id stringlengths 4 10 | text stringlengths 4 2.14M | source stringclasses 2
values | created timestamp[s]date 2001-05-16 21:05:09 2025-01-01 03:38:30 | added stringdate 2025-04-01 04:05:38 2025-04-01 07:14:06 | metadata dict |
|---|---|---|---|---|---|
974580995 | Default Dashboard Executive variable doesn't work when I assign it.
When I tried to change it from the default value of 'clap' by doing this
let g:dashboard_default_executive ='telescope'
it would ignore the command and continue to use the default 'clap' instead.
so I had to manually go and manipulate the plugins/dashboard.vim to 'telescope'
If nobody else is able to produce this bug then it might have just been me doing something wrong with my init.vim
nvm i figured out why
| gharchive/issue | 2021-08-19T11:51:01 | 2025-04-01T06:44:19.726112 | {
"authors": [
"BooleanCube"
],
"repo": "glepnir/dashboard-nvim",
"url": "https://github.com/glepnir/dashboard-nvim/issues/94",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1308564910 | Does NOT build on OSX 10.5 and 10.4 . (Builds on 10.3, 10.2 (ie prior releases)) Yea: I know "WONT FIX!!!!" FK you.
GLFW will not build on OSX 10.5 and 10.4, however it will build on 10.3 and 10.2, and 10.6 on and on.
Dear people who will "WONT FIX" this.
This does build on 10.3 and 10.2.
It does NOT build on 10.5 and 10.4. If you paid attention to your build bots you would know that.
I was told to use mac ports because you support all releases.
I was asked by a mac user to build a new OSX version of my project (chaos esque anthology).
It needs OpenGL.
Which means it probably needs glfw.
And Apple ADULTERATED strlcpy in any release after 10.5
So I need to build on 10.5.
And no: I'm not recoding the engine so it's written in "Apple C"
It works fine on grsecurity-linux in the most mem secure configuration. Works fine on BSD, FreeBSD, Solaris, Windows, Linux, etc. Only Apple has adulterated the defacto standards.
And I've contributed to opensource by programming for it and making opensource media (3d, 2d, music, etc) for 20 years so don't pull the "WELL WE DONT HAVE TO HELP ANYONE, PAY US" BULLSHIT.
NOO WE DOOONT SUPPORT MOR THAN 3 RELEASES
Good job being homebrew. FGGT DIPSHTS!
Mac ports "thing" is supporting all old stuff.
That's it's advantage vs homebrew.
GLFW is needed for most/all opengl apps.
Which now CANNOT build on OSX Mountain Lion and Lion.
Because YOU (you) don't pay any attention to the last non-adulterated version of C in OSX.
FIX IT.
And no I will not "be nice" to pro-tranny pro-fggt pro-women's rights pro-code of conduct people.
A code of conduct is an additional wrighting taking away rights given by the copyright holder; forbidden by the text of the GPL. You people, here on github, are violating the copyrights of the old contributors (who did not /TRANSFER/ their copyrights) by adding a code of conduct.
YHWH explicitly allows men to marry young girls in Book 5 of Moses, chapter 22, verse 28. In Hebrew, Greek, and Latin.
And if any should find the
[ child virgin], whoever was not espoused, and
using force on her, he went to bed with her, and he should be found,
[ shall give the man going to bed with her] to the
father of the young woman fifty double-drachmas of silver,
and she will be his wife, because he humbled her;
he will not be able to send her away at any time.
Ἐὰν δέ τις εὕρῃ τὴν παῖδα τὴν παρθένον, ἥτις οὐ μεμνήστευται, καὶ
βιασάμενος κοιμηθῇ μετ᾽ αὐτῆς καὶ εὑρεθῇ,
δώσει ὁ ἄνθρωπος ὁ κοιμηθεὶς μετ᾽ αὐτῆς τῷ πατρὶ τῆς νεάνιδος
πεντήκοντα δίδραχμα ἀργυρίου, καὶ αὐτοῦ ἔσται γυνή, ἀνθ᾽ ὧν
ἐταπείνωσεν αὐτήν· οὐ δυνήσεται ἐξαποστεῖλαι αὐτὴν τὸν ἅπαντα χρόνον.
chapter 22, verse 28)
כִּי־יִמְצָא אִישׁ *נַעַר בְתוּלָה אֲשֶׁר לֹא־אֹרָשָׂה
וּתְפָשָׂהּ וְשָׁכַב עִמָּהּ וְנִמְצָאוּ
וְנָתַן הָאִישׁ הַשֹּׁכֵב עִמָּהּ לַאֲבִי *הַנַּעַר
חֲמִשִּׁים כָּסֶף וְלוֹ־תִהְיֶה לְאִשָּׁה תַּחַת אֲשֶׁר עִנָּהּ
לֹא־יוּכַל שַׁלְּחָהּ כָּל־יָמָיו ס
si invenerit vir puellam virginem quae non habet sponsum et
adprehendens concubuerit cum ea et res ad iudicium venerit
dabit qui dormivit cum ea patri puellae quinquaginta siclos argenti et
habebit eam uxorem quia humiliavit illam non poterit dimittere cunctis
diebus vitae suae
I did not submit this "one hour ago" WTF.
Fking reddit bullshit.
You fking scum always chasing your "ohh I'm a better code of conduc supporter then you" clout.
FGGTS>
FCUKING FGGTS
We'll point you out when Rssa comes.
| gharchive/issue | 2022-07-18T20:32:12 | 2025-04-01T06:44:19.739248 | {
"authors": [
"MikeeUSA3"
],
"repo": "glfw/glfw",
"url": "https://github.com/glfw/glfw/issues/2157",
"license": "Zlib",
"license_type": "permissive",
"license_source": "github-api"
} |
121334777 | Classname removed from every but the last
When I explicitly pass a classname to a th tag it is stripped from every column but the last one. Is there a way around this?
ex:
...
Thanks.
the problem is right here: https://github.com/glittershark/reactable/blob/master/src/reactable/thead.jsx#L10-L19
for some reason when the child has an object for the children it will properly pass the props.. but if the children is a string it ignores them. Why not build the column object each time like:
columns.push({
key: th.props.column,
label: th.props.children,
props: filterPropsFrom(th.props)
});
seems to fix the problem.. but then again I didn't read much into the logic of the children, column === "string";
@mephux
I take you mean something like this (in JSX)?
<Table
className="table-hover"
id="myTable"
sortable={true}
props={filterPropsFrom(th.props.className)} >
But this doesn't work. Could you give an example of how to assign classnames to <Th> tags?
| gharchive/issue | 2015-12-09T20:43:50 | 2025-04-01T06:44:19.799055 | {
"authors": [
"int13h",
"mephux",
"paddotk"
],
"repo": "glittershark/reactable",
"url": "https://github.com/glittershark/reactable/issues/223",
"license": "mit",
"license_type": "permissive",
"license_source": "bigquery"
} |
506329227 | Translated Readme.md into English with proper grammar and punctuation.
Translated the existing readme.MD into English with correct spellings, grammar, and punctuations for ease of understanding and to make it reachable to a larger audience.
Closes #1
Thanks for your contribution, but we ended up using an internal translation. :+1:
| gharchive/pull-request | 2019-10-13T14:00:54 | 2025-04-01T06:44:19.828823 | {
"authors": [
"Sweta271097",
"mbenford"
],
"repo": "globocom/prettylog",
"url": "https://github.com/globocom/prettylog/pull/18",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1837822094 | 🛑 Nextcloud is down
In 3c21121, Nextcloud ($NEXTCLOUD_URL) was down:
HTTP code: 404
Response time: 457 ms
Resolved: Nextcloud is back up in 296aae3.
| gharchive/issue | 2023-08-05T16:01:35 | 2025-04-01T06:44:19.842493 | {
"authors": [
"gloriousDan"
],
"repo": "gloriousDan/Homelab-uptime",
"url": "https://github.com/gloriousDan/Homelab-uptime/issues/64",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1287628163 | Style NFT detail page
This styles the NFT detail page (/nft/[address]) page with a developer-y table of attributes and their values.
Solana addresses are copyable, and links are clickable. Traits (if present) are listed under a separate table.
Examples:
https://nftoken-git-l-5177-create-nft-detail-page.luma-dev.com/nft/uERDSnn8x7Hjv6zPMFpGXrbXEvGVwwBAaViFrDdjXr6
https://nftoken-git-l-5177-create-nft-detail-page.luma-dev.com/nft/GtmTbLRKJxPAo4ZREr8sWq928sbG9xJo845U5eQNmL2f
@vpontis Just curious, why did you choose to load the data both in getServerSideProps and on the client side with SWR? I can follow the same data loading approach with the collection page.
@vpontis Just curious, why did you choose to load the data both in getServerSideProps and on the client side with SWR? I can follow the same data loading approach with the collection page.
getServerSideProps — This allow us to have initial data when the page loads. That means there is no loading state and search engines can pick up our page.
swr — We want to refresh data automatically if you switch between tabs.
I’ve added a button that leads to the collection page but commented it out for now since the collection page doesn’t exist yet. Having another two levels of data (nft.collection attributes and maybe nft.collection.traits) on the NFT page felt like it would make the page too complex, so I’ll have a similar table of attributes/traits on the collection page a click away.
| gharchive/pull-request | 2022-06-28T16:49:57 | 2025-04-01T06:44:19.847418 | {
"authors": [
"benborgers",
"vpontis"
],
"repo": "glow-xyz/nftoken",
"url": "https://github.com/glow-xyz/nftoken/pull/155",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
2562807093 | Popover not working on mobile
Description
Popover component not working as expected on mobile
CodeSandbox/Snack link
https://gluestack.io/ui/docs/components/popover
Steps to reproduce
use https://gluestack.io/ui/docs/components/popover component on mobile
but it's not working
gluestack-ui Version
latest
Platform
[X] Expo
[ ] React Native CLI
[ ] Next
[ ] Web
[ ] Android
[ ] iOS
Other Platform
No response
Additional Information
No response
hi @boussaidev, could you please specify what exactly the problem you are facing in using Popover component in mobile.
+1 here, on mobile the backdrop appears with no content. I also noticed a regression with toasts too. All on V1, separate issue I know, toasts are throwing a native error.
