added
stringdate 2025-04-01 04:05:38
2025-04-01 07:14:06
| created
timestamp[us]date 2001-10-09 16:19:16
2025-01-01 03:51:31
| id
stringlengths 4
10
| metadata
dict | source
stringclasses 2
values | text
stringlengths 0
1.61M
|
|---|---|---|---|---|---|
2025-04-01T04:35:36.856544
| 2022-01-05T09:06:38
|
1094130086
|
{
"authors": [
"MTomBosch",
"simonbrowndotje"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11125",
"repo": "structurizr/cli",
"url": "https://github.com/structurizr/cli/issues/63"
}
|
gharchive/issue
|
Export using plantuml exporter does not work
On the Structurizr DSL demo page this workspace dsl leads to a plantuml error. Using the C4 plantuml exporter it works just fine.
I guess the reason is the incorrect location of the plantuml includes directive in the generated plantuml file.
workspace {
model {
user = person "User"
softwareSystem = softwareSystem "Software System"
user -> softwareSystem "Uses"
}
views {
systemContext softwareSystem {
include *
autolayout
}
properties {
"plantuml.includes" "https://gist.githubusercontent.com/simonbrowndotje/b84878f8b87af3b76753ed871611c700/raw/b659b7ab9ac02a04725606f59138f37d2e67c265/styles.puml"
"plantuml.legend" "true"
"plantuml.title" "false"
}
}
}
AddRelTag (in the included file) is a C4-PlanUML construct (see Custom tags/stereotypes support and skinparam updates), so it'll only work in conjunction the C4-PlantUML macros.
Ahh, I see.
Sorry for the inconveniences.
Will close it now.
|
2025-04-01T04:35:36.876680
| 2019-06-07T21:25:51
|
453699658
|
{
"authors": [
"kevinrobinson"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11126",
"repo": "studentinsights/studentinsights",
"url": "https://github.com/studentinsights/studentinsights/pull/2501"
}
|
gharchive/pull-request
|
Reader Profile: Moving parsing and segmenting IEP text to server
Who is this PR for?
K3 educators
What problem does this PR fix?
This is still just experimental, but moving parsing of IEP text to the server to run better data quality audits on it.
What does this PR do?
This is focused on just the dates first, but has experimental bits for the service grid and segmenting by sentences too.
Checklists
Which features or pages does this PR touch?
[x] Reader profile
Does this PR use tests to help verify we can deploy these changes quickly and confidently?
[x] Included specs for changes
[x] Manual testing made more sense here
selfie
|
2025-04-01T04:35:36.883962
| 2016-04-22T19:01:43
|
150447643
|
{
"authors": [
"alexsoble",
"erose"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11127",
"repo": "studentinsights/studentinsights",
"url": "https://github.com/studentinsights/studentinsights/pull/412"
}
|
gharchive/pull-request
|
Bar charts for behavior data
Moves the charts for behavior & attendance over to bar charts ("column" charts in Highcharts) from line charts. Mouseover on a bar for some month will bring up a list of the events that happened in that month.
Introduces a new React component, profile_bar_chart.js.
Fixes #343
Would love a code review and/or UI feedback!
Here's an alternate look, which I have on another branch.
Let me know if you prefer this one.
Thanks for tackling this, @erose!
Here is my feedback:
CODE
It would be good to write a spec for profile_bar_chart.js.
In general, any feature that adds significant complexity should get a spec. We've had two bugs that caused data to be shown incorrectly because of front-end bugs/mixups in the past couple of weeks: #363 and #415. Spec coverage for all our front-end features will help.
Eli, are you familiar with JS specs? If not, this is something we could pair on together at Code for Boston on Tuesday.
UI DETAILS
Noticed a couple of small possible improvements as I was playing around, offering as suggestions:
"Discipline Incidents, last 48 months" — The chart titles made me pause for a second to do mental math. Divide by 48 by 12: "oh, okay, last four years." Small thing, but easier for a viewer to mentally process number of years than number of months.
Date formatting in the tooltips could match the Full Case History formatting, i.e. "April 24th, 2016" instead of "04-24-2016".
MILESTONE
This is a great pull request because (among other reasons!) it resolves our last open issue in the Profile Page V2 milestone. That is a big deal: thank you for pushing this across the finish line, Eli!
🎉 🎉 🎉
I already pushed to staging and demo in the process of doing manual QA. I'll merge now and push to production, then I'd say our next order of business should be the spec.
Thanks for merging & QA'ing, Alex! Didn't even know this was the last one on the milestone. I agree with you on the UI stuff -- my fear was that "last n years" implies we cover the whole last n calendar years, start-to-finish, but that's just my hangup, no idea if it bothers our users or not.
Implemented and merged the UI fixes in #420.
Yeah let's pair on doing the spec, that sounds fun & morally improving!
@erose Indeed, like vegetables for your code!
|
2025-04-01T04:35:36.891994
| 2017-03-21T02:00:53
|
215607308
|
{
"authors": [
"alexsoble",
"jhilde",
"kevinrobinson"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11128",
"repo": "studentinsights/studentinsights",
"url": "https://github.com/studentinsights/studentinsights/pull/886"
}
|
gharchive/pull-request
|
Add graphing and filtering to student report
This is the consolidated pull request for phase 2 (graphs) and 3 (filtering) of the new student report.
Supersedes pull request #873
student_report.pdf
Pinging on this @alexsoble @kevinrobinson. Let me know if you have feedback.
Hi @jhilde!
I left a couple of in-line code comments about specific lines of code I want to understand a little better.
Also posting a GIF of what this new filtering UI looks like, below. I recommend posting screenshots and/or GIFs along with your PRs, it makes it clearer what the code you're merging in does.
GIF
What are your thoughts on the UI Justin?
Code duplication question
I'm noticing some of of the functions on student_profile_pdf.js are the same as functions on profile_bar_chart.js: monthKeys, defaultMonthKey, eventsToMonthBuckets. What do you think about extracting these methods out so that both of the files can share the same methods?
Cheers -- hope these questions and feedback are useful. Let me know what other ideas, thoughts, or questions are in your brain!
P.S. @jhilde let me know if you want to jump on a screenshare or hangout some time to talk about rebasing! We can use rebasing to get rid of some of the "noisy" commits that say things like Merge branch 'master' of https://github.com/jhilde/studentinsights and leave us an easier-to-read history.
@alexsoble Thanks for the helpful feedback! Those functions are exactly the same for that purpose. I'd be happy to factor them out. Thoughts about where they should belong?
And yes, I'd love to work through rebase and norms around verbosity on check in comments, etc.
And @alexsoble, what did you use to create the animated gif. Very cool.
@jhilde LICEcap! http://www.cockos.com/licecap/
@jhilde Let's figure out a good time to jump on a screenshare over email!
Hi @jhilde, sorry for the delay on reviewing, too much traveling the last two weeks. And thanks for jumping in @alexsoble so we can keep moving things forward. 🌮 👍
Moved shared graphing functions into a helper app/assets/javascripts/helpers/graph_helpers.js, removed code from app/assets/javascripts/student_profile/profile_bar_chart.js and pdf/student_profile_pdf.js, reworked the tests.
Nice job with the rebase @jhilde! Will merge in and push to prod early next week to avoid any Friday late afternoon push-to-prod jinxes!
Awesome! Thanks.
🚢 ing
💥 awesome @jhilde! 👍
|
2025-04-01T04:35:36.915326
| 2022-07-15T17:19:30
|
1306299894
|
{
"authors": [
"NicolaSaunders",
"simonrjones"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11129",
"repo": "studio24/amplify",
"url": "https://github.com/studio24/amplify/pull/9"
}
|
gharchive/pull-request
|
Add details on how to add Amplify into a project
As discussed at the last dev team meeting, these instructions explain how to get the Amplify front-end starter kit working in a project - please can you check it makes sense and that I haven't missed anything>
I've tested this on my own personal site & it seems to work well. I didn't need to copy the dist folder since running the npm build command generates all this for you.
See this other PR where I've added an automatically generated ZIP file we can attach to releases to enable people to easily download just the files they need. This can easily be expanded (e.g. an example standalone HTML file would be useful I think).
@maries24 Following Friday's dev team meeting, I have just pushed some more changes which removes the need for having two separate Webpack config files for handling JavaScript in Amplify, and have updated the getting started instructions accordingly.
@NicolaSaunders if you and @maries24 are happy with this, I'm happy for you to merge in the PR. If you want to talk through the versioning / ZIP file package stuff I'm happy to do so.
My only edit would be, there is no need to copy the dist folder if staff run the npm build command since it regenerates asset files.
|
2025-04-01T04:35:36.929392
| 2022-03-21T23:26:28
|
1176067540
|
{
"authors": [
"stv0g"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11131",
"repo": "stv0g/gose",
"url": "https://github.com/stv0g/gose/issues/3"
}
|
gharchive/issue
|
Support server-side encryption via SSE-C
We could add support for server-side encryption as supported by S3 SSE-C.
However this currently has the disadvantage that a user could not simple use the shortened URL anymore to download it via CURL/Wget, as SSE-C requires custom HTTP Headers for download.
This could be possibly worked around by a custom Javascript landing page which passes the headers via fetch() and then streams the response.
However, in this case we could already implement full end-to-end encryption which would make server-side encryption obsolete.
I am abandoning this feature in favour of full E2E encryption (#4)
|
2025-04-01T04:35:36.940439
| 2020-05-04T14:58:03
|
611941161
|
{
"authors": [
"MichaelDeBoey",
"jxnblk"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11134",
"repo": "styled-system/styled-system",
"url": "https://github.com/styled-system/styled-system/issues/1306"
}
|
gharchive/issue
|
[@styled-system/css] Performance issue due to Object.assign in _extends
Describe the bug
Just like @CompuIves described in codesandbox/codesandbox-client#4018, there's en enormous performance hit due to the fact that @styled-system/css is using Object.assign.
Replaces @styled-system/css with a version that doesn't use Object.assign, since that's very expensive for us. It improves render time 3 times for rendering components that use css.
Node 8 supports Object spreading, so I think this is something that can be fixed upstream and more people could benefit from this win.
Thanks! @styled-system/css is deprecated in favor of @theme-ui/css (not officially yet), but if you want to look into a fix, please see the Theme UI repo
|
2025-04-01T04:35:36.946398
| 2023-11-30T13:57:22
|
2018725777
|
{
"authors": [
"kristerkari",
"ybiquitous"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11135",
"repo": "stylelint-scss/stylelint-scss",
"url": "https://github.com/stylelint-scss/stylelint-scss/pull/888"
}
|
gharchive/pull-request
|
Fix function-no-unknown false negatives for functions with namespace
Since Stylelint 15.8.0, the built-in function-no-unknown rule has ignored SCSS functions with namespace.
For example, the following code unexpectedly passes with Stylelint 15.8.0:
a { color: color.unknown(#fff); }
Notes:
This change keeps backward compatibility. This means not to change peerDependencies.stylelint in package.json.
This change may bring performance penalty to keep backward compatibility.
Run npm i<EMAIL_ADDRESS>&& npm t to test this change with newer versions of Stylelint.
See also:
https://github.com/stylelint/stylelint/releases/tag/15.8.0
https://github.com/stylelint/stylelint/pull/6921
This Pull Request should resolve the CI failure in PR #873.
Thanks a lot! This is actually good to get fixed since it has been broken for quite a while already :)
|
2025-04-01T04:35:37.002348
| 2024-10-05T21:14:14
|
2568327171
|
{
"authors": [
"funkyhippo"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11136",
"repo": "subject-f/guya-status-page",
"url": "https://github.com/subject-f/guya-status-page/issues/251"
}
|
gharchive/issue
|
⚠️ Guya.moe Proxied has degraded performance
In a4c7016, Guya.moe Proxied (https://ice.guya.moe/) experienced degraded performance:
HTTP code: 200
Response time: 22153 ms
Resolved: Guya.moe Proxied performance has improved in 5b316b3 after 1 hour, 15 minutes.
|
2025-04-01T04:35:37.013010
| 2022-04-02T20:34:55
|
1190733427
|
{
"authors": [
"MANKEYYENAME",
"submada"
],
"license": "BSL-1.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11137",
"repo": "submada/btl",
"url": "https://github.com/submada/btl/issues/1"
}
|
gharchive/issue
|
Dub bug
The "unittest" configuration is built by default. Because of this, when exiting main, assert from btl is triggered
btl:autoptr 1.0.0-rc.4: target for configuration "unittest" is up to date.
btl:string 1.0.0-rc.4: target for configuration "unittest" is up to date.
btl:vector 1.0.0-rc.4: target for configuration "unittest" is up to date
I add new default(empty) configuration for all packages.
The triggered asserts from btl unittest configuration trigger when some memory is not freed. It can happend when string/smart_ptr/vector is inside gc allocated array (dlang bug for copy ctors), for value sequence on the stack (dlang bug for dtors) or some other bug. Can you create some example when this assert trigger in your code?
Thanks, I have a global associative array with weak pointers in my code. It will have to be forcibly removed.
|
2025-04-01T04:35:37.022588
| 2022-12-12T12:39:27
|
1491787523
|
{
"authors": [
"mkolesnik",
"vthapar"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11138",
"repo": "submariner-io/enhancements",
"url": "https://github.com/submariner-io/enhancements/issues/149"
}
|
gharchive/issue
|
Add support for nodeSelector and toleration scheduling for submariner pods
Epic Description
Submariner deploys many different agents as part of its core deployment. Except for Gateway and globalnet agent, that are deployed on designated gateway nodes, there is no way for users to control or limit where rest of the agent pods are deployed. Some customers want to run these agents on specific nodes for billing or networking reasons.
Acceptance Criteria
User will have option(s) to specify nodeSelectors and tolerations configurations for the following:
Submariner-operator
Submariner gateway
Submariner globalnet
Metrics Proxy
Lighthouse agent
Lighthouse CoreDNS
For this intial implementation, common selectors and tolerations that apply to all these agents would suffice.
Definition of Done (Checklist)
[ ] Code complete
[ ] Relevant metrics added
[ ] The acceptance criteria met
[ ] Unit/e2e test added & pass
[ ] CI jobs pass
[ ] Deployed using cloud-prepare+subctl
[ ] Deployed using ACM/OCM addon
[ ] Deploy using Helm
[ ] Deployed on supported platforms (for e.g kind, OCP on AWS, OCP on GCP)
[ ] Run subctl verify, diagnose and gather
[ ] Uninstall
[ ] Troubleshooting (gather/diagnose) added
[ ] Documentation added
[ ] Release notes added
Work Items
I think for these ones this doesnt make sense:
Submariner gateway
Submariner globalnet
Metrics Proxy
Gateway and globalnet are already only being deployed on nodes with label submariner.io/gateway=true which gives the users a lot of control already.
The metrics pods need to be deployed alongside the other pods they're proxying, so those ones don't make any sense in allowing this control either.
This issue should largely target the operator and the LH bits.
@mkolesnik Use case is user providing single set of nodeSelectors/Tolerations for all OCP pods
not just submariner pods. Considering this, what if user sets gateway label on the wrong node? Other scenario, case where we create a new GW node as part of cloud-prepare? In this case, should we set gateway label only on nodes that match the selector/toleration?
@mkolesnik Use case is user providing single set of nodeSelectors/Tolerations for all OCP pods not just submariner pods.
OK but I think some Subm pods behave in a very specific way.
We're more of an infra service, not a typical app, so expecting "all pods" is just not a realistic scenario.
Say like this, would you expect these selectors to affect every and all K8s/OCP components?
Considering this, what if user sets gateway label on the wrong node?
So they can fix the label quite easily, I don't see how this proposal is relevant to such a case in any way, they could be wrong here also, so I don't think its a valid scenario.
Other scenario, case where we create a new GW node as part of cloud-prepare? In this case, should we set gateway label only on nodes that match the selector/toleration?
IDK would we?
Currently we create the node and label it, otherwise what's the point in creating a dedicated gateway node in the first place?
|
2025-04-01T04:35:37.024302
| 2020-03-18T17:14:05
|
583881036
|
{
"authors": [
"mangelajo"
],
"license": "CC-BY-4.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11139",
"repo": "submariner-io/submariner-website",
"url": "https://github.com/submariner-io/submariner-website/pull/43"
}
|
gharchive/pull-request
|
Release process documentation
Signed-off-by: Miguel Angel Ajo<EMAIL_ADDRESS>Fixes-Issue: #31
It's a pitty we don't have a rendered version which is easy to use, I could may be teach the bot to render those so it's easier to preview, @tpantelis I think rendered output is some of the reasons for disagreement in comments.
|
2025-04-01T04:35:37.035200
| 2021-12-28T13:30:44
|
1089898482
|
{
"authors": [
"cwalterhk",
"nerdalert",
"nyechiel",
"sridhargaddam",
"yboaron"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11140",
"repo": "submariner-io/submariner",
"url": "https://github.com/submariner-io/submariner/issues/1642"
}
|
gharchive/issue
|
Auto-discovery of flannel CNI and ClusterCIDR is failing on K3s
What happened: Auto-discovery of CNI is failing on K3s. It does not detect the CNI in the cluster and uses "generic" route-agent.
✓ Showing Gateways
NODE HA STATUS SUMMARY
k3-subctl-w1 active All connections (1) are established
Discovered network details via Submariner:
Network plugin: generic
Service CIDRs: [<IP_ADDRESS>/16]
Cluster CIDRs: [<IP_ADDRESS>/24]
✓ Showing Network details
What you expected to happen: It should be able to detect flannel CNI.
