notebook editor, is now correctly computed as\nCreated new issue: , thanks for the steps"}
{"_id":"doc-en-vscode-37a8eb03e22c4baf5a6f878fe65fbc6ba70ac8b7416be5388390d42472f29ed3","title":"","text":"Issue Type: <bBug</bPerform a search In the results panel, right-click a filename Select 'Reveal in Explorer' File is highlighted in the 'Files' side panel. VS Code version: Code 1.39.2 (, 2019-10-15T15:35:18.241Z) OS version: WindowsNT x64 10.0. <details<summarySystem Info</summary </details<details<summaryExtensions (7)</summaryExtensionVersion ------ Bookmarks10.5.0 cucumberautocomplete2.14.1 vscode-svgviewer2.0.0 vscode-markdownlint0.31.1 vscode-eslint1.9.1 vscode-scss-formatter1.4.2 perforce3.2.0 (1 theme extensions excluded) </details<!-- generated by issue reporter --\nwhat you refer to as \"the 'Files' side panel\" is called Explorer. So the current behaviour of 'Reveal in Explorer' from a filename in Search results makes sense. But I can see it's confusing because when you then right-click on the filename that's been revealed in VS Code's Explorer, you get another 'Reveal in Explorer', and that one opens a Windows Explorer onto the folder containing the file and highlights the file. I guess on macOS the equivalent context menu will say 'Reveal in Finder'.\nExactly.\nSo maybe the option in the Search side bar's context menu should be recaptioned 'Reveal in Explorer Side Bar' (1 change): Alternatively, make the Explorer side bar context menu on Windows use the caption 'Reveal in Windows Explorer' (3 changes):\nfyi\nI'm still on Windows, I right-click the tab of an open file, and it offers and .\non my 1.39.2 on Windows the context menu from a file tab captions the options \"Reveal in Explorer\" and \"Reveal in Side Bar\" (not \"Open in ...\" as you wrote). So for consistency I think the context menu from the filename on the Search list should be changed on this line to become\nFair enough. My point, though, was obviously regarding the inconsistency in numerology across the application.\nWe talked about this at the time. The thing is, \"Explorer\" is used in a bunch of places to refer to the file explorer in vscode and it doesn't seem like a big deal to me."}
{"_id":"doc-en-vscode-b8f88dff3905219142e8b49557214bb4e778535bb8f59d3857a6c11392eb20f8","title":"","text":"However I realized that there is precedent for referring to it as \"Side bar\", the command \"File: Reveal active file in side bar\". I think that's kind of imprecise and also you are already looking at the search results in the side bar, so it doesn't make sense to \"reveal in side bar\" again. But anyway, I don't have a strong opinion.\nWe have been using the term \"side bar\" exactly for this reason: to have a way to distinguish between the Windows Explorer and the explorer in VSCode. We could in theory refer to it as \"Explorer View\", but that may cause frustration when users search for \"reveal explorer\" and find 2 results. I think no matter what we do, both commands should have a similar label. I don't think there is an issue calling the command in the search results \"Reveal in Side Bar\".\nThat'd be nice and clear.\nare you willing to take my PR for this? It failed CI on windows and darwin but as far as I can tell that wasn't caused by anything in my PR itself."}
{"_id":"doc-en-vscode-b1bf9acaba76bad875052994399425d0c43dac97e08f081c6a405dec2f3d9817","title":"","text":"[ ] Notebooks (.ipynb files) [X] Interactive Window and/or Cell Scripts (.py files with #%% markers) When rendering large output, I now get the option to at the bottom of the output in addition to the previous \"text editor\" view. However, when selecting this option, the full output is not rendered in the scrollable element. Instead, the ellipsis shown in the truncated view remain. The expected outcome would be that the content shown in the scrollable element would be the full output, not the truncated output. When changing the setting , the scrollable element is displayed immediately and shows the content as expected. To reproduce: Initial output: ! Output after clicking : ! Output when setting : ! 78.2 v2023.4. Python v3.8.13 No response No response Local"}
{"_id":"doc-en-vscode-763d9ef87493f1075a7570f415cbd6e6964578f3399133d8bada9b48495621d0","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.60.0 OS Version: MacOS 10.14.3 Steps to Reproduce: & paste into markdown file <kbd⌘</kbdand click the link .ext)\nThis is my first go at launching into the vscode code, but hopefully this information may be useful. Perhaps we could update the parsing of the URI schemes in general to strip out the angle brackets when editing markdown? See my findings below for why I'm suggesting this, but I'm also unsure of what problems and side effects doing that may cause. Main points of interest While the Command + Click doesn't follow the web link when you're a Markdown file, it does follow the link in the of the Markdown file. So I think markdown syntax is actually ok In debugging the project a little bit, I found that the link you try to follow that has angle brackets may be trying to or something similar (instead of open a web browser) In the function , I think the scheme of http(s) isn't recognized because of the < brackets in the link portion. See around line 18, where it tries to recognize the scheme, but doesn't recognize it when it tries calling In the file you can add a few new schemes, and then you're able to follow the link just fine. Though, this is probably not the final way the issue should be fixed"}
{"_id":"doc-en-vscode-1b2485d8fb892b28e2825e2abf16fbfbfc08e0b4ba97087ddaffabe5bff12b7c","title":"","text":"Looks like a one time fluke, given 30d statistics and 5000+ runs pass rate. <img width=\"1357\" alt=\"image\" src=\"\""}
{"_id":"doc-en-vscode-36e86d04ee85b9e59d33341b7c6828f573e270f56860304e91fbcfabf53e0fc8","title":"","text":"to search, then click the that expands the results. :bug: does not expand. Looking into this more, it is because the call for the node isn't working.\nI've also seen the same thing with not working. Unsure if this is related, but the repro is the same, but between step 2 and 3, expand the AI results.\nThis seems to also happen in when in tree mode.\nIt could be very expensive if expand all would actually expand all tree nodes in an async data tree. It's possible to have infinite deep async trees so in those cases this method would crash. My recommendation would be to create your own implementation for this in the search view. If that does not work the way you would like it to, we could consider alternatives. I think the bulk edit pane has a similar behaviour here:"}
{"_id":"doc-en-vscode-3325aa0ee008430635d723a33d9e393a412f617a33f43b9d75fe9f8d02781cea","title":"","text":"This: Now results in 1/2 the minimum contrast requirements which is a soft accessibility issue. It seems like these messages are special vscode messages which use special formatting in the real terminal:"}
{"_id":"doc-en-vscode-20205698d489a734bfb9c22b683dab480d6ff3ead83328d887203e1ab5509493","title":"","text":"There's no separator line between the two components. !\nSeeing the seperator: <img width=\"69\" alt=\"image\" src=\"\""}
{"_id":"doc-en-vscode-995b8de1ade90b3f1bd194d779618eb9624e5e936ca907b5c6ace89e354fd358","title":"","text":"over the last week I have seen zombie processes running at full CPU load when trying to debug them (via SIGUSR1) they die analysed a crash dump and sees that this is because the EH is already in the process of shutting down the previous point matches the observation that no window reports its EH as having crashed <img width=\"1290\" alt=\"Screenshot 2023-09-29 at 13 31 34\" src=\"\"\nCan you take a process sample from the OS activity monitor, I hope the native stack can give a clue to what is happening.\nJust happening. This is the command line of such an EH\nThis the process sample\nNot sure about the part. Does this mean we run the EH under development or not?\nIt looks like we set that for all local EH process launches. Just having inspect-port doesn't put the process in debug mode. What you can do is use the \"Debug: Attach to Process\" command and select that process. This will fail, since the debugger won't then know what port it's listening on (if --inspect-port isn't given in the args, it always listens on 9229, and js-debug assumes that is what happens). But, then you can . AFter doing so you could grab a performance profile using the edge/chrome devtools on edge://inspect, or by attaching to that using an attach-type launch config VS Code and using the \"take performance profile\" command -- if the spin sample you recorded above isn't enough for Deepak to figure out what's going on.\nDoes that use SIGUSR1? Asking cos I have tried that first and sending the signal kills the process. From the crash dumps we do know that the process already was shutting down\nAttaching the symbolicated sample Going from the top, indicates the EH process is already in the shutdown phase and has started to destroy the . At this point, connecting via a debugger is not possible since the agent host will stop receiving connections due to the following , so we cannot get perf profile from these processes. Moving further down most of the sample time is spent in which the environment gets deleted, this flushes the pending tasks in the libuv event loop but one thing to note is that entering JS via these tasks is not allowed due to the . But there are GC tasks being executed and which can enter JS."}
{"_id":"doc-en-vscode-ac62416af3a3c1215c13ec097de79e4b2486d645b00c72595891a7f4effd2005","title":"","text":"I think the error calls ` are from these callsites eventually leading to . At this point, I expect the process to abort and the main process should be able to reap the utility process but that is not happening. There is another data point looking back at the screenshot from , the zombie EH processes are titled and not in the process explorer. We resolve the pids to respective titles based on a map maintained in the main process, the process explorer gets it periodically via the following and the utility process related values are populated from . We default to when the pid map no longer has an entry for the process which can only happen when got invoked which inturn suggests that the main process did receive an exit or crash signal from these EH hosts. can you check the main process logs next time if it has any of the following entries for the corresponding zombie EH pid to validate this data\nI attempted to setup a similar flow based on the stack with base Electron app but the utility process are cleaned up as expected. I can make a change in the destruction phase of the utility process to signal termination of isolate so that tasks which require to enter JS will not do so and I hope that resolves the long running draintask. I won't be able to make a new insiders build with this change until I am back from vacation but I can provide a custom electron build for local testing with OSS. would that help to validate incase you face similar issue with OSS as well ?\nAh, my bad, skimmed over that part. Disregard. (Yes, it sends SIGUSR1 on unix)\nI am seeing this for which is currently spinning forever. Samples for this one\ni think i have the same issue on my mac. It happens from time to time and my workaround is simply killing vscode and then reopen it. But sometimes i don't realize it until my Mac has run out of battery, which can be quite annoying. <img width=\"963\" alt=\"CleanShot 2023-10-09 at 19 46 49 src=\"\"!\nHaving the exact same issue. utility-process is running at 100% all the time on M1 Max, very annoying.\nThis bug might also exist in Linux build of vscode: ! (6% 16 cores = 96%) Command line of \"utility-process\" is: eu-stack of main thread of this process: Noticed that is also inside sampling."}
{"_id":"doc-en-vscode-54389301523cd2cd2b5121b8ee40b1f65627941d40b70e07d4b368456e63ca4c","title":"","text":"Manually sampling with eu-stack running every second, for one minute. (And might related to and )\nSorry if I'm hijacking the thread but would uninstalling Node and installing Bun solve this problem? Doesn't sound like it's a VSCode bug, right?\nThanks for the logs , the stack does look related. I wouldn't be surprised if the issue is also present on Linux since the shutdown path is same for both.\nSimilar issue on my M1 Max, sometimes I observe even 3/4 zombie processes running in the background and have to kill them manually from process manager to stop them from draining battery. 2 more samples for such process:\nChecked log and it looks weird. : However, PID still exists and is running at 100%.\nAnother hang process today with log: \"Extension host with pid started\" \"Extension host with pid exiting with code 0\" \"Extension host with pid exited with code: 0, signal: unknown.\"\nIs there any temp workaround? any script or bash command that we can run to identify and kill those pid? I am interested to do it manually, because i have a M2 and i am rebooting my macbook after some hours because get soooo slow\nRestarting VS Code should already \"fix\" the problem because the extension host is a child process of the application. At least that does the job for me. Are you observing a different behavior? Note that \"Reload Window\" does not kill the zombie process (according to my observation).\nPlease help by posting the list of installed extensions. I am looking for similarities between our setups as this is possibly caused by an extension. This is my list To get this list run or . Please do the same and post your results here. Thanks\nThanks for looking into this! I am also facing this issue quite frequently. Here are my installed extensions:\nThanks everyone. Looks like the intersection is\ncode --list-extensions\nYep, I'm using those extensions too. Here's the full list:\nis not a very good workaround: what if the process exits, but another innocent process just takes this PID?\nIt probably cannot do that as the pid is still taken? I agree it's not a great workaround, but until the root cause is clear it does solve the issue where a cpu is occupied."}
{"_id":"doc-en-vscode-5e166690d63c4b0c575e4de6106a94dcb1a58be52bae29a178963b9d7f67a963","title":"","text":"off the hot exit and auto save settings an interactive window, run a cell vscode :bug: confirmation pop-up to save the IW"}
{"_id":"doc-en-vscode-f87c06c90d579180e0ab2f5975016282363fc172944c90e0e9911dbfcaacf195","title":"","text":"Type: <bBug</bIn previous VS Code version 1.78 when I pressed if inline suggest was visible with suggest widget it hides only inline suggest, if I press again same shortcut it then hides suggest widget. Now with 1.79 it hides both inline suggest and suggest widget. KeybindingService logs: My custom shortcuts to hide inline suggest & suggest widget: I even tried to add to clause and still it hides both. VS Code version: Code 1.79.0 (, 2023-06-07T14:31:03.317Z) OS version: Darwin arm64 22.5.0 Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (12)</summaryExtensionVersion ------ vscode-eslint2.4.0 gitlens13.6.0 prettier-vscode9.13.0 copilot1.88.132 vscode-graphql0.8.13 vscode-graphql-syntax1.2.0 symbols0.0.11 vscode-react-native1.11.0 vscode-thunder-client2.7.6 vscode-fileutils3.10.1 vim1.25.2 quokka-vscode1.0.543 (1 theme extensions excluded) </details<!-- generated by issue reporter --\nfull text search doesn't reveal the context key anymore. Has that been removed/renamed?\nIt is\nsorry I was writing that instead of copy-pasting. Already updated the issue.\nDoes it work when you swap the keybinding configurations? (Like literally exchange the positions in the file)\nStill the same issue even if I swap places :(\nSo, its either the or context key. Can you investigate with the F1 Inspect Context Key Tools to see if those context keys are set. Note that it will print context keys on the console of DevTools\n<details<summaryContext keys</summary</details\nas you can see is and is , but even though inline suggest was visible it returned false.\nSame in the monaco, if the suggest widget is visible, the inlineSuggestWidgetVisible will always be false no matter there is any ghost text or not\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected."}
{"_id":"doc-en-vscode-07d1e77cd87ad0b43c6d86735151c93f7973d480cefa08b7bc6ea794cef0e8eb","title":"","text":"If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nNow I see us hiding the suggest widget first and the inline suggest second, is that expected? This is true regardless of what order I declare the keybindings in (I do have the Vim extension installed)\nIt should hide first inline suggestions and second time you press it, it should hide suggest widget. Try to swap places of keybindings to see if it fixes it (order maybe matters?). Because previously this is how it worked (in 1.78, but regression was introduced in 1.79). EDIT: Oh I see you mentioned that order doesn't matter. Try to add in clause for command (I hope it works).\nis it fixed for you on the latest insider builds The following does not work for me either\nIt works! !\nGreat, thanks for confirming!"}
{"_id":"doc-en-vscode-aca8b59728e08f8eb51674028473b11296ab1c75427042bdd9f6bc151aba270f","title":"","text":"VSCode freeze on mouse wheel scrolling with lot of tests: open test explorer, ! right-click on the test button scroll tests with mouse wheel at least twice VS Code version: Code - Insiders 1.64.0-insider (, 2021-12-22T05:15:27.690Z) OS version: Linux x64 5.4.0-92-generic Restricted Mode: No <details<summarySystem Info</summary </details<details<summaryExtensions (20)</summaryExtensionVersion ------ vscode-eslint2.2.2 gitlens-insiders2021.12.2804 EditorConfig0.16.4 prettier-vscode9.0.0 themeswitch1.0.5 rpm-spec0.3.2 vscode-docker1.18.0 python2022.0.-dev vscode-pylance2021.12.3-pre.1 jupyter2022.1. jupyter-keymap1.0.0 jupyter-renderers1.0.4 remote-containers0.209.6 remote-ssh0.70.0 remote-ssh-edit0.70.0 code-stats-vscode1.0.18 code-spell-checker2.0.14 code-spell-checker-russian2.0.2 cron-explained1.3.1 material-theme3.13.8 </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nLooks like doesn't help.\nThanks (again) for the issue!"}
{"_id":"doc-en-vscode-52fd9bc8ef8ecee2ff4f976a580a29bf9897513540e5417d5e94eac3877d58ff","title":"","text":"Reminded of this by We currently have a context key (when clause) for the run menu (in the main menu). This means we don't know if the menu should be displayed on startup because it depends on extensions. The Terminal menu also has a context key but resolves very quickly so there is no delay. Should we remove both of these when clauses in favor of keeping them around with possibly disabled menu items. E.g. The \"Install additional debuggers\" entry should still be shown in the run menu if no current contexts are available.\nAssigning to who now owns the Debug UI. Rob let me know if I can help here.\nWhich menu are you talking about? How is the context menu involved?\nContext Menu should have been Context Key. Sorry about that, I'll edit the original post.\nLooks like has decided to go this route with the Terminal menu, so I think we should align here too.\nSo the terminal menu will be shown all the time in web? That's what will happen with Run. I guess it will show \"Install additional debuggers\" And should other items be disabled or missing? They currently use so are missing. It would look like this <img width=\"245\" alt=\"image\" src=\"https://user-\"I'm fine with that if it's consistent. Sort of gives false hope about what you're going to find, though\nActually, if the Run menu doesn't have a when clause, but all the subitems do and it is false, then the Run menu still doesn't show up. Would that be ok or is that still slow?\nWon't install additional debuggers always be visible?\nIt currently has a when clause to only show up when other debuggers are present, it would only show up if I remove that. And I think \"install additional debuggers\" usually won't give the user what they want.\nI see that entry as a way for users to discover how to get our web-based code execution extensions like Python. Maybe the text is not right? We could even have a web specific entry that filters the extension list to web-supported debuggers instead of all?\nYeah, I think that when those exist we should think about what the onramp looks like. For now I'll remove the when clause on the Run menu if you think that's the right thing to do."}
{"_id":"doc-en-vscode-ba75f891c286c705dd91b386dab618e74772f2755ca0f6b2b90d9c8a3d45de49","title":"","text":"sounds good, thank you\nOk, this actually isn't working the way I thought it did, and the result is that the Run menu never shows, at all. I must have tested it wrong, maybe you have to restart vscode for native menu changes to take full effect? The current state is that the Run menu has no when clause, and every command inside the menu does have a when clause. But the result seems to be: If the menu doesn't have a when clause, but initially doesn't have any enabled commands inside it, then it doesn't appear and will never appear, even if the commands inside it later get their when clauses flipped to true. The only way to get it to conditionally appear is to add a when clause to the menu itself. Does that sound like a menu bug But also I can just go back to having \"Install additional debuggers\" enabled all the time, I think that's fine.\nthat very well could be a menu bug. but thinking about it, I think having the one item always in the menu is necessary because otherwise we still result in the behavior where we don't know for some time on startup whether to show the menu or not which causes flickering.\nis it expected that all of these items in the run menu are active in web? <img width=\"163\" alt=\"image\" src=\"https://user-\"\nYou probably have some extension enabled for web which claims it can debug\nMock Debug and Python, thanks"}