+1 here. Popover Works on iOS, but doesn't work on Android.
please install the latest version of @gluestack-ui/popover.
| gharchive/issue | 2024-10-02T23:18:11 | 2025-04-01T06:44:19.896652 | {
"authors": [
"GiuSantos0",
"JavanPoirier",
"boussaidev",
"rajat693"
],
"repo": "gluestack/gluestack-ui",
"url": "https://github.com/gluestack/gluestack-ui/issues/2497",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
243293989 | Add classifiers to setup.py
Without language classifiers in setup.py this project gets marked as not supporting Python 3 when using a tool like caniusepython3.com to check Python 3 compatibility. I've added the basic classifiers, the most important ones being the language ones.
Coverage remained the same at 100.0% when pulling fd01b1f7b6bcea88967ef3d6890d7967032a9d0a on dsanders11:patch-1 into 3d8792e4a9d4ef29eec316e33a35af02b932f167 on glyph:master.
Thanks for noticing this oversight; looking forward to updating this in the next release.
| gharchive/pull-request | 2017-07-17T04:20:17 | 2025-04-01T06:44:19.971083 | {
"authors": [
"coveralls",
"dsanders11",
"glyph"
],
"repo": "glyph/automat",
"url": "https://github.com/glyph/automat/pull/74",
"license": "mit",
"license_type": "permissive",
"license_source": "bigquery"
} |
1766669905 | 🛑 i 學園 is down
In 63a78c5, i 學園 (https://istudy.ntut.edu.tw/mooc/index.php) was down:
HTTP code: 0
Response time: 0 ms
Resolved: i 學園 is back up in d487364.
| gharchive/issue | 2023-06-21T04:47:32 | 2025-04-01T06:44:19.995045 | {
"authors": [
"gnehs"
],
"repo": "gnehs/ntut-uptime",
"url": "https://github.com/gnehs/ntut-uptime/issues/797",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1268380497 | Upgrade to 5.18
Is it possible to upgrade the .deb kernels to 5.18.3 or higher?
Thanks
Ok great. Thanks
New kernel is built, but I need to test on my hardware and make sure everything is legit.
Ok great. Thanks. Let me know when is ready so I can test it as well.
@giovannicaligaris Deb packages are live in the 1.1.0 PR branch: https://github.com/gnickm/sof-essx8336-debian-fix/tree/version/1.1.0/dist
I installed and the sound card is detected. After running setup-pulseaudio.sh I can hear a crackling sound when the tablet powers on and thats about it. No sound after that. I tried disabling the pulseaudio setup and I get the same result.
Very close now, but I am not sure what it could be.
sudo dmesg | grep sof
[sudo] password for mobian:
[ 1.019790] PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
[ 1.019793] software IO TLB: mapped [mem 0x00000000694fe000-0x000000006d4fe000] (64MB)
[ 4.385777] xor: measuring software checksum speed
[ 5.096151] Modules linked in: rtsx_usb(+) btrfs blake2b_generic xor sd_mod t10_pi crc64_rocksoft crc64 crc_t10dif crct10dif_generic uas usb_storage raid6_pq zstd_compress libcrc32c crc32c_generic hid_generic usbhid hid i915 drm_buddy i2c_algo_bit drm_dp_helper spi_pxa2xx_platform dw_dmac dw_dmac_core cec mmc_block rc_core ahci ttm libahci drm_kms_helper libata xhci_pci xhci_hcd drm sdhci_pci scsi_mod usbcore cqhci sdhci i2c_i801 crct10dif_pclmul crct10dif_common crc32_pclmul crc32c_intel mmc_core i2c_smbus intel_lpss_pci scsi_common intel_lpss idma64 usb_common wmi battery video pinctrl_jasperlake button
[ 13.380855] sof-audio-pci-intel-icl 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if info 0x040100
[ 13.380983] sof-audio-pci-intel-icl 0000:00:1f.3: enabling device (0000 -> 0002)
[ 13.381255] sof-audio-pci-intel-icl 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if 0x040100
[ 13.381392] sof-audio-pci-intel-icl 0000:00:1f.3: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])
[ 13.388454] sof-audio-pci-intel-icl 0000:00:1f.3: use msi interrupt mode
[ 13.450141] sof-audio-pci-intel-icl 0000:00:1f.3: NHLT_DEVICE_I2S detected, ssp_mask 0x1
[ 13.450150] sof-audio-pci-intel-icl 0000:00:1f.3: hda codecs found, mask 4
[ 13.481723] sof-audio-pci-intel-icl 0000:00:1f.3: firmware: direct-loading firmware intel/sof/sof-jsl.ri
[ 13.481737] sof-audio-pci-intel-icl 0000:00:1f.3: Firmware info: version 2:1:1-3964a
[ 13.481739] sof-audio-pci-intel-icl 0000:00:1f.3: Firmware: ABI 3:21:0 Kernel ABI 3:19:1
[ 13.481741] sof-audio-pci-intel-icl 0000:00:1f.3: warn: FW ABI is more recent than kernel
[ 13.481749] sof-audio-pci-intel-icl 0000:00:1f.3: unknown sof_ext_man header type 3 size 0x30
[ 13.585078] sof-audio-pci-intel-icl 0000:00:1f.3: Firmware info: version 2:1:1-3964a
[ 13.585086] sof-audio-pci-intel-icl 0000:00:1f.3: Firmware: ABI 3:21:0 Kernel ABI 3:19:1
[ 13.585088] sof-audio-pci-intel-icl 0000:00:1f.3: warn: FW ABI is more recent than kernel
[ 13.670488] sof-essx8336 sof-essx8336: quirk mask 0x0
[ 13.670495] sof-essx8336 sof-essx8336: quirk SSP0
[ 13.722791] sof-audio-pci-intel-icl 0000:00:1f.3: firmware: direct-loading firmware intel/sof-tplg/sof-jsl-es8336-ssp0.tplg
[ 13.722813] sof-audio-pci-intel-icl 0000:00:1f.3: Topology: ABI 3:20:0 Kernel ABI 3:19:1
[ 13.722817] sof-audio-pci-intel-icl 0000:00:1f.3: warn: topology ABI is more recent than kernel
[ 13.723073] sof-essx8336 sof-essx8336: ASoC: Parent card not yet available, widget card binding deferred
[ 13.830283] sof-essx8336 sof-essx8336: hda_dsp_hdmi_build_controls: no PCM in topology for HDMI converter 3
[ 13.830290] sof-essx8336 sof-essx8336: hda_dsp_hdmi_build_controls: no PCM in topology for HDMI converter 4
[ 13.850061] input: sof-essx8336 Headset as /devices/pci0000:00/0000:00:1f.3/sof-essx8336/sound/card0/input18
[ 13.850124] input: sof-essx8336 HDMI/DP,pcm=5 as /devices/pci0000:00/0000:00:1f.3/sof-essx8336/sound/card0/input19
[ 13.850177] input: sof-essx8336 HDMI/DP,pcm=6 as /devices/pci0000:00/0000:00:1f.3/sof-essx8336/sound/card0/input20
[ 13.850228] input: sof-essx8336 HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:1f.3/sof-essx8336/sound/card0/input21
[ 13.722791] sof-audio-pci-intel-icl 0000:00:1f.3: firmware: direct-loading firmware intel/sof-tplg/sof-jsl-es8336-ssp0.tplg
[ 13.723073] sof-essx8336 sof-essx8336: ASoC: Parent card not yet available, widget card binding deferred
Ugh, same issue as in #2 . You have a board that isn't supported by the sof yet.
Found an interesting message just now about why ES8336 support sucks so bad in linux: https://lore.kernel.org/all/20220308192610.392950-6-pierre-louis.bossart@linux.intel.com/T/
As I suspected these chips were developed specifically for Windows.
@gnickm It is working on Manjaro for some reason. I wonder if they patched something
On Manjaro
sudo dmesg | grep sof
[sudo] password for manjaro:
[ 1.169687] PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
[ 1.169690] software IO TLB: mapped [mem 0x000000006997f000-0x000000006d97f000] (64MB)
[ 6.758550] sof-audio-pci-intel-icl 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if info 0x040100
[ 6.758723] sof-audio-pci-intel-icl 0000:00:1f.3: enabling device (0000 -> 0002)
[ 6.759208] sof-audio-pci-intel-icl 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if 0x040100
[ 6.759542] sof-audio-pci-intel-icl 0000:00:1f.3: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])
[ 6.768454] sof-audio-pci-intel-icl 0000:00:1f.3: use msi interrupt mode
[ 6.814964] sof-audio-pci-intel-icl 0000:00:1f.3: NHLT_DEVICE_I2S detected, ssp_mask 0x1
[ 6.814983] sof-audio-pci-intel-icl 0000:00:1f.3: hda codecs found, mask 4
[ 6.823771] sof-audio-pci-intel-icl 0000:00:1f.3: Firmware info: version 2:1:1-3964a
[ 6.823783] sof-audio-pci-intel-icl 0000:00:1f.3: Firmware: ABI 3:21:0 Kernel ABI 3:19:1
[ 6.823789] sof-audio-pci-intel-icl 0000:00:1f.3: warn: FW ABI is more recent than kernel
[ 6.823804] sof-audio-pci-intel-icl 0000:00:1f.3: unknown sof_ext_man header type 3 size 0x30
[ 6.924390] sof-audio-pci-intel-icl 0000:00:1f.3: Firmware info: version 2:1:1-3964a
[ 6.924408] sof-audio-pci-intel-icl 0000:00:1f.3: Firmware: ABI 3:21:0 Kernel ABI 3:19:1
[ 6.924416] sof-audio-pci-intel-icl 0000:00:1f.3: warn: FW ABI is more recent than kernel
[ 6.973812] sof-essx8336 sof-essx8336: quirk mask 0x0
[ 6.973825] sof-essx8336 sof-essx8336: quirk SSP0
[ 6.981739] sof-audio-pci-intel-icl 0000:00:1f.3: Topology: ABI 3:21:0 Kernel ABI 3:19:1
[ 6.981751] sof-audio-pci-intel-icl 0000:00:1f.3: warn: topology ABI is more recent than kernel
[ 6.982076] sof-essx8336 sof-essx8336: ASoC: Parent card not yet available, widget card binding deferred
[ 7.069796] sof-essx8336 sof-essx8336: hda_dsp_hdmi_build_controls: no PCM in topology for HDMI converter 3
[ 7.069809] sof-essx8336 sof-essx8336: hda_dsp_hdmi_build_controls: no PCM in topology for HDMI converter 4
[ 7.087922] input: sof-essx8336 Headset as /devices/pci0000:00/0000:00:1f.3/sof-essx8336/sound/card0/input13
[ 7.088065] input: sof-essx8336 HDMI/DP,pcm=5 as /devices/pci0000:00/0000:00:1f.3/sof-essx8336/sound/card0/input14
[ 7.088173] input: sof-essx8336 HDMI/DP,pcm=6 as /devices/pci0000:00/0000:00:1f.3/sof-essx8336/sound/card0/input15
[ 7.088272] input: sof-essx8336 HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:1f.3/sof-essx8336/sound/card0/input16
@giovannicaligaris Weird, I thought the Parent card not yet available meant it didn't work... What does aplay -l and arecord -l give you?
aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: sofessx8336 [sof-essx8336], device 0: ES8336 (*) []
Subdevices: 1/1
Subdevice #0: subdevice #0
card 0: sofessx8336 [sof-essx8336], device 5: HDMI 1 (*) []
Subdevices: 1/1
Subdevice #0: subdevice #0
card 0: sofessx8336 [sof-essx8336], device 6: HDMI 2 (*) []
Subdevices: 1/1
Subdevice #0: subdevice #0
card 0: sofessx8336 [sof-essx8336], device 7: HDMI 3 (*) []
Subdevices: 1/1
Subdevice #0: subdevice #0
Good news, I figured out!