How to reproduce it (as minimally and precisely as possible):
Follow the exact steps in URL https://submariner.io/getting-started/quickstart/k3s/. below is exact steps to build two clusters
Cluster a:
POD_CIDR=<IP_ADDRESS>/16
SERVICE_CIDR=<IP_ADDRESS>/16
curl -sfL https://get.k3s.io | INSTALL_K3S_EXEC="--cluster-cidr $POD_CIDR --service-cidr $SERVICE_CIDR" sh -s -
sudo snap install yq
sudo su
cp /etc/rancher/k3s/k3s.yaml kubeconfig.cluster-a
export IP=$(hostname -I | awk '{print $1}')
yq -i eval \
'.clusters[].cluster.server |= sub("<IP_ADDRESS>", env(IP)) | .contexts[].name = "cluster-a" | .current-context = "cluster-a"' \
kubeconfig.cluster-a
Cluster b:
POD_CIDR=<IP_ADDRESS>/16
SERVICE_CIDR=<IP_ADDRESS>/16
curl -sfL https://get.k3s.io | INSTALL_K3S_EXEC="--cluster-cidr $POD_CIDR --service-cidr $SERVICE_CIDR" sh -s -
sudo snap install yq
sudo su
cp /etc/rancher/k3s/k3s.yaml kubeconfig.cluster-b
export IP=$(hostname -I | awk '{print $1}')
yq -i eval \
'.clusters[].cluster.server |= sub("<IP_ADDRESS>", env(IP)) | .contexts[].name = "cluster-b" | .current-context = "cluster-b"' \
kubeconfig.cluster-b
Anything else we need to know?:
Environment:
Diagnose information (use subctl diagnose all):
walter@k3-subctl-m1:~$ subctl show all --kubeconfig kubeconfig.cluster-a
Cluster "default"
✓ Showing Connections
GATEWAY CLUSTER REMOTE IP NAT CABLE DRIVER SUBNETS STATUS RTT avg.
k3-subctl-w2 cluster-b <IP_ADDRESS> no libreswan <IP_ADDRESS>/16, <IP_ADDRESS>/24 connected
✓ Showing Endpoints
CLUSTER ID ENDPOINT IP PUBLIC IP CABLE DRIVER TYPE
cluster-a <IP_ADDRESS> <IP_ADDRESS> libreswan local
cluster-b <IP_ADDRESS> <IP_ADDRESS> libreswan remote
✓ Showing Gateways
NODE HA STATUS SUMMARY
k3-subctl-w1 active All connections (1) are established
Discovered network details via Submariner:
Network plugin: generic
Service CIDRs: [<IP_ADDRESS>/16]
Cluster CIDRs: [<IP_ADDRESS>/24]
✓ Showing Network details
COMPONENT REPOSITORY VERSION
submariner quay.io/submariner 0.11.0
submariner-operator quay.io/submariner 0.11.0
service-discovery quay.io/submariner 0.11.0
✓ Showing versions
walter@k3-subctl-m1:~$ subctl show all --kubeconfig kubeconfig.cluster-b
Cluster "default"
✓ Showing Connections
GATEWAY CLUSTER REMOTE IP NAT CABLE DRIVER SUBNETS STATUS RTT avg.
k3-subctl-w1 cluster-a <IP_ADDRESS> no libreswan <IP_ADDRESS>/16, <IP_ADDRESS>/24 connected
✓ Showing Endpoints
CLUSTER ID ENDPOINT IP PUBLIC IP CABLE DRIVER TYPE
cluster-b <IP_ADDRESS> <IP_ADDRESS> libreswan local
cluster-a <IP_ADDRESS> <IP_ADDRESS> libreswan remote
✓ Showing Gateways
NODE HA STATUS SUMMARY
k3-subctl-w2 active All connections (1) are established
Discovered network details via Submariner:
Network plugin: generic
Service CIDRs: [<IP_ADDRESS>/16]
Cluster CIDRs: [<IP_ADDRESS>/24]
✓ Showing Network details
COMPONENT REPOSITORY VERSION
submariner quay.io/submariner 0.11.0
submariner-operator quay.io/submariner 0.11.0
service-discovery quay.io/submariner 0.11.0
✓ Showing versions
walter@k3-subctl-m1:~$
Gather information (use subctl gather):
submariner-cluster-A-20211227124333.tar.gz
submariner-cluster-B-20211227124346.tar.gz
Cloud provider or hardware configuration:
VMWARE fusion to build K3S
Install tools:
NA
Others:
NA
Thanks for reporting this @cwalterhk. IIRC, despite the fact that the CNI is not being discovered properly, actual routing logic is in place and connectivity works as expected. Can you confirm? /cc @sridhargaddam
Thanks for reporting this @cwalterhk. IIRC, despite the fact that the CNI is not being discovered properly, actual routing logic is in place and connectivity works as expected. Can you confirm? /cc @sridhargaddam
Confirm that the routing logic is place. After I forced the clusterCIDR, everything works.
@nyechiel yes Nir, the routing logic would work fine. However, apart from CNI detection failure, the other problem is that Submariner operator was not detecting the clusterCIDR properly and wrong CIDR was getting configured in the cluster which was leading to the datapath failures.
IIUC, there are two separate issues here:
add support for flannel plugin discovery (currently flannel discovered as generic)
wrong pod cidr is retrieved when Submariner discovered flannel as generic plugin.
'2' is tracked at https://github.com/submariner-io/submariner-operator/issues/1730.
As per '1', I think we should handle it as a feature/enhancement request and not a bug.
Took a stab at Flannel discovery with in #1850 in operator
Thanks @nerdalert! As we are entering the final stages of the 0.12 cycle and this issue is not urgent AFAICT, is it ok if we defer this to 0.13? Just trying to make sure we meet our 0.12 priorities first. /cc @sridhargaddam @skitt FYI
Sounds good @nyechiel! Generic should tackle it. I mentioned in the PR for flannel discovery it is post v0.12. Maybe I should have put v0.13 instead.
|
2025-04-01T04:35:37.037518
| 2022-04-26T19:02:07
|
1216359830
|
{
"authors": [
"cohix",
"flaki"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11141",
"repo": "suborbital/docs",
"url": "https://github.com/suborbital/docs/pull/131"
}
|
gharchive/pull-request
|
fix(build): add build args to Dockerfile
This fixes build errors we're seeing in Cloud Run by setting the correct Docsearch env vars at build time.
TIL: ARG, but are those strictly needed here?
TIL: ARG, but are those strictly needed here?
As far as I can see from docs and StackOverflow, yes they're needed
|
2025-04-01T04:35:37.045669
| 2017-05-17T10:43:03
|
229310355
|
{
"authors": [
"michael",
"tobiassodergren"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11142",
"repo": "substance/substance",
"url": "https://github.com/substance/substance/pull/1038"
}
|
gharchive/pull-request
|
Collab rebase fix
Fix for rebase functionality, which seem to read from a property that was always undefined. Also, the version that was returned from the call to getChanges was missing in the callback
Thank you!
|
2025-04-01T04:35:37.065126
| 2023-10-18T17:22:00
|
1950275018
|
{
"authors": [
"fubhy",
"maoueh"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11143",
"repo": "substreams-js/substreams-js",
"url": "https://github.com/substreams-js/substreams-js/issues/40"
}
|
gharchive/issue
|
Migrate @bufbuild/connect (and relatives) dependency to @connectrpc
It appears that @bufbuild has moved over to a different organization and that new work is happening there. We would need to upgrade to that new version somehow.
Message I get when trying to install newest version(s):
$ yarn install
yarn install v1.22.19
warning package.json: No license field
warning<EMAIL_ADDRESS>No license field
[1/4] 🔍 Resolving packages...
warning<EMAIL_ADDRESS>Connect has moved to its own org @connectrpc and has a stable v1. Run `npx @connectrpc/connect-migrate@latest` to update. See https://github.com/connectrpc/connect-es/releases/tag/v0.13.1 for details.
warning<EMAIL_ADDRESS>Connect has moved to its own org @connectrpc and has a stable v1. Run `npx @connectrpc/connect-migrate@latest` to update. See https://github.com/connectrpc/connect-es/releases/tag/v0.13.1 for details.
[2/4] 🚚 Fetching packages...
...
Already in main!
|
2025-04-01T04:35:37.069369
| 2019-11-25T08:59:48
|
527941400
|
{
"authors": [
"subuk"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11144",
"repo": "subuk/vmango",
"url": "https://github.com/subuk/vmango/issues/54"
}
|
gharchive/issue
|
Better size selector
When specifying size of memory or disk it should be possible to specify unit, megabytes or gigabytes.
Done, now every size field has unit dropdown.
|
2025-04-01T04:35:37.071100
| 2018-12-03T10:36:15
|
386754647
|
{
"authors": [
"Almansherov",
"absidish"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11145",
"repo": "subutai-io/vagrant",
"url": "https://github.com/subutai-io/vagrant/issues/139"
}
|
gharchive/issue
|
Vagrant up gives error if pass disk as variable
SUBUTAI_DISK=300 vagrant up
successfully brought a peer up:
Fixed in master
Fixed in prod
|
2025-04-01T04:35:37.073535
| 2015-12-01T13:09:37
|
119714134
|
{
"authors": [
"AmyMcN",
"suedavis68"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11146",
"repo": "suedavis68/7_govuk_prototype",
"url": "https://github.com/suedavis68/7_govuk_prototype/pull/1"
}
|
gharchive/pull-request
|
Edit error messages to use plain language
Error messages were too wordy
@amymcn
I agree. Approved @suedavis68
|
2025-04-01T04:35:37.077612
| 2019-12-27T01:54:12
|
542713394
|
{
"authors": [
"Kiy4h",
"walterbender"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11147",
"repo": "sugarlabs/turtleart-activity",
"url": "https://github.com/sugarlabs/turtleart-activity/issues/74"
}
|
gharchive/issue
|
Suggestion: No delay when using the snail mode if the display doesn't change
It would be better when using the snail mode, blocks that doesn't cause the display to change doesn't delay.
Even better, if running the move/turn block with snail mode, the turtle smoothly glides/change direction.
What do you think?
Snail mode is really for debugging so you want to step through all the blocks, not just the graphics blocks themselves.
Re gliding, that would be a nice enhancement.
|
2025-04-01T04:35:37.108796
| 2021-10-08T20:16:17
|
1021432520
|
{
"authors": [
"andrewjbtw",
"peetucket"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11148",
"repo": "sul-dlss/happy-heron",
"url": "https://github.com/sul-dlss/happy-heron/issues/2072"
}
|
gharchive/issue
|
Embargoed items that will be released to the world have file-level rights that override the embargo
Description of the problem
When a self-deposit item is embargoed with settings that will release it to the world, the rights to the object should be citation-only and the files should not be available during the embargo period. Currently, the object rights are being set correctly to citation-only but the files are being given world rights, which means that they could be accessed prior to the end of the embargo.
Example druid: https://argo-stage.stanford.edu/view/druid:px464rn8992
The one file in this object has been given file-level world rights:
<access type="read">
<file>placeholder.txt</file>
<machine>
<world/>
</machine>
</access>
This means that the file could be downloaded directly from its Stacks URL
https://stacks-stage.stanford.edu/file/druid:px464rn8992/placeholder.txt
Fortunately, the Purl display does not expose the Stacks URL in this situation because it is set up to display the embargo information, so it makes it a little less likely that people have been able to access files that have been exposed.
Expected behavior
Embargoed objects should be citation only with files made inaccessible for the duration of the embargo.
This looks like the spot in the H2 code that is generating the cocina file level rights: https://github.com/sul-dlss/happy-heron/blob/main/app/services/cocina_generator/access_generator.rb#L34 and https://github.com/sul-dlss/happy-heron/blob/main/app/services/cocina_generator/access_generator.rb#L31-L37
I think we need to check to see at the object level if there is an embargo (like we do here: https://github.com/sul-dlss/happy-heron/blob/main/app/services/cocina_generator/access_generator.rb#L15 and https://github.com/sul-dlss/happy-heron/blob/main/app/services/cocina_generator/access_generator.rb#L31-L37) and then set the file level rights to none if so.
What is not immediately clear to me is how the embargo release process works ... is it smart enough to correctly set the file level rights once the embargo period is over?
See #2069 - similar problem
|
2025-04-01T04:35:37.111727
| 2024-09-10T14:32:52
|
2516651527
|
{
"authors": [
"ahafele",
"jermnelson"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11149",
"repo": "sul-dlss/libsys-airflow",
"url": "https://github.com/sul-dlss/libsys-airflow/issues/1194"
}
|
gharchive/issue
|
OCLC export - not receiving multiple oclc number email
Multiple OCLC Identifiers email was in place as of 8/27. No longer receiving this email or seeing report in UI when testing file of UUIDs with multiple OCLC numbers.
Example DAG run. https://sul-libsys-airflow-dev.stanford.edu/dags/select_oclc_records/grid?run_id=manual__2024-09-13T15:10:44.485233%2B00:00&execution_date=2024-09-13+15:10:44.485233%2B00:00&tab=graph&dag_run_id=manual__2024-09-13T15:10:44.485233%2B00:00
Receiving email now but not seeing the report in the UI.
@ahafele - Would there ever be a situation where multiple OCLC numbers would exist in more than one library? If so, we'll need to change the structure of the reports so that we can generate and save the reports that span multiple libraries.
@jermnelson do you mean a DAG run with 1 record from Law that has multiple OCLC numbers and 1 record from SUL that has multiple OCLC numbers? Yes, this can happen.
|
2025-04-01T04:35:37.126056
| 2015-03-17T10:05:01
|
62351801
|
{
"authors": [
"alexander-schranz",
"danrot"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11150",
"repo": "sulu-io/sulu",
"url": "https://github.com/sulu-io/sulu/issues/927"
}
|
gharchive/issue
|
Media content type overlay new media drop down hidden
@alexander-schranz Can't reproduce, please provide more information. Which browser, OS, Sulu version, ... Have you also tried other browsers?
@danrot Chrome Tested on 0.16.1 and actual develop.
To reproduce it you need to create many collection that the collection count is bigger then the window (not overlay) size.
@danrot I think the husky select component direction is on default (auto) we need to change it to bottom.
Fixed with complete new style of media assignment.
|
2025-04-01T04:35:37.127608
| 2015-05-29T08:45:00
|
82336195
|
{
"authors": [
"dantleech"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11151",
"repo": "sulu-io/sulu",
"url": "https://github.com/sulu-io/sulu/pull/1154"
}
|
gharchive/pull-request
|
Feature/refactoring strucure names
This PR renames the Component\Structrure\* classes to Component\Metadata\*Metadata.
This is to reduce the confusion about what these classes actually do.
We should also rename the ContentContainer to StructureContainer, and use Structure in general to refer to things derived from
@chirimoya @wachterjohannes @danrot fyi
|
2025-04-01T04:35:37.131218
| 2015-11-27T14:31:18
|
119204989
|
{
"authors": [
"wachterjohannes"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11152",
"repo": "sulu-io/sulu",
"url": "https://github.com/sulu-io/sulu/pull/1830"
}
|
gharchive/pull-request
|
Fixed load data in correct locale for excerpt
tasks:
[x] test coverage
informations:
q
a
Fixed tickets
none
Related PRs
none
BC breaks
none
Documentation PR
none
this issue happens when you use the excerpt categories in the website.
@danrot if the tests are green this is fixed (:
|
2025-04-01T04:35:37.134114
| 2015-12-11T09:11:19
|
121667546
|
{
"authors": [
"danrot",
"wachterjohannes"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11153",
"repo": "sulu-io/sulu",
"url": "https://github.com/sulu-io/sulu/pull/1869"
}
|
gharchive/pull-request
|
Removed hideChildrenAtBeginning option from datagrid
informations:
q
a
Fixed tickets
none
Related PRs
https://github.com/massiveart/husky/pull/589
BC breaks
none
Documentation PR
none
What is about this openToPath option? Shouldn't that also be the same on every screen?
@danrot fixed
|
2025-04-01T04:35:37.189798
| 2022-09-01T22:47:59
|
1359490559
|
{
"authors": [
"bdoublet91",
"lokize"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11154",
"repo": "sunlei/zsh-ssh",
"url": "https://github.com/sunlei/zsh-ssh/issues/4"
}
|
gharchive/issue
|
still working in 2022?
i using autocomplete plugin and when i try press space after ssh command i have this error:
fzf-complete-ssh:24: command not found: fzf
Yeah it's working
Make sure you have fzf installed.
sudo apt install fzf
if you are on linux
working good now thank you very much!
|
2025-04-01T04:35:37.192253
| 2024-01-06T14:19:43
|
2068653467
|
{
"authors": [
"sunny0826",
"yardenshoham"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11155",
"repo": "sunny0826/kubecm",
"url": "https://github.com/sunny0826/kubecm/pull/837"
}
|
gharchive/pull-request
|
Fix delete removes user even if another context is using it
I added a check to make sure that we don't delete a user or a cluster that is in use by another context than the one getting deleted.
I also added a unit test that deletes a context that references a user that is referenced within another context.
Fixes #814
Thanks a lot ! I will finish the Review as soon as possible
@all-contributors please add @yardenshoham for test.
|
2025-04-01T04:35:37.241059
| 2022-05-03T23:17:51
|
1224759164
|
{
"authors": [
"dshukertjr"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11156",
"repo": "supabase-community/gotrue-dart",
"url": "https://github.com/supabase-community/gotrue-dart/pull/75"
}
|
gharchive/pull-request
|
fix: inintialize completer before a new token refresh
This PR is a rework of https://github.com/supabase-community/gotrue-dart/pull/71 pretty much.
Instead of just retrying token refresh every 5 seconds, this PR will use a Completer to return a valid GoTrueSessionResponse upon successful token refresh after retry.
Haven't written tests for it, but am currently working on it. Also still working on some refactoring.
|
2025-04-01T04:35:37.248410
| 2023-11-01T05:44:16
|
1971742354
|
{
"authors": [
"coveralls",
"sweatybridge"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11157",
"repo": "supabase/cli",
"url": "https://github.com/supabase/cli/pull/1628"
}
|
gharchive/pull-request
|
fix: reformat functions list output
What kind of change does this PR introduce?
Bug fix, feature, docs update, ...
What is the current behavior?
Please link any relevant issues here.
What is the new behavior?
Feel free to include screenshots if it includes visual changes.
Additional context
Add any other context or screenshots.
Pull Request Test Coverage Report for Build<PHONE_NUMBER>
11 of 11 (100.0%) changed or added relevant lines in 1 file are covered.
5 unchanged lines in 1 file lost coverage.
Overall coverage increased (+0.008%) to 58.603%
Files with Coverage Reduction
New Missed Lines
%
internal/gen/keys/keys.go
5
12.31%
Totals
Change from base Build<PHONE_NUMBER>:
0.008%
Covered Lines:
5671
Relevant Lines:
9677
💛 - Coveralls
|
2025-04-01T04:35:37.250581
| 2022-01-18T08:24:02
|
1106606965
|
{
"authors": [
"alaister",
"joshenlim"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11158",
"repo": "supabase/storage-api",
"url": "https://github.com/supabase/storage-api/issues/88"
}
|
gharchive/issue
|
Add support for search in bucket
Feature request
Relevant PR: https://github.com/supabase/supabase/pull/4961
Searching for storage in the dashboard at the moment is a client side search - we do a name string search against what was already fetched from the API
As we're looking to introduce infinite scrolling to the dashboard, this means that we won't (be attempting to) pull all of the folder contents when the user clicks on a folder, hence we need to shift the search logic to the API level
Searching should also support pagination as well (we can just stick with offset limit as per the list items API)
Solved in #127
|
2025-04-01T04:35:37.257921
| 2022-05-14T18:28:04
|
1236092670
|
{
"authors": [
"MagnusHJensen",
"marcinkoziej",
"soedirgo"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11159",
"repo": "supabase/supabase-js",
"url": "https://github.com/supabase/supabase-js/pull/443"
}
|
gharchive/pull-request
|
Allow to override Supabase service URLs in client
Add 4 options to SupabaseClient that allow to set urls for Supabase
services: REST, Realtime, Auth, Storage, by option keys restUrl,
realtimeUrl, authUrl and storageUrl. The option should contain full
url (eg. https://foo.com/auth/v1 for GoTrue hosted at such url.)
What kind of change does this PR introduce? What is the current behavior?
Current SupabaseClient behaviour is to accept an url to Supabase instance, and construct sub-service urls by appending paths (/rest/v1, /auth/v1 etc).
This change allows to override the urls by passing them in options, when for example in self-hosting context such services run under different domains.
Please link any relevant issues here.
This solution helps to mitigate lack of custom domain support in Supabase.
Thanks to generously decoupled architecture, we will be able to just self-host GoTrue under our custom domain, and use SaaS Supabase. For this to work, we need to pass authUrl that points to our custom GoTrue instance.
What is the new behavior?
The new behaviour is same, unless options are used to override URLs for Supabase sub-services.
I just noticed this is a duplicate of PR #399 - almost identical but a bit more concise.
Hey Supabase team! This is a very small change, can you review?
I'll close this for inactivity, but for folks bumping into the issue mentioned by OP, my recommendation is to use an API gateway to route the /*/v1 endpoints. E.g. on our self-hosted setup: https://github.com/supabase/supabase/blob/c117ae67eb376e997a9e2d04e457a3c174e75df7/docker/volumes/api/kong.yml
@kangmingtay sorry for late reply.
We want to use Supabase (so all the services via SupabaseClient) but to host gotrue ourselves (so that we can use custom domain for Google OIDC and our email server).
So we want to point SupabaseClient to our gotrue, but use Supabase hosted Storage, PostgREST etc.
Because of how SupabaseClient is encapsulated using TS OO primitives, it is impossible to override only authUrl.
I could change the client.auth after creating the client, but the constructor has already added some listeners to a default gotrue client - which looks like there will be some conflicts in there.
I could subclass and change the constructor, copy-paste most of it and just change the auth url, but the constructor calls private methods so they will not be usable.