{"_id":"doc-en-vscode-142503b979b9f35fa3210689de9056c4b4acd6a330e6d9aebbcc1e3e8ff7a3a0","title":"","text":"Testing microsoft/vscode-python Following the steps, I first used the auto-generated launch configuration. This meant that it didn't exist in the I then repeated the steps by clicking the gear to create it in the the dropdown now has 2 entries with duplicate titles. I would expect that the non-file-backed configuration just disappears now that we use a Not sure if this is extension specific or the responsibility of core.\nHow did you trigger second time to get the dynamic one? after creating the After, , i don't see the debug welcome screen.\nUsing one of these methods\nThanks for the bug report. VSCode handles the auto debug configuration behavior, so I'll transfer the issue to their repository.\nThis still repros: ! Steps: with no file open, show all automatic: ! python -Get expected error: ! one shows up... great... now pick : ! hit the on : ! See two: !\nAdditionally... maybe a different issue, I noticed that if I do this: with no file open, show all automatic: ! python -Get expected error: ! one shows up... great... now pick : ! Hit to get rid of the quick pick: ! is still selected in the ! and the only way to get back to the Picker I escaped out of is to choose a different launch config in the drop down and then select again. Seems a bit odd. I would have expected the play button to give me the picker or selecting even if it's already chosen to re-trigger the picker.\nI fixed a different issue... I removed the duplicate from the quick pick, but not from the debug sidebar dropdown. Same data source, I think the dropdown just needs something else to refresh it.\nHandling these dynamic configs is tricky - the way they are represented, they look mostly the same as a real launch config with the same name, and there is probably another case where we fail to tell the difference\nFollowing the steps, I now get 2 different titles but still very similar. Not sure if this is breaking any checks that you do <img width=\"212\" alt=\"image\" src=\"https://user-\"\nHow did you get without the ? I tried again and don't see that."}
{"_id":"doc-en-vscode-ffe4d3aa56f164b49039f6a061cee42c763d4124d0dee3000d4fe73e9648d5e9","title":"","text":"Is that the name in your I have and <img width=\"173\" alt=\"image\" src=\"https://user-\"and I'm not sure how to get a different name in the saved config.\nThat is what was generated in my launch config, yes, I did not create it manually. I am on Windows.\nI can't repro it. If you want, you can show me. But, if this happens somehow, I don't think I can improve the situation anyway, because two configs with different names are just two different configs.\nI still managed to repro using SteVen's original steps in this issue <img width=\"248\" alt=\"image\" src=\"https://user-\"\nI can't repro the issue, are you sure you didn't end up with duplicate entries in If not can you open the quick access for and share a screenshot? Or just call me sometime so I can try?\nThis issue has been closed automatically because it needs more information and has not had recent activity. See also our guidelines. Happy Coding!"}
{"_id":"doc-en-vscode-f6125bcab3fc4b5ecbd7d7f2419507e1da84bc8f4ccacc6f76c88291b488a41a","title":"","text":"<!-- Do you have a question? Please ask it on For bug reports, please fill in the details below. For feature requests, please describe the desired feature. --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.23.1 OS Version: Windows 10 Enterprise Ver.1703 build .877 Im trying action. but, this action working only absolute path. is working code (absolute path) ex: - Emmet: Update Image size -not working code (root relative path) ex: - Emmet: Update Image size -its not working... <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: No\nThanks for reporting PRs are welcome to fix this issue. The code behind this feature is at\nHi, I'm interested in this bug. I noticed the person filing the bug gives as a example of a working path. I believe that path means that in the current working directory there should be folder called that has another folder named which has photo in it. However, is given as an example of the command not working. It is true that this path doesn't work but I believe that is intentional. at the beginning means it starts at the root and because of that it may be looking for in a different spot in the directory. For example, if my absolute path to a image is like this: and the location to the html file being editing is in here: I can't run the command with this code: But I can run it on this: and this: Hopefully this information is useful.\nThanks for the detailed analysis I can confirm your findings. Can you share what your folder structure is? Unless folder is directly under root, we cannot expect the command to work on Also see\nthx comments. I can be submitted sample folder structure. Sample folder structure is working code of path: is not working code of path but, code is Ctrl + LClick (img Preview) is working. because I think passed root path of VSCode.\nAre you on Windows, Linux or Mac?\nWindows\nCtrl+Click will navigate to the image file in the first 2 cases as expected in both Mac and Windows The works on all 3 cases in Mac, but only the first case in Windows. So, at the very least, we can look into if we can get the second case working to keep the experience consistent between the Ctrl+Click and the Emmet experience"}
{"_id":"doc-en-vscode-42aab202a16c1e1be0cf0d5c89863093fb6c37ba0dae9edcae80033f8b76310a","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.74.2 OS Version: Ubuntu 22.04 LTS (with all updates applied) Steps to Reproduce: in using a Active Directory account (Ubuntu must be joined into an Active Directory Domain) Visual Studio Code Terminal tab, click on [ + ] sign to open a new terminal, choosing zsh. A empty terminal tab opens and no shell is launched When I look in Output tab, section 'Shared', the following log lines were found By looking at message, it seems vscode is looking at /etc/passwd for some reason. And it may be failing because when I log in with an Active Directory account, no /etc/passwd entry is created for my user, because sssd/realm does not write entries on that. I opened a terminal Window and grab the following output for helping debugging:\nLooks like we need to try catch here:\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nVerified by code review (but would still appreciate your\nI confirmed now that it´s working as expected. Nice job from team! <img width=\"1009\" alt=\"image\" src=\"https://user-\""}
{"_id":"doc-en-vscode-d6b0fa58edd35c9b16e841e95e66d2e8620587a14325d8424653d5f1d3e852c2","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.7.2 OS Version: windows Steps to Reproduce: a file with error on `cue: line has an error\" settings the cursor to there\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.73.1. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!\nit still occurs with 1.73.1\nyeah I can reproduce this issue. is this expected?\nThis is not expected. Sounds should be played when cursor is navigated to the error line. worked on this initially and might have more details\nI also have the same problem, but this only occurs in the insider version, I have the latest version and no sound is played Issue:\nThey play in OSS, but not insider's. any idea why that would be?\nif that is the case then it is probably related to code bundling. Does that sound play when you F1 list audio cues?\nIt does play when I list audio cues in both OSS and insider's.\nIt also plays for terminal quick fixes - suggesting there's a bug in because I can also repro it not playing for other line feature types\nThis has been working for me lately, though it was definitely broken at one pt.\nI have set to I navigate to a line with an error-squiggle but there is no sound being played"}
{"_id":"doc-en-vscode-3c66fac27ab28eed0f48653761be81cc7186829dc02ac54c30c428d806e0845c","title":"","text":"Issue Type: <bFeature Request</bI never used the \"Contributions\" tab in extensions, thinking it was for recent github contributions to the extension. I accidentially pressed it once, and found it very useful. I think the name \"Contributions\" is confusing and that \"Details\", \"Internals\" , \"Technical Details\", or something similar would make it much clearer. VS Code version: Code 1.41.1 (, 2019-12-18T15:04:31.999Z) OS version: Linux x64 5.4.14-200.fc31.x86_64 <!-- generated by issue reporter --\nThese are in fact contributions to VS Code. How about ?\n<!-- -96bd-47a8-a0e8- --This feature request is now a candidate for our backlog. The community has 60 days to upvote the issue. If it receives 20 upvotes we will move it to our backlog. If not, we will close it. To learn more about how we handle feature requests, please see our . Happy Coding!"}
{"_id":"doc-en-vscode-b5387325536e9f1c70bc79e82f5f3913cdedf9b4c89aadb478db70df7f75859a","title":"","text":"It is possible with VSCode to follow a path printed in the terminal. When that path also contains the line and column number in the format ( ≙ line number; ≙ column number), VSCode currently doesn't interpret it completely and therefore doesn't also jump to the line/column in the opened file. Example: ` This path format indicates the path to the file but also the line (49) and column (46) number. Ctrl + clicking this path currently in VSCode will open the file but not jump to the line or the column. As this format is used by many linters and compilers (e.g. SASS), it would be a very useful and time-saving feature.\nCode link: I think here we'd want to require the exact (num)(colon)(num) to prevent problems with (link) (number) (text)\nnice feature for many front end build tools.\nI PR for this."}
{"_id":"doc-en-vscode-113e50b8f889a84353d64a9df7b4f3232537ad8fe9498b6bc591c57c29390ef9","title":"","text":"When updating cell metadata for a cell, users call which provides a function to alter the metadata of a cell. This uses defined in . When is called, the CellEditType is marked as , resulting in the metadata of the cell only being updated with the passed in data, rather than fully replaced. This makes it difficult to determine how to remove fields from a cell's metadata, a change needed for fixing . This can be worked around in this situation by setting the field to null. However, if a user calls , it appears that the functionality is a full replacement of initial metadata. There is a slight inconsistency here between the functionality of the two functions. Couple initial ideas: better document to reflect the behavior of . Add in a line stating that this is an iterative change, only altering the fields passed into this function - revise the functionality of either updateCell or updateNotebook metadata functions to align their behavior. This seems dangerous as it could likely break features leveraging this api. deprecate these functions, and rewrite two new ones with clearer functionality and documentation, based on updated needs from the notebook and jupyter teams\nAnalyzed all extensions using this API and most of them are already doing the right thing: clone the existing metadata, apply the change, and then call the API to update. However there are a few extensions didn't do so: there is one place where the metadata is not updated properly: PR . cc\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\nwhat is the best way to verify this?\nIt's a bit complex I would love either or to help verify."}
{"_id":"doc-en-vscode-7f3550b5a58f63799b1ebb92acbc86141e4b7e10e9d40de58e60f405d5da35db","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --The [NPM support for VS Code built-in extension][extension] (which automatically detects NPM scripts and provides a UI to run them in the sidebar) has an settting. Currently, this setting only supports the options or . This means it won't work with the [pnpm package manager][pnpm]. Adding a option would be really helpful to users who use pnpm rather than npm or yarn. Both [vscode-eslint][] and [typescript-tslint-plugin][] also already support using pnpm as the preferred Node package manager. I personally use pnpm because its reduced disk space/bandwidth footprint and its architecture both work better for me. I've experimented with just overwriting the value in with , and ignoring the warning from the linter. This seems to work (on macOS), so I'm guessing it might be as simple as adding to the array here: I just don't understand the extensions well enough to be sure that's correct / enough to properly add pnpm. [extension]: [pnpm]: [vscode-eslint]: [typescript-tslint-plugin]: [source]:"}
{"_id":"doc-en-vscode-8a5491b1f8015442288518724fff15ede5e85abed2a0477a4c5a582beec24d10","title":"","text":"uses quick chat and has found that going between the input and the response is cumbersome because of our underlying list structure. This is in particular an issue because we now have many hyperlinked tab stops. ! We should provide a way to go between the input and the focused response. AFAICT, there is not a way to do this. Pressing seems to focus the input (unexpectedly), but there's no way to get back to the response. Thoughts?\nMakes sense. Probably just need to handle the keybindings better\nthese probably just need logic for QC\nTyler figured out that there is a way to do this, but it's in the reverse direction from what one might expect based on tab order. Use to focus the input and to focus the last response.\nTalked to JooYoung and it's fine to have these reversed, we'll want to be sure that using doesn't still focus the input box.\nworking afaict - and for response -input and vice versa, tab works fine as well"}
{"_id":"doc-en-vscode-2788b7860a6fe84c04ff446d0b67506f64215cbf780377e8f0d6662eff0ba10b","title":"","text":"insiders, install remote SSH extension the Remote view container all the views via context menu menu on the title -the views are not there and they can not be brought back -the views are back -I would expect the hidden state to be preserved fyi !\nAssigning to as remote viewlet has special treatment.\nOriginal decision was that none of the views in the filter-view should be hide-able. Looks like I missed a spot!"}
{"_id":"doc-en-vscode-7af342caba10683e34814bd3ea27fd8020976b46e36844702028f7a9103da1bb","title":"","text":"Broken due to refactorings of the workbench CSS. Fix is simple (CSS only change)."}
{"_id":"doc-en-vscode-25bdf6104af2e1fe18d7d66c97ef98b3d2496a4e5e25d097d7777aee00da980b","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.44.1 OS Version: Linux 5.6.4-arch1-1 SMP PREEMPT x86_64 GNU/Linux Steps to Reproduce: a symbolic link to the VSCode starting script with a space in the name. For example: \"My VSCode\" from the terminal Error is printed to the terminal: All in the starting script () should be enclosed in double quotes to avoid this error. <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes\nI can take this!\ngreat!"}
{"_id":"doc-en-vscode-08d37d62cf7ccaeea9eafebb1b122b10f1345ded1c863f9b39bbd6bf34d3947c","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: Latest OS Version: Browser Steps to Reproduce: 'Cancel while enumerating through children' test case up above 'Root element should return children' (or use a test runner which randomized order. the test and notice that it fails. Notice how other variables in the same test use the method to make the test cases independent of each other."}
{"_id":"doc-en-vscode-f31e2bef103e8eaada48b8198ff67c68ce0604737fe46dadc29470610c2fce81","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: OS Version: Steps to Reproduce: Jupyter extension all cells Ctrl+Z in a cell icon in toolbar changes to No way to recover from this. Similar to However this reverts the Notebook Metadata (instead of cell metadata). <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.542 --- <!-- score: 0.483 --<!-- potentialduplicatescomment --\nValidated, works now, closing this issue"}
{"_id":"doc-en-vscode-cba559cfb20b234c348708a32ba3d2ff19a4c3f47d168757f0c465eac218bc0a","title":"","text":"Testing a file +enter to run cell in interactive window window GitHub Copilot classic installed and signed in something into the input box and ctrl+enter, this opens a third editor group with multiple synthesized Copilot solutions (feature from GitHub Copilot) close that editor :bug: now there is an empty third editor group which really shouldn't happen AFAIK, also there is screen cheese, it shouldn't really be possible for Copilot to cause this kind of layout issue <img width=\"1256\" alt=\"image\" src=\"\"<img width=\"1245\" alt=\"image\" src=\"\"\nI assigned you for grid layout and I assigned you for editor layout, please feel free to unassign if this isn't relevant to you\nempty editor groups are not a bug per-se, if you configure , groups will remain open when the last editor closes. Also, empty groups can explicitly be created, e.g. via . However, the fact that you cannot get rid of the empty group when you press the X button sounds like an issue to me. When you try to close, is there anything printed to the console?\nOhhhhhhhhhhh good bug! There is definitely splitview/gridview breakage here.\nWait. I could repro once... but not anymore. Can you still repro?\nYes I reproed twice\nI cannot see anything in logs when clicking on the X unfortunately (fwiw, I do not have that setting configured) When I repro this, here's what's in the log (this happens right after closing the last editor in a group that results in the group not also being closed)\nI wonder if this could possibly be a regression from the auto-sizing feature you did?\nWe just need to be able to repro on demand so we can attempt to repro on a version before my changes to confirm this. But I can't seem to repro no more. Can you ?\nI cannot reproduce this, I never get an empty editor group. This is with fresh user data dir and extensions dir.\nif you have a reliable repro that you can share, including your settings if any. Here is what I do: !\nI did some more messing around and was able to repro a similar problem ! <details</details\nOK, I have a trivial and consistent repro. TLDR it looks like the interactive window does not play nice with split view."}
{"_id":"doc-en-vscode-75cb77af712c2c5b142b24e8824cb62bd00a12e8b383ba6b8bb29b6962e60dfd","title":"","text":"an interactive window (you can even do this with the Jupyter: Create Interactive Window command) the Settings UI the Settings UI onto the right of the interactive window :bug: the split view is very unhappy Logs <details</details\nThis also repros for notebooks opened in the second group Right click a .ipynb file -open to the side open another editor and drag it to the right side of the first\nOh wow. So it looks like the call to throws an exception when done in the interactive window and that means the Grid will indeed be unhappy!\nfor surrounding some of the calls with .\nArguably, the grid/splitview should also protect itself...\nYeah, maybe. There could be other cases.\nI can't repro any of these now but it would probably help for to verify too\nI still do not have a functional cursor/trackpad so cannot drag/drop maybe you can verify?\nVerified it's working great"}
{"_id":"doc-en-vscode-7b043f6089ad33b5f43244f06a7aac7a421bb54dbd9949a1c752c5f07d1e6bfc","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes (all extensions other than the reproduction extension) <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.78.2 OS Version: macOS 12.5.1 Over at the Tailwind CSS IntelliSense repo we are seeing multiple reports of issues with colour decorators since the release of v1.78. Some are related to the new setting which I think is a separate issue, but we are also seeing an issue where document colour requests are not being sent to the language server at all ( is in this case). I was able to create a small reproduction based on the LSP extension sample. Please see below. Steps to Reproduce: instructions from reproduction readme: When the same project is opened in v1.76.2, the count is logged repeatedly when editing the document.\nThanks a lot for the great steps to reproduce. I pushed a fix for the issue."}
{"_id":"doc-en-vscode-33bd759abee14a07e7e2784b4aff991f12ab41360b8982c82c07d607a816e26d","title":"","text":"VS Code does not report errors in , and then silently drops the ball without telling the user. This is particularly annoying since some things might continue working, while others mysteriously stop. Errors in come in two flavours – actual syntactical errors, but also unrecognized new properties. In particular the latter is annoying, for example when TypeScript adds new features such as (see ). My suggestion would be to at least visibly and loudly report this, but preferably be more lenient about what's accepted (e.g. unrecognized properties).\nFYI\n+1 I had a similar issue (non existing compiler option) and I didn't get why my configured target=es6 got not accepted. Interestingly, it worked when compiling over command line but it failed when compiling over tsserver. See also the TS compiler code. Maybe one could get that error from tsserver back to VS code?\n+1 I've just had a similar issue that I've spent 3 hours working through. Adding a new typescript 2 option such as \"noUnusedParameters\" caused vscode to reference not , making some of my code have errors in the editor, but not when built! I've worked around by having a different that I used for building with.\nSee"}
{"_id":"doc-en-vscode-4e15164d0d9b9342372f821b2ae92c0e15d35906f23788741c1113f7f9fec61b","title":"","text":"Does this issue occur when all extensions are disabled?: Yes Version: 1.77.0-insider (Universal) Commit: Date: 2023-03-17T21:00:47.582Z Electron: 19.1.11 Chromium: 102.0.5005.196 16.14.2 V8: 10.2.154.26-electron.0 OS: Darwin arm64 20.6.0 Sandboxed: Yes Steps to Reproduce: Execute the following extension example: a text file and select a completion item , type , and select a completion item results are different https://user-\nThis is specific to additional edits that are after and on the same line as the suggestion. The newly typed characters (type after return the list of item and before selecting one) aren't taken into account"}
{"_id":"doc-en-vscode-0629382cbd075d991fd55465811b7f524c452dd0cb6472bf2340718df1be7480","title":"","text":"a notebook cell that will create a long stacktrace :bug: \"Output exceeds the size limit\" message is shown on a non-scrollable element"}
{"_id":"doc-en-vscode-9f4de7d213ff1e9f582213703bea3900a40687a7e59c28e27038deffe898c573","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.73.1 OS Version: macOS 13.0.1 Steps to Reproduce: a VS Code Remote SSH session a .hbs file selected language at the bottom right is \"Plain Text\" and when clicking it, Handlebars is not available in the list\nThe extension thinks it needs to be installed on the SSH remote: ! On it's own the handlebars extension doesn't need to be installed on the SSH remote, but the HTML languages features extension appears to be pulling it in with . It seems like this is related to The problem is easily fixed by adding this to handlebars: But that might break\nNice! Thanks for figuring this out! I initially believed this was a feature request to add a builtin handlebars grammar.\nThe handlebars could run on the renderer, but because it defines it is deduced to run on workspace. If were an internal contribution point, we could define a , but we don't have a way for contributions to declare a contribution point (a contribution contribution) So these are the possible solutions \"extensionKind\": \"ui\", \"workspace\"] as suggests as it overwrites the detected extension kinds Implement a contribution contribution point include the extension also in the server build I go with the first one for now. speak up if you disagree"}