For some reason your setup-sof.sh script messes something up with the sof folders.
I started from scratch and all I did was install your custom kernel and install firmware-sof-signed on Debian. Restarted and then on the terminal type alsamixer and with F6 selected the sof sound card. Increased all the volumes and now there is sound!
Thanks a lot man. Your kernel now makes my tablet usable.
apt-cache policy firmware-sof-signed
firmware-sof-signed:
Installed: 2.1.1-1
Candidate: 2.1.1-1
Version table:
*** 2.1.1-1 500
500 http://deb.debian.org/debian bookworm/non-free amd64 Packages
500 http://deb.debian.org/debian unstable/non-free amd64 Packages
100 /var/lib/dpkg/status
Actually, setup-alsa.sh has to be run at least once.
Can you show me how you built the kernel?
Can you show me how you built the kernel?
@giovannicaligaris https://github.com/gnickm/sof-essx8336-debian-fix/blob/version/1.1.0/build-kernel.sh
@gnickm which file did you add CONFIG_SND_SOC_INTEL_SOF_ES8336_MACH=m
| gharchive/issue | 2022-06-11T21:53:10 | 2025-04-01T06:44:20.028146 | {
"authors": [
"giovannicaligaris",
"gnickm",
"junocomp"
],
"repo": "gnickm/sof-essx8336-debian-fix",
"url": "https://github.com/gnickm/sof-essx8336-debian-fix/issues/1",
"license": "Apache-2.0",
"license_type": "permissive",
"license_source": "github-api"
} |
858908547 | Drain Account app
This PR adds a new Safe app to transfer all assets to another address.
I've also replaced Lerna with Yarn Workspaces because it was easier to get the tests run this way.
Closes #114
I have read the CLA Document and I hereby sign the CLA
Thanks Mikhail! I've fixed the issues and ran prettier on the app folder.
What would be the best way to make sure the transaction succeeded and the assets are transferred?
I've tried polling the /balances endpoint but it seems to take very long to update.
What would be the best way to make sure the transaction succeeded and the assets are transferred?
I've tried polling the /balances endpoint but it seems to take very long to update.
I think for now showing a message like "The transaction has been created, refresh the app when it is executed" is the most weighted solution between effort/quality 😃 Polling is also fine.
Multisig wallets UX is an ongoing research topic we have. For a regular wallet, you could just use promises to wait until the transaction is executed. For a multi-sig wallet, it's a little bit trickier because the transaction is created, but the execution can take some time (as other people could have to sign it)
Thanks for the detailed answer, @mikheevm! Sounds reasonable.
I'll add the message you suggested.
@lukasschor any critique/suggestions from your side?
Works very well and is definitely a valuable app! Few comments
The icon you used is open source / free to use, right?
The Safe address in the top-right is a bit redundant as it's already displayed in the sidebar. Is there a reason that it is again displayed there?
I would personally prefer a contained button at the bottom as it is the primary action and currently not really prominent. The style you used is usually applied to secondary actions in the app.
When I trigger a transaction and cancel the modal, the loading spinner will disappear and the input field is available again. Which is what I would expect. However, when I trigger a transaction and select one of the sub-transactions in the modal and only then close the modal, the loading spinner is spinning indefinitely. Not sure if this is an issue related to the app or that we do not communicate the closed modal via the SDK in case the user is in the detail view of one transaction (could be as we only launched this feature recently). Not that big of an issue though, as there is the cancel button.
Overall, nothing that I think necessarily needs to be addressed before merging this. Just some improvement suggestions.
Thanks Lukas!
The icon is from here: https://www.flaticon.com/free-icon/withdraw_1682382 – I'ved added an attribution in the SVG file. ✅
Good point about the button, fixed! ✅
Regarding the infinite loader after closing the popup from the transaction details, I'm also not sure why this isn't rejecting the send promise. Might need to investigate in the SDK.
Regarding the infinite loader after closing the popup from the transaction details, I'm also not sure why this isn't rejecting the send promise. Might need to investigate in the SDK.
This sounds like a bug in the integration on safe-react side. Could you please create a ticket so we don't forget about this issue?
Created https://github.com/gnosis/safe-react/issues/2178 👍
Feedback from @rmeissner: need to add some tests for the logic to make the app production-ready.
Thanks @dasanra! Done ✅
Thanks @dasanra! I've removed the console logs and cleaned up the dependencies. 👍
@rmeissner I've addressed your comments, thank you!
The test and lint jobs are failing with this error:
error https://registry.yarnpkg.com/lodash/-/lodash-4.17.20.tgz: Extracting tar content of undefined failed, the file appears to be corrupt: "ENOENT: no such file or directory, stat '/usr/local/share/.cache/yarn/v6/npm-lodash-4.17.20-b44a9b6297bcb698f1c51a3545a2b3b368d59c52-integrity/node_modules/lodash/fp/compact.js'"
Looks like something on the Yarn repo side? I'll try to rebuild later today.
@katspaugh , should it go through QA? It was not tested and approved by QA
Not approved by QA:
[ ] Decimals are not taken into account for the token value:
Drain account app :
Actual values on balance page where decimals are taken into account:
[ ] no default icon for the tokens without icon as on balance page :
Current result: __
Expected result:
[ ] What is the default sorting ?
@liliya-soroka I thought we were supposed to test individual tickets ourselves and you only to do the regression.
Anyway, thanks for testing so quickly! Could you create a separate issue for everything together?
Except for the last point with the sorting arrow – it's an issue with the Table component from SRC.
Verified
Thanks @liliya-soroka!
| gharchive/pull-request | 2021-04-15T13:56:53 | 2025-04-01T06:44:20.094520 | {
"authors": [
"katspaugh",
"liliya-soroka",
"lukasschor",
"mikheevm"
],
"repo": "gnosis/safe-react-apps",
"url": "https://github.com/gnosis/safe-react-apps/pull/120",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
2000216324 | Remove dappnode.io checkpoint sync provider
What
Remove https://checkpoint-sync-gnosis.dappnode.io/ from list of checkpoint sync providers as it is no longer available
Refs
Actual:
$ curl https://checkpoint-sync-gnosis.dappnode.io
{"code":404,"message":"NOT_FOUND","stacktraces":[]}
Expected:
The site to load similar to https://checkpoint.gnosis.gateway.fm
@unnawut , it is not down actually. Example: https://checkpoint-sync-gnosis.dappnode.io/eth/v1/beacon/genesis
| gharchive/pull-request | 2023-11-18T04:26:49 | 2025-04-01T06:44:20.098135 | {
"authors": [
"unnawut",
"zengzengzenghuy"
],
"repo": "gnosischain/documentation",
"url": "https://github.com/gnosischain/documentation/pull/548",
"license": "CC0-1.0",
"license_type": "permissive",
"license_source": "github-api"
} |
384062317 | Copy-paste woes
when trying to paste text I get this error: bracketed-paste-magic:zle:47: not enough arguments for -U
I am using zsh on Debian Sid if that helps
can you reproduce it with st or xst+other shell?
judging from error it is something to do with zsh code which is being executed at that moment
| gharchive/issue | 2018-11-25T09:22:06 | 2025-04-01T06:44:20.103281 | {
"authors": [
"actionless",
"superE03"
],
"repo": "gnotclub/xst",
"url": "https://github.com/gnotclub/xst/issues/59",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
2055774864 | Errors after launching go-architect in Ubuntu 20.04
Hi.
I tried to compile the project in Ubuntu with make install:
Wails
Version | v2.7.1
Package Manager | apt
System
┌──────────────────────────────────────────────────────────────┐
| OS | Ubuntu |
| Version | 20.04 |
| ID | ubuntu |
| Go Version | go1.19.13 |
| Platform | linux |
| Architecture | amd64 |
| CPU | 11th Gen Intel(R) Core(TM) i5-11400 @ 2.60GHz |
| GPU | unknown (Intel Corporation) - Driver: i915 |
| Memory | 32GB |
└──────────────────────────────────────────────────────────────┘
Dependencies
┌──────────────────────────────────────────────────────────────────────────┐
| Dependency | Package Name | Status | Version |
| *docker | docker.io | Installed | 24.0.2 |
| gcc | build-essential | Installed | 12.8ubuntu1.1 |
| libgtk-3 | libgtk-3-dev | Installed | 3.24.20-0ubuntu1.1 |
| libwebkit | libwebkit2gtk-4.0-dev | Installed | 2.38.6-0ubuntu0.20.04.1 |
| npm | npm | Installed | 8.19.4 |
| *nsis | nsis | Available | 3.05-2 |
| pkg-config | pkg-config | Installed | 0.29.1 |
└──────────────────────── * - Optional Dependency ─────────────────────────┘
Diagnosis
Optional package(s) installation details:
nsis: sudo apt install nsis
SUCCESS Your system is ready for Wails development!