My point is the style of how SupabaseClient is coded, does not really make its extension possible or even easy.
I think Object Oriented should mean that it's possible! Otherwise, lets use functional, not OO style of coding there.
@soedirgo yeah I could proxy all the api calls, but why? This will be less reliable and slower. i'd like only serve Auth api, rest should go straight to my Supabase project.
This a really simple change, and I don't see the problem with it.
As mentioned, you might want to use supabase overall, but self-host only one service and then it becomes tedious to run supabase-js and gotrue-js as seperate clients, instead of just being able to override the URL's.
|
2025-04-01T04:35:37.358430
| 2016-12-23T18:41:28
|
197412347
|
{
"authors": [
"supasate"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11160",
"repo": "supasate/connected-react-router",
"url": "https://github.com/supasate/connected-react-router/issues/8"
}
|
gharchive/issue
|
URL is out of sync in Time Travel mode
Redux DevTools works fine. Time Travel works fine except the URL bar is not changed when we rewind the state.
@dictions I made it as a new issue. We can leave it until there is a real need to use this feature or we can think of some better ways to do it.
I can think of a hacky way by breaking a uni-directional flow in this mode (so I don't want to use it yet). Assume we can check whether we are in time travel mode (not sure it is possible or not). When router state is changed, we push URL directly to history object if we are in the time travel mode. Then, in history.listen callback, we ignore calling props.onLocationChanged if we are in the time travel mode.
|
2025-04-01T04:35:37.431113
| 2021-01-28T07:47:24
|
795759085
|
{
"authors": [
"mariosant",
"supermacro"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11161",
"repo": "supermacro/neverthrow",
"url": "https://github.com/supermacro/neverthrow/pull/229"
}
|
gharchive/pull-request
|
Export factories directly
Hello! This PR is intended to start a discussion for ergonomics.
I have recently discovered this great library and introduced to a colleague of mine. I felt it was a bit awkward to use the available factories, so I am raising this PR to discuss about it.
This PR makes factories available as a top level export.
It will simplify the usage overall, as it will be more straightforward to use.
import { fromThrowable } from 'neverthrow'
// instead of
import { Result } from 'neverthrow'
const { fromThrowable } = Result
Hey @mariosant, thanks for opening this PR!
I agree with the proposal overall. However, I do not agree that existing methods should be moved / deprecated.
Instead what I ask is that you re-export these functions so that folks can chose how they interact with neverthrow. Some folks (like myself) can keep doing:
import { Result } from 'neverthrow'
// ...
Result.fromThrowable(blah)
While others (such as yourself) can import fromThrowable directly from neverthrow as you demonstrated above.
This, of course, also implies updating the README so that people are aware if their options.
Let me know what you think 🙂
Hey @mariosant, thanks for opening this PR!
I agree with the proposal overall. However, I do not agree that existing methods should be moved / deprecated.
Instead what I ask is that you re-export these functions so that folks can chose how they interact with neverthrow. Some folks (like myself) can keep doing:
import { Result } from 'neverthrow'
// ...
Result.fromThrowable(blah)
While others (such as yourself) can import fromThrowable directly from neverthrow as you demonstrated above.
This, of course, also implies updating the README so that people are aware if their options.
Let me know what you think 🙂
I think it's a great suggestion, @supermacro ! I will ping you when the PR gets updated.
I think it's a great suggestion, @supermacro ! I will ping you when the PR gets updated.
Hey @supermacro, I have updated this. Please take a look when possible!
Hey @supermacro, I have updated this. Please take a look when possible!
Subtle ping for @supermacro in case you missed the notification.
Subtle ping for @supermacro in case you missed the notification.
Hey @mariosant thanks for the contribution! Since there was a new beta release recently I have to decide when this upstream work will land in the npm default tag.
What I am thinking is that I create a new beta release under v4.1.0-beta.0 for now. I still don't yet feel comfortable moving v4.1.0 into the default npm tag.
Expect this new beta release to land sometime this week.
For future reference (and to anyone else reading this), I don't really pay attention to issues / pull requests on weekends 🙂
I also just started a new job today, so there may be longer delays on PRs / issues from now on.
Thank you for the update, @supermacro! I don't mind delays, thought you might have missed the notification or something.
Good luck with your new role!
|
2025-04-01T04:35:37.453689
| 2023-06-08T14:54:44
|
1748075170
|
{
"authors": [
"MaroLFC"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11162",
"repo": "superunitybuild/buildtool",
"url": "https://github.com/superunitybuild/buildtool/issues/119"
}
|
gharchive/issue
|
Can i run a prebuild scripts for a configuration without running the player build?
My Use case is the following. I create a pre-build script that builds the addressable for my project but sometimes I only need the addressable build and not the player build is there a way to abort the player build and continue to the other build configurations?
Thank you so much that is a great feature I will use it right away. Can I do the same for the build configurations? like control the Enable and disable of builds through a prebuild action.
Another question
I would like to contribute to the project with some of the Build actions I created. How can I do that?
Sorry for bothering you and thanks again
Another question also can i run the configure editor environment button using cli
|
2025-04-01T04:35:37.535399
| 2017-08-18T01:26:01
|
251115582
|
{
"authors": [
"dmitrykurmanov",
"haeyon098",
"tsv2013"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11163",
"repo": "surveyjs/surveyjs",
"url": "https://github.com/surveyjs/surveyjs/issues/570"
}
|
gharchive/issue
|
Can i use <script> with vue.js?
hi! I found surveyjs and i really wanna use it in my project.
I use vue.js. I'm not good front-end programmer. So, I can't use npm.
I want to use surveyjs with <script> like : https://jsbin.com/voqetuduri/2/edit?html,js,output
I found this context your Issues. But, I can't use it because of it made of knockout.
Can I use surveyjs with vue.js and <script>?
If you reply, it help me a lot.
Hi!
We created a surveyjs_vue_quickstart repo to demonstrate how to integrate SurveyJS and SurveyJS Editor into VueJS application. Could you check it?
Please feel free to contact us in case of any further questions.
Thanks, Serge
@tsv2013 Thank you for your answer.
I checked it already. Your surveyjs_vue_quickstart repo explains how to use it withnpm. But, I already told you, I can't use npm in my project.
I really wanna use surveyjs with <script>(or cdn) and vuejs.
Can I use it with <script>? If it possible, how can I use it?
In spite of SurveyJS editor is built with KnockoutJs, it can be used with almost any framework.
And it doesn't matter - how this framework compiled and bundled.
I'm a bit confused because of you are telling, that you have a VueJS application. Because probably you are writing Vue components and build and bundle them somehow and inject on the page.
Can I take a look at your project - as a github repo if it possible, and I'll try to help you.
Thanks, Serge
@tsv2013 Thank you!
At first, I don't understand about koockout.
I can use it now! Thanks to you!
Using surveyjs editor, I ask you other questions next time!!
@haeyon098
for library: if you need an example without npm you could open any our example for vuejs (http://surveyjs.org/examples/vue/questiontype-radiogroup/) and you will see all code (html/js/css/result tabs) for start or you may open it in sandbox (plunker).
for editor: we will create same possibility for editor for example https://plnkr.co/edit/oTG18BKp5AqQLCP3fHca?p=preview
@dmitrykurmanov Thank you about your information!
If you create example, Can I see your homepage? (http://surveyjs.org/index.html)
Will you create example page like library example page (http://surveyjs.org/examples/vue/questiontype-radiogroup/) ?
@haeyon098, hello! You can find source code of surveyjs.org in our repo: https://github.com/surveyjs/site. And you are right, we will create same examples for the editor like for the library.
|
2025-04-01T04:35:37.540657
| 2018-12-11T10:37:36
|
389697748
|
{
"authors": [
"dmitrykurmanov",
"willgdjones"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11164",
"repo": "surveyjs/widgets",
"url": "https://github.com/surveyjs/widgets/issues/95"
}
|
gharchive/issue
|
Include Bootstrap widget in build
Are you requesting a feature, reporting a bug or asking a question?
Requesting a feature
What is the current behavior?
I would like to use the Bootstrap Datepicker, which seems to be available in the widgets src.
https://github.com/surveyjs/widgets/tree/master/src
What is the expected behavior?
I can't seem to be able to import the bootstrap datepicker in the Survey Editor with e.g.
widgets.bootstrapdatepicker(SurveyKo);
It seems like it is not exported in the https://github.com/surveyjs/widgets/blob/master/webpack.config.js file - so should be a straightforward fix.
I've added the bootstrapdatepicker widget to the build.
As for your local error, could you please try to use "type": "bootstrapdatepicker",
Unfortunately using "type": "bootstrapdatepicker", doesn't change the error
I've created an example with latest widgets build and it works as expected for me :
https://plnkr.co/edit/ULr0gu2rXYD0uiq8yGNg?p=preview
Could you please check it?
Hi @dmitrykurmanov - I do see that this work well, but I don't immediately see how to translate that into my development environment. Also - there seems to be a large amount of external CDN scripts that makes it hard to see what I need to include myself. Is it possible to look at an example with codesandbox?
I will try to build it myself but it could take a while.
@willgdjones I am not sure, that bootstrapdatepicker works as es6 module. So could you please try to add it as script tag into your index.html?
|
2025-04-01T04:35:37.572129
| 2018-08-02T20:04:01
|
347154745
|
{
"authors": [
"aceresia",
"aceresia-bg",
"russalex83",
"sushilks"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11165",
"repo": "sushilks/kumojs",
"url": "https://github.com/sushilks/kumojs/issues/1"
}
|
gharchive/issue
|
cryptokeyFromAddress
I'm trying to create a python version of this for home assistant. Do you know what type of cryptology is being used in the function "cryptokeyFromAddress"? Is it a standard one or proprietary crypto from Mitsubishi?
You will need to reverse engineer the code for python conversion, The code uses the standard crypto library and some modification to do the hash.
I'm trying to create a python version of this for home assistant. Do you know what type of cryptology is being used in the function "cryptokeyFromAddress"? Is it a standard one or proprietary crypto from Mitsubishi?
Hi,
Did you figure it out yet? Thanks
Yes, Here is what I got
` @staticmethod
def __l2char(l):
r = ''
for c in l:
r += chr(c)
return r
@staticmethod
def __l2h(l):
result = ''
for i in l:
if i < 16:
result += '0'
result += i.toString(16)
return result
@staticmethod
def __h2l(h):
r = []
for i in range(0, len(h)-1)[::2]:
r.append(int(h[i:i+2], 16))
return r
@staticmethod
def __unicodeList(l):
result = []
for i in l:
result.append(ord(i))
return result
def __cryptokeyFromAddress(self, dt):
W = kumoZone.__h2l(self.__W)
p = base64.b64decode(self.__password)
hash_string = p.decode('latin1', 'replace') + dt
dt1 = hashlib.sha256(hash_string.encode("latin1")).hexdigest()
dt1_l = kumoZone.__h2l(dt1)
dt2 = ''
for i in range(0, 88):
dt2 += '00'
dt3 = kumoZone.__h2l(dt2)
dt3[64] = 8
dt3[65] = 64
dt3[32:64] = dt1_l
dt3[66] = self.__S
cryptoserial = kumoZone.__h2l(self.__cryptoSerial)
dt3[79] = cryptoserial[8]
dt3[80] = cryptoserial[4]
dt3[81] = cryptoserial[5]
dt3[82] = cryptoserial[6]
dt3[83] = cryptoserial[7]
dt3[84] = cryptoserial[0]
dt3[85] = cryptoserial[1]
dt3[86] = cryptoserial[2]
dt3[87] = cryptoserial[3]
dt3[0:32] = W
hash_string = kumoZone.__l2char(dt3)
result = hashlib.sha256(hash_string.encode("latin1")).hexdigest()
return result`
Yes Here it is
`def __l2char(l):
r = ''
for c in l:
r += chr(c)
return r
def __l2h(l):
result = ''
for i in l:
if i < 16:
result += '0'
result += i.toString(16)
return result
def __h2l(h):
r = []
for i in range(0, len(h)-1)[::2]:
r.append(int(h[i:i+2], 16))
return r
def __unicodeList(l):
result = []
for i in l:
result.append(ord(i))
return result
def __cryptokeyFromAddress(dt):
W = __h2l(__W)
p = base64.b64decode(__password)
hash_string = p.decode('latin1', 'replace') + dt
dt1 = hashlib.sha256(hash_string.encode("latin1")).hexdigest()
dt1_l = __h2l(dt1)
dt2 = ''
for i in range(0, 88):
dt2 += '00'
dt3 = __h2l(dt2)
dt3[64] = 8
dt3[65] = 64
dt3[32:64] = dt1_l
dt3[66] = __S
cryptoserial = __h2l(__cryptoSerial)
dt3[79] = cryptoserial[8]
dt3[80] = cryptoserial[4]
dt3[81] = cryptoserial[5]
dt3[82] = cryptoserial[6]
dt3[83] = cryptoserial[7]
dt3[84] = cryptoserial[0]
dt3[85] = cryptoserial[1]
dt3[86] = cryptoserial[2]
dt3[87] = cryptoserial[3]
dt3[0:32] = W
hash_string = __l2char(dt3)
result = hashlib.sha256(hash_string.encode("latin1")).hexdigest()
return result`
Just ported from node to python.
Yes Here it is
`
def __l2char(l):
r = ''
for c in l:
r += chr(c)
return r
def __l2h(l):
result = ''
for i in l:
if i < 16:
result += '0'
result += i.toString(16)
return result
def __h2l(h):
r = []
for i in range(0, len(h)-1)[::2]:
r.append(int(h[i:i+2], 16))
return r
def __unicodeList(l):
result = []
for i in l:
result.append(ord(i))
return result
def __cryptokeyFromAddress(dt):
W = __h2l(__W)
p = base64.b64decode(__password)
hash_string = p.decode('latin1', 'replace') + dt
dt1 = hashlib.sha256(hash_string.encode("latin1")).hexdigest()
dt1_l = __h2l(dt1)
dt2 = ''
for i in range(0, 88):
dt2 += '00'
dt3 = __h2l(dt2)
dt3[64] = 8
dt3[65] = 64
dt3[32:64] = dt1_l
dt3[66] = self.__S
cryptoserial = __h2l(__cryptoSerial)
dt3[79] = cryptoserial[8]
dt3[80] = cryptoserial[4]
dt3[81] = cryptoserial[5]
dt3[82] = cryptoserial[6]
dt3[83] = cryptoserial[7]
dt3[84] = cryptoserial[0]
dt3[85] = cryptoserial[1]
dt3[86] = cryptoserial[2]
dt3[87] = cryptoserial[3]
dt3[0:32] = W
hash_string = __l2char(dt3)
result = hashlib.sha256(hash_string.encode("latin1")).hexdigest()
return result
`
Just ported from node to python.
Great! Thanks a lot! Is there anyway you could share your complete project in python? like src, files, libs, etc. It would be greatly appreciated. I don't have whole bunch of experience so python is easier for me to work with. Thanks again for your time.
|
2025-04-01T04:35:37.585509
| 2024-08-09T05:54:46
|
2457161058
|
{
"authors": [
"sunya-ch",
"vimalk78"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11166",
"repo": "sustainable-computing-io/kepler-model-server",
"url": "https://github.com/sustainable-computing-io/kepler-model-server/issues/359"
}
|
gharchive/issue
|
Hidden power consumption of Kepler seems to significantly increases
What happened?
Data source: ec2 spot instance 5c.metal
This issue describes the significant different between power metrics collected on Feb and the power metrics collected on July. While CPU time from both are fair, the power consumption on July seems to much more increases from beginning even with a small load. The power of this machine seems to saturate around 450. These power number are from intel rapl directly.
Further investigation found that in July, the CPU instruction counter is highly increased compare to those in Feb.
previously (around Feb 2024)
current (July 2024)
What did you expect to happen?
Increment of CPU instruction used by Kepler should be explainable.
We should further investigate more metrics since CPU time is not enough for modeling.
How can we reproduce it (as minimally and precisely as possible)?
Run Kepler release in Feb separately from Kepler release in July.
Anything else we need to know?
No response
Kepler image tag
0.7.0 and 0.7.11
Deployment
[ ] Model server
[ ] Estimator
[ ] Online trainer
[ ] Offline trainer
[ ] Profiler
Kepler model server image tag if deployed
Kepler estimator image tag if deployed
Kepler online trainer image tag if deployed
Kepler offline trainer image tag if deployed
Kepler profiler image tag if deployed
Kubernetes version
$ kubectl version
# paste output here
Install tools
Kepler deployment config
For on kubernetes:
$ KEPLER_NAMESPACE=kepler
# provide kepler configmap
$ kubectl get configmap kepler-cfm -n ${KEPLER_NAMESPACE}
# paste output here
# provide kepler model server configmap if Kepler Model Server is deployed
$ kubectl get configmap kepler-model-server-cfm -n ${KEPLER_NAMESPACE}
# paste output here
# provide kepler deployment description
$ kubectl describe deployment kepler-exporter -n ${KEPLER_NAMESPACE}
For standalone:
put your Kepler command argument here
Feb build info:
{"kepler_exporter_build_info": [{"metric": {"name": "kepler_exporter_build_info", "container": "kepler-exporter", "endpoint": "http", "goarch": "amd64", "goos": "linux", "goversion": "go1.20.10", "instance": "kind-for-training-control-plane", "job": "kepler-exporter", "namespace": "kepler", "pod": "kepler-exporter-xbkgb", "revision": "unknown", "service": "kepler-exporter", "tags": "include_gcs,include_oss,containers_image_openpgp,gssapi,providerless,netgo,osusergo,gpu,libbpf,linux"}
July build info:
{ "kepler_exporter_build_info": [{"metric": {"name": "kepler_exporter_build_info", "arch": "amd64", "branch": "main", "container": "kepler-exporter", "endpoint": "http", "instance": "kind-for-training-control-plane", "job": "kepler-exporter", "namespace": "kepler", "os": "linux", "pod": "kepler-exporter-qp8cc", "revision": "bf1f62d8c580aa742d4ae90dedaff70044be9b78", "service": "kepler-exporter", "version": "v0.7.11"}
since this issue talks about kepler's power metrics, should this be a kepler issue or model-server issue?
We should further investigate more metrics since CPU time is not enough for modeling.
can you please elaborate this more? do we need to use more metrics provided by kepler? or kepler itself needs to produce more metrics to be used as new features in model?
|
2025-04-01T04:35:37.599718
| 2019-03-01T05:56:12
|
415953731
|
{
"authors": [
"suufi",
"thoricelli"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11167",
"repo": "suufi/noblox.js-server",
"url": "https://github.com/suufi/noblox.js-server/issues/2"
}
|
gharchive/issue
|
Visiting other api’s/not let it sign out of all accounts
Hello there,
I’ve been using this tool for some time now and I do really like it!
Theres a problem: Everytime I node “server.js” it signs me out of my account and I need to enter the 2FA code again!
I’d also like to know how you would be able to visit other API’s other than the ones given here.
Thanks!
Whoops! Sorry about the super long delay. This is a problem we can't really fix. We're advocating for better API support from Roblox here but until then your best shot is to use an alt.