{"_id":"doc-en-vscode-8201f28771ef260eab8ff2cf0eb4e8b7f845e4abe9b3ee3687fdfb32dcdcd557","title":"","text":"Testing a notebook with some content the replace widget to replace an existing string enter while focused in the replace box no effect, string should be replaced as it does in a text editor\nIt's a bad bug, but fixing it is not trivial. Considering it's not regression, I'll fix it next week."}
{"_id":"doc-en-vscode-416096d72ad920cb1af1f132058f092b9b0fac54b29765eec7c2d1b643d880b0","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --Version: 1.93.0-insider Commit: Date: 2024-08-29T05:04:03.909Z Electron: 30.4.0 ElectronBuildId: Chromium: 124.0.6367.243 20.15.1 V8: 12.4.254.20-electron.0 OS: Linux x64 6.9.3--generic Steps to Reproduce: a few editor groups in seemingly any split configuration one for text input or otherwise run to type: no result, as no editor has focus any more. This is especially frustrating when some of your other keyboard shortcuts rely on the editorFocus context, as they become unusable after using this feature.\nIn case this takes a while to fix, anyone looking for a workaround can rebind their preferred shortcut to automatically refocus the editor and then \"move\" the cursor to ensure if another window has now covered your cursor, the editor scrolls to reveal it again (also respecting scrolloff setting).\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!\n/verified"}
{"_id":"doc-en-vscode-5f8eca99c411d13aabd4ac9b123a0176524e47862ac82147d89e9af9eb55719f","title":"","text":"I see this when starting the panel chat voice action in dev tools. An easy way to know when this reproduces is to enable the audio cue, which then plays forever.\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!"}
{"_id":"doc-en-vscode-68f93141c64bcd17f2bf37b0a98bd07c368d65a632a074889ca152e7c0eff945","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --- Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- Microsoft Windows 11 Pro 10.0. N/A Build Not sure if this is intended behavior or not but I think if I use the button it should create and show the repository. If is set to , the repository is actually created but nothing is shown in VsCode I understand that this option is for auto-detect and it's kind of logical, but should it affect the action? !\nthank you very much for filing this issue. Setting to disables repository detection independent whether it already exists or it has been created using VS Code. While this is the intended behaviour, I agree that we could probably make this experience better."}
{"_id":"doc-en-vscode-01ee4f029461a1c38c07ba2f418beb83be0917d7afc94ce0c03a229ca553fcd2","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.81.0 OS Version: Windows_NT x64 10.0. Steps to Reproduce: In , set the content to then start VS Code with shell integration enabled and use Windows PowerShell. This bug will occur: The use of at: Needs to be guarded for Windows PowerShell (where it doesn't exist). /cc\nSharing from my profile:"}
{"_id":"doc-en-vscode-ce5280248152b12b0e44c6a58117b15b055ee3058a8b3bdf7871ae077efde470","title":"","text":"Testing I suspect results are not in right order <img width=\"1198\" alt=\"image\" src=\"https://user-\"Output result is shown first"}
{"_id":"doc-en-vscode-a73d5e91cae27ace3cc841e59b3b979df5a0743f77ec99e18a52241d9b5fccf2","title":"","text":"was broken for some time and I re-enabled it today. However eslint failures were reported in and I disabled them. I also moved the file into a test folder to avoid some other hygine failures. Please have a look and decide a proper place for the file and fix the corresponding errors."}
{"_id":"doc-en-vscode-aff256f71659ba1a20e442af69b12b57c1ba8e40da910c85e03c8dd0b4052ed0","title":"","text":"From We should add some clarifying comments to this parameter. It should be quite similar to that of , since the quickpick assumes the highlighting and fuzzy matching of the matches.\nadds a description, which will probably get cleaned up once we start cleaning up the APIs more."}
{"_id":"doc-en-vscode-5ceb0542473287b2908530cb3bbbf442bb58c099e5e51bed93999f251b37bbf9","title":"","text":"Issue Type: <bBug</bAfter updating to 1.43.0, when VSCode relaunched, it brought up around 80 editor windows from old projects I hadn't opened in quite some time. (spooky workspace necromancy) Perhaps obviously this made my computer behave quite poorly. I tried adding \"window.restoreWindows\": \"none\" to my with another text editor, but it didn't seem to be respected (might be my fault somehow). Note that it had been on \"folders\" previously, in case that helps. I also tried to muck about with file. VSCode seemed to ignore my changes to it (probably not a bug -- I was desparate and dunno how this file works), but I do have a backup of a from right after this happened. It shows all the windows it will try to open when launching, and maybe other stuff. I guess the system info probably has more PII anyway. I dunno how much this helps but: I think the main common feature of the workspaces it resurrected seemed to be that they had open files with unsaved changes -- but these are also the ones which can't be closed without additional UI confirmation, and I did try a couple things to close all windows in bulk (none seemed to work, but maybe I just didn't notice?) Relaunching without extensions also didn't help, but that doesn't mean this wasn't caused by an extension in the first place or somethign. I have more than is probably reasonable installed. That said, it happening immediately after updating to 1.43.0 makes me think it's vscode and not an extension. Anyway, ultimately I just bit the bullet and closed them manually -- VSCode would be usable for about 20 seconds after starting, during which time I could close a few before too many were restored and everything got too slow. So i just did this, quit, and repeated many times until there were no more left (this actually took a couple goes -- even when I'd get down to none left, save, and reopen, it would find some new ones to bring back... Guessing these just didn't make the cut for being restored the previous time, or something) P.S. it would be nice if there were a command line argument to launch without reopening anything. P.P.S."}
{"_id":"doc-en-vscode-20a2cc8d3703cbf25d892bde803e711fb24f2706f0f9e1ae6b549f52348ffd98","title":"","text":"In case it isn't clear: I've been using , , and somewhat interchangably here, just so you don't think I'm being precise about which I use when. VS Code version: Code 1.43.0 (, 2020-03-09T19:34:44.548Z) OS version: Darwin x64 18.7.0 <details<summarySystem Info</summary </details<details<summaryExtensions (55)</summaryExtensionVersion ------ code-gnu-global0.2.2 better-toml0.3.2 glsl-canvas0.2.8 bracket-pair-colorizer-20.0.29 dart-code3.8.1 xml2.5.0 scopes0.45.0 gitlens10.2.1 vscode-html-css0.2.3 tslint1.0.44 vsc-material-theme32.5.0 vsc-material-theme-icons1.1.4 prettier-vscode3.20.0 vscode-test-explorer2.17.0 vscode-peacock3.6.0 language-haskell2.7.0 center-editor-window2.3.0 chat0.31.0 restructuredtext120.0.0 fish-ide0.4.0 regionfolder1.0.8 Kotlin1.7.1 rust-analyzer0.1."}
{"_id":"doc-en-vscode-597b9106b0b0856da18f7996f8a46838bb54f8a2edadd41eec9688559acf8552","title":"","text":"Lisp0.1.10 ninja-syntax1.0.0 prettify-json0.0.3 vscode-json51.0.0 vscode-docker1.0.0 csharp1.21.13 sublime-keybindings4.0.6 vscode-typescript-tslint-plugin1.2.3 vsliveshare1.0.1711 vsliveshare-audio0.1.80 vsliveshare-pack0.3.4 debugger-for-chrome4.12.6 idl0.3.1 autotrim1.0.6 proto0.0.2 material-icon-theme4.0.1 polacode0.3.4 tcl0.1.0 crates0.4.7 trailing-spaces0.3.1 vscode-scheme0.4.0 fish-vscode0.2.1 shader1.1.4 rewrap1.9.1 shellcheck0.9.0 org-mode0.5.0 cmake0.0.17 vscode-lldb1.5.0 vscode-icons10.0.0 material-theme3.2.3 vscode-open-in-github1.3.6 tslint-snippets1.1.0 (11 theme extensions excluded) </details<!-- generated by issue reporter --\nFWIW this does seem to be respected now and I haven't changed anything, maybe indicating that it was being ignored for some reason during all this? Not sure if this is important, but worth noting I guess.\nHit the same issue here, VSCode opened up a few hundred projects after upgrading and my mac nearly fell over:\nYeah, I expected this to happen. Works now \"correctly\" after pushed This is expected hot exit behaviour to restore workspaces that have backed up files.\nHit this last night as well and it completely froze my machine up for minutes opening at least 40 windows. I get why this is an expected fix for a bug, but I it made my computer totally unresponsive and probably could have used a more progressive one-time rollout.\nSo it looks like this was a regression in 1.42 () and it's now fixed as it's designed. It's designed this way so that by default, backups don't go \"missing\" by accidentally closing window(s) and forgetting which one had your notes in it."}
{"_id":"doc-en-vscode-fbea0b747237c6bfd5255764346c75ec7dc68ea3b98951bbf868b1e05222e174","title":"","text":"I can attest to the fact that it's awful when many windows unexpectedly restore as it happened to me when testing settings sync. I'm not sure what the right recovery fix is but you can workaround this in the meantime by setting in your . To go back you would need to either delete the backup files manually ( from the 's folder) or open the workspaces and close unsaved files.\nJust to clarify, for anyone coming to this thread who's hosed with 40+ windows, it sounds like what you're saying is the workaround is set to so that we don't have to go through a ton of different \"do you want to save this?\" prompts when we try to close every window. set to (or another reasonable setting) so that VS Code doesn't try to open a ton of windows for every workspace with unsaved files. delete the unsaved buffers manually until things are bearable again. Is that the recommendation?\nIs all I needed to stop the window tsunami.\nif you just set you should be fine if you then open a new folder () and then close the window. This is actually how I've always used hot exit and recommend it if you haven't tried, it takes the rails off and was meant so that you can close windows arbitrarily even on macOS as well as not restoring all windows with backups whenever VS Code gets launched. If you want to go back to , deleting the buffers manually after reviewing is probably the best way to go.\nI'm struggling to find a way to properly fix this for users without just changing the behavior something since this change in behavior looks to have been in . Maybe some restriction on the maximum number of windows that can be opened at once, with a dialog if it's over 5 or something? do you know what the behavior was before? seems to add folders to a the active workspace, does this mean all workspaces with folders were getting restored into a single window, or did they get dropped somehow?\nI am quite sure we opened 1 window per backup directory with backups. So ended up always restoring every window that has backups. We decided with this approach to ensure we always bring back dirty files into their windows independent from any setting (other than the hot exit one at least). Btw untitled workspaces behave exactly the same for the same reasons because they may contain setting changes that are unsaved."}
{"_id":"doc-en-vscode-5da1a67dfa7ff117f5433c101bea84a184a46ddb4a434b83cc5eb7e6b40798c6","title":"","text":"I think you're describing the designed behavior, I mean to ask about the behavior to 1.43 so that it's clear how having caused the backups to behave and why backups stacked up not getting restored (was this because they weren't being saved to anymore?).\nI think can comment if there was some intention behind the change or just an oversight. To be clear: I would never expect a folder with backups to be to a workspace to turn it into a multi-root workspace.\nOne thing might be considered, as discussed with in Gitter, is to limit the number of windows/workspaces to be restored upon start, such as 10 at most, which might relieve this \"correction\".\nWe decided to fix this by reverting the change and no longer restore the folders with unsaved content. That was the behavior from 1.32 to 1.42 and as we didn't get any complaints, this seems to be practical. You can always open these folders from the recent menu, and unsaved content is restored. I created as a followup issue for the next milestone. We plan to also no longer restore workspaces (except unsaved/untitled workspaces) and we will add more information to the recent menu to show which folders/workspaces have unsaved content.\nTo verify: set open 2 windows, each opened on a folder. Make a unsaved change or add a untitled file with content close both windows. There should be no questions asked, exit VSCode open VSCode on an other folder (). Only one window should appear"}
{"_id":"doc-en-vscode-628657b2da39467893c306bb0c35ccb808da225fa971962dcd660589102139cc","title":"","text":"When a user opens a file extension that does not have a specific recommendation tied to it, we show the general notification that points users to the marketplace. This has the following flaws: Most popular file types are already covered by specific extensions, so what gets left in this experience is leftovers For most of the extensions there is no good extension to handle this experience. Thus opening the Extension page does not provide value This is adding un-needed pings to the Marketplace service Also we had a bug that this feature was not available in stable and only 3 users found it is not there Would love for us to get some telemetry on this actions to see if this is actually bringing value to users. If not, we should either: Remove this flow completely Have a pre-defined set of file extensions for which we show this. Checking with tags seems to be very faulty. In my example below there is not an extension that helps with PowerPoint files. !\nOnly of these notifications end up with users clicking on Search Marketplace ! When it is actually clicked here are the top extensions ! Thus I suggest that we: Remove this dialog Add explicit extension recommendation for and . Since those seem to be missing. what do you think?\nSounds good to me\nGreat. Ok, you remove the dialog when you find time, I will add the recommendations for csv and pdf. Or if you go on vacation, I can also remove the dialog. I am working in August. Thanks\nFeel free to remove it in August, otherwise I can also do it in September. Here it is\nTo verify: a file with some strange extension. For example -make sure no notification shows up a .py file or .go file. Make sure you get an extension recommendation for specific python / go extensions"}
{"_id":"doc-en-vscode-53ec8145a38fc06c1e7e1ce673af80597669c2df4bf7cf2cb9ba17f4308e20c9","title":"","text":"Testing Currently the \"Undo last commit\" command has not effect while the commit editor is open. It should either: disabled enabled and on execution close the editor and abandon the current commit I think option 1 is better !\nPart of disablement revamp effort:\nIt also shows up in command palette !"}
{"_id":"doc-en-vscode-3793a02ce136e8b0e214c677a1e81534315aae0b8de2756ae9050352d9d56467","title":"","text":"<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.41.1 OS Version: Windows 10 Steps to Reproduce: the \"Explorer\" view (e.g. with Ctrl+B) a file from the Windows Explorer to drop it in the Explorer icon in vscode. explorer does not automatically open, which would be somewhat helpful, in order to be able to drop the file in the folder of your choice directly. <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes\nThis is a fair feature request for the that on dragOver the corresponding view would get open. I think this feature reuqest makes sense in general, since in the future we might want to drag and drop views between containers, and then the corresponding container should expand. Thus also assigning to\nthis is a regression though. Testing I can move a file from the desktop over the (collapsed) activity bar and views expand properly and focus depending on which view is hovered. This was very useful to drag images to for writing release notes. Can we bring it back?\nRe-labeled it then, and assigned to Janaury so we try to fix it this milestone. hope that makes sense.\nThanks a lot for the fix"}
{"_id":"doc-en-vscode-a7d19ec1223ea15d252e6235e05ad3ab65973761eb46b081c9d7b3ae803f79db","title":"","text":"! 'void' should not as column num Version: 1.70.0-insider (user setup) Commit: Date: 2022-07-26T05:16:55.337Z Electron: 18.3.5 Chromium: 100.0.4896.160 16.13.2 V8: 10.0.139.17-electron.0 OS: Windows_NT x64 10.0.\nYour Insiders is outdated. Please update and retest.\nVersion: 1.71.0-insider (user setup) Commit: Date: 2022-08-12T05:17:00.323Z Electron: 19.0.11 Chromium: 102.0.5005.167 16.14.2 V8: 10.2.154.15-electron.0 OS: Windows_NT x64 10.0. Sandboxed: Yes still have problem\nsorry for mistake close\nTricky edge case, perhaps we should parse the col/row and remove col if invalid and col+row if they're both invalid?"}
{"_id":"doc-en-vscode-f5e9079860545d2f297cbb711c57d914420489068b403e02fafa232be518886d","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --Version: 1.78.1 (system setup) Commit: Date: 2023-05-04T09:48:08.683Z Electron: 22.5.1 Chromium: 108.0.5359.215 16.17.1 V8: 10.8.168.25-electron.0 OS: WindowsNT x64 10.0. Sandboxed: No Steps to Reproduce: a workspace at a UNC path that isn't set in security.allowedUNCHosts. \"Yes\" when prompted to temporarily allow working from that host. changes to one or more files without_ saving them. VSCode without closing the workspace (may require specific settings to automatically re-open the previous workspace). time, click the \"Learn More\" button when prompted to allow working from that host. Note that any unsaved editors remain, but the rest of the workspace is not loaded as though the user clicked the \"No\" button. I'd argue that \"Learn More\" doesn't constitute a decision from the user and shouldn't proceed as though it was. Depending on the nature of the changes, this may cause a problem for safely saving those changes because if the user tries to re-open the workspace, they are asked to either save or close these orphaned editors. Perhaps a link among the text in the dialogue box would be more appropriate than a button which closes the box.\nI agree that the dialog should not close and/or make a decision when you click \"Learn More\". Unfortunately the dialog closes when you press that button, so best I can do is to open it again asking the same question."}
{"_id":"doc-en-vscode-71d82df21e5ef92d963d0b6754e30265793e63780373d5b3db7ea2edc5d04821","title":"","text":"Markdown file on Plain Text on startup: ! Same with TS in another window: ! Reproduces by reloading the window. Not sure who the right person is for this Version: 1.93.0-insider (user setup) Commit: Date: 2024-08-08T05:03:51.428Z Electron: 30.1.2 ElectronBuildId: Chromium: 124.0.6367.243 20.14.0 V8: 12.4.254.20-electron.0 OS: Windows_NT x64 10.0.\nOn it"}
{"_id":"doc-en-vscode-150e8b460e2b012b96f8babdd6adb0dc16a3842042d7d01919a5996ec5b61d29","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Type: Bug Missing pricing parameter in extension manifest. The attribute is not included in the and so does not show up in IntelliSense. Related docs issue: I'm working on the fix. FYI: <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --"}
{"_id":"doc-en-vscode-ce799a6b083ea4adca8e057a225d0e43795cb075f0117f6a4c97c081d077f513","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes The behavior was correct few weeks ago. <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --Version : 1.65.0 Validation : Date : 2022-03-02T11:12:36.248Z Electron : 13.5.2 Chromium: 91.0.4472.164 : 14.16.0 V8 : 9.1.269.39-electron.0 Système d’exploitation : Linux x64 5.16.11-100.fc34.x86_64 Steps to Reproduce: an integrated terminal and type the following command : , alt + click on the printed link browser is open with url , but it should be opened with url ! Working test on native terminal : a native linux terminal (for my DE, it is gnome-terminal with zsh) the following command : , ctrl + click on the printed link browser is open with url , as this is the correct expected behavior !\nLooks like we're passing the right thing to the opener service !\nI can repro this with: I can copy and paste that URL into a browser and it will work. I can run the same code in regular terminal and it will work. It doesn't work when I click on the link from w/in the vscode terminal.\nThis has come up before where it was closed by design, originally I don't think it was possible to open a link like this because opener service always works on uris and they always encode like this. I notice though that the editor properly retains the in a link though so it should be possible now.\nOkay thanks!\nThank you"}
{"_id":"doc-en-vscode-0e13cd891ea8e4e7943f141eac9a02599149a85f1d62903e67276cbff931eeee","title":"","text":"Environment Details: VSCode Version : 1.24.1 OS Version : Win10 Additional Details: MAS Violated: MAS2.4.7 Repro Steps: VS Code Keyboard Shortcuts from the Settings icon on the bottom left bar. Dark+(default dark) theme and other dark themes and check the focus visibility. Actual: Keyboard focus is not clearly visible on Sort by precedence button when default dark theme is applied. Expected: Focus should be clearly visible and should be outlined by solid or dotted line. Recommendations: Refer below link which is repository of bug fixes code snippets: User Impact: User will not be able to see the focus properly. MAS Reference Attachment for Reference: ! Does this issue occur when all extensions are disabled?: Yes\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.593 --- <!-- score: 0.523 --- <!-- score: 0.457 --<!-- potentialduplicatescomment --\nSounds like colors and contrast, also assigning to\ncan you clarify if this also applies to our overall focus treatment? We use the same blue outline for most of our focus treatments: ! ! !\nYou can use any color for emphasizing the focus but it should be clearly visible to the user on all the color themes.\nPlease let me know when there are any changes to be done from my side.\nwill do, I think I'll need to play around with the offset of the outline color (and possibly opacity).\nThis issue has been closed automatically because it needs more information and has not had recent activity. See also our guidelines. Happy Coding!"}