♥ If Wails is useful to you or your company, please consider sponsoring the project:
https://github.com/sponsors/leaanthony
CGO_ENABLED=1 wails build -clean
Wails CLI v2.7.1
Build Options
Platform(s) | linux/amd64
Compiler | /home/userg/go1.19/go1.19.13/bin/go
Skip Bindings | false
Build Mode | production
Devtools | false
Frontend Directory | ./go-architect/frontend
Obfuscated | false
Skip Frontend | false
Compress | false
Package | true
Clean Bin Dir | true
LDFlags |
Tags | []
Race Detector | false
Warning: go.mod is using Wails '2.6.0' but the CLI is 'v2.7.1'. Consider updating your project's go.mod file.
Building target: linux/amd64
• Generating bindings: 2023/12/25 18:34:38 KnownStructs: comments.CommentsMetrics coupling.DependencyCoupling dependency.ModuleDependencyGraph dsm.DependencyStructureMatrix git.VCSAnalysisInfo gocyclo.GoCycloOutput instability.PackageInstability interfaces.InterfaceInfo interfaces.InterfaceMetrics loc.FileLOC loc.PackageLOC loc.ProjectLOC project.ProjectInfo repository.RepositoryInfo storage.HistoricalMetrics storage.Metrics storage.Project storage.ProjectList storage.SelectedProject types.PackageTypes types.ProjectTypes
storage.Metrics
Done.
• Installing frontend dependencies: Done.
• Compiling frontend: Done.
• Compiling application: Done.
• Packaging application: Done.
Built '/home/userg/projects/tools/go-architect/build/bin/go-architect' in 35.098s.
♥ If Wails is useful to you or your company, please consider sponsoring the project:
https://github.com/sponsors/leaanthony
make[1]: выход из каталога «/home/userg/projects/tools/go-architect»
Your Go-Architect executable was generated in current folder
After launch ./go-architect received the following errors:
Overriding existing handler for signal 10. Set JSC_SIGNAL_FOR_GC if you want WebKit to use a different signal
signal 11 received but handler not on signal stack
fatal error: non-Go code set up signal handler without SA_ONSTACK flag
runtime stack:
runtime.throw({0xb783bc?, 0xc000074000?})
/home/userg/go1.19/go1.19.13/src/runtime/panic.go:1047 +0x5d fp=0xc0000eebe0 sp=0xc0000eebb0 pc=0x43d95d
runtime.sigNotOnStack(0xb)
/home/userg/go1.19/go1.19.13/src/runtime/signal_unix.go:1024 +0x65 fp=0xc0000eec00 sp=0xc0000eebe0 pc=0x4541c5
runtime.adjustSignalStack(0xb, 0x13527a0, 0xc0000eec90)
/home/userg/go1.19/go1.19.13/src/runtime/signal_unix.go:581 +0x28b fp=0xc0000eec68 sp=0xc0000eec00 pc=0x452d4b
runtime.sigtrampgo(0xb, 0xc0000eef70, 0xc0000eee40)
/home/userg/go1.19/go1.19.13/src/runtime/signal_unix.go:469 +0x15b fp=0xc0000eece0 sp=0xc0000eec68 pc=0x4528bb
runtime.sigtramp()
/home/userg/go1.19/go1.19.13/src/runtime/sys_linux_amd64.s:359 +0x46 fp=0xc0000eed30 sp=0xc0000eece0 pc=0x470286
goroutine 17 [syscall, locked to thread]:
runtime.goexit()
/home/userg/go1.19/go1.19.13/src/runtime/asm_amd64.s:1594 +0x1 fp=0xc000084fe8 sp=0xc000084fe0 pc=0x46e661
goroutine 1 [running, locked to thread]:
goroutine running on other thread; stack unavailable
goroutine 2 [force gc (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:363 +0xd6 fp=0xc000070fb0 sp=0xc000070f90 pc=0x4405b6
runtime.goparkunlock(...)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:369
runtime.forcegchelper()
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:302 +0xad fp=0xc000070fe0 sp=0xc000070fb0 pc=0x44044d
runtime.goexit()
/home/userg/go1.19/go1.19.13/src/runtime/asm_amd64.s:1594 +0x1 fp=0xc000070fe8 sp=0xc000070fe0 pc=0x46e661
created by runtime.init.6
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:290 +0x25
goroutine 3 [GC sweep wait]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:363 +0xd6 fp=0xc000071790 sp=0xc000071770 pc=0x4405b6
runtime.goparkunlock(...)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:369
runtime.bgsweep(0x0?)
/home/userg/go1.19/go1.19.13/src/runtime/mgcsweep.go:278 +0x8e fp=0xc0000717c8 sp=0xc000071790 pc=0x42cdce
runtime.gcenable.func1()
/home/userg/go1.19/go1.19.13/src/runtime/mgc.go:178 +0x26 fp=0xc0000717e0 sp=0xc0000717c8 pc=0x421a86
runtime.goexit()
/home/userg/go1.19/go1.19.13/src/runtime/asm_amd64.s:1594 +0x1 fp=0xc0000717e8 sp=0xc0000717e0 pc=0x46e661
created by runtime.gcenable
/home/userg/go1.19/go1.19.13/src/runtime/mgc.go:178 +0x6b
goroutine 4 [GC scavenge wait]:
runtime.gopark(0xc000044070?, 0xf12930?, 0x1?, 0x0?, 0x0?)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:363 +0xd6 fp=0xc000071f70 sp=0xc000071f50 pc=0x4405b6
runtime.goparkunlock(...)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:369
runtime.(*scavengerState).park(0x13519a0)
/home/userg/go1.19/go1.19.13/src/runtime/mgcscavenge.go:389 +0x53 fp=0xc000071fa0 sp=0xc000071f70 pc=0x42ae73
runtime.bgscavenge(0x0?)
/home/userg/go1.19/go1.19.13/src/runtime/mgcscavenge.go:617 +0x45 fp=0xc000071fc8 sp=0xc000071fa0 pc=0x42b445
runtime.gcenable.func2()
/home/userg/go1.19/go1.19.13/src/runtime/mgc.go:179 +0x26 fp=0xc000071fe0 sp=0xc000071fc8 pc=0x421a26
runtime.goexit()
/home/userg/go1.19/go1.19.13/src/runtime/asm_amd64.s:1594 +0x1 fp=0xc000071fe8 sp=0xc000071fe0 pc=0x46e661
created by runtime.gcenable
/home/userg/go1.19/go1.19.13/src/runtime/mgc.go:179 +0xaa
goroutine 5 [finalizer wait]:
runtime.gopark(0x13527a0?, 0xc000007860?, 0x0?, 0x0?, 0xc000070770?)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:363 +0xd6 fp=0xc000070628 sp=0xc000070608 pc=0x4405b6
runtime.goparkunlock(...)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:369
runtime.runfinq()
/home/userg/go1.19/go1.19.13/src/runtime/mfinal.go:180 +0x10f fp=0xc0000707e0 sp=0xc000070628 pc=0x420b8f
runtime.goexit()
/home/userg/go1.19/go1.19.13/src/runtime/asm_amd64.s:1594 +0x1 fp=0xc0000707e8 sp=0xc0000707e0 pc=0x46e661
created by runtime.createfing
/home/userg/go1.19/go1.19.13/src/runtime/mfinal.go:157 +0x45
goroutine 6 [chan receive]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:363 +0xd6 fp=0xc0000726d0 sp=0xc0000726b0 pc=0x4405b6
runtime.chanrecv(0xc0001fc900, 0xc0000727a8, 0x1)
/home/userg/go1.19/go1.19.13/src/runtime/chan.go:583 +0x49b fp=0xc000072760 sp=0xc0000726d0 pc=0x40eafb
runtime.chanrecv2(0x0?, 0x0?)
/home/userg/go1.19/go1.19.13/src/runtime/chan.go:447 +0x18 fp=0xc000072788 sp=0xc000072760 pc=0x40e638
github.com/wailsapp/wails/v2/internal/frontend/desktop/linux.(*Frontend).startRequestProcessor(0xc0001fdbc0)
/home/userg/go/pkg/mod/github.com/wailsapp/wails/v2@v2.6.0/internal/frontend/desktop/linux/frontend.go:494 +0x5d fp=0xc0000727c8 sp=0xc000072788 pc=0xa026dd
github.com/wailsapp/wails/v2/internal/frontend/desktop/linux.NewFrontend.func3()
/home/userg/go/pkg/mod/github.com/wailsapp/wails/v2@v2.6.0/internal/frontend/desktop/linux/frontend.go:179 +0x26 fp=0xc0000727e0 sp=0xc0000727c8 pc=0xa01186
runtime.goexit()
/home/userg/go1.19/go1.19.13/src/runtime/asm_amd64.s:1594 +0x1 fp=0xc0000727e8 sp=0xc0000727e0 pc=0x46e661
created by github.com/wailsapp/wails/v2/internal/frontend/desktop/linux.NewFrontend
/home/userg/go/pkg/mod/github.com/wailsapp/wails/v2@v2.6.0/internal/frontend/desktop/linux/frontend.go:179 +0x53c
goroutine 7 [chan receive]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
/home/userg/go1.19/go1.19.13/src/runtime/proc.go:363 +0xd6 fp=0xc000072ec8 sp=0xc000072ea8 pc=0x4405b6
runtime.chanrecv(0xc0001fc8a0, 0xc000072fa8, 0x1)
/home/userg/go1.19/go1.19.13/src/runtime/chan.go:583 +0x49b fp=0xc000072f58 sp=0xc000072ec8 pc=0x40eafb
runtime.chanrecv2(0x0?, 0x0?)
/home/userg/go1.19/go1.19.13/src/runtime/chan.go:447 +0x18 fp=0xc000072f80 sp=0xc000072f58 pc=0x40e638
github.com/wailsapp/wails/v2/internal/frontend/desktop/linux.(*Frontend).startMessageProcessor(0x0?)
/home/userg/go/pkg/mod/github.com/wailsapp/wails/v2@v2.6.0/internal/frontend/desktop/linux/frontend.go:208 +0x59 fp=0xc000072fc8 sp=0xc000072f80 pc=0xa01219
github.com/wailsapp/wails/v2/internal/frontend/desktop/linux.NewFrontend.func4()
/home/userg/go/pkg/mod/github.com/wailsapp/wails/v2@v2.6.0/internal/frontend/desktop/linux/frontend.go:182 +0x26 fp=0xc000072fe0 sp=0xc000072fc8 pc=0xa01126
runtime.goexit()
/home/userg/go1.19/go1.19.13/src/runtime/asm_amd64.s:1594 +0x1 fp=0xc000072fe8 sp=0xc000072fe0 pc=0x46e661
created by github.com/wailsapp/wails/v2/internal/frontend/desktop/linux.NewFrontend
/home/userg/go/pkg/mod/github.com/wailsapp/wails/v2@v2.6.0/internal/frontend/desktop/linux/frontend.go:182 +0x57f
same issue
Thanks for sharing, I'm creating a similar environment (Ubuntu 20.04, Go 1.19.13, Npm 8.19.4) to reproduce the problem.