Hello, I kinda already fixed my own problem! What I did is: I changed the logout link with the presence one (This one: https://presence.roblox.com/v1/presence/users) and that seemed to work flawlessly! I don't need to log myself out just in order to get the X-CSRF-TOKEN!
|
2025-04-01T04:35:37.600664
| 2024-03-08T09:12:53
|
2175609558
|
{
"authors": [
"suvanbanerjee"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11168",
"repo": "suvanbanerjee/pycryp",
"url": "https://github.com/suvanbanerjee/pycryp/issues/1"
}
|
gharchive/issue
|
generate_password Function not working properly
if the length more than 26 then it is giving value error
Fixed 64bc8989f6ff164193ceb5a4b5e7d68bb2761823
|
2025-04-01T04:35:37.608364
| 2021-12-16T07:42:40
|
1081869287
|
{
"authors": [
"suzuki-shunsuke"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11169",
"repo": "suzuki-shunsuke/github-action-validate-envoy-proxy",
"url": "https://github.com/suzuki-shunsuke/github-action-validate-envoy-proxy/issues/2"
}
|
gharchive/issue
|
Screen shot
https://github.com/suzuki-shunsuke/github-action-validate-envoy-proxy/pull/1#issuecomment-995517990
https://github.com/suzuki-shunsuke/github-action-validate-envoy-proxy/pull/10#issuecomment-995644505
|
2025-04-01T04:35:37.624869
| 2016-09-09T16:45:55
|
176055221
|
{
"authors": [
"gidesa",
"harold-b",
"svaarala"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11170",
"repo": "svaarala/duktape",
"url": "https://github.com/svaarala/duktape/issues/960"
}
|
gharchive/issue
|
Missing functions in shared librarry
In the Duktape shared library (.dll, .so) are missing some functions reported in API guide.
These functions are very useful to a better comunication with the embedding program (C program, or whatever):
duk_peval_string
duk_eval
duk_error
duk_push_error_object
Those are defines as preprocessor macros so they wont be compiled into any object files. You either have to include duktape.h, of if you're loading the functions dynamically (which I assume is why your running into this issue) you'll have to copy over at least those macros into your project.
You should always include duktape.h: otherwise the compiler will be guessing the types involved which won't always work.
Thanks for your answers. Indeed I am using the Duktape shared library calling it from a Delphi Pascal project, so I cannot include duktape.h in the caller, or use directly the C macros.
The problem with macros is that they use filename and line number using standard C ambient variables (LINE , FILE). Of course they are inaccessible from any external program calling the shared library. So I cannot simply rewrite them inside the caller.
@gidesa This may be a duplicate to #87?
Re: __FILE__ and __LINE__ the intent is to ensure that the API provides variants with explicit file/line arguments for each such call so that the file/line "blaming" can be controlled by the application - #95. This is not yet the case however.
Yes, my question is a duplicate of #87 in some form, although it's more specific. I don't think that in shared library the macros are a serious problem in itself, because one could always rewrite them in the external language (in my case as functions/procedures in Delphi Pascal). The problem are variables internal to Duktape, or to C compiler, not exposed to external language.
Maybe there are other ways for caller to read the current file name and line of the script executed by Duktape. Or maybe future versions could add a specific API function for this task.
@gidesa Hmm. The calls you mention use a macro to automatically supply __FILE__ and __LINE__ to Duktape internals. For example duk_error() provides them so that the file/line information is recorded in the stack trace. But if a call was added to allow supplying these explicitly, say, duk_error_fileline(ctx, "myfile.xxx", 123, "connect failed"), would that solve your problem? (That's the goal of #95 and there are really only a few calls that lack explicit file/line argument variants.)
Maybe there are other ways for caller to read the current file name and line of the script executed by Duktape.
Do you mean here that when a Duktape/C function is called, you would like to find out the file/line of the calling function? There's some support for that via Duktape.act() but you're right in that there's no easy-to-use API for that specific purpose for now. Definitely an area of improvement.
Embedding it it's especially simple in Delphi, only a few lines of interfacing code and you have a complete JS interpreter ready inside your program.
Good to hear things are still working in your environment :)
Do you mean here that when a Duktape/C function is called, you would like to find out the file/line of the calling function? There's some support for that via Duktape.act() but you're right in that there's no easy-to-use API for that specific purpose for now. Definitely an area of improvement.
Yes, in some respect. Consider an ideal situation: the Duk interpreter at some point calls a function in the host environment (Delphi). Then here something goes wrong, and I would return my custom error message and code to Duk interpreter. Then Duk could decide to print the error, with message and error code and so on, adding the file name and line number, but instead could also decide to "eat" entirely the error, if a catch is present.
Actually I resolve to generate an error in Duk returning -1 from the called (host) function, so I can have the filename/line printed from standard Duk error message. But the host error message can only be directly printed to output in the called function, before it returns to caller Duk function. This is sufficient in many cases, but if a catch is present the host error message is printed anyway, because it's not managed by Duk interpreter at all.
So the duk_error macro does exactly that, pass a custom error message and code to Duk, and it decides what to do.
I have to complete the comment: if duk_error is a macro, then not exposed in shared library, one could rewrite them in the external (host) language, but is needed a real function in Duk API (exposed in library) to read the current JS file name and line number.
@gidesa Why would current JS file name and line number be needed? Note that duk_error() includes the C __FILE__ and __LINE__ information to the stack trace data - it's not related to JS file/line information. Any Ecmascript functions (functions within JS files) will already be included the stack trace data and their file/line information will be based on the bytecode executor state for each such function.
But maybe what you're after is that when you access .fileName and .lineNumber of an error object, those would reflect the Delphi call site? Having the ability to provide file/line explicitly would do that.
In some cases one would actually want to ignore the C call site file/line when assigning .fileName / .lineNumber (but they would still be included in a stack trace). See links in #458.
@svaarala Ok, I understand that __LINE__ and __FILE__ are not referenced to the JS script.
I just need the file/line of JS script, but they are just in some form in the stack trace. I have to see more on this, in the documentation.
Yes, I need the .fileName and .lineNumber directly in the called host function. For now I can read them only after the end of evaluation ( I use duk_eval_raw). If the host function return -1, Duk generate an error object with message "Unknown error" and JS filename and line. So I read them with:
duk_get_prop_string(ctx, -1, 'fileName');
duk_get_prop_string(ctx, -2, 'lineNumber');
duk_get_prop_string(ctx, -3, 'message');
Instead of the "Unknown error" message I would have the specific message from the host function. But I don't know how to pass that message to Duk evaluator.
So indeed there are two solution to my little problem:
read JS filename and number inside the host function, and print directly the complete message
or
pass an entire error message string from host function to Duk evaluator, other than returning the -1 code (better because evaluator can decide to stop or continue processing, based on try-catch blocks).
|
2025-04-01T04:35:37.633712
| 2016-12-02T00:53:41
|
193005030
|
{
"authors": [
"fatcerberus",
"svaarala"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11171",
"repo": "svaarala/duktape",
"url": "https://github.com/svaarala/duktape/pull/1128"
}
|
gharchive/pull-request
|
Add C API equivalent for Duktape.act()
[x] Rebase after #1125
[x] Add duk_inspect_callstack_entry()
[x] API documentation
[x] Website changes
[x] Testcase coverage
[x] Releases entry
Fixes #777.
One naming idea:
duk_inspect_callstack(ctx, level) and correspondingly duk_inspect_value(ctx, idx) for the Duktape.info() equivalent
Agree, "inspect" makes it clear that you're using some heavy introspection, rather than "get" which sounds more typical.
duk_inspect_value() and duk_inspect_callstack_entry() are the API call names now. Pretty verbose, but on the other hand these are very rarely called so it's good their names are descriptive.
Good to have this merged, I took advantage of it right away. There were a few places in minisphere where I did stuff like this (this being written early in development before I thought about duk_eval_string() or even duk_get_global_string()):
duk_push_global_object(ctx);
duk_get_prop_string(ctx, -1, "Duktape");
duk_get_prop_string(ctx, -1, "act");
duk_push_int(ctx, -4);
duk_call(ctx, 1);
// interpret callstack entry data
It was nice to replace all that with a single call, with the added bonus that I could choose to leave out the Duktape built-in in the future and callstack inspection would still work. Or if some game script happens to modify it destructively. That's important since I rely on it for error blaming, and error handling needs to be robust for obvious reasons.
One enhancement, if I could suggest, would be for duk_inspect_callstack_entry() to return a boolean indicating whether the callstack entry could be successfully retrieved while still leaving undefined on the valstack in that case, similar to the semantics of duk_get_prop_xxx(). For example my blaming code does this:
dukrub_inspect_callstack_entry(ctx, -1 + blame_offset);
if (!duk_is_object(g_duk, -1)) {
dukrub_inspect_callstack_entry(ctx, -1);
duk_replace(ctx, -2);
}
It would be nice if I could move the duk_inspect_callstack_entry() call itself into the if-condition, saving a line of code.
That sounds good - maybe duk_inspect_value() could have a similar return value.
@svaarala Just wanted to mention, I've been using Duktape.act() for a long time in this handy "blamed" duk_error() variant:
https://github.com/fatcerberus/minisphere/blob/master/src/engine/api.c#L293-L327
It's really useful for throwing argument validation errors from Duktape/C functions, since I can blame the error on the Ecmascript activation just below. Now that callstack inspection is a first-class citizen in Duktape, maybe this "blamed throw" call is something to consider for the public Duktape API?
Letting the caller select the "blamed" function for file/line assignment is one possibility - but I think the more robust approach is to allow functions to indicate if they're "infrastructure" (not blamed) or application (blamed).
Often a function can be called both by an application but also by another infrastructure function, so that it's difficult to know what level in the callstack to blame. For example, a target function may be called directly or via .forEach() and it would be incorrect to blame the .forEach() for the error.
|
2025-04-01T04:35:37.641580
| 2017-01-06T01:06:33
|
199100386
|
{
"authors": [
"fatcerberus",
"svaarala"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11172",
"repo": "svaarala/duktape",
"url": "https://github.com/svaarala/duktape/pull/1272"
}
|
gharchive/pull-request
|
Fix exponentiation operator behavior on Linux gcc 4.7+ -O2
For some reason exponentation operator behaves incorrectly on Linux with gcc 4.8.4 -O2 (-Os used for commit regression tests is fine). Symptom example:
duk> 2 ** 10
= 1024
duk> x = 2
= 2
duk> x ** 10
= 1 // <-- ?
duk> Math.pow(2,10)
= 1024
Exp at compile time works while runtime results are incorrect. The mysterious thing is that in both cases the internal function used to compute the exponentiation is the same. Also other arithmetic operations, handled by the same arith helper, work.
This issue is also GCC version specific. Gcc 4.6.4 -O2:
duk> x = 2
= 2
duk> x ** 10
= 1024
Gcc 4.8.4 -O2:
duk> x = 2
= 2
duk> x ** 10
= 1
So it's most likely related to some optimization decision (aliasing assumptions, union assumptions, for example) that GCC 4.7+ make with -O2.
Maybe an optimizer bug?
Could be, but at the outset a Duktape bug would be more likely IMO.
dukd compiled with gcc-4.8.4 and forced -O2 exhibits the issue, and managed to get this debug dump:
duk> x = 2
= 2
duk> x ** 10
D0 duk_js_arith.c:69 (duk_js_arith_pow): POW: 2.000000 0.000000
= 1
Disabling fastint support but otherwise the same:
duk> x = 2
= 2
duk> x ** 10
D0 duk_js_arith.c:69 (duk_js_arith_pow): POW: 2.000000 10.000000
= 1024
So the issue may be fastint related. Indeed, adding back fastint support and ensuring that the exponent is not a fastint:
duk> z = 0.1
= 0.1
duk> z = 10
= 10
duk> x ** z
D0 duk_js_arith.c:69 (duk_js_arith_pow): POW: 2.000000 0.000000
= 1
duk> x = 2
= 2
duk> y = 10
= 10
duk> x ** y
D0 duk_js_arith.c:69 (duk_js_arith_pow): POW: 2.000000 0.000000
= 1
duk> y += 0.1; y -= 0.1 // <-- remove fastint status
= 10
duk> x ** y
D0 duk_js_arith.c:69 (duk_js_arith_pow): POW: 2.000000 10.000000
= 1024
So, the issue is most likely in converting a fastint into an IEEE double which does use union tricks. No idea why it would manifest here but not in other arithmetic though.
I think I found the issue. The executor arith helper has a fastpath for both arguments being fastints. That fastpath handles SUB, MUL, DIV, MOD (ADD has a separate helper) but not EXP. The case clause for the fastint check has a default clause with DUK_UNREACHABLE() which is incorrect because the default clause is reached with EXP. The case is concretely:
default:
DUK_UNREACHABLE():
goto skip_fastint;
Because DUK_UNREACHABLE() compiles to __builtin_unreachable(), the compiler is justified in not making that code path work. Interestingly it only affects -O2 - but that's par for the course for this kind of bugs.
Removing the DUK_UNREACHABLE() fixes the issue.
Running the test set manually nothing else fails with gcc 4.8.4 -O2 after this fix.
This was my bad, I saw that the default case jumped to skip_fastint but somehow missed the DUK_UNREACHABLE() hint... :)
These are pretty hard to catch because they only manifest in only certain conditions (like compiler versions, register pressure, whatever). Here also if I added debug prints the problem went away because the debug prints affected register handling.
Realistically this could have been caught with an updater matrix compile test that would run testcases with different compiler versions, different compiler options, and different Duktape options. I had that tooling for the Duktape 1.x config model but not yet for the Duktape 2.x config. It also needs a bit more testing hardware, and I do have 4 dedicated physical hosts now for continuous testing.
|
2025-04-01T04:35:37.743660
| 2023-03-20T13:58:49
|
1632173109
|
{
"authors": [
"skovalyova"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11173",
"repo": "sveinungf/spreadcheetah-samples",
"url": "https://github.com/sveinungf/spreadcheetah-samples/issues/1"
}
|
gharchive/issue
|
Spreadsheets are empty when opening in "Spreadsheet compare" Office tool
Hello! First of all, thank you for the amazing tooling, metrics are really great and so performant. I'm able to generate large xlsx files using .NET.
However, I have faced the issue when trying to compare the file generated with SpreadCheetah, with other xlsx files using the Spreadsheet Compare 2016 tool (part of Office). The file generated using SpreadCheetah is treated as empty for some reason by this tool, so I can't compare it with others. At the same time, the file in Excel looks good and has the data. I had no similar issue when opening files created using NPOI package.
So, is it possible to make the SpreadCheetah-generated xlsx files compatible with Spreadsheet Compare?
This is how the issue looks like in Spreadsheet Compare (left file - SpreadCheetah, right file - created in Excel):
This is how it looks like in Excel:
Minimal code:
using var tempStream = new MemoryStream();
await using var spreadsheet = await Spreadsheet.CreateNewAsync(tempStream, null, cancellationToken);
await spreadsheet.StartWorksheetAsync("Report", null, cancellationToken);
var columnNames = new List<string> { "column1", "column2" };
var values = new List<string> { "value1", "value2" };
var headerRow = new List<Cell>();
foreach (var columnName in columnNames)
{
headerRow.Add(new Cell(columnName));
}
await spreadsheet.AddRowAsync(headerRow, cancellationToken);
var valuesRow = new List<Cell>();
foreach (var value in values)
{
valuesRow.Add(new Cell(value));
}
await spreadsheet.AddRowAsync(valuesRow, cancellationToken);
await spreadsheet.FinishAsync(cancellationToken);
Environment:
Spreadcheetah 1.9.0
.NET 6
Sorry I accidentally created it here, in samples repo. Moved the issue to the main repo: https://github.com/sveinungf/spreadcheetah/issues/16
|
2025-04-01T04:35:37.745185
| 2020-07-09T17:36:30
|
654233494
|
{
"authors": [
"stephane-vanraes"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11174",
"repo": "svelte-society/site",
"url": "https://github.com/svelte-society/site/pull/32"
}
|
gharchive/pull-request
| |
2025-04-01T04:35:37.783471
| 2024-02-17T19:06:36
|
2140471137
|
{
"authors": [
"andresgutgon"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11175",
"repo": "sveltejs/kit",
"url": "https://github.com/sveltejs/kit/issues/11866"
}
|
gharchive/issue
|
Convoluted way of using @svelte/package inside a monorepo
Describe the problem
👋 Hi, how are you? 😃 I hope everything is all right.
We are developing a monorepo with pnpm workspaces and Turbo repo.
This is our monorepo file structure
/
└── fun-with-monorepo
├── apps
│ └── server (SvelteKit main app
└── packages
└── svelte-ui
NOTE: apps/server and packages/svelte-ui both are SvelteKit apps.
We have several goals with @fun-with-monorepo/svelte-ui package
We want to have a great dev experience in the monorepo by not having a build
step inside this package and let our main app in ./apps/server compile the
.svelte and .ts files in this package. Way faster, feels like developing
in the app.
We want to have the easiest Storybook setup. And that's by being this package
another SveltKit app.
We want to use the official way in Svelte to author npm packages. And that's by
using their packaging npm module called @svelte/package.
We want our package published in npmjs.com
So as you can points (1) and (2) are about our dev experience as monorepo
developers. And points (3) and (4) are about shipping this package as an npmjs
UI library other people can use.
The problem?
Right, the problem. Here starts the funny part. Before 2023 svelte-package was
modifing your package.json and Svelte maintainers felt (I think) that was not a
good idea (and I agree). Rich Harris can explain better than I do. But long story short we want 2 different things while we're developing this package inside the monorepo and when we build it for others to use.
The difference resides in the package.json exports field. For comparison
here is what we want in dev and production build:
Development
// package.json
"exports": {
".": {
"types": "./src/lib/index.ts",
"import": "./src/lib/index.ts"
}
}
Production
// package.json
"exports": {
".": {
"types": "./dist/index.d.ts",
"import": "./dist/index.js"
}
}
As you can see in dev we point directly to the .src/lib source code because we
know our Storybook or our other SveltKit app (the main app) in ./apps/server are doing
the work of transpiling all the *.{ts,svelte} files. Nice!
On the other hand on production we want to point exports field in
package.json to ./dist folder that was produced by svelte-package.
Describe the proposed solution
No solution, at least nothing I can think that can be applied. But I'm here to listen to alternatives. : )
Alternatives considered
Weeell, as always with code is all about trade-offs. We would't had this problem
if this package was not a SvelteKit app and we didn't use svelte-package. We
could do our own rollup plugin and ship it this way. But I think that by staying with the official
way we're more aligned with svelte ecosystem.
Sooo, what's the solution?
The solution is to put in git the development version of exports pointing to ./src/lib so this way dev experience is fantastic. And when we do the building we modify on runtime the package.json with a prebuid script that you can see in this same folder. This script is used in the scripts field in the package.json like this:
// package.json
"scripts": {
"build:vite": "vite build",
"prebuild": "node ./scripts/hack-svelte-package/prebuild.js",
"buildAndPackage": "pnpm run build:vite && pnpm run package",
"build": "pnpm run prebuild && pnpm run buildAndPackage",
}
All is perfect for one little thing. After we run pnpm build our package.json gets modified
"./*": {
- "types": "./src/lib/ui/*/index.ts",
- "import": "./src/lib/ui/*/index.ts",
- "svelte": "./src/lib/ui/*/index.svelte"
+ "types": "./dist/ui/*/index.d.ts",
+ "svelte": "./dist/ui/*/index.js"
}
And this is bad, very bad. So the solution for this is to check if exports
field contains ./dist in it. It never should. We acomplish that with a custom
eslint rule you can find in ./eslint-local-rules.cjs
I hope you enjoyed as much as I did this little fairy tale 😹
Importance
would make my life easier
Additional Information
I'm not sure if the goal of svelte-package is to place nice with monorepos. I'm not asking for an official solution. I'm just adding here my 2cents to see if someone have a better solution of if this solution can help others.
We ended for now running the build command on watch mode. I think is not ideal but it works
|
2025-04-01T04:35:37.799311
| 2024-02-24T18:58:40
|
2152444928
|
{
"authors": [
"HalfdanJ",
"eltigerchino",
"konstantinov90"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11176",
"repo": "sveltejs/kit",
"url": "https://github.com/sveltejs/kit/issues/11896"
}
|
gharchive/issue
|
redirect after invalidate creates unwanted record in history
Describe the bug
Hi!