{"_id":"doc-en-vscode-65fc933ed0946819c6467f64384e073c1be0e6e5067b41d748a40e478e289749","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: OS Version (in docker): building and running the following dockerfile gives the following: changing the command in the docker file to gives the following: creating a non-sudo user in the docker container and running it from that works, but that leads to a headache of problems with other parts of my docker image. ideally, the fix suggested by code in the first example would solve this but it doesnt.\nrelated to\ndoesn't apply to . That's an option for electron, and doesn't run electron. Just omit that argument and run the command.\nomitting just makes it complain about being unexpected. no combination of arguments fixes this as far as i can tell causes causes causes causes\nWhat code print?\ni put the output of at the top of my issue, but here it is again:\nwould it be possible to reopen this? it seems like there is no solution based on whats been said so far\nLooks like a workaround would be setting when you call . But the better solution is just to use the CLI () instead of the full desktop install. This will be fixed when the desktop install uses the native CLI as its primary entrypoint.\nWorks with . The is important, otherwise it will be misinterpreted.\nI still get the same error: adding CANRUNAS_ROOT=1 did not make a difference.\nI cannot repro, can you share the exact command you ran?\nAll is good, sorry, I tested against an old version!\nThis issue will be closed because it appears to be verified"}
{"_id":"doc-en-vscode-f055dfa9d8c52f05b6bda20b8d56f23f4365be7860b6ba2728332009dae108f5","title":"","text":"Type: <bBug</bCC a terminal in editor area in the terminal alt+f2 to open Accessible View alt+UpArrow and alt+DownArrow to see if it moves between executed commands. Notes: alt+DownArrow unexpectedly opens a new terminal view. VS Code version: Code - Insiders 1.83.0-insider (, 2023-09-21T05:34:38.227Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (91)</summaryExtensionVersion ------ android-dev-ext1.3.2 aiprm-lang0.0.2 Bookmarks13.4.1 openscad1.2.1 spellright3.0.118 zoterolatex0.4.1 mermaid-markdown-syntax-highlighting1.5.2 doxdocgen1.4.0 vscode-markdownlint0.51.0 vscode-eslint2.4.2 vscode-quick-select0.2.9 vscode-deno3.23.1 gitlens14.3.0 EditorConfig0.16.4 prettier-vscode10.1.0 vscode-google-translate1.4.13 codespaces1.15.3 copilot1.115.430 copilot-chat0.8. remotehub0.60.0 vscode-github-actions0.26.2 vscode-pull-request-github0.72.0 overleaf-workshop0.2.1 cslpreview0.2.2 easy-snippet0.6.3 path-autocomplete1.25.0 latex-workshop9.14.0 lilypond-syntax0.1.1 scheme0.2.0 better-cpp-syntax1.17.2 google-search0.0.1 vscode-lua-format1.3.8 lilypond-formatter0.2.3 lilypond-pdf-preview0.2.8 lilypond-snippets0.1.1 vslilypond1.7.3 zotero0.1.10 git-graph1.30.0 vscode-docker1.26.1 black-formatter2023.4.1 flake82023.6.0 isort2023.11."}
{"_id":"doc-en-vscode-a2a08b87baa90cbaee8f4069b8f2df8286a977a1a9223b9b0bb66c3e43af2b01","title":"","text":"python2023.16.0 vscode-pylance2023.9.20 jupyter2023.8. jupyter-keymap1.1.2 jupyter-renderers1.0.17 vscode-jupyter-cell-tags0.1.8 vscode-jupyter-slideshow0.1.5 remote-containers0.311.0 remote-ssh0.106.4 remote-ssh-edit0.86.0 remote-wsl0.81.4 vscode-remote-extensionpack0.24.0 azure-repos0.36.0 cmake-tools1.15.31 cpptools1.17.5 cpptools-extension-pack1.3.0 js-debug-nightly2023.9.2017 powershell2023.6.0 remote-repositories0.38.1 vscode-github-issue-notebooks0.0.129 vscode-selfhost-test-provider0.3.18 vscode-serial-monitor0.10.0 vsliveshare1.0.5883 autodocstring0.6.1 pandocciter0.10.3 shiny-python0.1.4 shinyuieditor0.4.3 quarto1.98.0 r-debugger0.5.4 java1.22.1 vscode-xml0.26.1 r2.8.1 multi-command1.6.0 vscode-deepl1.0.6 abc-music0.4.0 lua3.7.0 latex-utilities0.4.10 cmake0.0.17 errorlens3.13.0 intellicode-api-usage-examples0.2.8 vscodeintellicode1.2.30 vscode-arduino0.6.0 vscode-java-debug0.54.0 vscode-java-dependency0.23.1 vscode-java-pack0.25.14 vscode-java-test0.39.1 vscode-maven0.42.0 markdown-all-in-one3.5.1 grammarly0.22.1 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nI could repro this yesterday, but it's working for me in today's insider's. Pls let me know if it's not for you.\nStill does not work. Tested on Windows 11.\nI had forgotten this is an issue only when the terminal is in the editor area."}
{"_id":"doc-en-vscode-550215f2e2ef2a2b0613cb718b295537256b04566839fc51ee255fd6bb0fa403","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No Does this issue occur when all extensions are disabled?: Yes/No VS Code Version: 1.77.1 OS Version: Ubuntu 22.10 Steps to Reproduce: !\nThe bot never kicked in, can you check that it is fixed in the latest Insiders build of VS Code?\nI can confirm, it is fixed!"}
{"_id":"doc-en-vscode-9ddd322842e082aae3c1fd1bcef6aac150743b12650ec73aacdd2dbc0a8cd694","title":"","text":"I don't know the codicons by heart. So I went with . Here's what I see: ! ! Can you validate if is valid? Or somehow show a better default? I'd prefer validation though. Testing"}
{"_id":"doc-en-vscode-69d8ba1b2b6352b8a4a45673abfe2dee86eb9b4485b98e6225ef3795de600427","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.77.3 OS Version: MacOS 13.3.1 Steps to Reproduce: Run the following commands Trust the workspace, then start the script from the \"NPM Scripts\" panel. The script runs, and the name of the script appears to the right of the output panel, together with a spinner showing that it's running. Now run the script again. A popup window appears saying that it's already running, and do you want to terminate or restart it (just close it without doing either). This is all expected behavior. Now (with the task running) run Cmd-Shift-P \"Developer: Reload Window\". The window reloads, and the script continues to run (you can tell it's the same script because it's printing its pid, and it doesn't change). The name is still in the right hand pane, but now the spinner has gone, as if vscode no longer knows that it's running. Now run the script again, and although it's already running, it starts a second copy. This is the bug. It's especially painful for \"runOnOpen\" tasks, because every time you reload the window (eg because an extension needs updating) they all continue to run AND a new copy of each one starts. I thought there might be another bug in the recently (since that was supposed to fix exactly this issue - ), and took a look at that."}
{"_id":"doc-en-vscode-ddbdeab34f6289d2693c15941b41610786e27475d6c3daecece49fd8666b1c7f","title":"","text":"There does appear to be a pretty serious bug in that code: It adds to the at line 1839, then starts the task at line 1853, and then removes at line 1855 without waiting for the task to finish. So in practice the list is almost always empty, and doesn't actually do anything. I think the at line 1855 should actually be in a clause attached to the on line 1856, so that it doesn't run until the task completes. But fixing that doesn't actually fix this issue, because the set isn't preserved (or reconstructed) across reloads. In addition it breaks the above behavior where trying to run an already running task asks what you want to do with it (because now that inProgressTasks is actually working, it aborts the attempt to re-run the task before it gets to the point where that dialog is presented). It also breaks tasks that are explicitly marked as allowing more than one instance. So since it isn't currently doing anything, and fixing it breaks things, I would suggest removing all the associated code ( I'm not sure what the actual fix should be though. There seems to be code to try to reconnect tasks to terminals: but it doesn't seem to work (I've stepped through the code with my example, and it goes into but doesn't actually find any tasks to reconnect). So maybe the bug is there. Or maybe the bug is that the tasks shouldn't be preserved across reload (many are not; a terminal that isn't running anything will be killed by reload, for example). But I've reached the limits of what I can debug, unless anyone has suggestions for where to look.\nthanks for looking into this. the reason it's not in a is this was meant to (and did) fix an issue that was super hard to reproduce where duplicate tasks would get started. the only tasks that get reconnected to are those with and have problem matchers. you can disable this with .\ntask reconnection works for me https://user-\nOk. So there's a race that happens between deciding to launch the task and actually launching it? In that case, the solution seems kind of misleading. Shouldn't it be something like rather than ? but also, instead of why not just do And why do you need the in since it was already deleted right after it launched?"}
{"_id":"doc-en-vscode-f43d588c75026e4cb7dba3132bf570d5611c524079456ffe09596ef9e16da423","title":"","text":"But all this is incidental; I was hoping this code could prevent the bug I reported, but clearly it can't.\nOk... but the problem is that my task is left running across the reload, but it doesn't get reconnected (apparently intentionally), so attempting to start it again succeeds, even though it's supposed to only allow one instance. https://user- I'm not sure what that .mov is supposed to be showing me; I can't see anything relating to tasks being reconnected (but I'm not sure what Im looking for). But anyway, I'm not claiming it's broken; just that it doesn't reconnect my task (or similar ones). And if, as you say, thats intentional, then the bug is elsewhere. But as far as I can tell, either my task shouldn't be left running, or it should be reconnected.\nAnother data point. After reloading, the task is clearly running, but tells me \"No running tasks\"\nSo when I set this to false, my task is killed on window reload, which does mitigate the problem (in fact, for my use case, I think this does solve the problem - although I don't know yet whether it introduces new ones). But that still leaves us with the issue when is . The task is left running, but not (properly) reconnected - so although it's there, and its output is going to.a terminal, the abstractTaskService, and terminalTaskSystem aren't aware of it (so that \"Show Running Tasks\" says there are none, and attempting to start a second instance succeeds).\nI won't be looking into this any time soon just fyi. So if it bothers you, feel free to keep digging. Otherwise, just keep the setting disabled.\nI run into this problem all the time too, which is very frustrating. I have to reload often due to SSH reconnection etc, this reattach issue together with ssh agent forwarding on mac os breaking, are really detrimental to the work flow. Using works around some of the issue, but exposes a new one; one of my processes doesn't get killed and blocks the new one from starting. That process appears to be a nodemon process, perhaps related to\nYeah that is unrelated to this feature as you mentioned.\nas noted in the pull request, this doesn't fix everything, but it makes my workflow (with several auto-run tasks) much cleaner."}
{"_id":"doc-en-vscode-6be88b04ede3f6e3bd5aae6594f7b4be5ed98fba7bba8427c89a48ac17a98bd9","title":"","text":"Those tasks will be properly reconnected on reload, rather than an additional set being kicked off on each reload. For tasks that don't auto-run on reload, the only change is that the first attempt to run a new instance of that task will simply attach to the existing process; and subsequent attempts will fail the same way they do before the reload. Other than that, there should be no change in behavior, so I think this should be pretty safe. I spent some time trying to figure out how to actually reconnect all the running tasks, rather than just marking them for reconnection, but I kept running into issues where needed to know about the internals of ( in particular). Or I needed to extend the ITaskSystem api - which I'm reluctant to do without guidance. I also tried a completely different approach - namely, including non-background tasks in . That almost worked, but the problem was that gave a different result when the task is launched manually vs when it's run from (when run manually, its a , but when run from its a , which gets resolved to a ). As a result, the tasks did get reconnected properly, the various commands (eg ) know they're there and do the right thing; but now trying to run them manually uses a different key, and still starts a new copy...\nthanks for looking into this. I commented on your PR\nTo verify, run a non background task. Reload the window. Verify that you can run that task again"}
{"_id":"doc-en-vscode-7e4bd1a8054eafe3c714a70a69382bfc7e73aca2497d56227263854dc90b7d4a","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.88.0-insider OS Version: Darwin arm64 23.2.0 Steps to Reproduce: folder text View as Tree View as Tree ! View as List ! Mysteriously, some of them do show up when run in a project with many folders...\nDid not reproduce in\nSame for me: Version: 1.88.0-insider (user setup) Commit: Date: 2024-02-29T22:51:22.369Z Electron: 28.2.2 ElectronBuildId: Chromium: 120.0.6099.276 18.18.2 V8: 12.0.267.19-electron.0 OS: Windows_NT x64 10.0.\nAlso for non-tree view it doesn't show the single result: 3 results shown: ! 1 result isn't shown: !\nwatching :\nThere is no way to set aiResultsVisible and it always returns false now. The hasDownstreamNonAIResults called here should also only contain NonAIResults, so it is false and should not be in the if statement, but it is now. Upon deeper investigation, we found that it occurs under the following conditions. It seems that the deepest part of the TREE is not displayed in this way. This is because the folderMatchesIterator is retrieved, but the deepest point is always [] and [].some() =false. You also need to look into fileMatches iterators.\nApologies for the inconvenience, and thanks for finding the fix!"}
{"_id":"doc-en-vscode-e60fd1f7a3daf0f8b3de0fcbb76b58ae0f6f8679377a2c8aedd704b78f460f7f","title":"","text":"It runs the correct intent, but renders rather than ! ! This does not happen in panel chat\nYou implemented this, right?\nYes, the problem is here"}
{"_id":"doc-en-vscode-66a20aadb10c43057269e3174d1120b4c2fa6c30e5b351de11baf1331cb0f038","title":"","text":"The code in question looks like: This doesn't actually do anything, and is likely a bug. There are a few cases in ."}
{"_id":"doc-en-vscode-8f5298fca6d291caa8e1752157288f565c72927d7acafdf4a88dccfa3d0abc12","title":"","text":"Issue Type: <bBug</bFold a region and observe it. Unfortunately this went away after a window reload: VS Code version: Code - Insiders 1.66.0-insider (, 2022-03-23T05:27:08.615Z) OS version: Darwin arm64 21.4.0 Restricted Mode: No <!-- generated by issue reporter --\nThere haven't been any changes from mein that area recently. Likely a Chromium glitch. Please reopen if you have it reproducible.\nI still don't know why this happens sometimes, but it shows up regularly. In the dev tools it shows that chromium has loaded the stylesheet in a wrongly encoded way. I hope to avoid this by escaping the ellipses unicode character."}
{"_id":"doc-en-vscode-dcbdd8e4c7391ac6ba53aea50af07a3b64d4f71112c04060bfbf1cb8f2244f7a","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.92.0 Commit: Date: 2024-07-31T23:26:45.634Z Electron: 30.1.2 ElectronBuildId: Chromium: 124.0.6367.243 20.14.0 V8: 12.4.254.20-electron.0 OS: Linux x64 6.9.3--generic Steps to Reproduce: theme to Light High Contrast a Javascript file final bracket is black, not red as in other themes\nI believe this may be the core cause: The value for hcLight here is just an empty string\nThanks for investigating, looks like this color got forgotten when creating the Light HC theme. I took the same color as used in the overview ruler"}
{"_id":"doc-en-vscode-2390845ecc5c6d11d808b8f52a52ff460c7e20f9b1bc4278839c3b1a515c2767","title":"","text":"with the remote repositories and codespaces extensions installed the remote explorer on under prompt (see ) escape to close and cancel Bug A url is still opened Version: 1.68.0-insider Commit: User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X ) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/101.0.4951.64 Safari/537.36 Embedder: <!-- generated by web issue reporter --\nI'm not seeing the section that Matt describes in the linked issue. !\nThe remote explorer is in the sidebar: !"}
{"_id":"doc-en-vscode-89663826d01a6c2c5033e83f10f46332f72106d3703b44f540b1d653dcefa9fa","title":"","text":"When I disable the \"Show raw keys\" option (showKeys) in the VS Code settings, the editor no longer shows the shortcuts either. If the \"showKeybingings\" option is enabled it should show the shortcuts. This is not behaving the way expected. !\nWhen I set these settings: <img width=\"652\" alt=\"image\" src=\"\"I still see the command if I hit . You don't see it?\nUnfortunately it doesn't work for me in this same configuration. Version: 1.85.1 Commit: Date: 2023-12-13T09:47:11.635Z Electron: 25.9.7 ElectronBuildId: Chromium: 114.0.5735.289 18.15.0 V8: 11.4.183.29-electron.0 OS: Linux x64 6.5.0-14-generic\nI found something strange. Some commands works fine, like \"Ctrl + N\", \"Ctrl + P\", \"Ctrl + Shit + P\", \"Ctrl + Tab\". But it still doesn't work with commands like: \"Ctrl + C\", \"Ctrl + V\", \"Ctrl + A\", \"Ctrl + X\", \"Alt + Arrows\", \"Ctrl + Arrows\". Could you please test the commands when editing code?\nSadly this is this: A big chunk of our internal commands simply do not have good descriptions. It requires a large change over the code base. Feel free to send a PR my way which adds descriptions to the few that you are dependent on, if this is important.\nOk, but I would like it to show commands such as \"Ctrl + C\" even without a description and that is not happening. If I set it to show only the commands it continues to behave the same way, without showing the commands in the editor part. I remember that when I sent the commit that fixed some problems in the screencast, it was working perfectly, but now it no longer works. ! It still doesn't work in this configuration.\nYou simply closed the issue and didn't even understand the problem I brought up. I see that these are two different problems. This is very frustrating...\nYou're right, sorry for jumping the gun there, and thanks for the additional explanation. Fix coming."}
{"_id":"doc-en-vscode-60766d38b8541fc2d321439c930b20e5f5e24e5b13718414fad91ec91d8996fe","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: It requires at least the remote SSH extension to be enabled due to the nature of the bug, all other extensions can be disabled/uninstalled. <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.75.0 OS Version: Local Windows 10 21H2 / Remote Fedora 29 Steps to Reproduce: -up VS Code with Remote SSH Python Extension in to Remote SSH error message and get asked to download manually opens link: the following error text: it's because the extension hasn't been published for 2023.2.0 for Linux yet so try installing older version same error and browser instead opens this link: very similar error text: I have had to revert to 1.74.3 where I am not getting this error so can I continue to work, but if you need me to test things again I can install 1.75.0 again and retest.\nI am having the same issue since upgrading to VSCode 1.75.0. When trying to update the Python extension on the remote server I get an with the log Trying to download manually brings me to the as reported above. I am getting the same errors for the Jupyter extension. I could successfully update the extensions locally. VSCode: 1.75.0 OS: local openSUSE Tumbleweed / remote Red Hat Enterprise Linux 8.5 (Ootpa)\nI have the same problem.\nSame issue. Downgraded VSCode to 1.74.3 to restore functionality."}
{"_id":"doc-en-vscode-b249d6f56d684c598450373b22127255ff5ff0a173fa138e5fc69dd8a4b8cc76","title":"","text":"VS Code Version: 1.75.0 OS Version: Local: macOS Monterey 12.6.3 Remote: Amazon Linux 2 w/ Kernel 5.4\nTriaging this issue I am unsure if the initial post with the error message of is the same as the error with the message . Therefore all conversation on the issue should be moved to this other issue: Thanks\nThis issue is related to target platform and not related to sign verification.\nYes, I am initial poster, I also get the signature untrusted error. I assumed the two errors had the same underlying cause, i.e. when VS Code made the web request it got the above error which does not have the signature of the extension and therefore throws a red herring of \"Untrusted: Signature\", then when VS Code prompts you to download it manually you can see the actual underlying error. This is why I posted the latter error and not the former error. But if you are convinced they are two separate errors then it makes sense to track them separately. But be aware I got both errors, and everyone else who has posted seems to have also got both errors.\nThanks for the info It seems that the root cause is signature error for you too. Being said that the manual download shall work. Can you please share the link you are using to download the VSIX?\nFrom the marketplace web page. Version History -2023.3. -Download Manual download and install work for me.\nAs per my original post the link that VS Code to manually download the extension is: Which gives this error:\nIt seems VS Code is generating an invalid download link. Updating this issue for fixing the update link. For signing related, please refer to\nI'm not sure how to get this, are there repro steps?\nTo verify Open extensions view Browse for eslint extension (any non platform specific extension) Disable network connection Click on install, you shall see a prompt to Download Manually Enable network connection and click on that Download Manually button Make sure the VSIX is downloaded/"}
{"_id":"doc-en-vscode-9b9dac0c2f8b0f8d1a1c3cc3e7ea582cf5048d149d9f095218054cfa9ccf3cc1","title":"","text":"Moving this over from After the backups are not restored if the workspace directory doesn't exist anymore. The workspace is no longer tracked but the backups are kept around in the backup directory should they be needed later.\nTricky.... I would expect an empty window to open with those files dirty. This might require some work on the area of restoring on empty workspaces. Have not thought about this longer yet.\nMarking as important because this means data loss if the folder is deleted on disk.\nmaybe this should be done on the main side on startup similarly to If a folder no longer exists, convert it into an empty workspace backup and let it open as empty window.\nOn test plan:"}