As soon as I can reproduce the issue I will try to find out the reason.
Thanks for sharing, I'm creating a similar environment (Ubuntu 20.04, Go 1.19.13, Npm 8.19.4) to reproduce the problem. As soon as I can reproduce the issue I will try to find out the reason.
thanks having same problem
Guys, could you check if the ubuntu-2004-error branch fixed the problem?
git clone -b ubuntu-2004-error https://github.com/go-architect/go-architect.git
cd go-architect
make install
./go-architect
or
cd go-architect
git pull
git checkout ubuntu-2004-error
make install
./go-architect
So, not on ubuntu, but on manjaro i tested c93eca7c, and at first sight - everythink is ok - i could add my project and get metricts for it, thanks a lot!
Thanks @Gorbovskyfallz , can you share in this issue your environment (Manjaro version, Go version and NPM version)?
@fdaines
npm: 10.2.5
wails: 2.7.1
manjaro: 6.1.69-1-MANJARO #1 SMP PREEMPT_DYNAMIC Thu Dec 21 12:29:38 UTC 2023 x86_64 GNU/Linux
go version : 1.21.5
Guys, could you check if the ubuntu-2004-error branch fixed the problem?
git clone -b ubuntu-2004-error https://github.com/go-architect/go-architect.git
cd go-architect
make install
./go-architect
or
cd go-architect
git pull
git checkout ubuntu-2004-error
make install
./go-architect
thanks i can run ./go-architect without any problems and can the other metrics and depencies graph
| gharchive/issue | 2023-12-25T16:34:28 | 2025-04-01T06:44:20.187579 | {
"authors": [
"Gorbovskyfallz",
"fdaines",
"lechuhuuha",
"userg3003"
],
"repo": "go-architect/go-architect",
"url": "https://github.com/go-architect/go-architect/issues/19",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
578997958 | TLS. #minor
Add a way to specify extra certificates to use to connect to the chef server.
Include #153
| gharchive/pull-request | 2020-03-11T04:13:56 | 2025-04-01T06:44:20.189326 | {
"authors": [
"MarkGibbons"
],
"repo": "go-chef/chef",
"url": "https://github.com/go-chef/chef/pull/156",
"license": "apache-2.0",
"license_type": "permissive",
"license_source": "bigquery"
} |
1078963261 | refactor: Refactor test
Checklist:
[x] Tests for the changes have been added (for bug fixes / features / refactoring)
Codecov Report
Merging #189 (bf9f31f) into main (50c77fb) will increase coverage by 0.58%.
The diff coverage is 79.56%.
@@ Coverage Diff @@
## main #189 +/- ##
==========================================
+ Coverage 58.30% 58.89% +0.58%
==========================================
Files 10 10
Lines 343 343
==========================================
+ Hits 200 202 +2
+ Misses 135 132 -3
- Partials 8 9 +1
Impacted Files
Coverage Δ
internal/faker/boolean.go
100.00% <ø> (ø)
internal/faker/uuid.go
100.00% <ø> (ø)
internal/server/server.go
0.00% <0.00%> (ø)
internal/server/handler.go
27.27% <35.71%> (+1.81%)
:arrow_up:
internal/faker/internet.go
85.71% <85.71%> (ø)
internal/faker/person.go
95.65% <91.66%> (ø)
internal/faker/faker.go
95.60% <94.59%> (ø)
... and 3 more
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 19d766b...bf9f31f. Read the comment docs.
| gharchive/pull-request | 2021-12-13T20:18:52 | 2025-04-01T06:44:20.212118 | {
"authors": [
"alexander-melentyev",
"codecov-commenter"
],
"repo": "go-dummy/dummy",
"url": "https://github.com/go-dummy/dummy/pull/189",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
679577374 | Add OpenID/OAuth2 Role Mappings
Description
I have my Gitea instance setup with OIDC and it works well. However at the moment anyone with an account in the realm Gitea accesses also has access to Gitea, regardless of their groups/roles/permissions. It's not a "break" for me as I don't mind providing Gitea access to all users, however I would like to be able to control it and especially control admin access or even organizational access from within my OIDC provider (keycloak in this case).
A lot of clients that use OpenID/Oauth2 allow to create roles and role mappings so you can assign a user a role in the identity manager and they automatically receive that role in the client application.
I am assuming that this would be a new feature request, as I haven't found this feature in Gitea, the docs, my web searching, or the code I've looked through thus far.
I believe this is a dupe of https://github.com/go-gitea/gitea/issues/10016
Closed as duplicated.
| gharchive/issue | 2020-08-15T13:43:53 | 2025-04-01T06:44:20.229599 | {
"authors": [
"Morriz",
"lunny",
"pard68"
],
"repo": "go-gitea/gitea",
"url": "https://github.com/go-gitea/gitea/issues/12493",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1000223308 | [Feature Request]: Decentralize, one gitea account, involve in the activities amoung the gitea instance all over the world.
Feature Description
Just inspired by Matrix Protocol, https://matrix.org/
Each server hold a bunch of accounts, but the accounts in different server can also interact each other.
Currently, Github is the center of git-based open source development. But itself is not open source at all. Github might be influenced by politics or other non-tech related issue.
Gitea is completely open source, and the gitea instance is distributed all over the world.
It might be a large list about what exactly the feature in details include.
But it would be nice to make exist gitea instance form a large network. And make users of those instance feel that they are in a gap-less Gitea decentralized network.
All the activities in a single server can be done between servers.
For example, user A @A@gitea.hello.world might clone a project gitea.nice.peace:/B/project_sample from @B@gitea.nice.peace. A's fork will display the original repo position, and he can easily submit new merge request back.
If a user on gitea.nice.peace [at] A, the server gitea.nice.peace will send a request to server gitea.hello.world, and A will receive the notification. While this is just an example, more inter-server communication should be included.
Screenshots
No response
With the power of decentralization, it have potential to become the large ecosystem just as Github has done.
@lunny @zeripath what is your opinion?
By the way, the feature of real-time cross-server mirroring could be added.
This is not just git-mirroring, but include all the activities, issues, merge request, milestone, and wiki. It is not sync at specific time,
When ever new activities take place, the mirroring server will sync each other, and make all the nodes always up to date.
The open source development won't be affected if one of those server down.
It can make open source software development always available to every one in the world without affected by politics and situation.
You might be interested in the work around FedeProxy and forgefed, as well as all tickets tagged federation.
Please have a look at https://github.com/go-gitea/gitea/issues?q=is%3Aopen+is%3Aissue+label%3Atheme%2Ffederation.
I think your proposal is already known.
If I understood the conversation correctly, one of the main features for Gitea 1.16 or 1.17 that are currently in development is exactly enabling federation between different (Gitea) instances.
Especially interesting for that might be #16827.
Please go to known issues to discuss further.
| gharchive/issue | 2021-09-19T06:22:42 | 2025-04-01T06:44:20.237312 | {
"authors": [
"delvh",
"evan0greenup",
"lunny",
"mscherer"
],
"repo": "go-gitea/gitea",
"url": "https://github.com/go-gitea/gitea/issues/17087",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
349965726 | Extra entry for previous commit on dashboard with "git push --mirror"
Gitea version (or commit ref): 1.5.0
Git version: 2.18.0
Operating system: Debian GNU/Linux 9.5
Database (use [x]):
[ ] PostgreSQL
[ ] MySQL
[ ] MSSQL
[x] SQLite
Can you reproduce the bug at https://try.gitea.io:
[x] Yes: https://try.gitea.io/markshep/duplicate-dashboard-entries
[ ] No
[ ] Not relevant
Log gist:
?
Description
When a client/downstream repo has a post-commit hook that invokes "git push --mirror origin" then whenever a push happens two new entries will appear on the dashboard. There's the expected one for the commit that just happened, but then an extra one that refers to the previous commit, but with with "refs/remotes/origin/master" listed as the branch rather than just "master".
Screenshots
Closing as this seems to be fixed now. I think the fix happened somewhere between 1.6.4 and 1.7.0.
(Dunno why this issue was marked as a question - it isn't!)
| gharchive/issue | 2018-08-13T10:06:00 | 2025-04-01T06:44:20.243497 | {
"authors": [
"markshep"
],
"repo": "go-gitea/gitea",
"url": "https://github.com/go-gitea/gitea/issues/4691",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
229617282 | Correct grammar in APIEmpty documentation
As title
LGTM
Sorry for the mistake. LGTM.
Sorry, I forgot to ask, can you please also re-generate swagger.json ? (make generate)
Done
| gharchive/pull-request | 2017-05-18T10:01:59 | 2025-04-01T06:44:20.245040 | {
"authors": [
"cez81",
"lafriks",
"sapk"
],
"repo": "go-gitea/gitea",
"url": "https://github.com/go-gitea/gitea/pull/1748",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1322475941 | Add missing Tabs on organisation/package view (Frontport #20539)
Frontport #20539
Hotfix https://github.com/go-gitea/gitea/issues/20106
.
| gharchive/pull-request | 2022-07-29T16:47:18 | 2025-04-01T06:44:20.246207 | {
"authors": [
"6543"
],
"repo": "go-gitea/gitea",
"url": "https://github.com/go-gitea/gitea/pull/20540",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
737739249 | Update data from CLDR v32 to v34.
Here's an attempt to do #33 in pieces. Instead of going to v38 we first go to v34. I tried v35.1 but it didn't generate cleanly.
There was one failing test, ru-RU.go:
--- FAIL: TestFmtTimeFull (0.00s)
ru_RU_test.go:858: Expected '9:05:01 Восточная Америка, стандартное время' Got '09:05:01 Восточная Америка, стандартное время'
--- FAIL: TestFmtTimeLong (0.00s)
ru_RU_test.go:889: Expected '9:05:01 EST' Got '09:05:01 EST'
--- FAIL: TestFmtTimeMedium (0.00s)
ru_RU_test.go:915: Expected '9:05:01' Got '09:05:01'
--- FAIL: TestFmtTimeShort (0.00s)
ru_RU_test.go:941: Expected '9:05' Got '09:05'
FAIL
Fixed in a followup commit. The build is now green.