There is a cryptic bug in svelte kit routing, which goes as follows
Say you have an app behind authorization. Whenever an unauthorized user tries to access the app, he is redirected to the login page. And likewise, whenever authorized user tries to access login page (for example from browser history, or with redirect param) he is forcefully redirected to the main page or another page according to redirect param. This logic should work both on client and server regardless
Such behaviour could be implemented in many different ways, but I find it natural to perform redirects via load functions and invalidation on the client, so that all the redirection rules are stored in one place (load functions)
Having an app with such routes structure, lets define scenarios, starting with welcome screen and describing expected browser history
/ (welcome screen, no auth)
/signin (login page)
/main (behind auth)
Authorized client
-> / (link to /main) -> /main --- history (1 /, 2 /main)
-> / (link to /signin?redirect=/main) -> /main --- history (1 /, 2 /main)
-> /signin?redirect=/main -> /main --- history (1 /main) (server side redirect with 302)
-> /main --- history (1 /main)
Unauthorized client
-> / (link to /signin?redirect=/main) -> /signin?redirect=/main --- history (1 /, 2 /signin)
-> / (link to /signin?redirect=/main) -> /signin?redirect=/main -> (client inputs password) -> /main --- history (1 /, 2 /main) STATE REPLACED BUG HERE
-> / (link to /main) -> /signin?redirect=/main --- history (1 /, 2 /signin)
-> / (link to /main) -> /signin?redirect=/main -> (client inputs password) -> /main --- history (1 /, 2 /main) STATE REPLACED BUG HERE
Thus a rule is described - authorized client should not have /signin page in his history, because this page is inaccessible at all when authorized but it redirects to /main page instead. If this rule is broken and /signin page is present inside browser history, should a client click "go back" button in his browser he becomes stuck in redirect loop - /main -> /signin -> /main
Returning finally to the bug at hand - in described circumstances such a redirect loop indeed does emerge, in the place which is marked with BUG HERE - route state is not replaced, when navigated via invalidate
This bug is kind of hard to describe in text, so please consider trying a repro
PS
Reproduction
I made a branch with a solution to this bug
https://github.com/sveltejs/kit/pull/11895
reproduction scenario can be found in the basics test app (/redirect/app-with-auth)
please let me know, if I should provide any other form of reproduction
Logs
No response
System Info
System:
OS: macOS 14.2.1
CPU: (8) arm64 Apple M1
Memory: 243.41 MB / 16.00 GB
Shell: 5.9 - /bin/zsh
Binaries:
Node: 20.11.0 - /usr/local/bin/node
Yarn: 1.22.17 - /opt/homebrew/bin/yarn
npm: 8.4.1 - /opt/homebrew/bin/npm
pnpm: 8.15.3 - /opt/homebrew/bin/pnpm
bun: 1.0.0 - ~/.bun/bin/bun
Browsers:
Brave Browser: <IP_ADDRESS>
Chrome: 113.0.5672.92
Safari: 17.2.1
Severity
serious, but I can work around it
Additional Information
No response
I can't reproduce the issue based on the test from the PR. Please provide a minimal reproduction separately.
I just hit this same issue of redirects creating an infinite loop of redirects for a login page, specifically because we have SSR disabled, so the redirect doesnt cause a true 307 or 302 response, but is instead redirected clientside.
I think the issue here is that a true server side redirect would not be recorded in the history, so sveltekit should imitate the same behavior when doing client side redirects based on data from +page.server.ts
so sveltekit should imitate the same behavior when doing client side redirects
That is exactly what I had in mind! Unfortunately I haven't had the opportunity to get back to this issue at the time, maybe you could help with the repro @eltigerchino
Oops, I meant to tag @HalfdanJ
-> / (link to /signin?redirect=/main) -> /signin?redirect=/main -> (client inputs password) -> /main --- history (1 /, 2 /main) STATE REPLACED BUG HERE
Shouldn't this have three history entries (one for each page)? And I think it does at the moment.
@eltigerchino in the PR it works correctly because I fixed it, you should run the test on the main branch, then you would see the redirect loop
-> / (link to /signin?redirect=/main) -> /signin?redirect=/main -> (client inputs password) -> /main --- history (1 /, 2 /main) STATE REPLACED BUG HERE
Shouldn't this have three history entries (one for each page)? And I think it does at the moment.
No, it does not in my opinion, because the "go back button" makes you stuck between main and signin, which is not the case with the server redirects
Ok, I will have to rework this PR
I suggest we close it for now, seeing that this PR has got minimal attention from community
|
2025-04-01T04:35:37.811133
| 2024-06-15T11:18:45
|
2354761578
|
{
"authors": [
"Conduitry",
"eltigerchino",
"jacobbogers"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11177",
"repo": "sveltejs/kit",
"url": "https://github.com/sveltejs/kit/issues/12358"
}
|
gharchive/issue
|
implement "Upgrade" as a function in a api route to support websocket
Describe the problem
One can easy implement GET, PUT in an api route by a function (see below)
import { json } from '@sveltejs/kit';
export function GET() {
const number = Math.floor(Math.random() * 6) + 1;
return json(number);
}
Describe the proposed solution
If we implement UPGRADE Protocol_upgrade_mechanism we could implement websockets aswell
this file is in src/route/chat-app/server.js connect with new WebSocket('/chap-app')
// if this file is in src/route/chat-app/server.js
import { json } from '@sveltejs/kit';
// called when http call wants to upgrade to websocket protocol
export function UPGRADE() {
// this function will be called after successfull upgrade
return function connection(ws: WebSocket) {
ws.send('hello');
ws.onmessage = (event) => {..});
// other handlers
}
}
### Alternatives considered
_No response_
### Importance
would make my life easier
### Additional Information
Currently, customers expect some kind of live real time interaction, a clean solution integrating with sveltekit would be a huge advantage
Would the fallback method handler fit your needs? https://kit.svelte.dev/docs/routing#server-fallback-method-handler . Our motivation for adding this was so that sveltekit wouldn't need to check for all methods (we only focus on the most commonly used ones)
It wouldn't really help, no. Web Sockets are a two-way stream, and you can't make them fit into a Request/Response framework, which is what you'd still have to do with the fallback method handler.
This is basically #1491. It's a hard problem to come up with an API for, and we don't have a solution for it.
Would a conditional in the fallback method handler fit your needs? https://kit.svelte.dev/docs/routing#server-fallback-method-handler . Our motivation for adding this was so that sveltekit wouldn't need to check an indefinite list of named exports in the +server file (we only focus on the most commonly used ones)
That is not how websocket works (read the link in the OP), the UPGRADE request is not a HTTP method, I suggest this handler name because to me it fits more how the kit uses api routes.
It wouldn't really help, no. Web Sockets are a two-way stream, and you can't make them fit into a Request/Response framework, which is what you'd still have to do with the fallback method handler.
actually websockets are designed to piggy back on request/response framework, how we expose things to the developer is either we create a specific context per websocket connection and a singler handler function or a function factory that creates a PERMANENT lived function context for the duration of the websocket connection (my proposed solution)
Fallback would not work because a websocket initial request is just a GET (with some special headers). and normal GET handlers have precedence in sveltekit. the only thing skit should not do is physically close the socket after the request.
I looked at kit code and you can add specific handler names if you want (ofc you need to change skit code for this).
So basically
if kit receives GET + websocket upgrade headerss -> call Upgrade handler
if kit receives GET (no specific websocket headers) -> call the usual GET handler
Thanks for all the information on websockets, especially since I don't have much understanding of it yet.
Reading through the main websocket issue, it seems there are a few workarounds the community has come up with, such as this one that also handles upgrading the GET requests.
Is it a good idea for the framework to handle these upgrades and for the developer to simply export a websocket function to handle these connections? But, reading the MDN docs on this it seems like more control is needed depending on the raw request or just the Upgrade header?
Might be a good idea to move this discussion to https://github.com/sveltejs/kit/issues/1491 so that we can collect these ideas and brainstorm an API for this.
But, reading the MDN docs on this it seems like more control is needed depending on the raw request or just the Upgrade header?
No the upgrade header is what makes this piggy back on existing http transports (remember the billions of routers and proxies that dont need to be configured to handle websockets). other then this you enter the websocket space and ofc you have full duplex control mechancism and framing protocol you can read about full websocket here rfc6455
From Server perspective you need to:
initiate sending data to client at any time
close connection
receive data from client
From Client perspective:
create connection
send data
receive data from server
close connection
there is no crazy magic going on.
|
2025-04-01T04:35:37.822333
| 2023-02-06T17:03:12
|
1572941335
|
{
"authors": [
"Rich-Harris",
"dummdidumm",
"n00ki"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11178",
"repo": "sveltejs/kit",
"url": "https://github.com/sveltejs/kit/issues/8909"
}
|
gharchive/issue
|
Form Resets after Second Submit
Describe the bug
I'm noticing a strange behavior on subsequent submits when using progressively enhanced form actions.
From the second form submission onwards, when the form elements' values are bound to the ActionResult data - the form resets all values (even if values were returned from the action successfully and exist in the PageData store).
In the attached video, you'll be able to see the responses to the first and second form submissions (passwords do not match in both cases). as email & password values are returned from the action, they remain in the form after calling form.reset()& update(). On the second submission onwards, though, they disappear.
Someone suggested that the reason for this behavior is the fact that SK doesn't update the form fields when the values don't change (the same actual values are returned from the server).
I want the form to reset and then get populated by data coming from the server, as it does on the first failed submission. It feels like the right flow to me (as it mimics the browser req-res default behavior).
I'm probably doing something wrong, yet beyond merely solving the issue I'd like to get a better understanding of the internal mechanism and flow.
https://user-images.githubusercontent.com/42724748/217035819-b3735fce-69ef-4420-89d6-9cb7a6c04355.mov
Reproduction
https://gist.github.com/n00ki/ed50a7fdc6a96c077f3f32f8f8268557
Logs
No response
System Info
System:
OS: macOS 13.0
CPU: (12) x64 Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz
Memory: 115.29 MB / 16.00 GB
Shell: 5.8.1 - /bin/zsh
Binaries:
Node: 18.12.1 - /usr/local/bin/node
npm: 8.19.2 - /usr/local/bin/npm
Browsers:
Chrome: 109.0.5414.119
Firefox: 107.0.1
Firefox Developer Edition: 109.0
Safari: 16.1
npmPackages:
@sveltejs/adapter-auto: ^1.0.2 => 1.0.2
@sveltejs/kit: ^1.3.9 => 1.3.9
svelte: ^3.55.1 => 3.55.1
vite: ^4.0.4 => 4.0.4
Severity
annoyance
Additional Information
No response
Probably a duplicate of #8513
Probably a duplicate of #8513
Probably.
Can we expect the suggested fix to be submitted any time soon?
definitely a dupe — closing. it'll be fixed when someone fixes it :)
|
2025-04-01T04:35:37.826851
| 2022-01-20T17:12:35
|
1109560127
|
{
"authors": [
"Rich-Harris",
"ignatiusmb"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11179",
"repo": "sveltejs/kit",
"url": "https://github.com/sveltejs/kit/pull/3454"
}
|
gharchive/pull-request
|
Update 01-routing.md
Supersedes #3453
Please don't delete this checklist! Before submitting the PR, please make sure you do the following:
[ ] It's really useful if your PR references an issue where it is discussed ahead of time. In many cases, features are absent for a reason. For large changes, please create an RFC: https://github.com/sveltejs/rfcs
[ ] This message body should clearly illustrate what problems it solves.
[ ] Ideally, include a test that fails without this PR but passes with it.
Tests
[ ] Run the tests with pnpm test and lint the project with pnpm lint and pnpm check
Changesets
[ ] If your PR makes a change that should be noted in one or more packages' changelogs, generate a changeset by running pnpx changeset and following the prompts. All changesets should be patch until SvelteKit 1.0
A bit late, but this is also covered in #3455
|
2025-04-01T04:35:37.834914
| 2022-08-02T20:01:22
|
1326341157
|
{
"authors": [
"Rich-Harris",
"chvanam",
"j4w8n",
"theetherGit"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11180",
"repo": "sveltejs/kit",
"url": "https://github.com/sveltejs/kit/pull/5792"
}
|
gharchive/pull-request
|
Added cli for adding new route 'svelte-kit add route --name svelte' cli in svelte-kit
I have added svelte-kit add route --name svelte for future changes as @Rich-Harris mentioned in discussion 5748. We going to need a cli to create routes so we don't have to add three files again and again.
Example: svelte-kit add route --name routeName
routeName is your routes name that you wanna create.
---
PR Requirements:
[x] It's not a discussed issue or feature when i saw discussion 5748, I wanted to have a cli for creating routes without any issues.
[x] This might not solve a bigger problem but it's going to ease our issue related to adding three files over and over in every route so why not use cli for that.
Tests
[x] Run the tests with pnpm test and lint the project with pnpm lint and pnpm check
Changesets
Added a new cli command in svelte-kit cli.
This can be released after discussion 5748 is completed.
Would it be useful if the created route files contained some boilerplate functionnalities ?
For example a GET handler in +page.server.js showing how to get prop data in +page.svelte with some comment about other handlers or an example load function in +page.js 🤔
The future documentation itself will undoubtly make page creations straight forward, just opening the discussion here.
@chvanam it is possible. I wanted this to be approved first so in future i can add more things to cli and boilerplate code too.
For what it's worth, I deployed an npm package that does this.
Has some boilerplate code for the files @chvanam
Supports typescript
Has a config file, so you can set defaults like creating .ts files instead of .js, whether you want the boilerplate code (I called it 'codekit'), and if the default route is a page or server.
You can't create named layouts with it yet, but there's an option to say that a page route should use a named layout.
https://www.npmjs.com/package/mkrt
@j4w8n wow man that's a perfect module. I was already working on those additions and i got more ideas from your work.
Hey — thank you for the PR, but this is really something that needs an issue first. It's probably not something we're going to add, but if it's a very popular feature request then we would need to figure out the answers to a ton of questions (what about layouts? what about +server.js files? do we automatically add all three page files even though most routes will only need two or one? what methods do we create handlers for? is typescript auto-detected? does it work relative to config.kit.files.routes or the current working directory? etc etc etc) before attempting an implementation.
Realistically, answering all those questions would leave us with a process that wasn't really any easier than just creating the files manually.
Thank You for your reply, I already had some of the questions before starting it, like we always not going to need all 3 files and relative to config.kit.files.routes or the current working directory. I was thinking about them and have some notes but that is for some other time. And for other queries i already have layout or code on my local. Again thank you so much.
|
2025-04-01T04:35:37.912794
| 2023-02-24T17:22:10
|
1599026010
|
{
"authors": [
"BeeMargarida",
"masagatech"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11181",
"repo": "svelteuidev/svelteui",
"url": "https://github.com/svelteuidev/svelteui/pull/323"
}
|
gharchive/pull-request
|
Component/drawer
Also, you want, you can implement your changes on top of the main branch. We are discontinuing the next-minor branch and starting to work on top of main.
I'm gonna close this, since it does not have the code for Drawer component, if you could then open another PR I would greatly appreciate!
|
2025-04-01T04:35:37.919842
| 2024-01-31T06:24:34
|
2109376456
|
{
"authors": [
"svenjacobs"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11182",
"repo": "svenjacobs/reveal",
"url": "https://github.com/svenjacobs/reveal/issues/98"
}
|
gharchive/issue
|
Add Metalava to project
https://android.googlesource.com/platform/tools/metalava/
@Goooler I think this is obsolete due to your PR. What do you think?
|
2025-04-01T04:35:37.930320
| 2019-05-25T01:15:45
|
448397177
|
{
"authors": [
"mesqueeb",
"strarsis"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11183",
"repo": "svg/svgo",
"url": "https://github.com/svg/svgo/issues/1111"
}
|
gharchive/issue
|
Replace with hard coded
Is there an option to, whenever <use> appears, and the <def> is available, to replace that <use> with the actual content?
Related?: https://github.com/svg/svgo/issues/563
Yeaaap. I don’t need checking if the def is used once or multiple times though.
Just the feature to change all <use> with the actual def contents inline.
Sent with GitHawk
PR ready: https://github.com/svg/svgo/pull/1279
No recursion yet, but svgo multipass feature should handle this.
|
2025-04-01T04:35:37.941125
| 2017-11-22T20:13:07
|
276192164
|
{
"authors": [
"GreLI",
"alexjlockwood"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11184",
"repo": "svg/svgo",
"url": "https://github.com/svg/svgo/issues/851"
}
|
gharchive/issue
|
Confusion about a couple JsApi methods/documentation
I have been analyzing the way svgo works for the past few days... and I have noticed some inconsistencies with the documentation for a few methods.
I want to clarify how each method is intended to behave before I submit a pull request to update the documentation.
Question 1
The hasAttr method takes a string as its first argument: https://github.com/svg/svgo/blob/ceccf1ffdcdcba0f422a49685c2f60520579e46a/lib/svgo/jsAPI.js#L134-L152
However, the removeAttr method calls hasAttr and in some cases passes an array as its first argument: https://github.com/svg/svgo/blob/ceccf1ffdcdcba0f422a49685c2f60520579e46a/lib/svgo/jsAPI.js#L235-L258
So my question is which of the two methods should be updated? Should the hasAttr documentation be updated, saying that it can take an array as an argument as well? Or should the removeAttr method be updated so that it never passes an array to the hasAttr method?
Question 2
The documentation for the computedAttr method says that it should return an object or undefined, which makes sense: https://github.com/svg/svgo/blob/ceccf1ffdcdcba0f422a49685c2f60520579e46a/lib/svgo/jsAPI.js#L215-L233
However, as you can see on line 128, it is possible for the method to return elem.hasAttr(name, val), which is a boolean: https://github.com/svg/svgo/blob/ceccf1ffdcdcba0f422a49685c2f60520579e46a/lib/svgo/jsAPI.js#L228
Should the documentation be updated to indicate that a boolean value can be returned as well? Or should the method be updated so that it is not possible for a boolean value to be returned?
hasAttr is just a method to check for a certain attribute. removeAttr was introduces for convertShapeToPath, and it's just a convenient way to remove a bunch of attributes in one line. Second parameter was added just for the sake of completeness. I don't think it's much use for an array in hasAttr.
computedAttr is used to check for an attribute existence or a certain value, so boolean is a valid response, need to update the description.
So to clarify, should it be possible to pass an array to hasAttr? Because currently it is possible for an array to be passed as an argument.
I don't think it worth it. Just can't think even a case where it could be useful. Unlike removeAttr, where removing a bunch of attrs is one of the primal cases.
What do you mean by “it's possible for an array to be passed”?
Right now if you call removeAttr(name) (where name is an array of attr names), the method will end up calling this.hasAttr(name) on line 248, right?
Ah, yeah, it should have return earlier.
OK, great... that is kinda what I thought but wasn't sure. :)
I'll file a pull request to fix that.
Actually wait... I have one more question...
On line 246 you call name.forEach(this.removeAttr, this): https://github.com/svg/svgo/blob/ceccf1ffdcdcba0f422a49685c2f60520579e46a/lib/svgo/jsAPI.js#L246
According to the documentation for Array#forEach, it takes a function callback(currentValue, index, array) as its first argument.
On line 246, you pass this.removeAttr, which doesn't seem right (because the second argument for this.removeAttr is a string, but the second argument for the callback passed to Array#forEach is an index number.
Is this also a mistake?
Yep, that's why on line 250 it checks for the 3rd parameter recursive—it would be an array in case of forEach.
OK, thanks.
I created a pull request here with what I assume should be the fix... LMK if I should change anything though: https://github.com/svg/svgo/pull/855
|
2025-04-01T04:35:37.948434
| 2017-03-23T09:06:20
|
216355261
|
{
"authors": [
"DamienCassou",
"sviridov"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11185",
"repo": "sviridov/undercover.el",
"url": "https://github.com/sviridov/undercover.el/issues/26"
}
|
gharchive/issue
|
undercover-file-handler never called
Hi,
I'm setting up undercover in https://github.com/DamienCassou/hierarchy/pull/9. As you can see in the log, undercover can't find any coverage info:
UNDERCOVER: No coverage information. Make sure that your files are not compiled?