{"_id":"doc-en-vscode-ddcf568d39ee1c2c2b6c29652e19a40c388896d18079144e19ebff47110bcf4b","title":"","text":"Testing scenario in Codespaces web UI Opening a Python file in a fresh codespace Web UI shows a notification: ! Clicking on Yes shows ! VSCode desktop is opened Python calls VSCode should intercept that URI and handle it by directly opening the extensions viewlet and not make an external call\nUnassigning myself as it seems url handling related. Please let me know if there is something I have to do in extensions area.\nWhat do you think? Should we perform the fast-track to handle these URIs internally?\nI can't repro this in desktop Code. I can repro in web\nI couldn't get the Python extension to show that dialog again, so I verified that with my own extension that does That code is inspired by That code works and, on codespaces now directly opens the extension viewlet. -verified However, if you do '();' it opens the external open dialog again. In the first case the URL has a relative path (), in the second one an absolute (). Relative paths are kind of rare. Is that all as designed? Or would it be better to also accept the absolute path?\nPaths are not required to be rooted in , or even separated by characters: they can be arbitrary strings within the allowed character set. We've long defined extension URIs to simply be . Not sure if you've seen the format anywhere in our docs? If so we gotta fix it.\nI will have to revert the fix for this, since it broke existing functionality. With the fix, a web extension can no longer actually open a URI in desktop Code. When looking at the breakage, I no longer this this is true. If is invoked, it should always be handled externally, never internally, unless it's absolutely obvious (running in Desktop). The right way to move forward here is to have the Python extension run the instead of calling ."}
{"_id":"doc-en-vscode-79f36c906d6185205fc9ffcb4b05edf8fdd225f5a26c912b1ecf14436b4bbbec","title":"","text":"VSCode Version: 1.37.1 OS Version: Linux x64 5.1.14--generic Steps to Reproduce: a search term (make sure it actually exists in the current document) the replace term empty the match case box Ctrl-Alt-T the error ! Does this issue occur when all extensions are disabled?: Yes\nWhat does do here? I don't think I have that keybinding.\nI think I meant Ctrl+Alt+Enter, the shortcut for replacing all. I am not sure if it is the correct shortcut (I don't have access to my laptop right now), but it is not specific to it. I am guessing replace or replace all will trigger the same bug.\nI can't reproduce it. It would help if you can try it again and get the full error stack from the developer tools (Help Toggle Developer Tools)\nThis issue has been closed automatically because it needs more information and has not had recent activity. See also our guidelines. Happy Coding!"}
{"_id":"doc-en-vscode-742468c8ba115a13289fe235c5a58211da6ff3e65698b544cf6f1acac7e7a574","title":"","text":"Tracks updating the TS 5.4.4. recovery build\nActually may wait for TS 5.4.5 with a fix for auto imports\nMerging 5.4.4 for now"}
{"_id":"doc-en-vscode-8769dc54a0a579febc881ce092bc8904756e61d40dd7bcfead6c8722222a37b8","title":"","text":"The \"Open in Terminal\" does nothing when the current window has no folders. Instead, if there are no folders, it should open a Terminal window in the same directory as the current file.\nthis works for me, how are you running \"Open in Terminal\" exactly?\nI'm using while I have a saved file focused in VSC. !\nThat command opens the active workspace folder currently. Maybe we should fallback to the active file if there is no folder open? Open to PRs\nI would like to work on this as my first bug. Could I get some advice on how I should go about fixing this?\nGreat! All the external terminal code is located in the \"execution\" part You change will be in here: Utilizing the (see for how to pull it into that file)."}
{"_id":"doc-en-vscode-6369c75e2ac28ce4ceb26ef68e35b11e7b8e5e6a144b429b29169492b6cc809c","title":"","text":"Type: <bBug</bo meu terminal do vscode ficou todo preto derrepente, ele esta normal e quando fui abri-lo estava preto, nao consigo resolver e nao posso voltar a minha rotina por causa disso. My vscode terminal suddenly went all black, is that normal and when I opened it was black, I can't resolve it and I can't get back to my routine because of this. ! VS Code version: Code 1.81.1 (, 2023-08-09T22:22:42.175Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (8)</summaryExtensionVersion ------ turbo-console-log2.9.6 gitlens14.2.1 auto-rename-tag0.1.10 copilot1.105.350 prettify-json0.0.3 vscode-language-pack-pt-BR1.81. vscode-thunder-client2.10.5 vscode-icons12.5.0 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nPlease write the pn in english.\n/duplicate\ncan you share your file?\nhere it is\nConfirmed that is disabled by default. I have now enabled terminal gpu acceleration and images, though I'm not sure whether I'll run into the issue on this machine."}
{"_id":"doc-en-vscode-f286bd6bbe43d825bbe39309fab2e0797abf6ede47160fdd6b60172a1851c794","title":"","text":"Currently there are only tests for the groupview but not for the model. Add some test to make sure there will not be any regressions in the future."}
{"_id":"doc-en-vscode-de76302416dd4316ec5961386298f01d204c9cb015fb6e7cd85bed663456a74e","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --The vscode shell integration currently needs to be turned off or disabled for anyone with complex prompt commands, or any trap DEBUG. Please consider supporting some kind of chaining model so we can work with the shell integration as needed. Even better - use the defacto package that has already been written. VSCode is the 3rd script that wants to mess with prompts and traps - but it's the first to completely ignore Please consider shipping this along side and sourcing/using it. If that is unacceptable, consider at least detecting this module has been loaded as documented in the repo - and using it: fwiw: iterm2 has very very similar functionality to the vscode shell integration, but I've been able to work with it. Although they embedded the script directly, they kept the compatible function names and mechanisms and interop fine.\nI believe we have already fixed this: Adding to make sure\nWell, line 100 breaks my prompt. Adding the line as the 1st line of vscpreexec shows why. I think perhaps there needs to be 1 more state 'not in prompt' that causes preexec to bail until all promptcommand processing is done. Alternatively, if this is all to send vsccommandoutputstart - it seems like using PS0 on bash = 4.4 would be a lot cleaner?\nHi This was on 1.67.2 with dropped in place.\nWe've made some big changes in the past few days around debug trap and PS1 handling in bash, I believe the issues here are fixed. We may at some point hook into similar to but for now that will just increase the number of cases we would need to test for (contributions very welcome).\nActually I'll keep this open with help wanted to track integrating with bash-preexec.\nOK. Pulling the - Things look better, but not quite right. The problems I see (only when DEBUG trap is being used): Still seems to be not preserving $? for promptcommand Seems to execute the trap twice. repro with setup instructions below: Do 'cd' into some directory. The command '..' should then pop one directory up, but it will try to process twice and pop twice."}
{"_id":"doc-en-vscode-659d754e31112ab26a9d503e9b66fc4f01747e748efe739ecd0ac8caaccda999","title":"","text":"Instead of 'success dots' in the gutter, all commands seem to think there are fail exit status. Here are my dotfiles: and a branch to help debug: The debug branch just build an array of what commands were that triggered the trap and print them at the start and end of building the prompt. It does show the trap being called twice when shell integration is enabled - and once otherwise. The repro environment/steps: Repros in OSX and WSL Ubuntu 22.04 for me. It uses dotbot. In a scratch user directory you can run ./install to link everything into ~/. Pass to ./install to avoid installing extra tooling it finds missing (pygments/tmux/coreutils)There is a script that get's sourced . This is the root of all evil. It's on the hacky side. The function is small and is (I believe) the only relevant part. I should stop using it, but it's sooooo_ ingrained into my muscle memory. To verify: Removing from makes everything work great with none of the above mentioned problems. One hack, that seems irrelevant is that uses debug trap directly, but also detects and uses the pre-exec infrastructure if it has been loaded (to not break iTerm2's shell integration). I appreciate the time and effort going into this feature. It's becoming a shockingly advanced embedded shell environment.\nthanks again, planning on tackling this in June\nI use just merged in a fix for this. It would be great if someone could help verify in tomorrow's insiders build as I'm not normally a user of bash-preexec.\nStill some problems around . Very simple repro: clean environment. My bash and vscode versions are: use this .bashrc: For me (In WSL, repros in plain OSX): ! Without ! If extdebug works, then this would be the next simplest test (I think). Same clean environment. The following\nAre there verif steps for this?\n<!-- AUTHORVERIFICATIONREQUEST --This bug has been fixed in the latest release of ! you can help us out by commenting if things are now working as expected. If things still don't seem right, please ensure you're on version of Insiders (today's or later - you can use in the command palette to check), and leave a comment letting us know what isn't working as expected. Happy Coding!"}
{"_id":"doc-en-vscode-16b8b83d796ad8473fd4b47c326495a7bf55a3ac486d3ea61e7ff1ace596e1e4","title":"","text":"With the more complicated scenario, the decorations don't seem to show blue and red anymore. ! I realize that in the screenshot I forgot to add , but that didn't make a difference.\nwhat is your exact debug trap when that happens?"}
{"_id":"doc-en-vscode-bc341a9b84fd9ed67a0bc061c6de97971a1efc7afd62449d938b3aeb43ea8333","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --Currently, it is treated as . (I have disabled all my extensions and confirm this behavior) It does support as comment like does. A simple workaround now is to set But it's nice to have it built in supported.\nClone the repository Open the repository in VS Code Insiders Rename to Open the file: Confirm that the file type is detected as \"Ignore\" Confirm that the comments in the file are coloured correctly"}
{"_id":"doc-en-vscode-6f55ac2165c2261d2d5b847fd73fc16a32a974556ae1345e8cbea83097295bbf","title":"","text":"Type: <bBug</bCC CC I create a terminal in VSCode, press shift+alt+f1 to display the accessibility help, and then press the esc key to dismiss the displayed information. I believe that the focus should be moved back to the terminal but it doesn't. I need to press shift+tab to move the focus back to the terminal. I tested using orca but I believe the same happens using nvda, I'll test later. VS Code version: Code - Insiders 1.76.0-insider (, 2023-02-09T05:23:03.109Z) OS version: Linux x64 6.1.9-200.fc37.x8664 Modes: Sandboxed: Yes <details<summarySystem Info</summary </details<details<summaryExtensions (37)</summaryExtensionVersion ------ doxdocgen1.4.0 vscode-markdownlint0.49.0 vscode-chatgpt3.5.9 rest-client0.25.1 better-cpp-syntax1.17.2 azure-pipelines1.208.0 vscode-docker1.23.3 csharp1.25.4 vscode-kubernetes-tools1.3.11 remote-containers0.278.0 remote-ssh0.97. remote-ssh-edit0.84.0 azure-account0.11.3 cmake-tools1.13.45 cpptools1.14.1 cpptools-extension-pack1.3.0 makefile-tools0.6.0 powershell-preview2023.2.0 remote-explorer0.3. vsliveshare1.0.5828 fabric8-analytics0.3.6 java1.15. vscode-xml0.25. vscode-yaml1.11.0 sonarlint-vscode3.14.0 cmake0.0.17 intellicode-api-usage-examples0.2.7 vscodeintellicode1.2.30 vscode-boot-dev-pack0.2.1 vscode-spring-boot1.44.0 vscode-java-debug0.48.0 vscode-java-dependency0.21. vscode-java-pack0.25. vscode-java-test0.37. vscode-maven0.40."}
{"_id":"doc-en-vscode-233679a3ae28b2a6dd2cc25b77def6c83cce981b8f55958d12518aa52959ee48","title":"","text":"vscode-spring-boot-dashboard0.10.1 vscode-spring-initializr0.11.2 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --"}
{"_id":"doc-en-vscode-7ee8a689957c63b06c39376634bb67232c9f4ed6b6d322dd895723c543110640","title":"","text":"! It could be that I caused this, but I doubt it (injected text has nothing to do with the gutter, maybe they shared a CSS rule).\nI can reproduce but I'm not aware of any changes in that area from the debug side.\nI would like to work on this issue!\nBefore fixing anything we should do a \"git bisect\" to find the commit that broke this.\ncommit . it should\nthanks for pointing out the commit! If you like, you can provide a PR, that would be awesome!\nPR"}
{"_id":"doc-en-vscode-a26d27615b4417b4c2ddc8fdddeaa7e15d67401a7f7698c2685cf11c7ed7c4ab","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: No. Related to extension api <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.79.0-insider OS Version: Macintosh; Intel Mac OS X 1340 Getting a session for a Microsoft account for the default tenant, and also getting a session for the account but for a specific tenant results in two items for the same account appearing in the Accounts menu. <img width=\"546\" alt=\"image\" src=\"\"Steps to Reproduce: First call with the following scopes: . And another call to with these scopes: I want to point out that the account objects on the returned sessions are actually different: <br<details<summaryMicrosoft Authentication provider logs</summary<p</p</detailscc:\nThese accounts aren't duplicate, they are sessions that exist for each signed-in tenant. I think the best solution would be to change the account label to . This would match what the Azure portal does. <img width=\"201\" alt=\"image\" src=\"\"I think this is an important change to make because this really doesn't look good and is confusing to users. Especially when users are trying to sign out. The screenshot below is after I sign in with a single Microsoft account that has 4 tenants. Having 4 tenants is not uncommon. <img width=\"515\" alt=\"image\" src=\"\"\nYeah this goes back to that \"technically they're different accounts because they're in different tenants\" thing. I'm inclined to agree with you and dedupe per-label.\nAlthough that would present better, I think deduping on the label wouldn't be great unless you made the label more standard across tenants."}
{"_id":"doc-en-vscode-deafd245b22b7357646c0a189d2c02f55f5c25dba4ab68979dc98a750a3a763d","title":"","text":"In that case, I think changing the label to be just email would be best. Then all of these sessions in my example screenshot could be grouped as one. If you deduped per-label with the current labelling it would show two accounts, one that represents 3 tenants, and another representing another single tenant. That would still be pretty confusing. IMO there's two options: Dedupe by email and hide all tenant related stuff. Signing out of that account removes all sessions associated with that email, regardless of tenant. Don't dedupe at all, instead include the tenant display name in the label as .\nsays this works!"}
{"_id":"doc-en-vscode-b69c0690c2519f95aab030c47bf64a8d481c1199112ca95b1da195b8ed08c5c0","title":"","text":"Testing ! I would have expected it to be centered.... though this could just be this codicon.\nIs this happening with all codicons or just this one?\nI can see it for codicons. Fixing it"}
{"_id":"doc-en-vscode-1e4c917e6b1cb72aa6a698545d5c8c97a1c5a403ce683007a9828b93d63e0579","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.91.0 (Universal) OS Version: Darwin arm64 23.5.0 Steps to Reproduce: new profile test to new profile in Devcontainer sample container is loaded without extensions. Opening in DevContainer with the default profile works correctly. Works correctly with the previous version of vscode 1.90.2\nSame For Me\n+1 extensions from are not enabled in the container when using a custom profile. Although the container logs does report that the extension was successfully installed, and I can see it is installed in the container, in . It's just not enabled when using a custom vscode profile."}
{"_id":"doc-en-vscode-6562ac2a17bc3871d78c90db2b9762721408237a9356828d98df21afefe4ac19","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.83.0 OS Version: macOS Sonoma(14.0) Steps to Reproduce: the color theme to high contrast any theme to extension search (I have attached a picture below which you will see there) will see no border-right edge of the install button !\nI'll start trying to fix this weird border-right"}
{"_id":"doc-en-vscode-ba49afe23f37bda4e961ea8156c63947eac1cc552c122434397864872f95ba08","title":"","text":"Similar to but for libraries... Here are the docs for : <!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. -- (, ) (, Needs version bump)\nThe HTML/CSS/JSON services are now all using adds the code to configure the servers with location of the localization bundle"}
{"_id":"doc-en-vscode-fe2888a0e6dd707d62d28436e1f26b16cd4ce6c51750e419f4213d15fa8d5fb3","title":"","text":"Repro: Open DevTools: !\nThis started occurring recently due to It is now possible to get this warning from various parts of the product.\nYes. That's on purpose and the goal is to get rid of invalid URIs in our system. So, when your code causes such warnings, you need to change it ;-). I can consult. The background is that an URI must have a scheme but we never enforced that. The warning I have is the first step into that direction.\nThe warning is valid. We're passing a path to , which is wrong. Git remote addresses are not always uris, they can be paths; even relative paths.\nSo the consumers of should beforehand check if what they are passing is valid uri or not? Doesnt that fail the purpose of ? is used to parse json content. We can pass non json to it as well. It throws an error. Can we not expect something similar from ? If we are worried that throwing errors will break existing code, then can we have a stricter that we can use?\nYes, parse will throw an error but it never used to do that for missing scheme. I didn't think it's a good idea to now start throwing errors at first because not everyone encloses in a try catch block and also often people confuse and ."}
{"_id":"doc-en-vscode-4be85bf0c3b07ce93f03c7b385753f00306c44f562b7731e0f42ed814a8a97c3","title":"","text":"The new notifications pop open at the lower right corner of the editor. Some command palette actions pop these up with a button that I need to click on to proceed. In these cases, I have to move my mouse from the top center of the window down to the bottom right corner of the window. One such example is the \"Git: Delete branch\" command. I have to click on the branch I want to delete from the drop down selector, then click the notification button far away in the corner to confirm the delete. If I am deleting multiple branches at a time (because I'm finally cleaning up my local state ) I have to move the mouse back and forth a ton of times so now I use a terminal for this task instead. It would be nice to have an option that lets me decide where I want the notifications to appear. I would choose the top right corner to reduce mouse movement.\nI'll narrow this issue down to make the delete branch popup a modal one, which it should be anyways.\nI'm trying to repro, but don't get any such popup after selecting a branch to delete. The branch is instantly deleted. I'm on Mac OS X 10.11.16, vscode version 1.21.1"}
{"_id":"doc-en-vscode-772ed2e446a297096144d13529c3601b160c766c5558f712dff4fecfd98a88cb","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --I work on an extension that has a custom tree view. Some of the items don't have meaningful extra information to show in a tooltip but setting the field in the to or results in the label being used for the tooltip which looks odd as it provides no additional information. We would like it if there was some way to disable tooltips for a tree item. It might be hard to used for that now as that would be a breaking change but perhaps there is some other approach that would be possible. I would note that if the tooltip is required for accessibility in some way then I'm happy that it stays but I think that angle is covered by the aria-label attribute and the tooltip is something else entirely. I have searched and I can't see reference to this from an extension API point of view in the current issues. I am not well versed in the VS Code code base but I believe the issue comes from these lines:\nHere is an example screenshot of the issue: ! The supported features are programmatically detected and we don't have a reliable way of providing any additional useful information for the tooltip so being able to avoid it entirely would be desirable.\nI think that allowing extensions to have no tooltip is a good feature request. Form an API perspective, do you think setting to empty string is an ok way to accomplish this? I don't think we should use since that would be a big disruption to the current behavior (where the is used).\nYeah, that makes sense to me.\nTo verify, modify an extension that provides a tree view so that the of it's tree items is an empty string. Verify that no hover is shown on those tree items. You can use the sample for this:"}
{"_id":"doc-en-vscode-25c19be7a7d9e19284763931193ab19bd1f613805391f148bc0209d71d37e41e","title":"","text":"contribute command that uses enablement assign a keybinding, via :bug: enablement isn't honoured\nWhen users configure keybindings we allow them to change preconditions as well, e.g rename requires an editable document so I don't plan to have a different strategy for contributed commands\nI don't see the nice \"Command 'Hello World' is not enabled in the current context.\" info toast when I try the keybinding though. I only see that message when I try to run it from the command palette. The command correctly doesn't run when not enabled though.\nYeah, that's the design. Keybindings don't show messages"}
{"_id":"doc-en-vscode-aae5b5dd4311d1106d38249de23af37d169e3318c0310de71db3421dc46fb9e7","title":"","text":"Type: <bBug</bstep 1 - Run any command or project whose lines does'nt fit in single line step 2 - Your terminal can be full sized or small size does'nt matter step 3- Presss Alt + z to wrap line in terminal Step 4 - Resize your terminal window Example - If initially terminal window height is 100px from the bottom then word wrap using Alt + z and after that increase or discrease the size of terminal window. You will find that content is still showing in 100px of height VS Code version: Code 1.91.0 (, 2024-07-01T18:52:41.959Z) OS version: Linux x64 5.15.0-113-generic Modes: <!-- generated by issue reporter --\n!\nThanks for fixing the issue One more questions Is there any way to off wrap word in terminal parmanent. I have to use Alt + z each time when I run my project\nthere is not\nCan you make it happen like Eclipse Please\ni haven't seen it in eclipse but I'm guessing they're using the old console api on windows which isn't really compatible with the pty based system that we use.\nWorks, but I never see a horizontal scrollbar with word wrap disabled, should I expect to? I see it in the screenshot above."}