Huh. v35.1 didn't work, but v36.1 did. Closing this PR in favor of #35.
| gharchive/pull-request | 2020-11-06T12:58:57 | 2025-04-01T06:44:20.281187 | {
"authors": [
"bojanz"
],
"repo": "go-playground/locales",
"url": "https://github.com/go-playground/locales/pull/34",
"license": "mit",
"license_type": "permissive",
"license_source": "bigquery"
} |
191858871 | Fixed Readme typo ZRangeBy
Correct me if I'm wrong, but it should by ZrangeBy not ZRangeByScore?
Yeah, thanks
| gharchive/pull-request | 2016-11-27T12:30:25 | 2025-04-01T06:44:20.286682 | {
"authors": [
"nathan-jones",
"vmihailenco"
],
"repo": "go-redis/redis",
"url": "https://github.com/go-redis/redis/pull/428",
"license": "bsd-2-clause",
"license_type": "permissive",
"license_source": "bigquery"
} |
766217793 | run cmd question
in terminal i run “interface print stats” ,
in this lib ，what write ？
"/interface/print/stats" return faslse
it's fine
| gharchive/issue | 2020-12-14T09:50:52 | 2025-04-01T06:44:20.287962 | {
"authors": [
"Circleee"
],
"repo": "go-routeros/routeros",
"url": "https://github.com/go-routeros/routeros/issues/15",
"license": "mit",
"license_type": "permissive",
"license_source": "bigquery"
} |
1814652173 | Prompt Template Improvements
Description
Lays some of the groundwork for LLAMA2 compatibility as well as other future models with complex prompting schemes.
Started small refactoring in pkg/model/loader.go regarding template loading. Currently still a part of ModelLoader, but should be easy to add template loading for situations other than overall prompt templates and the new chat-specific per-message templates
Adds support for new chat-endpoint-specific, per-message templates as an alternative to the existing Role: XYZ sprintf method.
Includes a temporary prompt template as an example, since I have a few questions before we merge in the model-gallery side changes (see )
Minor debug logging changes.
Big thanks to @tmm1 who helped work out the proper LLAMA2 prompt text as well as helping with testing!
Notes for Reviewers
This obviously isn't 100% compatibility with LLAMA2. We'll need some upstream changes for the larger models, as well as the proper BOS/EOS support... but we might as well get started now.
Signed commits
[ ] Yes, I signed my commits.
Thanks for your contribution. I suggest rebasing 13 commits to several reasonable commits. It will be easy for us to split features.
I was planning on squash merging the PR to a single commit, but I can rewrite history into a new branch and merge as several if that's the style we prefer.
@dave-gray101 that's fine, just squash them during merge
| gharchive/pull-request | 2023-07-20T19:17:05 | 2025-04-01T06:44:20.292519 | {
"authors": [
"dave-gray101",
"mudler"
],
"repo": "go-skynet/LocalAI",
"url": "https://github.com/go-skynet/LocalAI/pull/782",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
162850873 | suffix validator only model generation with _validator
closes #578
This change is
Reviewed 3 of 3 files at r1.
Review status: all files reviewed at latest revision, all discussions resolved.
Comments from Reviewable
| gharchive/pull-request | 2016-06-29T06:34:22 | 2025-04-01T06:44:20.297606 | {
"authors": [
"casualjim"
],
"repo": "go-swagger/go-swagger",
"url": "https://github.com/go-swagger/go-swagger/pull/579",
"license": "Apache-2.0",
"license_type": "permissive",
"license_source": "github-api"
} |
1566519057 | Local support for --set-env secrets
local support for --set-env secrets (#309 ). Goblet will now get the secret from secrets manager and set as env variable. If permission is not granted, then the secret will be skipped locally.
local support for --set-env and stage for cloudrun jobs (#312 )
Codecov Report
Base: 88.20% // Head: 88.22% // Increases project coverage by +0.01% :tada:
Coverage data is based on head (b7c2b3a) compared to base (f8db644).
Patch coverage: 90.90% of modified lines in pull request are covered.
:mega: This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more
Additional details and impacted files
@@ Coverage Diff @@
## main #318 +/- ##
==========================================
+ Coverage 88.20% 88.22% +0.01%
==========================================
Files 50 50
Lines 4222 4262 +40
==========================================
+ Hits 3724 3760 +36
- Misses 498 502 +4
Impacted Files
Coverage Δ
goblet/config.py
79.24% <ø> (ø)
goblet/decorators.py
85.51% <ø> (ø)
goblet/infrastructures/alerts.py
84.35% <ø> (ø)
goblet/infrastructures/vpcconnector.py
46.80% <ø> (ø)
goblet/resources/jobs.py
77.94% <ø> (ø)
goblet/revision.py
88.23% <ø> (ø)
goblet/tests/test_app.py
90.54% <ø> (ø)
goblet/tests/test_routes.py
98.44% <ø> (ø)
goblet/backends/cloudrun.py
87.17% <81.81%> (-0.68%)
:arrow_down:
goblet/backends/cloudfunctionv1.py
94.20% <83.33%> (-2.35%)
:arrow_down:
... and 4 more
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.
:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.
| gharchive/pull-request | 2023-02-01T17:28:14 | 2025-04-01T06:44:20.349244 | {
"authors": [
"anovis",
"codecov-commenter"
],
"repo": "goblet/goblet",
"url": "https://github.com/goblet/goblet/pull/318",
"license": "Apache-2.0",
"license_type": "permissive",
"license_source": "github-api"
} |
316178550 | Mutationstests in S# implementieren
Notwendige Implementierungen in S#, um das Mutationstesten durchzuführen
Warte auf #77
Ziel: Mutationstests müssen ausgeführt werden können
Tests können ohne spezielle Anpassung ausgeführt werden
| gharchive/issue | 2018-04-20T08:40:54 | 2025-04-01T06:44:20.351062 | {
"authors": [
"gobo7793"
],
"repo": "gobo7793/ssharp",
"url": "https://github.com/gobo7793/ssharp/issues/69",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
353992904 | Use stringer to reduce duplicate constants in bytecode instruction.go
Finally I found the way to use stringer to remove duplicate constants.
Updated the Make file to add make generate (I wonder there might be a better way)
If this is OK, I'd try using stringer against token.go.
I think this is a great attempt but not seeing as much benefit as the complexity it adds. @eliothedeman what do you think about this?
Code generation is pretty standard practice for interpreters, so I'm not generally opposed. I'll have to think about it a bit more though.
@hachi8833 thanks for the effort! But I think at this stage, maintainability is more important than optimization. And since it's very likely that we'll still add/remove instructions, I think we shouldn't do it now. So I'm closing it now, can revisit in the future.
| gharchive/pull-request | 2018-08-25T07:50:14 | 2025-04-01T06:44:20.353640 | {
"authors": [
"eliothedeman",
"hachi8833",
"st0012"
],
"repo": "goby-lang/goby",
"url": "https://github.com/goby-lang/goby/pull/725",
"license": "mit",
"license_type": "permissive",
"license_source": "bigquery"
} |
281841865 | Object Translation not referring as expected
Operating system or device, Godot version, GPU Model and driver (if graphics related):
Windows 10 64 bit, Godot 3 B1
Issue description:
These codes below wont work:
var p = get_parent().get_node("Player").translation #--------> this works!
translate(Vector3(0,0,p.z-2)) #--------> this works!
p.z -= 2 #--------> this WONT WORK!
have to do this below:
var p = get_parent().get_node("Player")
#we skip this line coz it works anyway
p.translate.z -= 2
That's expected, once you do var p = $Node.translation you're getting a copy of the vector, not a reference to it. The second part works because nodes comes as a reference.
i see.. got it.. pretty confusing there at once.. Thanks vnen.. so we can pull this one out?
| gharchive/issue | 2017-12-13T17:57:49 | 2025-04-01T06:44:20.455026 | {
"authors": [
"AkatDeen",
"vnen"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/14631",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
287841340 | Custom selection region for a node
Godot version:
3 beta 2
Issue description:
Is it possible to define custom region usually its a rectangle to a node that will be used to select node in editor by mouse click ?
My custom node is inside tileset and has unique shape that overlap others. Its hard to select them easily in editor.
Do we have to add one node more to achieve that effect ?
I'm also working on a plugin. My thought is that if there is this push to not bloat the engine and that we should extend by use of plugins, then the plugin interface needs the ability to allow for node selection as it does for built-in node types.
Feature and improvement proposals for the Godot Engine are now being discussed and reviewed in a dedicated Godot Improvement Proposals (GIP) (godotengine/godot-proposals) issue tracker. The GIP tracker has a detailed issue template designed so that proposals include all the relevant information to start a productive discussion and help the community assess the validity of the proposal for the engine.
The main (godotengine/godot) tracker is now solely dedicated to bug reports and Pull Requests, enabling contributors to have a better focus on bug fixing work. Therefore, we are now closing all older feature proposals on the main issue tracker.
If you are interested in this feature proposal, please open a new proposal on the GIP tracker following the given issue template (after checking that it doesn't exist already). Be sure to reference this closed issue if it includes any relevant discussion (which you are also encouraged to summarize in the new proposal). Thanks in advance!
https://github.com/godotengine/godot-proposals/issues/5289
| gharchive/issue | 2018-01-11T16:24:53 | 2025-04-01T06:44:20.460170 | {
"authors": [
"Bladum",
"Calinou",
"Giwayume",
"remorse107"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/15599",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
343841166 | Creating a PhysicalBone crashes the editor
Godot master 779c9d638e7046f285b301aa79f7cf50aa4c2f1c
Windows 10 64 bits
Create a new scene with a root Node
Create a child node of type PhysicalBone: editor crashes.
CC @AndreaCatania
| gharchive/issue | 2018-07-24T00:26:27 | 2025-04-01T06:44:20.462136 | {
"authors": [
"Zylann",
"bojidar-bg"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/20397",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
351797199 | Add volume multiplier to audio stream players.
While tinkering with audio streams I found the controls a bit cumbersome to do simple things. I compared with Unity and UE4, they both have a simple 0-1 volume multiplier (UE4 even calls it "volume multiplier"), all my audio hw and sw I use for recording also have a simple volume control. It would help a lot to have it in the node editor.