What am I doing wrong please?
I don't have anything compiled file (*.elc).
I reproduced the problem locally by:
deleting undercover.elc
adding message everywhere in its source code (poor-man debugger)
forcing the coverage with (setq undercover-force-coverage t)
What I see is that at the end of undercover--edebug-files, the file-name-handler-alist variable value is:
((/\(?:hierarchy\.el\)$ . undercover-file-handler) (\.gpg\(~\|\.~[0-9]+~\)?\' . epa-file-handler) (\(?:\.dz\|\.txz\|\.xz\|\.lzma\|\.lz\|\.g?z\|\.\(?:tgz\|svgz\|sifz\)\|\.tbz2?\|\.bz2\|\.Z\)\(?:~\|\.~[-[:alnum:]:#@^._]+\(?:~[[:digit:]]+\)?~\)?\' . jka-compr-handler) (\`/[^/]*\' . tramp-completion-file-name-handler) (\`/[^/|:][^/|]*: . tramp-autoload-file-name-handler) (\`/: . file-name-non-special))
but, undercover-file-handler never prints the message I put there.
Here are the changes I applied to undercover.el.
--- .cask/25.2/elpa/undercover-20161114.819/undercover.el 2017-03-23 10:13:50.820551685 +0100
+++ .cask.bak/25.2/elpa/undercover-20161114.819.bak/undercover.messages.el 2017-03-23 10:09:42.207369720 +0100
@@ -89,6 +89,7 @@
(defun undercover--load-file-handler (file)
"Handle `load' FILE operation."
+ (message "Loading file: %s" file)
(let ((edebug-all-defs (undercover--coverage-enabled-p))
(load-file-name (file-truename file))
(load-in-progress t))
@@ -104,6 +105,7 @@
(defun undercover-file-handler (operation &rest args)
"Handle `load' OPERATION. Ignore all ARGS except first."
+ (message "undercover-file-handler %s %s" operation args)
(if (eq 'load operation)
(condition-case nil
(undercover--load-file-handler (car args))
@@ -114,9 +116,11 @@
(defun undercover--edebug-files (files)
"Use `edebug' package to instrument all macros and functions in FILES."
+ (message "undercover--edebug-files: %s" files)
(when files
(let ((regexp (->> files (regexp-opt) (format "/%s$"))))
- (add-to-list 'file-name-handler-alist (cons regexp 'undercover-file-handler)))))
+ (add-to-list 'file-name-handler-alist (cons regexp 'undercover-file-handler))
+ (message "undercover--edebug-files: %s" file-name-handler-alist))))
(setf (symbol-function 'undercover--stop-point-before)
(lambda (before-index)
@@ -400,10 +404,12 @@
(defun undercover-safe-report ()
"Version of `undercover-report' that ignore errors."
(ignore-errors
+ (message "Reporting...")
(undercover-report)))
(defun undercover-report-on-kill ()
"Add `undercover-safe-report' to `kill-emacs-hook'."
+ (message "Adding kill hook")
(add-hook 'kill-emacs-hook 'undercover-safe-report))
;;; Main functions:
@@ -462,10 +469,12 @@
If running under Travic CI automatically generate report
on `kill-emacs' and send it to coveralls.io."
- `(undercover--setup
- (list
- ,@(--map (if (atom it) it `(list ,@it))
- configuration))))
+ `(progn
+ (message "Measuring coverage...")
+ (undercover--setup
+ (list
+ ,@(--map (if (atom it) it `(list ,@it))
+ configuration)))))
(provide 'undercover)
;;; undercover.el ends here
The output is:
$ cask exec ert-runner
Measuring coverage...
Adding kill hook
undercover--edebug-files: (hierarchy.el)
undercover--edebug-files: ((/\(?:hierarchy\.el\)$ . undercover-file-handler) (\.gpg\(~\|\.~[0-9]+~\)?\' . epa-file-handler) (\(?:\.dz\|\.txz\|\.xz\|\.lzma\|\.lz\|\.g?z\|\.\(?:tgz\|svgz\|sifz\)\|\.tbz2?\|\.bz2\|\.Z\)\(?:~\|\.~[-[:alnum:]:#@^._]+\(?:~[[:digit:]]+\)?~\)?\' . jka-compr-handler) (\`/[^/]*\' . tramp-completion-file-name-handler) (\`/[^/|:][^/|]*: . tramp-autoload-file-name-handler) (\`/: . file-name-non-special))
............................
Ran 28 tests in 0.015 seconds
Reporting...
Report files: nil
UNDERCOVER: No coverage information. Make sure that your files are not compiled?
Hi @DamienCassou, this bug happens because of your .ert-runner config. You load hierarchy.el in it.
So when you run undercover this file is already loaded and (require 'hierarchy) does nothing.
Replace .ert-runner content with -L . and everything will work.
Btw, you can test undercover locally by using TRAVIS=true in your terminal
Sure, feel free to do it! :)
done
|
2025-04-01T04:35:37.959144
| 2020-03-31T09:02:14
|
590926470
|
{
"authors": [
"sisou",
"svub"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11186",
"repo": "svub/nimiq-voting-app",
"url": "https://github.com/svub/nimiq-voting-app/pull/1"
}
|
gharchive/pull-request
|
Fix production problem by using Nimiq from CDN
Use the Nimiq object as it's supposed to and update the CDN to use JSDelivr, as the Nimiq CDN is blocked from third-party websites (The CDN will work fine if the app gets deployed on a nimiq.com domain).
Nice. :)
|
2025-04-01T04:35:37.987854
| 2024-04-04T00:43:08
|
2224181797
|
{
"authors": [
"sullis"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11187",
"repo": "swagger-api/swagger-parser",
"url": "https://github.com/swagger-api/swagger-parser/pull/2076"
}
|
gharchive/pull-request
|
maven-surefire-plugin 3.2.5
https://maven.apache.org/surefire/maven-surefire-plugin/
ready for review
|
2025-04-01T04:35:37.989299
| 2017-07-14T16:20:30
|
243044232
|
{
"authors": [
"fehguy",
"mmosemmle"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11188",
"repo": "swagger-api/swagger-parser",
"url": "https://github.com/swagger-api/swagger-parser/pull/481"
}
|
gharchive/pull-request
|
parser change.
Issue found by running custom written query on lgtm.com (as discussed by email with @fehguy: https://lgtm.com/query/2023830455/project:24760076/lang:java/
Many thanks! Tests look good.
|
2025-04-01T04:35:38.006074
| 2019-05-07T12:21:09
|
441202126
|
{
"authors": [
"Nerzal",
"lsattem",
"ubogdan",
"zimpy"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11189",
"repo": "swaggo/swag",
"url": "https://github.com/swaggo/swag/issues/391"
}
|
gharchive/issue
|
External objects don't work for @Success {object}?
Describe the bug
Added //Success 200 {object} calendar.Event to an endpoint.
The calendar.Event object exists in the "google.golang.org/api/calendar/v3" package.
I get the error can not parse response comment "200 {object} calendar.Event "event was updated""
To Reproduce
1: Create an endpoint with swagger doc.
2: Set return type for success to "calendar.Event" from "google.golang.org/api/calendar/v3".
2: Run swagger generation
Expected behavior
I expect no error to be thrown, and the calendar.Event object to be described in the swagger.json file.
Your swag version
v1.5.0
Your go version
v1.12.2
Desktop (please complete the following information):
OS: ubuntu 18.10
Browser: N/A
Version: N/A
I also need this feature.
Any hint's on how to tell the ast to parse external packages?
have you tried
import (
calendar "google.golang.org/api/calendar/v3"
)
//Success 200 {object} calendar.Event
func handler() {
}
Don't forget to do use --parseDependency when you do swag init. And also --parseVendor if you use a vendor folder
|
2025-04-01T04:35:38.012452
| 2024-05-14T04:55:21
|
2294402825
|
{
"authors": [
"hasmithagunda123",
"swapnilsparsh"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11190",
"repo": "swapnilsparsh/30DaysOfJavaScript",
"url": "https://github.com/swapnilsparsh/30DaysOfJavaScript/issues/1332"
}
|
gharchive/issue
|
[DOCS] add installation and requirements
Description
assign it under gssoc
Screenshots
No response
@hasmithagunda123 Thank you for providing your insight but it is not required as of now.
|
2025-04-01T04:35:38.033811
| 2024-07-02T23:42:36
|
2387346826
|
{
"authors": [
"CLAassistant",
"dp-152"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11191",
"repo": "swc-project/plugins",
"url": "https://github.com/swc-project/plugins/pull/324"
}
|
gharchive/pull-request
|
Feature/missing import export types
Closes https://github.com/swc-project/plugins/issues/323
This pull request adds support for namespace exports, and default and namespace imports, which were previously not supported by the plugin.
Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.You have signed the CLA already but the status is still pending? Let us recheck it.
|
2025-04-01T04:35:38.045038
| 2022-11-12T16:15:39
|
1446510609
|
{
"authors": [
"JeffGuKang",
"coadtan",
"devcloser",
"kdy1"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11192",
"repo": "swc-project/swc",
"url": "https://github.com/swc-project/swc/issues/6414"
}
|
gharchive/issue
|
[1.3.15] File extension is added as ts after compilation and makes error Cannot find module
Describe the bug
File extension is added as ts after compilation.
Expectation
const _app = _interopRequireDefault(require("./app"));
Result<EMAIL_ADDRESS>const _app = _interopRequireDefault(require("./app.ts"));
Error: Cannot find module './app.ts'
Input code
No response
Config
{
"$schema": "http://json.schemastore.org/swcrc",
"jsc": {
"parser": {
"syntax": "typescript",
"tsx": false,
"dynamicImport": true,
"decorators": true
},
"transform": {
"legacyDecorator": true,
"decoratorMetadata": true
},
"target": "es2020",
"externalHelpers": false,
"keepClassNames": true,
"loose": false,
"minify": {
"compress": false,
"mangle": false
},
"baseUrl": "src",
"paths": {
"@/*": ["*"],
"src/*": ["*"]
}
},
"module": {
"type": "commonjs"
}
}
Playground link
No response
Expected behavior
const _app = _interopRequireDefault(require("./app"));
Actual behavior
const _app = _interopRequireDefault(require("./app.ts"));
Error: Cannot find module './app.ts'
Version
1.3.15
Additional context
It makes huge error for our production server. 😭
No response
Use .js extension
Use .js extension
Could you explain more about it?
It was working well until 1.3.14 with typescript parser.
It was a bug and it's fixed.
Use .js extension for import statments
Related to https://github.com/swc-project/swc/pull/4620
I know, but problem is your code, not swc
I know, but problem is your code, not swc
Thank you for reply.
I am only using typescript files. So I cannot understand how to use js extension.
I tried to call app.ts from @/app from server.ts file.
server.ts
`import App from '@/app'` // src/app.ts
.swcrc
...
"baseUrl": "src",
"paths": {
"@/*": ["*"],
"src/*": ["*"]
}
...
What did i miss?
in .swcrc in you set
"paths": {
"~/*": ["*"]
}
I got app.ts (with .ts extension) import in my compiled .js
causing node execute failed subsequently
However, if the paths is set to
"paths": {
"~/*": ["src/*"]
}
I got "./src/app" (without .ts extension) import in my compiled .js
But I want to use ["*"] and I don't need .ts extension in my compiled .js files.
Any advice on this, please...
Use .js extension instead
@kdy1 Could you clarify further please? What should I do to use .js extension instead?
Use .js for import statements
Related to https://github.com/swc-project/swc/issues/6432
|
2025-04-01T04:35:38.051078
| 2022-11-11T03:15:03
|
1444860901
|
{
"authors": [
"alexander-akait"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11193",
"repo": "swc-project/swc",
"url": "https://github.com/swc-project/swc/pull/6402"
}
|
gharchive/pull-request
|
fix(css/parser): bug with scope in block
Description:
refactor code + fix bugs with block scope
BREAKING CHANGE:
No
Related issue (if exists):
No
@kdy1 Rebased and ready for review :+1:
|
2025-04-01T04:35:38.085268
| 2021-08-30T16:27:23
|
982976965
|
{
"authors": [
"allydog",
"swharden"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11194",
"repo": "swharden/FSKview",
"url": "https://github.com/swharden/FSKview/issues/45"
}
|
gharchive/issue
|
password using FSKView 1.1.7
I am having a problem dealing with the password in FSKView. I see the one in the settings.xml but it does not work, what am I doing wrong. It seems the only one you can use is what is in the settings.xml, correct ? Yet when I do use that PW, all it says is INCORRECT USERNAME/PASSWORD.
I had other issues, but got them figured out, but that this one.
Looks like I got the answer in groups.io , qrssknights.
Looks like I got the answer in https://groups.io/g/qrssknights
Welcome to Knights Al.
You don't need to use those settings unless you decide to upload your grabs to a website.
If at some time you think that you'd like to become an online grabber, then you can upload
each grab automatically, to your own website.
Many of us just open an qsl.net account and upload the files to it.
If a grabber seems to be doing good public service then one of us will add it to the
AJ4VD compendium.
73/4 de Andy
|
2025-04-01T04:35:38.104232
| 2022-07-25T15:05:02
|
1316974707
|
{
"authors": [
"adam-fowler",
"aelam"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11195",
"repo": "swift-server/vscode-swift",
"url": "https://github.com/swift-server/vscode-swift/issues/373"
}
|
gharchive/issue
|
the package.swift in subfolder can trigger a lot of events
Describe the bug
A clear and concise description of what the bug is.
My test case is the workspace folder doesn't have a Package.swift but subfolders have Package.swift in each of them.
when I tap one folder, the watcher starts to work
https://github.com/aelam/vscode-swift-test-packages/tree/main/
To Reproduce
Steps to reproduce the behavior:
Go to '...'
Click on '....'
Scroll down to '....'
See error
Expected behavior
A clear and concise description of what you expected to happen.
Only Package.swift in Root folder can trigger events? not sure current behavior is the expected one?
Environment
OS: [e.g. Ubuntu 20.04]
Swift version (output of swift --version)
Visual Studio Code version:
vscode-swift version: 0.7.0
Additional context
Add any other context about the problem here.
Current behaviour is Package.swift files in subfolders (note, not when they are subfolders of a SwiftPM project), will be treated as swift projects, when you open files in them. This is the intended behaviour
does it mean the .build folder will be generated.
Get Outlook for Androidhttps://aka.ms/AAb9ysg
From: Adam Fowler @.>
Sent: Tuesday, July 26, 2022 7:22:48 PM
To: swift-server/vscode-swift @.>
Cc: Ryan Wang @.>; Author @.>
Subject: Re: [swift-server/vscode-swift] the package.swift in subfolder can trigger a lot of events (Issue #373)
Current behaviour is Package.swift files in subfolders (note, not when they are subfolders of a SwiftPM project), will be treated as swift projects, when you open files in them. This is the intended behaviour
—
Reply to this email directly, view it on GitHubhttps://github.com/swift-server/vscode-swift/issues/373#issuecomment-1195297627, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AABYXEKNWFDOFHPGCUKI6ITVV637RANCNFSM54ST5HNA.
You are receiving this because you authored the thread.Message ID: @.***>
It will run a swift package resolve on the project so yes.
sorry my question is not clear enough
the folder structure like this
Root
A/B/C
D
A B C D are packages and they are under the same folder
B C are the depencies of A
open the folder by vscode
if I open C, does it mean C will be revolved?
follow-up question, how does the dependence lost show
Get Outlook for Androidhttps://aka.ms/AAb9ysg
From: Adam Fowler @.>
Sent: Tuesday, July 26, 2022 7:27:42 PM
To: swift-server/vscode-swift @.>
Cc: Ryan Wang @.>; Author @.>
Subject: Re: [swift-server/vscode-swift] the package.swift in subfolder can trigger a lot of events (Issue #373)
It will run a swift package resolve on the project so yes.
—
Reply to this email directly, view it on GitHubhttps://github.com/swift-server/vscode-swift/issues/373#issuecomment-1195302228, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AABYXEMHBIRLXNXBLLLBA53VV64R5ANCNFSM54ST5HNA.
You are receiving this because you authored the thread.Message ID: @.***>
yes it will resolve C. The extension supports SwiftPM projects inside sub-folders as they may be part of a bigger project which involves multiple languages or they may be part of a suite of projects like a collection of lambdas. Also by resolving C you do get the advantage that it will allow you to run tests from C.
It does not parse the folder structure in advance to find the SwiftPM projects as this may be an expensive operation based on how many subfolder projects there are. So each subfolder project is treated independently.
how does the dependence lost show
Not sure what you mean
I got your idea
how does the dependence list show
Not sure what you mean
if I tap C/package.swift, will the dependency tree for C shown in VSCode?
if I tap C/package.swift, will the dependency tree for C shown in VSCode?
yes
is there a case that
I modify A/Package.swift
I modified B/Package.swift
currently, I'm on B
B responded faster than A
the final dependency list will be for A?
The dependency list is based on the currently viewed file
|
2025-04-01T04:35:38.112749
| 2024-08-06T22:18:32
|
2451776271
|
{
"authors": [
"JannThomas",
"NachoSoto",
"realityworks"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11196",
"repo": "swiftlang/swift",
"url": "https://github.com/swiftlang/swift/issues/75733"
}
|
gharchive/issue
|
Swift <IP_ADDRESS>.6 is extra safe to the point it's unusable
Description
Consider this class hierarchy:
class A {}
@MainActor class B: A {}
Because of polymorphism, A's methods overridden by B need to be nonisolated. But this leaves a pretty large hole in the language without an actual solution.
Reproduction
@MainActor
final class A: NSObject {
var n: String = ""
}
extension A {
override var debugDescription: String {
// Main actor-isolated property 'n' can not be referenced from a nonisolated context
n
}
}
Expected behavior
I suppose this is working as intended, the diagnostic is correct. But how are we supposed to handle this?
Environment
swift-driver version: 1.113 Apple Swift version 6.0 (swiftlang-<IP_ADDRESS>.6 clang-16<IP_ADDRESS>)
Target: arm64-apple-macosx14.0
Additional information
No response
cc @hborla
I wonder if there is a solution here. Consider this example:
class A {
func f() {}
}
@MainActor
final class B: A {
private var a: String = ""
// Implicitly `nonisolated`
override func f() {
// Main actor-isolated property 'a' can not be referenced from a nonisolated context
print(a)
}
}
B is implicitly Sendable since it's actor-isolated, but A is not. So even though A.f() is non-isolated, instances of A cannot be sent, and instances of B can only be instantiated in the @MainActor. So wouldn't it be possible to allow B.f to have a different isolation than A.f?
Another approach that I wonder if it would work is something like @isolated(any) as suggested by @mattmassicotte. Methods from non-isolated classes like A.f (or NSObject) would be implicitly @isolated(any) instead of non-isolated?
We have reverted all actor code and are sticking with locks. Easier to manage ourselves instead of trying to work around syntax like this.
This is clearly a huge bug. We thought this would be fixed in time for Xcode 16 since a lot of people encountered issues with this.
Factually correct and clearly concurrency-correct code makes it impossible to compile under Swift 6. This is unacceptable and should be fixed as soon as possible.
Please take a look at this @hborla
|
2025-04-01T04:35:38.113979
| 2024-08-17T05:34:09
|
2471353443
|
{
"authors": [
"tshortli"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11197",
"repo": "swiftlang/swift",
"url": "https://github.com/swiftlang/swift/pull/75936"
}
|
gharchive/pull-request
|
Sema: Refactor unavailable declaration diagnostics
Consolidate code related to determining which kinds of diagnostics to emit and platform names to display so that it is not duplicated between regular declaration diagnostics and conformance diagnostics.