{"_id":"doc-en-vscode-fcffc2d6e5a2d8de0eefd13a90f7c2c8012740e15c2212703bcb79070f51df83","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --- VSCode Version: 1.47.2 (user setup) OS Version: Windows_NT x64 10.0. Steps to Reproduce: a file named in any directory. the file with VS Code. file is opened with the \"JSON\" language mode (instead of the expected \"JSON with Comments\" mode). Other file names, e.g. , are also opened in \"JSON\" language mode. In contrast, is opened in \"JSON with Comments\" mode even if the file name contains suffixes, e.g. . <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes\nThis is easily fixable, look at the relevant code: Submit a PR\nThank you. Is a local build/test warranted, or would a direct PR suffice? Asking because I don't have a local build environment right now (and setting up one would take some time)"}
{"_id":"doc-en-vscode-043929804b9f336c474350f156c651600a42c1f0be9697500d620d312db92f51","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --VS Code currently associates the extension with the Make language. is another common file extension for files included by s. It's associated with Make by , i.e. files are recognized by , and there are a few dozen public . VS Code should recognize this extension automatically.\ncould you point me to some documentation about please?\nHi I've updated the issue description with some additional information. Notably, , which I consider to be a fairly canonical source. I've not any of the other file extensions recognized by GitHub Linguist for Make, as I've not seen them in use. Anecdotally, my employer has been using the extension internally since at least 2001.\nHere are some additional examples which demonstrate inclusion of files by s:\nThat's good enough for me, thank you!\n/verified"}
{"_id":"doc-en-vscode-1aaf54c2e0d2e56d4ff1d1268c62bb4e31ac0cac31605366977f83c80f5fb9ca","title":"","text":"Type: <bBug</bipynb notebook bug In dev tools, see: VS Code version: Code - Insiders 1.70.0-insider (Universal) (, 2022-07-26T05:17:37.262Z) OS version: Darwin x64 21.5.0 Modes: <details<summarySystem Info</summary </details<!-- generated by issue reporter --\nLooks like the directory is excluded from publish: You likely want to have the notebook code go to a different directory, such as instead of the dir\nJust changed that inside esbuild, didn't realize that out wasn't included. I had been having an issue with needing to run while using OSS, could that be related? I'll make a PR to adjust and"}
{"_id":"doc-en-vscode-228de066cf1ca99d343f4ea20d147de0153b104004ba723ddbf9503ff25a0c09","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.70.0 (Universal) OS Version: Darwin x64 19.6.0 Steps to Reproduce: to delete a file with nested files When you right click and hit \"delete\", vscode does nothing and the file and its nested files remain. You must open the parent and delete each nested file and then you are able to delete the parent file.\nThanks for creating this issue! It looks like you may be using an old version of VS Code, the latest stable release is 1.70.1. Please try upgrading to the latest version and checking whether this issue remains. Happy Coding!\nAlso, I think it would be a good idea for an option to rename all nested files when the parent is renamed.\nbased on the setting description for file nesting, I expected a file I created in a folder to be nested under the that is in that same folder. Is that the wrong expectation here?\nThat is correct. I'm seeing that here <img width=\"457\" alt=\"image\" src=\"https://user-\""}
{"_id":"doc-en-vscode-f1a105499515cc2bd444c8bdfd613580e88589f4654d6ba5479772920aa16b75","title":"","text":"From : Error: TreeError [FileExplorer] Data tree node not found: ExplorerItem:\nAny idea here because there's no explorer stack involved this might be a tree issue\nYou can try this. Enable fileNesting and select a file with nesting, control+c, the error would appear. The chosen file should be in preview.\nthird line in the stack frame:\nThis also happens when you try to open a file with enter or space keys. ! Pressing delete instead of enter will display the same error. (related issue: ) Once the nested files are expanded, this error does not occur until the next window reload, perhaps because the node is to the tree.\nThanks for the repro"}
{"_id":"doc-en-vscode-56f0c166bab049bc4f9a72a7166073a99d6e5fd04fcddbef9664ba3ed3c92984","title":"","text":"On Windows: clone open workspace look at the workspace settings: all files are associated with schema . Note that this is a relative path to the workspace file open both files and check that code completion gives back property . You can also click on the JSON language selector ( next to JSON in the status bar) to see the schema that is associated :bug: No code code completion or schema association\n76 candidate PR:"}
{"_id":"doc-en-vscode-2b499af95876e452c2c65385cd2751a639b0286565f8feb0ca892468d1ac7e29","title":"","text":"! Testing\nI'll try to deal with this\nI'm sorry, I would like to ask a simple question. If I want to use this accessibility mode, do I need a software like NVDA to make it work? Or if I only have VS Code, can I still trigger this functionality?\nHey you can enable this with the command - a status bar indicator will appear\nI used to be able to trigger , but now when I trigger it, it will show up in the lower left corner, but it won't have any function, which is a bit strange, are and the same function, I didn't find\nYes that's what I had meant. Hm, maybe a recent regression? Last time I tried this, it worked for me. I'll investigate tomorrow\nAlso, re the question of do you need to use a screen reader, it depends on what you're trying to do/ test. With that mode on, certain audio cues will (should) be enabled by default for example.\nI have used this function once before, and it is true that the audio will play after it is turned on, but this time there is no response, I will check again to see if it is a problem with my settings\nyou can test this - and it's working for me with: screen reader mode status bar item appears in the terminal hear sound screen reader mode status bar item disappears no sound\nThanks! This work for me\nI started trying to solve the problem"}
{"_id":"doc-en-vscode-94ff3f958f41c70833a1c5cfd665b2bee74f62456e2105d8f440803b4f9ddd8f","title":"","text":"Description Currently, when editing a file, the command <kbdctrl</kbd+<kbdshift</kbd+<kbda</kbd(toggle block comment) generates the following comment placeholder . I wonder if it might be also wise to include this as an auto-closing pair so that when users attempt to write the closing is automatically like other languages (ex: Typescript).\nThis issue is caused by an extension, please file it with the repository (or contact) the extension has linked in its overview in VS Code or the for VS Code. See also our guidelines. Happy Coding!\nI'm not quite sure as to who or where the owner is for the extension. Looking at the I noticed multiple past updates to support additional features. Isn't the proposed feature request present in the issue similar to those past updates?\nOh, you are right. This part is us, not the extension. Sorry for the confusion...\nLooks like a good change to me. Feel free to make a pull request.\nNot able to reproduce ctrl+shift+a not having any effect.\nI assume you are editing a file, but sorry the keybinding I mentioned is for Linux, on macOS for example the keybinding is . The easiest way to test is to simply run the command from the menu 🡒 :)\nGot it, in windows it is Shift+Alt+A.\nI forgot to mention I won't be able to submit a PR myself due to my company policy. If anyone would like to take on the issue, please feel free to do so :)\nIn that case, I am more than happy to make a pull request for this. Thanks for the information. I will submit one soon.\nthank you!"}
{"_id":"doc-en-vscode-23ba006552f54fb9c746cc88a21ddffd774e64d41e9d63f1dd4b243bd9ab2785","title":"","text":"VSCode Version: 1.36.0 OS Version: Windows 10 / Mac OS Steps to Reproduce: any c++ code inside of a .cpp/h/hpp etc file. \"/\" at the begging of of a line that has already been written \"/\" gets automatically afterwards, but before the code, so nothing actually gets commented out Simple example, where \"|\" denotes the cursor position: \"/*\" shouldn't auto-add the end symbol of the comment block by default, should be setting controlled, and should be able to recognize when it's been entered at the start of a line. Does this issue occur when all extensions are disabled?: Yes Originally reported here: They forwarded me to you folks, so I'm assuming this was to the \"C/C++ Language Basics\" built-in extension without a setting to control it. Please for the love of god fix this. It's super annoying.\nFound a temporary workaround: to: (User Folder)AppDataLocalProgramsMicrosoft VS Coderesourcesappextensions \"cpp\" to (User Folder).vscodeextensions it something like \"cpp.language-basics-custom\" to avoid any later confusion. \"language-\" & comment out (or remove) this line: the trailing comma in the previous line (might not be necessary) VS Code. The custom version should take precedence over the built-in version. Comment blocks will no longer auto-close.\nAlso, of note: Neither javascript nor typescript contains that in their language- files, but uses the same style of block comments. I think it's safe to say it should just be removed from the cpp language-\nIt looks like C# and json use the as an auto closing pair, but java, javascript, php, and typescript use the . For C languages we whould use since it is more consistent.\nFix will be available in next insiders if you want it ASAP. It installs side by side with stable, so you can have both:\nthe C/C++ extension already registers a custom , including for . Please don't add autoClosingPairs for comments since there is no way to disable them in vscode. There are only settings for and . I seem to recall these all being controlled by a single setting at some point, but now there are only 2 out of 3 available."}
{"_id":"doc-en-vscode-79b2dd71a07865cd5b1bc88f27d6fe9bcb3854259272a059b8975ce987c6b107","title":"","text":"Visual Studio Code lacks support for automatic comment-block ending in C and C++. If you were to start a comment-block with , an automatic will not be as it would be in the traditional Visual Studio. When I type , it should automatically be turned into and the text cursor should be between the two ``. It isn't that big of a deal but it would be really nice to have this feature in Visual Studio Code.\nUgh. I currently have the exact OPPOSITE problem: I want to DISABLE this feature. This is happening for me and I can't figure out how to turn it OFF. There are a large number of developers who just don't ever want their text editor to ever, under any circumstances, insert a character that did not correspond exactly 1:1 to a specific keystroke. I push \"A\", I get \"A\". And so on. \"Surprises\" break my flow worse than any other kind of interruption. Obviously, there are a lot of developers in the other camp. But the upshot is that the fact that both camps exist, and both camps are large, means that when features like this exist, it needs to be possible to disable them. I have encouraged many projects for years to consider adding a \"dumb as a brick\" switch that will just globally disable all of these. In the absence of a master switch, we at least need configuration switches that can shut these things off-- preferably arranged hierarchically, so that we can shut off entire branches. But the absolute lowest rung on the ladder is that individual switches must exist to disable these things.\nLet's track in"}
{"_id":"doc-en-vscode-6006ac8e2ae7b29e96d83fb323faa116ddfe72c88ecbeb7707ab0be2c05ddf8e","title":"","text":"Issue Type: <bBug</bmultiple brackets \"\" Expected: All cursors at brackets become selections starting where the original cursor was and and ending at its matching bracket. Actual: Only the first cursor becomes a selection, the rest dissapear. VS Code version: Code 1.21.0 (, 2018-03-07T10:54:22.798Z) OS version: Darwin x64 17.3.0 </detailsReproduces without extensions <!-- generated by issue reporter --\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.531 --- <!-- score: 0.527 --- <!-- score: 0.498 --- <!-- score: 0.498 --<!-- potentialduplicatescomment --"}
{"_id":"doc-en-vscode-8d896e69fc6227c8fecbb9e1beb5c80565e6a0456d65060569359c93cb29b6f4","title":"","text":"Issue Type: <bBug</bThe preview in the minimap slider is wrong when: The minimap is NOT overflown (doesn't need to scroll internally when the editor is scrolled) is (default value) When is set to , the preview is shown as expected. Reproducible without extensions: Yes ! The slider preview ends at the beginning of the block comment, while it should contain a portion of it. VS Code version: Code 1.33.1 (, 2019-04-11T08:27:14.102Z) OS version: Windows_NT x64 10.0. <details<summaryExtensions (15)</summaryExtensionVersion ------ gitstash2.1.0 bracket-pair-colorizer-20.0.28 vscode-eslint1.8.2 todo-tree0.0.129 svg0.1.4 tag-rename0.2.1 vscode-simple-icons1.11.4 dotenv1.0.1 vscode-docker0.6.1 vscode-javascript-booster0.10.2 code-settings-sync3.2.9 markdown-preview-enhanced0.3.13 code-spell-checker1.7.0 language-pde1.3.0 markdown-all-in-one2.2.0 (1 theme extensions excluded) </details<!-- generated by issue reporter --\nA new observation: It's not about the state of the window (maximized/restored), it's about weather the minimap is overflown or not (needs to scroll internally when the editor is scrolled). I edited the issue accordingly.\nAnother observation: has no effect on the issue either. The problem is that the minimap height is not calculated correctly when is , the extra lines that this setting adds are not accordingly to the minimap height, only when it's NOT overflown (i.e. when )."}
{"_id":"doc-en-vscode-f595391ad8f4e81dafb8add7ef9ff268f09766dc3d6014c707b1a080123b8c17","title":"","text":"In Markdown preview, inline formats in single backticks (`) seems small sized in heading. Headings of different sizes have this problem in common. !\nfyi\nYes, this collides with our font-size rule for the code block which is not ideal. I am happy to accept PRs for making markdown CSS better. The CSS is defined in\nStill happens with our new markdown extension: !"}
{"_id":"doc-en-vscode-ab3c8aba33170badc330ac161249ab76b31e17829d207cf58c2d005e686502f9","title":"","text":"Testing <img width=\"348\" alt=\"image\" src=\"https://user-\"\nExtension fails to launch child_process during initialization Will check if the issue is isolated to this extension with a minimal example using module\nWas able to isolate the issue, firstly as pointed above the extension makes an attempt to launch a child process via the helper which basically emulates options and passes them eventually into Since variant relies on when node executable path is not provided, this will default to use the Electron executable that is used to create the extension host. Now there are two problems with this usage, (i) execa module will not add the env variable, so the exectuable cannot be launched as intended. (ii) We also disable calling Electron executable as unless flag is provided in the call site. For the (ii) we automatically add the flag at both and callsites internally in the runtime, so this should not be a concern for the extension authors. As for (i), if the extension author used the vanilla then the env variable would be set internally by the runtime. But since the call is done via a 3rd party module, I think it is upto the extension authors to ensure the env variable is provided if they rely on the Electron executable. I feel like there is no action required from runtime/VSCode core for this issue, but we should loop in the extension author to fix this on their end. Thoughts ?\nSo what exactly is different in the utility process vs. child process today? I did not fully understand.\nWhen extension host is created via , internally the runtime would launch the process with env variable set and we never remove this variable explicitly in the extension host after launch (i-e) , this would indirectly help above scenarios since the env gets inherited by extensions. This is not the case with utility process since there is no runtime induced env variable, the extensions have to explicitly add them if they rely on the electron executable. I would say the previous behavior was a hidden bug and is visible now with the utility process api.\nOh yeah understood! I think I somewhat agree that we need to just document this to extension authors.\nbtw, since this is a public issue, I've shared it with the dendron extension author directly to track\nhey, dendron author here. thanks for looping us in on this. so as I understand, we have two options: either use vanilla or set when using the module."}
{"_id":"doc-en-vscode-ee2781e6e193c3eee121d6d365ba8b1c78a1b2581f54d47ad86f4e10abf9fda2","title":"","text":"is this an accurate understanding?\nyup that is correct. To clarify, only when ends up using on the desktop extension host.\nGot it. What's the timeline for us patching this? Is this already live in insider?\nThe change is not default in insiders yet, it is currently behind an environment variable , can give more context on the timeline.\nafter discussing with the team, we have decided to push a workaround in core to avoid breaking the extension community with the new process type for extension host. You don't need to make the adjustments outlined in"}
{"_id":"doc-en-vscode-23d1400f951469b1a41a8585c241b55214a5df211ebdd8c686aa2f01f0722c39","title":"","text":"see gif have the debug hover above the cursor expand some objects the hover grows over the current line/cursor !\nYeah, this is a consequence of the change to not make it jump. Since in this scenario it does not have enough space above to grow, it can only grow over the line and below. Alternative would be to show it down, which would cause a jump. Or to limit the size, that it does not grow over the current line. That I think makes sense. However will not tackle during endgame.\nHow about not growing and use the scrollbar sooner? Just measure the max size above the cursor and don't go over that\nI think this makes sense. My current max height is 70% of the editor height. Idealy as you suggest the max height should be the size above the cursor if I am placed above, or max size below the cursor if I am placed below. However I am not aware of editor API that gives me height above or below the cursor. Are you already using something like this for the suggest?\nI am using and some math:"}
{"_id":"doc-en-vscode-55c253e4f3118a1d6b0da371d3a99e14bf75b732818c7a7cb31a50321c92e033","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.85.2 OS Version: Windows 11 Enterprise (Windows_NT x64 10.0..3007) Description During an accessibility test, we found the contrast ratio between the highlight color of the focused button in \"More Actions...\" in notification popup and its background is not high enough under \"Default Dark Modern\" theme. Highlight Color: Background Color: Contrast Ratio: 1.346:1 Required Contrast Ratio: 3:1 !\nI think this refers to the background color of the gear icon in the toolbar of a notification relative to the background color of the notification.\nThe bug reported on my side particularly specified the \"Manage Extension\" button highlight color and the background color of that actions popup under settings icon in the notification.\nLooks like it's about the blue on black bg, in that case this applies to all the web menus in VS Code.\nyes, it's about the blue highlight and black background.\nA11yMAS; #A11yTCS; #A11ySev2; #DesktopApp; #Win32; #Visual Studio Code Client; #BM-VisualStudioCodeClient-Win32-Aug2023; #WCAG1.4.11; #Non-text Contrast;\nHi, is there any update on the plan of fixing this accessibility bug?\nputting this on the March plan.\nHi, may I check if the fix is included in v1.88.0?\nHi, I have checked v1.88.0 but it seems the active color is still not the one you changed to in your PR. Did I miss something?"}
{"_id":"doc-en-vscode-bc60442640b3fbb7f087d6464aa30d379356503b05f26f717b8671384e1ae446","title":"","text":"Verified with the Dark Modern theme, but do we also have to fix this with the Dark+ or Dark (Visual Studio) themes?\nThanks, I think it's fine we only fix it in default dark theme for now. The test team doesn't require that in other themes so far."}
{"_id":"doc-en-vscode-9c60595bb46183f8518375d247b68ea54b0582d654be1d6ceae8bc85a62c809b","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.73.1 OS Version: macOS Ventura 13.0.1 Steps to Reproduce: VoiceOver a command that uses a Quick Pick List where the items have a as well as a title or description (demoed with Azure Extension \"Create Resource\") list a command that uses a Quick Pick List where the items have no detail, e.g. \"Select Display Language\" or even command palette reader will read the title of the new item with the detail of the item in the previously displayed quick pick list https://user- Same bug with the command palette reading details for items: https://user-\nThis issue is caused by the item still containing the values from the previous list.\nVerification steps: the issue steps are good. If you want a non-Azure extension command that shows a quick pick with details, use"}
{"_id":"doc-en-vscode-bba691d1490003211d348bea77737eea0c6a0c3242f5fc0aba7a4d1212b82f66","title":"","text":"Issue Type: <bBug</bCompared to input boxes in other viewlets, the search box in the extensions viewlets looks slightly different with a little padding of a different color inside the box. VS Code version: Code - Insiders 1.27.0-insider (, 2018-08-15T05:19:24.200Z) OS version: WindowsNT x64 10.0. <details<summarySystem Info</summary </details<!-- generated by issue reporter --\nThis appears to be because the sets the background color to a transparent color (), if that is changed to , the look is the same but without the border. This seems like incorrect behavior of the renderer, as the type of color shouldn't affect what gets colored.\nActually, looks like the transparency is showing up differently on the different backgrounds. If you increase the alpha, it starts to look like this: ! I'll look in to making the transparency affect both equally"}