As I'm working with some audio streams in Godot right now I'm noting down some further enhancements for audio controls in addition to the multiplier, specifically some much needed loop controls.
Move the loop checkbox from the import to the inspector and make it per node instead of per resource.
Indicate the loop start and end visually in the audio resource editor.
Make the start/end indicators interactive so you can just drag the start and end.
Indicate the time for the sample offsets somewhere. Sometimes it's useful to know the exact offset of a sample, but often as humans we prefer the time in seconds.
| gharchive/issue | 2018-08-18T06:52:39 | 2025-04-01T06:44:20.464267 | {
"authors": [
"spongeboburu"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/21138",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
381145539 | Godot Snap distribution can't export to Android
Godot version:
Godot v3.0.6
OS/device including version:
Ubuntu v18.04.01
OpenJDK 8
Issue description:
The snap store Godot version by using Godot file explorer can't access the root folder therefore you can't access the "jarsigner" file. If you enter the path by hand it still says "OpenJDK 6 jarsigner not configured in editor settings".
The quick fix is to download the Godot from official website because that version has an access to the root folder. But it has no icon and you can't add it to favourites bar :(
P.S. I was trying to solve an issue but didn't fount the solution anywhere. The closest one was with Godot v3.0.5 suggesting to solve by installing 3.0.6 version and OpenJDK 8, but I've already had proper versions.
Steps to reproduce:
Install OpenJDK 8 > Install Godot > Try to set jarsigner path> in export options check if there is "OpenJDK 6 jarsigner not configured in editor settings" error message.
Minimal reproduction project:
Install OpenJDK 8 > Install Godot > Try to set jarsigner path> in export options check if there is "OpenJDK 6 jarsigner not configured in editor settings" error message.
Godot needs access to /usr/bin/jarsigner if you want to use a system-installed OpenJDK, so you should report this to the Snap package maintainers. If the snap prevents you access to /usr but still exposes jarsigner in the PATH, you could try to set only jarsigner as path for the option.
I'm not sure what's the proper location to report issues with a given snap in the snapcraft ecosystem. CC @popey who did the original Godot snap: https://forum.snapcraft.io/t/godot-engine-call-for-testing/4997
@popey I guess we need your help to provide the file system access of your snaps. Can you set classic confinement for it?
The snap is just unusable right now
Thanks for the ping @michaellee8. Sorry it's not working as you'd expect, I'll take a look!
@popey it seems that there are no such permission to access /usr/lib for a snap, I tried devmode but I still didn't work, maybe you can try with classic confinement?
To provide some context on what system access Godot needs:
To export to Android, we need OpenJDK (or Oracle JDK)'s jarsigner, as well as adb (either from the SDK or standalone packaged version). So users need to be able to access /usr/bin/jarsigner and /usr/bin/adb (they provide the path themselves in EditorSettings), or any arbitrary path where they might have those installed (e.g. /home/foo/Dev/android-sdk/platform-tools/adb).
To export to Windows, users can optionally use rcedit.exe that they download themselves, and which, being a Windows application, needs Wine on Linux. So they'd need access to /usr/bin/wine and also the possibility to access a random, user-readable path (e.g. /home/foo/Dev/rcedit.exe). This is not a mandatory feature (it's used to embed the game icon and PE metadata in the .exe), but would be nice to have when using the snap too.
I'm experiencing this issue as well - I have both jarsigner and adb installed to /usr/bin/ and can access both from the command line. Unfortunately, even when I manually put the paths into the editor settings Godot doesn't see the files and therefore won't permit exporting to Android.
This should now be fixed with the newer releases. We switched to classic confinement, so you'll need to snap refresh godot --classic --stable to get the new update as the Snap Store don't allow you to go from strict to classic without explicitly specifying.
Late to the party, but had the same issue. the --classic --stable version solves the issue.
@Elyukeyey Thanks for confirming it :slightly_smiling_face:
I'll close this issue, as it's not related to an official Godot distribution. Please report Snap-related issues on https://forum.snapcraft.io/t/godot-engine-call-for-testing/4997.
| gharchive/issue | 2018-11-15T12:45:44 | 2025-04-01T06:44:20.475249 | {
"authors": [
"Calinou",
"Elyukeyey",
"Mandriausias",
"akien-mga",
"davidholiday",
"michaellee8",
"popey"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/23725",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
389561951 | Add Jack audio I/O support
Hi devs,
I look to use Godot for artistic audio and video/3d things, and it can it a big benefit to directly play audio from godot and process it after via jack and other softwares.
Is it a way to enable the jack part of rtaudio ?
This can permit to get multiple independents audio buses at low latency for professional usages.
Thanks a lot for any information.
A first step here https://github.com/jpcima/godot/tree/rtaudio-jack
Edited the title as we likely won't be using RtAudio for that, but implementing a proper driver as done for MIDI drivers. Then it should also be done for all supported platforms (at least desktop ones).
Why don't use RtAudio, as it already support Jack ?
(it's really a question :)
Feature and improvement proposals for the Godot Engine are now being discussed and reviewed in a dedicated Godot Improvement Proposals (GIP) (godotengine/godot-proposals) issue tracker. The GIP tracker has a detailed issue template designed so that proposals include all the relevant information to start a productive discussion and help the community assess the validity of the proposal for the engine.
The main (godotengine/godot) tracker is now solely dedicated to bug reports and Pull Requests, enabling contributors to have a better focus on bug fixing work. Therefore, we are now closing all older feature proposals on the main issue tracker.
If you are interested in this feature proposal, please open a new proposal on the GIP tracker following the given issue template (after checking that it doesn't exist already). Be sure to reference this closed issue if it includes any relevant discussion (which you are also encouraged to summarize in the new proposal). Thanks in advance!
@wargreen, @unfa, do either of you mind to open the GIP?
I'm not able to answer such questions as "Describe the project you are working on" or further details, because I don't have any; but I'm sure that the both of you surely do.
@jpcima Done
| gharchive/issue | 2018-12-11T01:47:53 | 2025-04-01T06:44:20.481511 | {
"authors": [
"akien-mga",
"clayjohn",
"jpcima",
"wargreen"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/24274",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
422592107 | Exported project with utf8 chars in resource name fails to load scene, works fine in editor
Godot version:
3.1-stable
OS/device including version:
Windows 10 64 bits
Ubuntu 18.04
Issue description:
What happened :
When importing a resource containg utf8 chars in it (like 'é' or 'è'), the scenes using the resources works fine when used in the editor, but when the project is exported, these scenes fail to load.
The error looks like #21524 but the cause is not the same.
What was expected:
Either the scenes fails to load in the editor so it's possible to know before exporting that there is an issue with utf8 file names,
or the scene loads in the exported binary.
Steps to reproduce:
Create a godot project
Import a resource called "carré.png"
Add a sprite to the scene that displays the texture "carré.png"
Launch the project on the editor (it works fine)
Export the project for any platform
Launch the binary (the scene fails to load)
I was unable to reproduce the issue, on Arch (Kernel 4.20.15 64bit). Were you using a 3D or 2D sprite?
For Android, it does not support utf8 file name for assets, buy only ascii.
That's a my bad from me.
When looking back at my projects structure : the problem seems to be quite different. I modified the name of a file used in a tileset then exported the project.
So the issue is quite different from what i thought.
I'm closing this issue as it is no longer relevant.
| gharchive/issue | 2019-03-19T08:04:48 | 2025-04-01T06:44:20.486860 | {
"authors": [
"AzariasB",
"jdbener",
"volzhs"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/27225",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
507170269 | Array variables use a common array instance in the editor
Godot version:
v3.1.1.stable.official, downloaded from the Godot main page as Godot_v3.1.1-stable_x11.64.zip
OS/device including version:
OS: KDE neon 5.16 Kernel: x86_64 Linux 5.0.0-31-generic CPU: Intel Core i5-4670K @ 4x 3.8GHz GPU: AMD/ATI Ellesmere [Radeon RX 470/480/570/570X/580/580X], AMD/ATI Ellesmere [Radeon RX 470/480/570/570X/580/580X]
Issue description:
Filling arrays in editor mode (tools script) does not work correctly.
In the attached example project, when you click on the start_test variable, two arrays are automatically filled. It was expected that it would work on only one node, and not on all.
Steps to reproduce:
Open Godot.
Open the attached array_bug project.
Open Main.tscn.
Choose any node.
Set any value from 1 and above for the array_size variable.
Click on start_test variable.
Click on any node and you will see that its arrays have changed.
Minimal reproduction project:
array_bug.zip
Other observations:
If you try to delete the nodes except the very first one, reset the parameters, duplicate the nodes and perform the operation of filling the arrays again, you will notice that only one node is filled. However, if you exit the project and follow all the steps that were written above, you will get the same result - filling occurs in all nodes, except what you completed the last time.
I think the problem is somehow related to export(Array, ...) var ...
Is it possible that the engine allocates a common memory area for such variables?
Already fixed in 3.2. It's broken in some other cases, but there are already issues for these.
| gharchive/issue | 2019-10-15T11:01:36 | 2025-04-01T06:44:20.493325 | {
"authors": [
"AntonSalazar",
"KoBeWi"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/32850",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
541447153 | UniformTexture in VisualShader Not Being Applied / Saved in Some Situations
Godot version: 3.2 beta 4
OS/device including version: Windows 10
Issue description: UniformTexture in VisualShader is not being applied / saved in some situations. It is shown in the viewport, but running the game has no effect and when I reload the project no longer appears the texture placed. I noticed that only when saved the first time saves successfully, but if I have several meshs with material, when changing them (are unique) the others do not save
Steps to reproduce:
Add a InstanceMesh
Load a mesh and create a VisualShader to surface of the mesh
Make a UniformTexture and put in a Channel (in my case metal and roughness) then Save Material
Duplicate InstanceMesh and load material, make unique
Change UniformTexture of the InstanceMesh copy and run
On editor shows the change, but in game keep with old texture or empty
When i reload project, on InstanceMesh copy has no changes
Every time i make these steps happens this.
Minimal reproduction project:
Ok, i notice uniform only save on first save after opened godot, after this every change even showing on editor, doenst affect on running, after loading anything has changed on second time save
I believe I can reproduce this (or something like this) on 3.2.2.stable.