NFC.
@swift-ci please smoke test
|
2025-04-01T04:35:38.116566
| 2024-11-15T19:42:29
|
2662967855
|
{
"authors": [
"xedin"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11198",
"repo": "swiftlang/swift",
"url": "https://github.com/swiftlang/swift/pull/77653"
}
|
gharchive/pull-request
|
[CSBindings] Revert changes in BindingSet::isViable
This change although correct cases performance issues in some scenarios.
@swift-ci please test
Reverts https://github.com/swiftlang/swift/pull/76487 and https://github.com/swiftlang/swift/pull/77153 but leaves their tests.
@swift-ci please smoke test
@swift-ci please smoke test
|
2025-04-01T04:35:38.120488
| 2024-05-30T18:30:43
|
2326302693
|
{
"authors": [
"adam-fowler",
"ahoppen",
"plemarquand"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11199",
"repo": "swiftlang/vscode-swift",
"url": "https://github.com/swiftlang/vscode-swift/pull/830"
}
|
gharchive/pull-request
|
Add Swift: Capture VSCode Swift Diagnostic Logs command
Adds a new command that users can use to help generate bug reports for the extension itself. The Capture VSCode Swift Diagnostic Logs command will create a new folder in a temporary directory that contains:
Swift version and path information
Extension logs
Any diagnostics in the Problems pane
Users can copy the folder path or, if they're on macOS, open it in Finder. From there they can zip and attach these files to a GitHub issue. We'll want to update the New Issue Template with instructions on how to capture and attach these logs.
I have virtually never needed the diagnostics in the problems page when someone is submitting an issue. Things that are useful are swift log (which already includes swift version and paths), sourcekit-lsp log, user settings.
@adam-fowler I've added sourcekit-lsp logs and user settings to the diagnostics. Sourcekit-lsp logs will only be captured if the user has the trace level turned on in their settings.
@ahoppen it looks like it should be pretty straightforward to watch the work done progress events. However, there is only title and message on the work progress messages, meaning I'd have to check if title === 'SourceKit-LSP: Restoring functionality', which seems brittle. I guess we can use it if we can guarantee that string wont change
I can totally add an ID for this to the work done progress that we can check for. Just wanted to verify that it’s possible on the VS Code side before I extend LSP. Let’s do this in a follow-up PR.
@swift-server-bot test this please
|
2025-04-01T04:35:38.121911
| 2014-11-04T05:44:59
|
47680024
|
{
"authors": [
"clayg",
"stensonb"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11200",
"repo": "swiftstack/vagrant-swift-all-in-one",
"url": "https://github.com/swiftstack/vagrant-swift-all-in-one/issues/9"
}
|
gharchive/issue
|
error on streaming download to stdout
I'm sure this is a bug upstream, but here's what I get:
vagrant@saio:/run/shm$ swift download test test -o -
Error downloading object 'test/test': 'Error downloading test: read_length != content_length, 0 != 8'
this one finally got merged
https://review.openstack.org/#/c/144899/
|
2025-04-01T04:35:38.132095
| 2023-08-10T23:00:30
|
1845997427
|
{
"authors": [
"afolson",
"ajay-gov"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11201",
"repo": "swimos/swimos-docs",
"url": "https://github.com/swimos/swimos-docs/issues/79"
}
|
gharchive/issue
|
docs: Introspection reference guide
[ ] Show an app with the MetaKernel in the server.recon
[ ] Generate data into a few Web Agents and demonstrate the introspection APIs i.e get all node types, nodes of a given type, lanes for a give node, pulse etc. with the swim-cli
[ ] Also show swim-cli reflect command
Should this be a Reference Guide or a How-To-Guide. Any thoughts @afolson @PiggottDev @brohitbrose ?
@ajay-gov This one feels like a how-to guide as written, but we probably need a reference guide at some point as well.
@afolson i think so too. Maybe we should change it to the How-To-Guide. I feel that the how-to guide in this format is easier to do and more usable. We can add a comprehensive introspection reference guide later.
|
2025-04-01T04:35:38.158597
| 2019-11-10T16:11:24
|
520621426
|
{
"authors": [
"barneyb"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11202",
"repo": "switzerb/CookBook",
"url": "https://github.com/switzerb/CookBook/issues/94"
}
|
gharchive/issue
|
Labels should disallow the slash character
Labels should disallow the slash character, so that they may safely be used in URLs (and other path-like structures) without ambiguity.
Done!
|
2025-04-01T04:35:38.180727
| 2024-01-29T19:40:14
|
2106278599
|
{
"authors": [
"redclif43",
"xCENTx"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11203",
"repo": "swordbluesword/PalWorld-NetCrack",
"url": "https://github.com/swordbluesword/PalWorld-NetCrack/pull/60"
}
|
gharchive/pull-request
|
Additional Features & Optimizations
FEATURES
Teleport Pals to crosshair
Waypoints ( add by name & rendering )
Death Aura ( distance, dmg amount , scalar, vFx )
Debug print engine globals
attempt restore default values for some toggles
HELPERS
Get Party Pals
Get class pointer by blueprint name
Forge Actors ( set position , height, angle )
FIXES
Fixed modify attack
Fixed Uninject DLL crash
include SDK solution to base project
debug esp and teleport pals crash when actived
|
2025-04-01T04:35:38.183709
| 2017-08-11T18:47:52
|
249717549
|
{
"authors": [
"omirbod",
"pjreed",
"ramboto"
],
"license": "bsd-3-clause",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11204",
"repo": "swri-robotics/mapviz",
"url": "https://github.com/swri-robotics/mapviz/issues/501"
}
|
gharchive/issue
|
multires_image with transparent regions
Hello,
When I try to overlay an image on a tile map in mapviz, I notice the transparent regions of the image get colored black(I am using mapproxy with google map layer). If I put the same image on the google-earth application (with kml file), the transparent region is displayed correctly(I see the google map layer). Are there any ideas on how I can resolve this?
Are you drawing the image with Mapviz's image plugin? I suspect that alpha blending may not be enabled when it's drawing the image; we probably haven't tested it with images that have alpha channels. I'll do that sometime when I have the time for it, unless somebody else gets there first...
Ah sorry I think I wasn't specific in my description. I'm not using the Image plugin(sensor_msgs::Image), but the Multi-res Image plugin. I created a geo file giving the path to my image that I want to overlay onto a tile_map plugin. When I run mapviz and choose the multi-res image plugin, I see the image being overlaid correctly onto the tile_map plugin, but the transparent regions of my multi-res image turn out black.
If I use the same image on google earth the transparency is displayed correctly(or I can see the map layer underneath where the image is transparent)
Ok, I took a look at this and was able to reproduce it, and I think it's because the multires_image plugin is internally storing GL textures without an alpha channel. Should be easy to fix.
Great thank you.
|
2025-04-01T04:35:38.225890
| 2023-05-09T17:58:25
|
1702486004
|
{
"authors": [
"GODLiangCY",
"nikowonde",
"sxzz"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11208",
"repo": "sxzz/unplugin-vue-macros",
"url": "https://github.com/sxzz/unplugin-vue-macros/issues/362"
}
|
gharchive/issue
|
betterDefine doesn't seem to transform keyof typeof X correctly.
Describe the bug
Hi, I've encountered this problem with betterDefine that it doesn't seem to transform keyof typeof X correctly while X being an object with key and value pairs being strings.
I've provided quite minimal repro in which Vue throws following warning:
[Vue warn]: Invalid prop: type check failed for prop "trigger". Expected Null | Array, got String with value "hover".
Reproduction
https://stackblitz.com/edit/nuxt-starter-xgkq5y?file=components/Test.vue
System Info
-
Used Package Manager
yarn
Validations
[X] Follow our Code of Conduct
[X] Read the Contributing Guide.
[X] Check that there isn't already an issue that reports the same bug to avoid creating a duplicate.
[X] Check that this is a concrete bug. For Q&A, please open a GitHub Discussion instead.
[X] The provided reproduction is a minimal reproducible of the bug.
I think you may have written the wrong type declaration out of carelessness, the correct way to write it should be
const props = defineProps<{
trigger: keyof typeof Trigger | Array<keyof typeof Trigger>; // works well
}>();
instead of
const props = defineProps<{
trigger: keyof typeof Trigger | Array<keyof typeof trigger>;
}>();
I think you may have written the wrong type declaration out of carelessness, the correct way to write it should be
const props = defineProps<{
trigger: keyof typeof Trigger | Array<keyof typeof Trigger>; // works well
}>();
instead of
const props = defineProps<{
trigger: keyof typeof Trigger | Array<keyof typeof trigger>;
}>();
As a matter of a fact, you are correct in a sense that there was a typo with the lowercase t in Trigger in the reproduction. But I am still getting that same error [Vue warn]: Invalid prop: type check failed for prop "trigger". Expected Null | Array, got String with value "hover". so I don't think that typo was the root cause of this problem.
Fixed in<EMAIL_ADDRESS>
Great stuff, works like a charm. Thank you very much.
|
2025-04-01T04:35:38.231364
| 2020-11-30T12:52:18
|
753453460
|
{
"authors": [
"jeanfabrice",
"syberalexis"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11209",
"repo": "syberalexis/linky-exporter",
"url": "https://github.com/syberalexis/linky-exporter/pull/2"
}
|
gharchive/pull-request
|
Jeanfabrice/updated metrics
This PR adds support for 'Base' and 'Tempo' Enedis plan
It is also a proposal to switch to Prometheus recommended naming and labelling practice.
Thank's to your contribution.
It's a good PR :)
Can you just rebase the commit "Fix typo" and rename your variables in camelCase please ?
|
2025-04-01T04:35:38.320649
| 2019-08-08T13:20:35
|
478467465
|
{
"authors": [
"lyrixx",
"tristanbes"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11229",
"repo": "symfony/monolog-bundle",
"url": "https://github.com/symfony/monolog-bundle/pull/312"
}
|
gharchive/pull-request
|
add InsightOps support
closes #311
Tested on my project: it works 👍
Could you add a note in the CHANGELOG?
It will be part of 3.5.0
Thanks
yes i'll add this tomorrow.
is that ok for you @lyrixx ?
Thanks for your work on this new feature!
|
2025-04-01T04:35:38.322716
| 2017-11-09T09:37:14
|
272496307
|
{
"authors": [
"apfelbox",
"fabpot"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11230",
"repo": "symfony/profiler-pack",
"url": "https://github.com/symfony/profiler-pack/pull/3"
}
|
gharchive/pull-request
|
Add debug-bundle
Related to https://github.com/symfony/debug-pack/pull/4
Not sure about this one. Why would you expect debug bundle to be installed here?
For me it would be the inverse argument as for debug-pack:
If I use dump() via debug-pack I would expect to automatically be displayed in the webprofiler toolbar if it is installed.
And if I am already using the webprofiler toolbar, I would expect that a simple c req symfony/var-dumper would enable it inside my already installed toolbar.
What do you think?
|
2025-04-01T04:35:38.337053
| 2015-11-30T21:50:15
|
119588747
|
{
"authors": [
"iagomelanias",
"javiereguiluz",
"ogizanagi",
"pounard"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11231",
"repo": "symfony/symfony-installer",
"url": "https://github.com/symfony/symfony-installer/pull/206"
}
|
gharchive/pull-request
|
Fix Symfony3 is now the latest release
Symfony 3 is now the latest release. :tada:
Fix the following issue:
symfony new test
Downloading Symfony...
4.93 MB/4.93 MB ▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ 100%
Preparing project...
PHP Warning: require(/Users/Ogi/test/app/SymfonyRequirements.php): failed to open stream: No such file or directory in phar:///usr/local/bin/symfony/src/Symfony/Installer/DownloadCommand.php on line 265
Warning: require(/Users/Ogi/test/app/SymfonyRequirements.php): failed to open stream: No such file or directory in phar:///usr/local/bin/symfony/src/Symfony/Installer/DownloadCommand.php on line 265
PHP Fatal error: require(): Failed opening required '/Users/Ogi/test/app/SymfonyRequirements.php' (include_path='.:/usr/local/php5/lib/php') in phar:///usr/local/bin/symfony/src/Symfony/Installer/DownloadCommand.php on line 265
Fatal error: require(): Failed opening required '/Users/Ogi/test/app/SymfonyRequirements.php' (include_path='.:/usr/local/php5/lib/php') in phar:///usr/local/bin/symfony/src/Symfony/Installer/DownloadCommand.php on line 265
Are you using the most recent release of the installer?
@javiereguiluz : I'm currently using the master branch of the repository.
OK. I'm going to merge and release a new version. Thanks for fixing this issue @ogizanagi.
Same problem here. I also can't run the command symfony new [name] 3.0.
@javiereguiluz : tests will need to be updated, but I guess this could be done after a new release.
@iagomelanias try excuting the command again and it should work now.
Don't know if this is related, but I had similar but not exactly the same kind of errors, trying to ugprade. It seems that the generated composer.json specifies app/Kernel.php in autoload files, it's being loaded prior to the bootstrap.php.cache file generation, so most of the class chain required by the kernel is declared while the Symfony\Component\ClassLoader\ClassCollectionLoader is trying to generate the cache file. Because those class are declared, they are being ignored and not dumped: on next run, an incomplete bootstrap.php.cache file is loaded and generate PHP fatal errors due to missing interfaces.
|
2025-04-01T04:35:38.450263
| 2012-02-28T09:54:55
|
3415815
|
{
"authors": [
"brendo",
"michael-e"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11232",
"repo": "symphonycms/symphony-2",
"url": "https://github.com/symphonycms/symphony-2/issues/1036"
}
|
gharchive/issue
|
Fix for issues with IE8's image/x-png mime type.
This fixes an issue with IE8 sending a strange mime type when uploading an image. Adding the mime type to the imageMimeTypes array will fix the size detection feature of the upload field.
For this IE behaviour see http://msdn.microsoft.com/en-us/library/ie/ms775147(v=vs.85).aspx:
…In some cases, the detected MIME type can differ from the generally accepted value for backward compatibility, as shown in the following table:
Standard MIME Type FindMimeFromData Returns
image/jpeg image/pjpeg
image/png image/x-png
That must of been a fun one for you to debug :)
Actually it was rather simple, cause my application uses frontend publishing all over, so I could simply use ?debug to see the XML data of my uploaded files.
:-)
Nice :)
|
2025-04-01T04:35:38.601741
| 2024-10-30T21:12:23
|
2625354166
|
{
"authors": [
"richcooper95",
"woody1872"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11233",
"repo": "syntasso/kratix",
"url": "https://github.com/syntasso/kratix/issues/264"
}
|
gharchive/issue
|
Bug: secret "gitea-credentials" not found
On the guide to install Kratix on KinD there appears to be an issue on the Set up State Store step. It fails to setup Gitea because the secret "gitea-credentials" doesn't exist.
The gitea-install.yaml file references the "gitea-credentials" secret but it doesn't yet exist.
There are no instructions which tell a user to create this secret, so I'm assuming it should be created for them?
Error:
$ kubectl --context=$PLATFORM get pods -n gitea
NAME READY STATUS RESTARTS AGE
gitea-5c965c46cb-5xnst 0/1 Init:CreateContainerConfigError 0 34s
gitea-create-repository-77n5l 0/1 Init:0/4 0 24s
$ kubectl --context=$PLATFORM describe pod gitea-5c965c46cb-5xnst -n gitea
...
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 53s default-scheduler Successfully assigned gitea/gitea-5c965c46cb-5xnst to platform-control-plane
Normal Pulled 3s (x6 over 52s) kubelet Container image "gitea/gitea:1.21.11-rootless" already present on machine
Warning Failed 3s (x6 over 52s) kubelet Error: secret "gitea-credentials" not found
Hey @woody1872, thanks for bringing this to our attention! We've just fixed this and https://github.com/syntasso/kratix/issues/263 - you can find updated steps in the installation instructions.
The instructions now have an extra step to install the Gitea namespace and credentials:
bash <(curl -s https://raw.githubusercontent.com/syntasso/kratix/refs/heads/main/scripts/generate-gitea-credentials.sh)
Note that you will need to install the gitea CLI for this to work (required because the certs need to be generated dynamically).
Hope this helps!
Closing this as done, but please do let us know if you have any further issues @woody1872!
|
2025-04-01T04:35:38.615428
| 2023-11-13T15:14:25
|
1990868815
|
{
"authors": [
"JuliaRegistrator",
"syoshida1983"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11234",
"repo": "syoshida1983/AngularSpectrumMethod.jl",
"url": "https://github.com/syoshida1983/AngularSpectrumMethod.jl/issues/2"
}
|
gharchive/issue
|
Register v0.1.0
@JuliaRegistrator register
Register the first release of AngularSpectrumMethod.jl
Registration pull request created: JuliaRegistries/General/95248
Tip: Release Notes
Did you know you can add release notes too? Just add markdown formatted text underneath the comment after the text
"Release notes:" and it will be added to the registry PR, and if TagBot is installed it will also be added to the
release that TagBot creates. i.e.
@JuliaRegistrator register
Release notes:
## Breaking changes
- blah
To add them here just re-invoke and the PR will be updated.
Tagging
After the above pull request is merged, it is recommended that a tag is created on this repository for the registered package version.
This will be done automatically if the Julia TagBot GitHub Action is installed, or can be done manually through the github interface, or via:
git tag -a v0.1.0 -m "<description of version>" 34839c2b31a9fcc3d91f576d04f896da6c0ce396
git push origin v0.1.0
|
2025-04-01T04:35:38.830818
| 2024-05-28T21:08:11
|
2321924993
|
{
"authors": [
"jobelenus"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11235",
"repo": "systeminit/si",
"url": "https://github.com/systeminit/si/pull/3870"
}
|
gharchive/pull-request
|
Fix: dont confetti unless on head
Confetti was firing on a changeset when deleting a component that was added in the changeset.
bors merge
bors merge
|
2025-04-01T04:35:38.838687
| 2019-03-04T07:22:15
|
416664105
|
{
"authors": [
"JayjeetAtGithub",
"ivotron"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11236",
"repo": "systemslab/popper",
"url": "https://github.com/systemslab/popper/issues/510"
}
|
gharchive/issue
|
cli: run all actions in a stage in parallel
A stage is a set of actions that a subsequent action depends on. A stage is currently executed serially. For this issue, we will add the ability of running it in parallel, with options that control the level of parallelism.
@ivotron Can we use a ThreadPoolExecutor for this ?