{"_id":"doc-en-vscode-2f2bbfc1647bd30c397a57cbf08e4b5562d5291f276f1b5ac7634e0fdc79edab","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.81.1 OS Version: win10 Steps to Reproduce: new version vscode after 1.78.2 remote connection before vscode 1.78.2 in win 10 right click win10 taskbar show remote connect like ! but recently I update my vscode to 1.81.1, the information for remote node disapear. ! This bug is quite inconvenience when I have many projects with the same name in different remote nodes.\nSame Issue Here\nDo you own this part of the label service Aware of any changes?"}
{"_id":"doc-en-vscode-d9e07d809278ecea60e94946f4892461db3400d7b85b4aa8da61efe5fcdd5162","title":"","text":"Not much padding between it and the word: ! !\nPassing to as this is a file icon but in the icon we actually have it designed at 14px scale (and offset to avoid pixelation) and i know the file icons use different scaling styles <img width=\"568\" alt=\"CleanShot 2022-06-17 at 12 33 56 src=\"https://user-\"\nThe trust editor uses a product icon (codicon) as editor icon here: I think this needs to be adjusted there."}
{"_id":"doc-en-vscode-5bed4cde3a3094f1d1343561d4690b64a7d2954b345e75cb8033e59d71913427","title":"","text":"I've had a couple of people report debug adapter crashes recently. One of them I'm unable to reproduce which means I need to get more information from the user. Currently, when a debug adapter crashes, the error information that appears in the console is very vague: ! The actual exception that was thrown is not shown. Since I can't repro this, the only ways I can think of to debug are: try/catch around all debugger methods that handles errors and passes them back nicely (and remember to do this for all future methods) the user to run my project from source and debug the debug adapter to catch the error (whilst not the hardest thing in the world, it's a bit involved for an end user) Is it possible to get better information from the error back into the console without having a debugger attached to debug adapter?\nI don't know if the text \"ERR\" in the console was actually supposed to be my error; but it isn't. My error was \"AAAAAARGH\".\nthis request makes sense. Not sure though if I am getting your error when your process crashes. Can you be so kind and set a breakpoint make your server crash and check if your message is coming through as a or as something else. Thanks\nDo you mean the text? I also don't get that now when trying to repro, only the \"terminated unexpectedly\" message. If you still want me to try that breakpoint, can you explain how to add a breakpoint there. I did try breaking in my adapter and navigating the callstack (that file wasn't in there anywhere) and also checking the filenames of the files that were (they showed ).\nI mean let's try to see if your error \"AAAAARGH\" surfaces on our side. You should be able to just run regular vscode, F1 open developer tools, wait some time and put a breakpoint by openning a . Also try to put a breakpoint and check if the error stream spits up something.\nThe breakpoint is not hit, which is perfectly normal since the exception occurs before the socket connection is created. You should be able to reproduce by adding require('unknown_module') to the top of any debugger extension .js file\nOk, I got breakpoints in both and . First wasn't called and the second was called but is and is . So seems like the message isn't there."}
{"_id":"doc-en-vscode-451686c569606774358e81742898f6b408de7df501a49caca61f0767db2a1d59","title":"","text":"For testing, all I've done is put inside my debug adapter. Though now I think about it; the error isn't occurring in code that's been executed directly by Code - it's during the response from my debugger. In my debug adapter I spawn a process and I add a handler for . When something comes over and my callback runs, that's where the crash is occurring. I don't know if this changes things (for example, I don't know if this is on another thread?).\nThat one (in ) didn't seem to be hit either.\nActually, I think I can repro this really easily by just putting in the constructor of the ? Also gets .\nThis issue has the needs-more-info tag though I think it's easily reproducible with my last comment. Let me know if that's not the case!\nI have removed tha tag and plan to look into this issue this week. Thanks\nInvestigated and the issue here is that I am not getting any exception data from the spawned process. I simply get the code while all the other error chanels are blank. I am not sure what would be the proper solution here. Suggestions are very welcome. Also are all your debug adapters launched using node? If yes then it might be that our  launch is somehow eating up the error if it happens on startup. If this also happens for adapters which are not realying on node to be launched then is not to blame.\nplease try to disable the launch for node () and see whether this makes a difference. DAs should work independent from . But make sure that the node version used is roughly the same as Electron's version (i.e. 7.9.0).\nYes, using stdfork is the cause here. If I do not use it, I nicely get the error popping up in the extensions output channel. My weekend is starting if someone is feeling smart code review and let me know if you find something fishy about stderr not being wired through.\nThen you might want to involve/assign\nI am going on a vacation tomorrow evening, and this is like this since forever so pushing to April to investigate with Alex what is going on behind the curatans here."}
{"_id":"doc-en-vscode-9292d147bd48d4a74297ba32f1f3b6dc6086e4772de5e4da8a558a7b28794ecc","title":"","text":"Because both colors are blue they become hard to read and fail the color contrast ratio. Would expect for the highlighted text in these selected states to become white. Outline Breadcrumb cc\ncan you also attach the look for quick open after this change? and would this change be for the light theme only or all themes?\nI can't find an example that uses selection for quick open, do you know of one? And this change would be the new default for the breadcrumb/outline view, so it would affect all themes that use the \"light theme\" defaults in those two views. It's currently broken and was hard to see in our previous color as well: !\nyeah, none of the pickers use selection color, I was more referring to how quick open would look like if you change highlight color to white. My understanding is that this would have an impact pretty much everywhere where highlighting is being used (independent from the tree or not)\nTo clarify: this would affect all light themes, even the ones in the market place? or only our light theme\nsorry if I wasn't clear, my intention is to only change the highlight color when it's on a selection row. So I'm not intending to change the default highlight color. In the example below, the highlight color remains the same except when it's selected: ! I still need to investigate how to do this, but my suspicion is that we already make the text lighter on a selected row, so I'd explore doing something like while it's selected. yes, this would affect all light themes.\nbut there is only , we do not distinguish the highlight color based on the type of selection currently. that sounds like it could break many themes because the change from our current blue to white is quite a different color. if this was an issue for any of the themes in the marketplace, they would have fixed this already I suppose.\nme do some investigation to figure this one out.\nI think the point is the we don't have (and want) a highlighted-and-selected-foregournd-color but that the selection color always wins.\nThe highlight color on the dark theme also fails the color contrast ratio (3.2:1): !"}
{"_id":"doc-en-vscode-dab1c7d458a208b99ee55e135ae4e49798f69a5b49384bd61af504269cbfb97d","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --Version: 1.58.0-insider (user setup) Commit: Date: 2021-07-01T09:30:57.217Z Electron: 12.0.13 Chrome: 89.0.4389.128 14.16.0 V8: 8.9.255.25-electron.0 OS: Windows_NT x64 10.0. Steps to Reproduce: error I think only occurs using Workspaces. A single project, the terminals seem to work fine. However, with multiple projects, the environment gets confused. Terminal (ctrl+shift+) Terminal I get this error message: 'The termial process failed to launch: Starting directory (cwd) \"homeusernamemyproject\"'. The first terminal works, but it's the second or subsequent terminal that creates the error. This is using the VSCode for Windows using WSL Remote. This error has now existed for at least a week in the Insiders edition. I am surprised this has not already been brought up. I think the error is indicating that its confusing a windows path for a linux path. Because this is using WSL-Remote, the path should be using forward slashes not backslashes for windows. This happens when \"terminal.integrated.splitCwd\" is inherited or initial. The workaround is to set this to workspaceRoot, then it works. However, it's annoying not to have inherited or initial working. Also, nowhere in my settings is \"cwd\" assigned anywhere."}
{"_id":"doc-en-vscode-5fc031dcc89d35c003e8c28a3970d79df1886fdbe8820cb32c1a19b73da097d6","title":"","text":"Also there is , but I suspect the issue on MacOS is different. Jae\nThe bottom line: Only while using Workspaces, the first or default terminal is properly run in WSL2. However, any subsequent new terminals are being recognized as a Windows environment, not Linux. Somehow the combination of Workspace+WSL is messing up the terminal environment.\nI just discovered there is a command called \"Create new Integrated terminal (In Active Workspace)\"; where as the default \"New Terminal\" calls \"Create new Integrated terminal\". This is another possible workaround. However, you have to make sure the active file is where you want the new terminal to be created. You lose the ability to select which folder to create the new terminal in, unfortunately.\nCan repro: ! multi-root workspace in WSL +shift+ +shift+5\nAh this is happening because we're getting a instead of a here:\nI'm also getting this issue when using inherited for workspaces and remote ssh from a windows to a linux host. If I reload the window I am able to split the terminal for terminals that were created before the reload. Version: 1.58.2 (system setup) Commit: Date: 2021-07-14T22:10:15.214Z Electron: 12.0.13 Chrome: 89.0.4389.128 14.16.0 V8: 8.9.255.25-electron.0 OS: Windows_NT x64 10.0.\nSame here.\nAs a workaround for this issue in version 1.58, I was using this setting: But now in version 1.59.0, the issue happens with any values .\nAwesome! Just got the fix in insiders! Thank you!\nVerified fixed in"}
{"_id":"doc-en-vscode-7d5f9129b780881ea615b9b269b3ababd8cf3761ad082c57c7eff4710bc614f1","title":"","text":"We want to discourage extension authors from using activation. We should consider adding a diagnostic warning similar to the one we have for implicit events.\nIs it not enough that provides the same prompt in terminal when packaging/publishing an extension?\nat that stage in the development lifecycle, authors are more likely to just ignore the warning from vsce. Additionally, they might be publishing in CI and might not see it there either.\nverification: in an extension, put in activation events see this: !"}
{"_id":"doc-en-vscode-2edc22b4d9ceacf73568353d96b82b73c6e65d38f13f6f35690234d188aa2ed3","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --Version: 1.75.0 (Universal) Commit: Date: 2023-02-01T15:24:42.903Z Electron: 19.1.9 Chromium: 102.0.5005.194 16.14.2 V8: 10.2.154.23-electron.0 OS: Darwin arm64 22.3.0 Sandboxed: No Steps to Reproduce: SCM tab at the left side at the border of the Commit & Push combo box The inconsistency is more visible in non-default themes like \"Bearded Theme Monokai Reversed\" <img width=\"717\" alt=\"vscodescmcp1\" src=\"https://user-\"but also slightly in the \"Default Dark+ Experimental\" for example <img width=\"656\" alt=\"vscodescmcp2\" src=\"https://user-\""}
{"_id":"doc-en-vscode-961eb61283f1e1b1ed01816f319a8a297ead584ac148de2a6d3bd875b00ffb0d","title":"","text":"This is to track all the issues surrounding the terminal switcher for the August iteration. [x] move keyboard focus [x] add create new terminal option [x] add kill terminal actionbar item to entries and [x] fix tab focus for action items [x] fix focus on arrow keys [x] fix spacing between actionbar icons"}
{"_id":"doc-en-vscode-5737fae945476611e403912a27bc9156e4e145f02b5ba371c95c969cf9b5b4d1","title":"","text":"Similar to GitHub style or Keyboard Shortcuts GUI: <kbdCtrl</kbd"}
{"_id":"doc-en-vscode-4b4676ffc5541813b4369586c025a8d0f5c06a9b3a6a6c77d2794b8f86d89bfd","title":"","text":"Type: <bBug</bCC (alt+F9) does not stop at the last comment. It cycles back to the first comment once it reaches the end. This is especially confusing for screen reader users because pressing alt+F9 multiple times does not always guarantee the last comment position. Please make stop at the last comment position. Of course, (alt+shift+F9) has the same issue. ! VS Code version: Code - Insiders 1.93.0-insider (, 2024-08-20T08:04:15.567Z) OS version: WindowsNT x64 10.0. Modes: <details<summarySystem Info</summary </details<details<summaryExtensions (128)</summaryExtensionVersion ------ vscode-openapi4.28.1 zotenote1.0.1 android-dev-ext1.4.0 aiprm-lang0.0.2 Bookmarks13.5.0 openscad1.2.2 spellright3.0.136 zoterolatex0.4.1 mermaid-markdown-syntax-highlighting1.6.0 external-pdf1.2.0 doxdocgen1.4.0 vscode-markdownlint0.53.0 vscode-eslint3.0.10 vscode-quick-select0.2.9 vscode-deno3.38.2 gitlens14.6.1 EditorConfig0.16.4 prettier-vscode10.1.0 figma-vscode-extension0.3.5 vscode-firefox-debug2.9.10 shell-format7.2.5 vscode-google-translate1.4.13 codespaces1.17.2 copilot1.223.1058 copilot-chat0.20. remotehub0.62.0 vscode-github-actions0.26.2 vscode-pull-request-github0.95."}
{"_id":"doc-en-vscode-f5ce6f21e2cab1e103108699747725e2a33337042b65e5ef2853da4bdfd125b4","title":"","text":"cloudcode2.16.4 overleaf-workshop0.13.2 cslpreview0.2.2 path-autocomplete1.25.0 latex-workshop10.1.0 lilypond-syntax0.1.1 scheme0.2.0 better-cpp-syntax1.17.2 commitlint2.6.0 language-julia1.109.1 google-search0.0.1 vscode-lua-format1.3.8 lilypond-formatter0.2.3 lilypond-pdf-preview0.2.8 lilypond-snippets0.1.1 vslilypond1.7.3 language-matlab1.2.5 zotero0.1.11 git-graph1.30.0 azure-dev0.8.3 vscode-azureappservice0.25.2 vscode-azurecontainerapps0.6.1 vscode-azurefunctions1.15.2 vscode-azureresourcegroups0.8.3 vscode-azurestaticwebapps0.12.2 vscode-azurestorage0.16.0 vscode-azurevirtualmachines0.6.5 vscode-cosmosdb0.22.0 vscode-docker1.29.2 vscode-edge-devtools2.1.5 black-formatter2024.3. debugpy2024.11. flake82023.13. isort2023.13. python2024.13. vscode-pylance2024.8.101 jupyter2024.8. jupyter-keymap1.1.2 jupyter-renderers1.0.19 vscode-jupyter-cell-tags0.1.8 vscode-jupyter-slideshow0.1.5 remote-containers0.381.0 remote-ssh0.114. remote-ssh-edit0.86.0 remote-wsl0.81.8 vscode-remote-extensionpack0.25.0 azure-account0.12.0 azure-repos0.38.0 cmake-tools1.18.44 cpptools1.21.6 cpptools-extension-pack1.3.0 js-debug-nightly2024.8.1517 powershell2024.3.2 remote-explorer0.5. remote-repositories0.40.0 remote-server1.6."}
{"_id":"doc-en-vscode-910fedcaff27efa0aa365f6c8e62383bb6c8a55d14995fff48d37aceafe35a04","title":"","text":"vscode-github-issue-notebooks0.0.130 vscode-node-azure-pack1.2.0 vscode-selfhost-test-provider0.3.25 vscode-serial-monitor0.12.0 vscode-speech0.10.0 vscode-speech-language-pack-en-ca0.4.0 vscode-speech-language-pack-en-gb0.4.0 vscode-speech-language-pack-ko-kr0.4.0 vsliveshare1.0.5936 windows-ai-studio0.4.0 autodocstring0.6.1 pandocciter0.10.4 publisher1.1.5 shiny1.0.0 shinyuieditor0.5.0 quarto1.114.0 r-debugger0.5.4 java1.33.0 vscode-xml0.27.1 vscode-yaml1.14.0 r2.8.4 multi-command1.6.0 AudioQ0.0.2 vscode-deepl1.1.1 abc-music0.4.0 lua3.10.5 latex-utilities0.4.14 cmake0.0.17 vscode-terminal-here0.2.4 windows-terminal0.7.0 errorlens3.16.0 intellicode-api-usage-examples0.2.8 vscodeintellicode1.2.30 vscode-conventional-commits1.25.0 vscode-arduino0.7.0 vscode-gradle3.16.4 vscode-java-debug0.58.0 vscode-java-dependency0.24.0 vscode-java-pack0.29.0 vscode-java-test0.42.0 vscode-maven0.44.0 zotero-cite0.8.6 markdown-all-in-one3.6.2 grammarly0.25.0 (1 theme extensions excluded) </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\ni'm wondering if we were to announce \"comment 1 of 3, 2 of 3, 3 of 3\" etc., that would be a sufficient workaround?\n-- Adding the index number would be really helpful! However, that is a separate feature enhancement and I would prefer not to wrap once it reaches the end. At least, a simple bell sound needs to be played when it wraps."}
{"_id":"doc-en-vscode-3a92ad595f536f78a5bb3c8459c7ff221be34fc83ee25a6564ef714a46724ea4","title":"","text":"Toggling the command center works, but the title bar height is stuck until the window is resized. I also noticed some issues where I had to click twice to open the command center once enabled. The command center input then disappears while the command center quick pick opens.\nMaybe an electron update regression"}
{"_id":"doc-en-vscode-457d48af04968a9aa8ace4eb55a6a5bcb68304b6ff6cf53e8a2e29b486e48011","title":"","text":"Steps: begin with multi tabs showing open an editor hide tab bar actions move to the top close editor open same editor actions are not visible anymore, there is just a \"...\" <img width=\"1348\" alt=\"image\" src=\"\"\nWhen creating the editor actions we rely on the event to tell us when actions change. However, this event is not set when there is no active pane. In that case we should have a different event trigger the onDidChange event such as onDidChangeActiveEditor for example"}
{"_id":"doc-en-vscode-2a5b635c2295465c7e3cd7efad06a042c0a413e890c2939578d2fb4445b8ccc9","title":"","text":"<!-- Do Not Delete This! bugreporttemplate --<!-- Please read our Rules of Conduct: --<!-- Read our guide about submitting issues: --<!-- Search existing issues to avoid creating duplicates. --<!-- Test using the latest Insiders build to see if your issue has already been fixed: --<!-- Instead of creating your report here, use 'Report Issue' from the 'Help' menu in VS Code to pre-fill useful information. --<!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes <!-- If you answered No above, use 'Help: Start Extension Bisect' from Command Palette to try to identify the cause. --<!-- Issues caused by an extension need to be reported directly to the extension publisher. The 'Help Report Issue' dialog can assist with this. --- VS Code Version: 1.69.0-insider OS Version: Windows 10 Steps to Reproduce: Two split terminals + one standalone terminal, select them all then join. !\nthis works now, must've been fixed\nI tried it in the latest insider build, still fails. Instead of selecting the terminal tabs one by one, can you try to select all, then join them? If you can reproduce, please reopen it."}
{"_id":"doc-en-vscode-fc56ab8d92248d18667a559a7c0c1ef645fc70ff575a434de16b2f2d543791f9","title":"","text":"(originally reported in ) Extensions that are enabled both in the show the text . This text can be confusing. asked me to open an issue Te proposal is thus to remove the text. To be honest though, I am unsure what the text even means. Look at this image: ! That extension is clearly installed locally. But I do not find it installed in on the remote. It is not there in the remote part of the Extensions sidebar. VS Code version: Code 1.88.1 (, 2024-04-10T17:41:02.734Z) OS version: WindowsNT x64 10.0."}
{"_id":"doc-en-vscode-0afb01540dd254562832bf36daed528f435b5ee97c5cc152e25b35bbf23a5edc","title":"","text":"Modes: Remote OS version: Linux x64 5.4.0-177-generic <details<summarySystem Info</summary </details<details<summaryExtensions (40)</summaryExtensionVersion ------ yang0.1.3 jupyter-keymap1.1.2 remote-containers0.354.0 remote-ssh0.110.1 remote-ssh-edit0.86.0 remote-wsl0.88.0 vscode-remote-extensionpack0.25.0 remote-explorer0.4.3 remote-server1.5.1 vscode-openapi4.25.3 lit-html1.11.1 markdown-preview-github-styles2.0.4 vscode-fish1.0.38 ruff2024.20.0 vale-vscode0.19.0 vscode-markdownlint0.55.0 vscode-eslint2.4.4 gitlens14.9.1 prettier-vscode10.4.0 code-runner0.12.2 go0.41.4 path-autocomplete1.25.0 restructuredtext190.3.0 vscode-azureresourcegroups0.8.6 vscode-cosmosdb0.21.0 vscode-docker1.29.0 vscode-kubernetes-tools1.3.16 debugpy2024.4.0 python2024.4.1 vscode-pylance2024.4.1 azure-account0.11.6 vscode-thunder-client2.21.15 vscode-yaml1.14.0 vscode-direnv0.0.2 code-spell-checker3.0.1 vscode-stylelint1.4.0 even-better-toml0.19.2 shellcheck0.37.1 simple-rst1.5.4 vscode-proto30.5.5 </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --"}
{"_id":"doc-en-vscode-bd739be92ff3c056242b416f8ffea6aadf57526e8165418db020f1ebc6248e38","title":"","text":"Type: <bBug</bIt's a UI fault in the merge editor. I have written the steps to reproduce it and also attached a screen recording for better understanding. Note: What I mean by , and in this issue ! First things first i) Need a code base which have conflicts You can clone this repository with both & branches and merge branch into to produce merge conflict. ii) Open the conflicted file in the Merge Editor And then: (Please watch the video for better understanding) in the Result tab, place the conflicted code block's widget aligned with 'Complete Merge' button in bottom-right corner in a way that the conflict actions bar is intersecting with the button (as shown in the video below). code from any stream, incoming or current doesn't matter. try to hover over same straight horizontal line as conflict actions bar towards the 'Complete Merge' button, You will notice that the cursor does not become a pointer, and you cannot even click the button. Now, If you reload the window, the issue will be fixed. However, as soon as you select any action from the actions bar, such as 'remove incoming/current', the issue will occur again. From my understanding, what is happening here is that after we select any action from the actions bar, the actions bar is overlapping the 'Complete Merge' button in the z-index's stacking context. I have tried to fix this on my local setup, and I believe I have solved the issue. If the maintainers want, I can raise a PR! VS Code version: Code 1.78.2 (, 2023-05-10T14:39:26.248Z) OS version: WindowsNT x64 10.0."}