@JayjeetAtGithub yeah, that would work. We need to take care of 2.7 though, which I believe is by installing the futures module
|
2025-04-01T04:35:38.858243
| 2021-03-06T03:16:09
|
823547575
|
{
"authors": [
"elenners"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11237",
"repo": "syuilo/misskey",
"url": "https://github.com/syuilo/misskey/issues/7308"
}
|
gharchive/issue
|
Cannot update to latest version
I have updated my server, and tried multiple times and get the same error each time I try to update Misskey. My server is Debian 10 x64 (buster) with 2gb of ram. My whole instance is broken now. seaside.cafe
I somehow fixed it
|
2025-04-01T04:35:38.872955
| 2023-04-30T17:09:24
|
1689999658
|
{
"authors": [
"szepeviktor"
],
"license": "Unlicense",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11238",
"repo": "szepeviktor/wordpress-website-lifecycle",
"url": "https://github.com/szepeviktor/wordpress-website-lifecycle/issues/9"
}
|
gharchive/issue
|
Architecture of WooCommerce (lifecycle overview)
Trim down WC, "WooCommerce lite"
Load WooCommerce class from a modified copy of includes/class-woocommerce.php
replace __FILE__ with dirname(WC_PLUGIN_FILE).'/includes/class-woocommerce.php'
compare class version to plugin version during deploymentversion_compare(\WooCommerce::instance()->version, \Composer\InstalledVersions::getVersion('wpackagist-plugin/woocommerce'), '=');
WC Features: src/Internal/Features/FeaturesController.php
WC Admin features: filter:woocommerce_admin_features includes/react-admin/feature-config.php
WC Blocks: src/Packages.php
wget https://github.com/woocommerce/woocommerce/releases/download/7.6.1/woocommerce.zip
sed -e "s#\\b__FILE__\\b#dirname(WC_PLUGIN_FILE).'/includes/class-woocommerce.php'#" includes/class-woocommerce.php
Legacy features
var_dump(array_keys(Automattic\WooCommerce\Utilities\FeaturesUtil::get_features(true, true)));
analytics
new_navigation
product_block_editor
cart_checkout_blocks
marketplace
order_attribution
hpos_fts_indexes
custom_order_tables
respond with HTTP/410 on trashed products
https://gist.github.com/szepeviktor/dedfeb7d4a5d2a1d1df1139cf48935f5
|
2025-04-01T04:35:38.901718
| 2021-12-13T09:15:12
|
1078267536
|
{
"authors": [
"LostOnTheLine",
"ZoltrixGFC",
"gui28347",
"pakshuang",
"rackey12",
"t0mer"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11239",
"repo": "t0mer/broadlinkmanager-docker",
"url": "https://github.com/t0mer/broadlinkmanager-docker/issues/47"
}
|
gharchive/issue
|
Broadlink Manager Windows see invalid device, docker version sees nothing?
Hi,
I have an Broadlink RM4 Pro S. Tried both the Windows Broadlink Manager software and the docker version.
The Windows version reports:
Looking for devices...
Found device : Unknown [<IP_ADDRESS>]
Writing compatible device not detected!
The docker version finds nothing when doing a scan.
I can confirm the docker version is setup on the same network, so unsure why it isn't picking anything up?
Any advice?
same problem for me, RM4 PRO and RM4 Mini
Hi,
I stopped working on the desktop version long ago.
I can try add the supported devices from the docker version but I'm not sure it will work.
@t0mer would be great if you could add them, but why do you think it won't work?
I'm a bit stuck at the moment, as Home Assistant won't scan one of my remotes (scans others ok). I posted the issue, but no response yet: https://github.com/home-assistant/core/issues/63578#issuecomment-1007112478
I have the same issue
i have an RM4 Mini (RM4C i think)
and an S3 hub but the docker is not picking up anything
Yep still the same issue here.
Same. RM4C Mini and nothing shows up on the scan. Home Assistant Broadlink integration works but the docker broadlink manager can't find the remote :(
I think @t0mer is on leave, not sure when he is due back.
I think @t0mer is on leave, not sure when he is due back.
Hi @ZoltrixGFC
I'm not on leave (:
Working on other projects and also on a big update to the docker version of broadlink manager.
Hopefully I'll get back to the Windows desktop version by mid-April.
Same. RM4C Mini and nothing shows up on the scan. Home Assistant Broadlink integration works but the docker broadlink manager can't find the remote :(
Hi @pakshuang,
What is the model of your RM Device?
Hey @t0mer,
It says RM4C mini on the device, which is 0x62be in your docs.
Windows version of Docker (if on WSL2 at least) does not allow net=host without going through a complicated process each time it's run to port forward to a changing on boot IP for the WSL2. That's why it sees nothing.
If there was a way to NOT do the net=host, even if it requires forwarding a bunch of ports, That'd be a big help. Does anyone know which ports it's using? As an application-level tool, that doesn't need to be running except when needed, I wouldn't mind forwarding a bunch of ports since it'll only be using them while I run the tool.
I didn't see a windows version, just the Docker, I know everyone says it doesn't work but where can I find the Windows version?
@LostOnTheLine, @t0mer said he updated the Windows version mid-April, did you get the latest version?
|
2025-04-01T04:35:38.931104
| 2023-02-10T03:45:23
|
1578935911
|
{
"authors": [
"bai",
"c-ehrlich",
"juliusmarminge"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11240",
"repo": "t3-oss/create-t3-app",
"url": "https://github.com/t3-oss/create-t3-app/pull/1191"
}
|
gharchive/pull-request
|
feat: Use ~/* import alias
✅ Checklist
[x] I have followed every step in the contributing guide (updated 2022-10-06).
[x] The PR title follows the convention we established conventional-commit
[x] I performed a functional test on my final commit
Changelog
Use ~/* import alias like the one recently added to Next.js.
💯
Thanks for feedback. My thinking was that ct3a is opinionated so it'd be fine to have one more opinion. I personally have no preference on whether it's @/ or ~/ or something else as long as it resolves the inconvenience of counting/keeping-in-mind ../'s.
I certainly would prefer not to "Allow the user to define the alias" since it'd add more mental overhead to everyone involved.
I certainly would prefer not to "Allow the user to define the alias" since it'd add more mental overhead to everyone involved.
It's not that much overhead, Next does this fairly straightforward:
https://github.com/vercel/next.js/blob/7bbd239fe7e6d72ef467484619b978c11f5fbb70/packages/create-next-app/templates/index.ts#L85-L104
Allowing either with/without aliases would be difficult though
Thanks @juliusmarminge, that looks pretty straightforward - happy to implement this if you think it's fine to have aliases used within ct3a 🙏🏼
I thought about this and I think I'm leaning towards "default to a sensible alias, and let the user override".
I'm not a fan of @ as many packages also use this. My vote for the default would be either ~ (create-next-app uses this) or ~~ (nuxt uses this).
@c-ehrlich I wanted to confirm with you something: been meaning to switch from @/ to ~/ and while going through create-next-app, I noticed that they actually use the former by default:
Since I have no preference on the alias and you mentioned that "we're more closely related to create-next-app", I wanted to double-check with you that you're preferring ~/ instead of create-next-app's default? 🙏🏼
Since I have no preference on the alias and you mentioned that "we're more closely related to create-next-app", I wanted to double-check with you that you're preferring ~/ instead of create-next-app's default? 🙏🏼
Yes ~/* by default
Alright, this seem to be working as expected.
After packages are installed, I find-and-replace the default (~/) with the alias entered by a user. I initially used the glob package like create-next-app does but then replaced that with a function.
Could you please take a look? And apologies for a bit of commit spam, this is my first contribution to ct3a and I'm very likely doing something incorrectly 🙏🏼
@c-ehrlich adding "typescript.preferences.importModuleSpecifier": "non-relative" fixes the issue for me.
Note that there are two settings:
@c-ehrlich adding "typescript.preferences.importModuleSpecifier": "non-relative" fixes the issue for me.
Note that there are two settings:
oh yeah makes sense that it's not in javascript 😅
nonetheless this is not great dx as we are now making everyone use absolute imports, but the default editor setting is to ignore this. i guess this brings us back to the debate of whether we want to ship a .vscode with a bunch of stuff in it, i'm still leaning towards no personally but the reasons to do so seem to keep increasing.
i just checked and create-next-app has the same issue, so at least we're not alone.
let's get this merged, and we can think about the other stuff later
Thanks for reviews and helping me ship this ❤️
Re .vscode config dir, one more reason to have one would probably be Next.js' TypeScript Plugin that requires configuring VS Code to use local TypeScript SDK. From what I've seen while playing with app dir (experimental-app), create-next-app will actually generate .vscode dir for you.
|
2025-04-01T04:35:38.935165
| 2023-08-22T10:27:31
|
1861161996
|
{
"authors": [
"JackMcBride98",
"andreaTarascioSophany"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11241",
"repo": "t3-oss/create-t3-app",
"url": "https://github.com/t3-oss/create-t3-app/pull/1541"
}
|
gharchive/pull-request
|
docs: Update next-js.md to add an example on how to use getServerSideProps anyway
Closes #1539
✅ Checklist
[x] I have followed every step in the contributing guide (updated 2022-10-06).
[x] The PR title follows the convention we established conventional-commit
[x] I performed a functional test on my final commit
Changelog
Update next-js.md to add an example on how to use getServerSideProps anyway
Screenshots
💯
Thanks for doing this :)
|
2025-04-01T04:35:38.941137
| 2022-09-01T20:38:21
|
1359380248
|
{
"authors": [
"intagaming"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11242",
"repo": "t3-oss/create-t3-turbo",
"url": "https://github.com/t3-oss/create-t3-turbo/issues/11"
}
|
gharchive/issue
|
Seems like EAS iOS building is having problem
I'm trying to get the Expo app built using EAS, building for iOS (simulator).
Here's what I've done:
Follow what's in here until I get the step that builds for iOS Simulator
Here's my build profile:
"iosPreview": {
"ios": {
"simulator": true
}
},
I build with eas build -p ios --profile iosPreview
The first time it fails, it's complaining about the entry point not being index.js. It is defined as src/_app in package.json. I provided the index.js as follow:
import { registerRootComponent } from "expo";
import App from "./src/App";
registerRootComponent(App);
I've changed src/_app.tsx to src/App.tsx, remove the registerRootComponent and go export default App.
It seems to build now. However I'm getting this Xcode Logs error in EAS Build after failing:
/Users/expo/.nvm/versions/node/v16.13.2/bin/node /Users/expo/workingdir/build/node_modules/react-native/cli.js bundle --entry-file index.js --platform ios --dev false --reset-cache --bundle-output /Users/expo/workingdir/build/apps/expo/ios/build/Build/Products/Release-iphonesimulator/main.jsbundle --assets-dest /Users/expo/workingdir/build/apps/expo/ios/build/Build/Products/Release-iphonesimulator/iHaveTime.app
warning: the transform cache was reset.
Welcome to Metro!
Fast - Scalable - Integrated
error Unable to resolve module @babel/runtime/helpers/interopRequireDefault from /Users/expo/workingdir/build/apps/expo/index.js: @babel/runtime/helpers/interopRequireDefault could not be found within the project or in these directories:
It seems that it needs @babel/runtime, but I'm not sure. I tried to npm i -D @babel/runtime and build again to see what happens. I'll come back with the result. Meanwhile it seems like the current Expo settings in this monorepo is not enough to build EAS iOS by following the Expo's guide. There's something missing and I'm trying to figure out which exactly is missing.
Update 1: Not working, same error. I'm trying again with dependencies instead of devDependencies. Don't know if a fresh Expo app will be able to build EAS without error (I'm trying one). If it's indeed the monorepo's error, metro.config.js might need some tweaks. https://docs.expo.dev/guides/monorepos/
I just realized something: by placing eas.json inside apps/expo, maybe the thing in the packages/ folder are not included. So it seems the monorepo guide is relevant by telling me that I need eas.json at the root of the project. Will see about this.
Yeah it's fixed. The above metro.config.js is the solution. No @babel/runtime required.
|
2025-04-01T04:35:39.008240
| 2024-10-16T13:29:36
|
2591899749
|
{
"authors": [
"Arunn2287",
"cesarmr-github",
"tabarra"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11243",
"repo": "tabarra/txAdmin",
"url": "https://github.com/tabarra/txAdmin/issues/995"
}
|
gharchive/issue
|
[BUG] The txadmin UI takes a long time to appear.
txAdmin/FXServer versions:
v7.3.1
Describe the bug
When opening /tx, in the TXADMIN section, it takes a while to appear. I don't know if it's something with the machine or with txadmin.
To Reproduce
Steps to reproduce the behavior:
Go to /tx
Click on txadmin
See error
Expected behavior
It should appear without having to wait so long
Screenshots
Additional context
Node: v16.9.1
OS: Linux 5.15.0-121-generic
We have the same issue.
We tryed with the new 10367 linux atrifact, but the same problem.
This has been solved in 93f8d305b94c55bb0071d5c7d7966c3aed7e3aa3.
A new update will be available tomorrow, thanks for the report.
Fixed, update to v7.3.2.
Thanks again for the report!
|
2025-04-01T04:35:39.009433
| 2022-09-04T07:29:46
|
1361068459
|
{
"authors": [
"Mats4k",
"tabarra"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11244",
"repo": "tabarra/txAdmin",
"url": "https://github.com/tabarra/txAdmin/pull/672"
}
|
gharchive/pull-request
|
added missing german translations
Hey,
i added a few missing translations to stuff that was still in english.
The whole file should now be translated.
(Not using translator, native german)
Thank you for the PR.
|
2025-04-01T04:35:39.010629
| 2021-12-17T00:32:28
|
1082759136
|
{
"authors": [
"fantasai",
"tabatkins"
],
"license": "CC0-1.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11245",
"repo": "tabatkins/bikeshed",
"url": "https://github.com/tabatkins/bikeshed/issues/2200"
}
|
gharchive/issue
|
Don't allow forless links for descriptors
With Contain 3 taking on width/height especially, we're getting a lot of "ambiguous property/descriptor" errors. It might be time to disallow them for descriptors and default to properties.
Yeah, I think we're long overdue to consider that a failed remnant of the old preprocessor. Some specs now have link failures I need to go clean up, but Bikeshed-wise the change is done now: all for-less propdesc autolinks are now property only.
|
2025-04-01T04:35:39.019692
| 2024-08-27T03:40:08
|
2488189359
|
{
"authors": [
"jacalata",
"jorwoods"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11246",
"repo": "tableau/server-client-python",
"url": "https://github.com/tableau/server-client-python/pull/1449"
}
|
gharchive/pull-request
|
fix: Pager typing
Pager Protocols were missing the generic flags. Added those in so the Pager correctly passes through the typing information. Also removes the kwargs from the function signature of the Endpoint.get protocol to make the Workbook endpoint match.
Adding a return annotation of None to the tests is very important because it is what enables static type checkers, like mypy, to inspect those functions. With these annotations now in place, users of TSC should more transparently be able to carry through typing information when using the Pager.
Not sure I understand your comment about adding the return type to the tests. How does that make it easier for users to get typing information in their code?
Making the protocols generic is what allows the pager to retrieve the typing information from the endpoint provided to Pager and surface that to users.
Adding a return type on tests is what gives mypy the ability to inspect the test functions. Since test functions somewhat mirrors how people would use the package, contributors to TSC should annotate test functions returning None to verify that typing information is available to the package consumers.
In this case, when I added typing to the Pager object in #1390, I missed annotating the functions as returning None. Since those test functions were not annotated, mypy falsely reported that there were no errors in typing.
|
2025-04-01T04:35:39.028210
| 2015-03-18T16:38:58
|
62737400
|
{
"authors": [
"jeremybmerrill"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11247",
"repo": "tabulapdf/tabula",
"url": "https://github.com/tabulapdf/tabula/issues/293"
}
|
gharchive/issue
|
factor out setTimeout based image resolution picker
https://github.com/tabulapdf/tabula/blob/feature/newUI/webapp/static/js/pdf_view.js#L656-L668
the file info API should hold this information;
alternatively, only do the setTimeout on the page, not on individual pages (otherwise lazyload can break)
fixed in 7a5e565
|
2025-04-01T04:35:39.041192
| 2018-05-24T06:47:55
|
325985288
|
{
"authors": [
"az360ct",
"jeremybmerrill"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11248",
"repo": "tabulapdf/tabula",
"url": "https://github.com/tabulapdf/tabula/issues/862"
}
|
gharchive/issue
|
trying to fill out fields of spread sheet with extracted data from pdf
Tow Ticket (4).pdf
thats the pdf
paperless statement.xlsx
here is the spread sheet
Hi @az360ct, I don't understand what you're asking. Can you please describe your question in greater depth?
i dont know how to use the program to move information from there
respective fields on the pdf invoice to the proper fields on the
spreadsheet template
On Mon, May 28, 2018 at 9:59 PM, Jeremy B. Merrill<EMAIL_ADDRESS>
wrote:
Hi @az360ct https://github.com/az360ct, I don't understand what you're
asking. Can you please describe your question in greater depth?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/tabulapdf/tabula/issues/862#issuecomment-392631808,
or mute the thread
https://github.com/notifications/unsubscribe-auth/Alv1vIIhJ9wniJBXinmI59T4ipj2k4Iwks5t3KuHgaJpZM4ULp-2
.
Hi @az360ct: Tabula only extracts the data into a spreadsheet for you. It is not in scope for the project to place it into another spreadsheet's template (though you could likely accomplish this wish some Excel scripting that pastes, for instance, cell A1 of the Tabula output into the relevant cell of the template).
thats what i was thinking, will tabula batch all the pdf i have in a folder
according to template i have saved into one spread sheet or multiple
On Tue, May 29, 2018 at 3:20 PM, Jeremy B. Merrill<EMAIL_ADDRESS>
wrote:
Hi @az360ct https://github.com/az360ct: Tabula only extracts the data
into a spreadsheet for you. It is not in scope for the project to place it
into another spreadsheet's template (though you could likely accomplish
this wish some Excel scripting that pastes, for instance, cell A1 of the
Tabula output into the relevant cell of the template).
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/tabulapdf/tabula/issues/862#issuecomment-392909026,
or mute the thread
https://github.com/notifications/unsubscribe-auth/Alv1vO9JfIH05Up3qzsCM0u8FNWs1Q1Oks5t3Z-bgaJpZM4ULp-2
.
is there a way to clean up the data from tabula output
On Tue, May 29, 2018 at 3:26 PM, Azimuthllc llc<EMAIL_ADDRESS>wrote:
thats what i was thinking, will tabula batch all the pdf i have in a
folder according to template i have saved into one spread sheet or multiple
On Tue, May 29, 2018 at 3:20 PM, Jeremy B. Merrill <
<EMAIL_ADDRESS>wrote:
Hi @az360ct https://github.com/az360ct: Tabula only extracts the data
into a spreadsheet for you. It is not in scope for the project to place it
into another spreadsheet's template (though you could likely accomplish
this wish some Excel scripting that pastes, for instance, cell A1 of the
Tabula output into the relevant cell of the template).
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/tabulapdf/tabula/issues/862#issuecomment-392909026,
or mute the thread
https://github.com/notifications/unsubscribe-auth/Alv1vO9JfIH05Up3qzsCM0u8FNWs1Q1Oks5t3Z-bgaJpZM4ULp-2
.
https://github.com/tabulapdf/tabula-java can process multiple identically-formatted PDFs into multiple spreadsheets. (tabula-java is the command-line version of the extraction engine that powers Tabula). I think there's been some movement about putting templated processing of multiple PDFs into mainline Tabula, but we're not there yet.
|
2025-04-01T04:35:39.085974
| 2020-10-17T16:37:38
|
723798372
|
{
"authors": [
"roschaefer",
"tillprochaska"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:11249",
"repo": "tactilenews/100eyes",
"url": "https://github.com/tactilenews/100eyes/pull/434"
}
|
gharchive/pull-request
|
415 configurable settings
close #415
merge this after #445
@tillprochaska could you give me a review on this?
@tillprochaska agree with what you said. I already committed a couple of suggestions. Feel free to pluralize SettingForm and setting_params if you want.
@roschaefer I changed a few UI related things and also added the UI texts for the onboarding success page to the settings. (Will fix those rubocop errors later.) It’s quite long already, at some point we should probably think about splitting the sections up into separate pages (similar to the way settings are organised e.g. in the GitHub UI).
This is what the page now looks like. Maybe either @FrauCsu or @drjakob could have a look at it and check if everything that needs to be configurable by editors is in there?
@roschaefer I changed a few UI related things and also added the UI texts for the onboarding success page to the settings. (Will fix those rubocop errors later.) It’s quite long already, at some point we should probably think about splitting the sections up into separate pages (similar to the way settings are organised e.g. in the GitHub UI).
This is what the page now looks like. Maybe either @FrauCsu or @drjakob could have a look at it and check if everything that needs to be configurable by editors is in there?
I don't see all text fields yet, that's because you didn't push them, right?
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.