{"_id":"doc-en-vscode-c588a7dee2ee8d4b905b61b6dd5e56f2b3892f03483fa1cd57c16a244e0188bf","title":"","text":"Modes: Sandboxed: Yes <details<summarySystem Info</summary </details<details<summaryExtensions (26)</summaryExtensionVersion ------ vscode-tailwindcss0.9.11 vscode-eslint2.4.0 docs-markdown1.0.1 docs-preview1.0.1 es7-react-js-snippets4.4.3 auto-rename-tag0.1.10 live-sass6.0.5 githd2.3.3 vscode-text-pastry1.3.1 vscode-peacock4.2.2 vscode-selfhost-test-provider0.3.15 material-icon-theme4.28.0 vscode-react-refactor1.1.3 prisma4.15.0 vscode-thunder-client2.7.5 reactree1.0.8 LiveServer5.7.9 mongo-snippets-for-node-js1.3.12 svg-preview2.8.3 vscode-fileutils3.10.1 vscode-wakatime24.0.11 console-ninja0.0.141 vscode-import-cost3.3.0 JavaScriptSnippets1.8.0 pretty-ts-errors0.4.1 html-css-class-completion1.20.0 </details<details<summaryA/B Experiments</summary</details<!-- generated by issue reporter --\nThanks for this detailed bug report! That would be awesome!\nHey I have raised the PR"}
{"_id":"doc-en-vscode-26e2a534d91150591a689a12276ebb849d91883d82c1a6de10e1dc6382f53669","title":"","text":"Checking our drop (e.g. ) noticed that some layered code has sneaked in: !"}
{"_id":"doc-en-vscode-12ae12681afad73968b5c0ae35d458a4ec7844ca7c694aae6871150df5efa802","title":"","text":"Was verifying an issue Disabled settings sync Reenabled it but on the Stable service Changed profiles a few times Disabled Settings sync Got an error poupup, here is the stack from the window log\nDo you have settings sync views enabled?\nTurn on Settings Sync Open Synced Data views (F1 Show Synced Data) Turn off Settings Sync Above error is shown It is a regression and after looking at the root cause, it seems this can also impact other settings sync scenarios. Hence tagging this as candidate.\nCan you please verify this on latest insiders?"}
{"_id":"doc-en-vscode-208ffda02178461a598c70cef0c16994b13081b16a95262e7613ac95b0c291af","title":"","text":"Sometimes inlay hints are off for a short amount of time: ! It happens randomly and fixes itself quickly: ! I cannot reproduce it in isolation or in the monaco editor playground, but I see it very often when working in the VS Code repo. I think this is what happens: I type somewhere to trigger a delayed update I wait until the delay completes and the provider is asked I add new lines The provider comes back, but does not consider the edits in between The inlay hints are off I'm not sure if this is an issue of the inlay hints feature or typescripts implementation.\n/duplicate of\nThanks for creating this issue! We figured it's covering the same as another one we already have. Thus, we closed this one as a duplicate. You can search for . See also our . Happy Coding!\nI still do think is buggy behavior and not a problem of delays. The initial screenshot should not be observed, regardless of delays. I'm not sure though if this is a typescript bug or inlay hint bug. And I can still reproduce it by doing random edits (took me some minutes though): !\nthis is still a duplicate of\nRepro steps: Wait for 4 seconds, notice the inlay hint afer Press enter in line 1 Wait for 1 second Press enter in line 1 every 100 ms Notice that after 3 seconds there are invalid inline suggestions ! Validation: In the playground, click on \"Compare with latest dev\" Verify that there are no invalid inline suggestions after doing the steps again"}
{"_id":"doc-en-vscode-baf3677221b3913de7ba0d61f7304d6cf2b90ddd3abbf2757b5ffa633415348a","title":"","text":"<!-- Do you have a question? Please ask it on --<!-- Use Help Report Issue to prefill these. --- VSCode Version: OS Version: Steps to Reproduce: a new file with a name starting with a space (using MacOS Finder) to rename the file by removing the space at the beginning of the file name (using the VS Code explorer). <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: Yes/No\nConfirmed on Windows too.. Edit: problem is . trims file names, but (in this case? in all cases?) should only remove trailing spaces."}
{"_id":"doc-en-vscode-3bb976adad717d6e0424ebe4fa8476bc8d29fa4e091dbe1af278d4f620ed3648","title":"","text":"We have received feedback that inlay hints blend in with the code too much. We aren't eager to tackle that with changing font sizes because that can \"harm\" readability. Instead we should tweak the colors, esp for our default themes\nLooking at a combination of slight color changes and some minor padding/border radius tweaks to make them look like a visibly different element and less like highlighted regular code. An example for dark+: <img width=\"337\" alt=\"CleanShot 2022-07-22 at 13 45 31 src=\"https://user-\"And for light+: <img width=\"280\" alt=\"CleanShot 2022-07-22 at 13 48 40 src=\"https://user-\"\nAdding to the list to make sure we look at the current line highlight color in relation to the inlay hints: !\nDid you push these changes or are these changes anywhere to be applied for August? Otherwise I'll move this to September\nMoved to September—I have the draft PR but I didn't consider the compact settings yet.\nping\nMade changes to all four base theme types. The most noticeable changes will be in the high contrast and light themes. The dark had decent contrast already but I tweaked it a bit. Let me know what you think."}
{"_id":"doc-en-vscode-41f17fb04d8fb6ee6090333d73557228f786756b4ed1270d43eb14da117e6cde","title":"","text":"I would expect the info box to remain open and not close and open on each keystroke: ! /cc\nduplicates\nThanks for creating this issue! We figured it's covering the same as another one we already have. Thus, we closed this one as a duplicate. You can search for existing issues . See also our guidelines. Happy Coding!"}
{"_id":"doc-en-vscode-605ae7c8ffd53bbe7be3d1396ed9a52dbbaf968a00dc2076eadc208e3ef7a547","title":"","text":"Issue Type: <bBug</bTesting . 1) Start creating a new folder in the explorer 2) Type 'folder1/' -once the '/' is in the input box every keystroke afterwards causes a full flicker of the information widget shown below the input box. This also happens on other platforms. VS Code version: Code - Insiders 1.21.0-insider (, 2018-02-27T06:17:53.785Z) OS version: Linux x64 4.4.0-59-generic Reproduces without extensions <!-- generated by issue reporter --\nWe need to debounce\nFixed by via"}
{"_id":"doc-en-vscode-01b44df2fda414757b9a06d4c7dbcef5467a7e99ce08b56d0c1f016ecd7bca76","title":"","text":"Issue Type: <bBug</bAssume If a FileSystemProvider extension implements more than one scheme, and if a file is opened using a scheme that is not used by the root folder(s) of the workspace, then when focus is switched to that file a call is issued on a folder resource that cannot possibly contain the file resource (because the scheme is different). In some scenarios a stat() call on a root folder can be costly, meaning it is important not to do it unnecessarily. This is not a new 1.37 issue. I will add information about how to demonstrate the issue using a variant of fsprovider-sample, and I am also willing to supply a PR to fix it. VS Code version: Code 1.37.0 (, 2019-08-08T02:33:50.993Z) OS version: Windows_NT x64 10.0.\nSteps to reproduce: Use 1.38.0-insider, or a 1.37 build that contains the fix for branch at fsprovider-sample in VS Code a logpoint in the stat() method of (line 58) with the following message: that the 'Launch Extension' debug configuration. the Extension Development Host window appears, run the following commands in it: MemFS: Setup Workspace MemFS: Add Second Root MemFS: Create Files to Explorer view. on under the second root (MemFS2 - Different Scheme) the first root doubleclick on at least two files. the Debug Console of your initial VS Code instance, observe the stat messages as you switch between file tabs on the EDH, e.g. context menu of second root (MemFS2) to Remove Folder from Workspace. between tabs of the files and observe how when you click on the opened from memfs2 there is an additional stat of the first root (memfs)\nFixed via can you just let us know tomorrow once the new insiders is out if the issue is indeed fixed for you. Thank you\nconfirmed good in today's Insider. Thanks!\nThanks, adding verified label."}
{"_id":"doc-en-vscode-2f3d21ca2ebe663527263985ce896c981f9ffe57c5fb26009541e00702335cf5","title":"","text":"from Windows to Linux where zshell is installed, with shell integration turned on a new (zsh) terminal. Terminal remains empty a bash terminal. That works shell intergation and create a zsh terminal. That works Left side was created with shell integration on, right side was create  Let me know if there's any addition info to get.\nIt would be helpful if you could turn on trace logging and provide the logs"}
{"_id":"doc-en-vscode-9a878a5589f6cfec7579a3a4b3e72d40210b1c8b0b9048b91563916fcede5078","title":"","text":"Testing on run continue on clone locally I see a browser-native dialog to open desktop I see a custom message that just reads \"We tried opening Visual Studio Code - Insiders on your computer.\" which is confusing Unsure if we have to show the message or not but its wording needs to be improved. I mean tried and succeed in opening desktop so don't confusing me with this message. If we cannot prevent it I'd suggest to rephrase it and assume that it worked and give me a close option. Today, there this only covers the \"didn't work cases\" despite success https://user-\nWe cannot tell from the browser whether opening the protocol url succeeded, and the current wording for this dialog is based on a similar dialog from Office which also unconditionally appears in the browser: Maybe renaming 'Continue here' to 'Cancel' or 'Close' would help? Some other ideas for wording--when you launch a Zoom meeting from a browser in the desktop client, it says !\nI can accept that but the wording should be positive/happy. Like: \"This session now continues in VS Code Desktop. You can close this window, all your changes have been carried along. Try again if VS Code Desktop didn't start yet.\""}
{"_id":"doc-en-vscode-c185a4e1045ea416992cb4f46b7f4e20f175ab270437c4829b39cedefd51b820","title":"","text":"Window Sync entry is shown again Version: 1.39.0-insider (user setup) Commit: OS: Windows_NT x64 10.0.\nCan confirm it happens on macOS with VSCode version 1.38.0"}
{"_id":"doc-en-vscode-2d3f4b7c472facde2b74eecbfb1fc0a022148015f3d135105ffbe130b1be187b","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --I am not happy with ctrl-shift-f. My steps: hit ctrl-shift-f enter the search term see several search results. want to go to the first match : There is no easy way to get to the first search result. After that I want to use up/down keys to go through the search results. The focus should not be in the source code. Screenshot of state of Step 5: ! Goal: Up to now I use the mouse and click on the first search result. If I use the mouse, I am at this state: ! After thinking about this some days, I think the best solution would be to map the ENTER to \"Go to first search result\". But not like F4, since F4 moves the cursor to the file, and not to the search result list. The ENTER should give the first item in the search result the focus. Then I can navigate to the matching result, and the next ENTER will bring me to the source code line. This would be great for all people who can't or don't want to use a mouse. Related:\nHi again! I'm just a bit confused on this part: For me, F4 moves to the first match, not the first file. Is this not what you're seeing? !\nI think right now this might be the best you can do: is a better command to consider than what you were using. The problem is that <kbdEnter</kbdis used to re-run the search if you make any changes to the inputs. But the above keybinding does focus the first result without moving the cursor, and then <kbdEnter</kbddoes move the cursor.\nthank you very much. If I add they keybinding via \"Open Keyboard Shortcuts (JSON)\", it works. But not always. After alt+enter I would like to use the up/down arrow keys and see the preview in the files. The strange thing: sometimes up/down shows me the places in the files (and the current entry in the result list is highlighted in blue color), and sometimes up/down does not show me the preview in the files."}
{"_id":"doc-en-vscode-564b7e42b23dfe9c88cd69fae3dbb46c6f6c6e37ef34b8c68961a89addcc9d23","title":"","text":"We are on the right track, but not there yet. Thank you very much for your kind support.\nFor me, F4 moves to the first match. If I press up/down keys, the cursor moves in the file. But I would like to move up/down in the search result list and see the preview on the right side (in the window of the source code editor).\nHmm, that's interesting, I'm also seeing that the files aren't previewing after using . will look into this more.\nI think I can get reproducible steps to the state where the file doesn't open when you have it highlighted in the tree view a long search it is searching, run using the up/down arrows to browse the search results, the file preview isn't triggering :bug: tree node is focused, but not selected :bug: This bug might be related to This is weird in that it only happens sometimes."}
{"_id":"doc-en-vscode-1fe3b765d107a2a47c233e4f75d334afe6dfcdbc04dd9fb0717af3b34cf83b7d","title":"","text":"Look after is merged in Dark Modern: !\nI see that it's currently using via Changing it to works, but I'm wondering what exactly broke this and why it was using in the first place?\nIt might have just always been out of sync which is common between the editor and simple find widget\n! left : version 81 right : insiders Maybe it's related to background-color ?\n! test, --vscode-panelInput-border color effect\nhi , I don't know if it's some new issues.\nis the find bg color meant to look like that?\nYes—changes made in to make editor widget and quick input backgrounds lighter to more easily distinguish against the editor background.\nYes, it's normal.Terminal simple find widget just not synchronized with the editor."}
{"_id":"doc-en-vscode-0cd9955c01655a7828bb2a7056b93bc9e6dc311213d3fd98b044c458a671d115","title":"","text":"Issue Type: <bBug</bwhen clicking the percentage at the bottom-right of the state bar,zooming function won't work, and the aspect ratio changes to abnormal value. VS Code version: Code 1.28.1 (, 2018-10-11T18:13:53.910Z) OS version: WindowsNT x64 10.0. <details<summarySystem Info</summary </details<details<summaryExtensions (9)</summaryExtensionVersion ------ intellij-idea-keybindings0.2.27 vscode-language-pack-zh-hans1.28.2 mssql1.4.0 java0.32.0 vscode-java-debug0.14.0 vscode-java-pack0.4.0 vscode-java-test0.10.0 vscode-maven0.11.1 markdown-all-in-one1.6.2 </details<!-- generated by issue reporter --\n(Experimental duplicate detection) Thanks for submitting this issue. Please also check if it is already covered by an existing one, like: <!-- score: 0.47 --<!-- potentialduplicatescomment --"}
{"_id":"doc-en-vscode-f6ca914ef0bc793c084832b9c5b3a8294cde271aea5cf8699261d2f3ca8ba931","title":"","text":"I use quite a lot and have noticed that sometimes earlier queries are restored after runs. AFAICT it might be because when we run , the restores the input value from the saved view state, which can be stale\nBlocked on verifying by\nyou can work around that issue by manually typing /clear into the input box, that issue only applies to when /clear is selected from the suggest widget"}
{"_id":"doc-en-vscode-a7071cbdb722acbd380ecd81657fc7e879141078efe319c80bfed5c86aac5979","title":"","text":"The issue is that when I resize the page to see it along with other content, I can see some text over the scroll page area. I think it wasn't supposed to work like this, that's why i'm reporting it here. ! ! .\nThe scrollbar is intentionally transparent. We have a color that is used only when the minimap is enabled. We could potentially use it always if set to allow defining opaque scrollbars.\nOh. Thank you for explaining :) In this case, sorry for bothering! Best, Alvaro\nTo verify, configure the color and that will be applied and will make the overview ruler opaque. e.g."}
{"_id":"doc-en-vscode-4e74b4d4d56c05ad5ca93c57c94c176aaa1e74cb4d751e768e29db8d3e427398","title":"","text":"<!-- Please search existing issues to avoid creating duplicates. --<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: --<!-- Use Help Report Issue to prefill these. --microsoft/monaco-editor VSCode Version: Monaco 0.17.0 Browser: Crhome Steps to Reproduce: a Link Provider for monaco-editor 2 .Click on the link on the monaco editor. results with the following error: <!-- Launch with to check. --Does this issue occur when all extensions are disabled?: N/A Solution: Looks like this is an error in Line 301: It compares the rejected promise errors to 'invalid' and 'missing', but in it rejects an Error not a string Line 61: So instead of the warning in throws an error. The code in should be:"}
{"_id":"doc-en-vscode-fc17320e54b4416db78cdc0043489034bd25d918b3d1f1c8eb6cd43c4e2df438","title":"","text":"<!-- Do Not Delete This! featurerequesttemplate --<!-- Please read our Rules of Conduct: --<!-- Please search existing issues to avoid creating duplicates. --<!-- Describe the feature you'd like. --Several shells (, , ) include vi mode keybindings. It's common for users to rebind vi escape to something like , , or . When interfacing with the shell, VS Code behaves as if it were a user typing keys, which may trigger vi escape if file names or other text input to the shell contains the escape sequence. To show the problematic behavior (assuming the shell is ): Include these lines in your In VS Code, try to run a Python file named . VS Code enters this file name into the shell, which triggers the escape sequence and causes the shell to interpret following the as vi normal mode commands. If VS Code input text to the shell with bracketed paste, this would not happen.\nNot sure if this is related, but I use Spectacle to move and resize my windows. It used to work perfectly, but now, since today's update, I cannot use my shortcut to resize my VSCode instance. The shortcuts I used were control+option+command+(left/right/down/up) to move and resize, and option+command+f to toggle fullscreen(not fullscreen but resized as fullscreen). These shortcuts do nothing in VSCode and are not triggered anymore through Spectacle\nCan you clarify how you're running the file? I think I understand the problem, sending text to the terminal via extension API or run active file, etc. are triggering the keybinding but wrapping them in bracketed paste would fix it. It should be safe to wrap in the bracketed paste sequences provided the mode is enabled in the terminal. that sounds like a different issue, maybe related to Electron.\nThank you, I'm gonna post this as a new issue.\nI'm running the file using Run -Start Debugging, selecting 'Python File' as the debug configuration. VS Code then enters this into the shell: The file is supposed to be called but my shell sees (my escape sequence) and then stops at ."}
{"_id":"doc-en-vscode-ad4a56d80df0bfd2cd72e04e3dd5770f3076cb5fbbfaf12b12bc259940f7f86a","title":"","text":"Thanks for confirming, we should be able to include the fix in 1.70 (too late for 1.69 unfortunately)\nThis caused a regression unfortunately so reopening The actual fix is more involved than I thought, we will need to be able to pass in a flag for whether to wrap in bracketed paste mode into the sendSequence command, internal API, extension API. An extension could wrap what it's sending in bracketed paste mode now but that won't work when the shell isn't in bracketed paste mode and the extension is not able to determine that.\nWe closed this issue because we in the foreseeable future. If you disagree and feel that this issue is crucial: we are happy to listen and to reconsider. If you wonder what we are up to, please see our and . Thanks for your understanding, and happy coding!"}
{"_id":"doc-en-vscode-a38f22c80796a1ba9c2a99777a2ba235466f8fc177fe0f90d5dfd9761e51aa3a","title":"","text":"Type: <bBug</ba new profile from empty click on activity bar \"Move Primary Side Bar Right\" ! VS Code version: Code - Insiders 1.72.0-insider (, 2022-09-29T08:07:23.017Z) OS version: WindowsNT x64 10.0. Modes: Sandboxed: Yes Remote OS version: Linux x64 5.15.62.1-microsoft-standard-WSL2 Remote OS version: Linux x64 5.15.62.1-microsoft-standard-WSL2 <details<summarySystem Info</summary </detailsExtensions: none <!-- generated by issue reporter --"}
{"_id":"doc-en-vscode-2e5a9ffeccc2d2387dfdc88c8830c93af714621ca6140ac110dd39aae39b60ab","title":"","text":"I'm using Light+. ! Could this be caused by bundling issues?\nI also see this in oss. Seems like a somewhat recent regression"}
{"_id":"doc-en-vscode-79dfcbe2fdb0fde8bacb9af26f144ac10e7411c28d5ea53912aa8f0325f773d5","title":"","text":"Testing a GH/Azure Repos repository on and file, and open the timeline view that both Git and Local History entries are shown a change to the file and save Local History entry is to the top of the list Local History entry is then moved under \"Uncommitted Changes\" entry\nJust in case: this is not specific to web version and can be reproduced in Desktop as well.\nI think this works as designed: the \"Uncommitted Changes\" from Git is probably always updating when you save and is more recent than the actual save event. would you agree this works as intended?\nRoughly as designed, but could be smoother. Maybe if we had a way to denote at an API level that an event is \"current\" rather than historical. Or maybe we shouldn't even have the \"uncommitted changes\" entry now that we have local file history.\nI agree, I don't think \"uncommitted changes\" should be a history entry at all. It's not a historical event. You might have uncommitted changes, but itself is not a change like everything else the timeline shows. Perhaps have a \"diff\" action in the Timeline title instead.\nI really like the idea to not have \"Uncommitted Changes\" at all. It seems to originate from the git extension: Especially now that we have local history enabled, there will always be an entry for local history above the last committ, so the information is redundant. Lets do this.\nnot sure whether we want to remove those for remotehub extension as well:\nRemoved the Uncommitted Changes timeline entries for RemoteHub. I kept the Staged Changes timeline entry though, since that is a historical event and it looks like we also kept those for the git extension.\nThanks!\nThis change has actually caught me off guard. While I like the idea for the \"Local History\" in general, It's sometimes too granular and I'd like to view all the changes since the last commit. Imho, the \"Uncommited Changes\" entry should be shown when \"local history\" is disabled, or it should be configurable by settings.\nCan you report a new feature request for this?"}