When I follow the steps, at step 4 I make the material unique, but I do not make the visual shader unique. So, when I edit the shader in the new material I am changing it for the old as well.
If I reload the scene, both materials now have unique visual shaders. I think this change should instead be happening either at the same time as making the material unique, or not at all. It looks like a bug.
If on step 4 I make both the material and visual shader unique, then the two materials are properly independent of each other.
I don't seem able to reproduce the "different in editor and game" effect mentioned.
video: https://youtu.be/Y-Vw8zjh2Fk
note: on step 5 ("change uniform") I instead put connections through a oneminus node, as changing the textureuniform wasn't making a visible difference (haven't used visualshader before, I guess you're supposed to supply an actual texture but that wasn't in the steps so...) I don't think it matters but just so you know.
| gharchive/issue | 2019-12-22T12:52:42 | 2025-04-01T06:44:20.500106 | {
"authors": [
"marcosn41",
"tom-jk"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/34540",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
547111217 | CenterContainer blocks parent from receiving drag and drop
Godot version:
Godot (Mono*) 3.2 Beta 5
*The reproduction project uses GDScript to demonstrate the issue
OS/device including version:
Manjaro Linux x86_64
Issue description:
When the mouse drag is released over a CenterContainer it prevents its parent from receiving the drag data.
Steps to reproduce:
See reproduction project + explanation
Minimal reproduction project:
DragAndDrop.zip
When dragging on the grey Panel the number 42 is given to Godot as the drag data. When it is released on the left side of the screen the HBoxContainer receives this and prints it. However when released on the right side of the screen the CenterContainer instead blocks it. The VBoxContainers on both sides of the screen do not exhibit this behavior.
That's because your CenterContainer has mouse_filter set to Stop. When it's set to Pass (same as BoxContainers), it works correctly.
Although I wonder why CenterContainer's default is different. It's still a container.
Although I wonder why CenterContainer's default is different. It's still a container.
mouse_filter defaults to MOUSE_FILTER_STOP in Control, which Container and thus CenterContainer inherit.
BoxContainer and GridContainer seem to have local overrides to MOUSE_FILTER_PASS. The discrepancy is probably not intended.
| gharchive/issue | 2020-01-08T21:26:54 | 2025-04-01T06:44:20.505141 | {
"authors": [
"ForLoveOfCats",
"KoBeWi",
"akien-mga"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/34933",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
646728478 | BoneAttachment node does not work in one of the Skeleton bones
Symptom: one of the BoneAttachment nodes, with red material, does not work. MeshInstance don’t follow bone movement with SkeletonIK node. It was working in Godot 3.2.1.
Operating system: Windows 10
Godot version: 3.2.2 stable official
Steps to reproduce: run the provided minimal project in Godot 3.2.2. BoneAttachment node, in red material, does not follow bone rotation.
minimal project.zip
CC @AndreaCatania @TwistedTwigleg
I can confirm this is still an issue with Godot 3.2.3 Beta 1 on Linux. From what I can tell when running this project, the first bone in the SkeletonIK chain does not appear to be taken into account when solving.
| gharchive/issue | 2020-06-27T18:12:08 | 2025-04-01T06:44:20.508704 | {
"authors": [
"Leocesar3D",
"TwistedTwigleg",
"akien-mga"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/39893",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
744519184 | How to set font for Hint/Tooltip?
Godot version:
Latest release: 3.2.3 stable official
OS/device including version:
Windows10/Windows7/Linux Manjaro
Issue description:
I set the font for a specific button, and the text on it uses that font without problems, but the hint(Tooltip) isn't using it, i tried changing the editor font but it didn't affect it. I said button but i also tired textedit, I think its general
Normally this isn't a problem but I was experimenting with using Kurdish-Sorani characters and they don't appear at all without font
By the way
How is the progress on BiDi/RTL support?
Another question please
I tried fixing the flipped&disjoined Kurdish (or Arabic /Persian/etc...) characters by writing a script that converts them on the fly, Like this:
writing سڵاو in the editor results in واڵس
So when printing, my script first flips it سڵاو
Then based on context replace the standalone characters with their respective joined versions ﺎوﺳ here is the problem, that ڤ (It isn't he only one) character version which my browser cant render is Private use (according to windows Character Map), it should look like
No problem though since I just need to use the correct font
BUT My Actual Problem is that godot does not accept this character at all, i try putting it in the code editor and it throws (Parser Error: Parse error: Unknown character)
That character is a 3 byte utf-8, is there a fundamental problem here? Godot doesn't accept the full Unicode characters?
Thanks
See https://godotengine.org/article/complex-text-layouts-progress-report-1 and https://godotengine.org/article/complex-text-layouts-progress-report-2.
This work is being done for Godot 4.0 and can't be backported to the 3.2 branch.
You can try @bruvzg's libgdtl for the 3.2 branch, which supports and subset of what's being implemented for Godot 4.0, and can be used as a C++ module (if you build the engine yourself) or GDNative plugin (check the "Releases" page).
Regarding the question on setting the font for tooltips, you have two options:
Change the font theme item for TooltipLabel in your Theme resource. See the recently written documentation about this (not online yet): https://github.com/godotengine/godot/blob/d0a5ef9c7a79e24ec04643ee629f952b4077e6fe/doc/classes/Control.xml#L848-L859 (see Label documentation for a list of its theme properties, which are the same for TooltipLabel).
Make your own custom tooltip with _make_custom_tooltip:
https://github.com/godotengine/godot/blob/d0a5ef9c7a79e24ec04643ee629f952b4077e6fe/doc/classes/Control.xml#L60-L85
That being said, this is an issue tracker for bug reports, and not for user support. Such questions would be best asked and answered on the Godot Q&A.
| gharchive/issue | 2020-11-17T08:23:12 | 2025-04-01T06:44:20.518392 | {
"authors": [
"akien-mga",
"hunar1997"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/43609",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
970543308 | [4.0] Creating a Node from "Recent" list in "Create New Node" dialogue creates Node2D instead of Node
Godot version
v4.0.dev.custom_build [3a48474c4]
System information
Windows 10, 1080 Ti
Issue description
Creating a Node from "Recent" list with "Create New Node" dialogue returns Node2D instead of Node:
https://user-images.githubusercontent.com/2698825/129388525-8d8a6e90-0b89-40e1-b7e8-af38644d05b1.mp4
Steps to reproduce
Open "Create New Node" dialogue (Ctrl + A)
Select "Node" in Recent list
Minimal reproduction project
N/A
@Xartorx Can you reproduce this in 3.3.2 or 3.4 beta3?
@Calinou nope, only on master branch.
Working on a fix for this along with something else, this issue had a more interesting reason behind it than I imagined at first.
| gharchive/issue | 2021-08-13T16:13:33 | 2025-04-01T06:44:20.522541 | {
"authors": [
"Calinou",
"Xartorx",
"hinlopen"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/51632",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
160516738 | Variant class-reference
Operating system or device - Godot version:
Linux 4.5.0-kali1-amd64
git master 367aabf03080da8e4e72c9db56b0b2097c86d6b4
Issue description (what happened, and what was expected):
Class reference has many references to Variant(as a return type or argument), yet after clicking on this type in script editor reference, it opens an empty tab. This should be fixed, replaced by a page explaining what a Variant is.
Steps to reproduce:
Link to minimal example project (optional but very welcome):
While we're at it, the built-in docs could use a sort of glossary that explains what a lot of the more unique argument types are (variant, parameters, what, ect...). Either that or change the argument type in the docs. to something more descriptive as to what is expected.
Then there's the fact that some classes still remain without any sort of documentation (like the one for the proximity node).
@Ace-Dragon:
While we're at it, the built-in docs could use a sort of glossary that explains what a lot of the more unique argument types are (variant, parameters, what, ect...).
I think you meant "argument names"? Except for Variant, those are not "types" they are the "names" which follow the type. Usually the arguments to pass is explained in the function description (unless it's obvious). And, as always, some specific examples would help.
About the empty class documentation, it's just a matter of someone writing them.
This issue is however more than just a missing class reference, it is about exposing the notion of what a Variant is to doctool/bindings, and the broader issue of letting a Variant return another Variant from a method as would be needed by(Array.pop_* #3763).
@J08nY yes, maybe it wasn't clear but I was talking about the other comment citing other empty classes (which we are aware: #3488). This Variant issue is another thing.
I made the docs generate Variant, you can document that it's a wildcard type and not meant to be used directly.
BTW EditorExportPlatform (used in Editor{Import,Export}Plugin) suffers the same problem, but I'm not sure if it should be solved the same way.
| gharchive/issue | 2016-06-15T20:31:41 | 2025-04-01T06:44:20.528729 | {
"authors": [
"Ace-Dragon",
"J08nY",
"reduz",
"vnen"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/5237",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
993106951 | Godot's custom theme property makes Godot crash on startup.
Godot version
3.3.3 stable
System information
Linux
Issue description
After I set custom theme in Godot interface settings, editor should save and restart the editor but it closes and doesn't start automatically. After that Starting Godot manually all projects seems to crash on start up. It only works till project manager shows up.
Steps to reproduce
Open Godot with any project.
Create new theme resource.
Create theme from currant editor theme. (This step is necessary. Using theme file with default theme works fine. That means new theme files created aside from defaults are missing something or messing up editor somehow.)
Save the theme.
Go to Editor Settings > Interface > Theme. Then set the saved theme as custom theme.
Click on save and restart popup at bottom.
Start Godot again.
Minimal reproduction project
No response
Related to #48477.
This is fixed in 3.x by https://github.com/godotengine/godot/pull/51648.
| gharchive/issue | 2021-09-10T10:33:13 | 2025-04-01T06:44:20.532874 | {
"authors": [
"Calinou",
"autofool",
"pycbouh"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/52530",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1063905233 | Exported bool variable resets when changing script file
Godot version
v4.0.dev.custom_build [a6a2e0feb]
System information
Any OS (tested on Windows 10 and Arch Linux)
Issue description
A script contains an exported variable of type bool that defaults to true. The script is attached to a node in the scene. On that node, the variable is changed to false. Whenever the script file is modified, the value on the node is reset to true as soon as the script is saved.
Steps to reproduce
Open the reproduction project. If the main scene main.tscn is not loaded, load it
Click the node MyNode and in the inspector, uncheck the variable Test Variable, coming from script my_script.gd
Save the scene or don't, it doesn't matter
Move to the Script tab and open my_script.gd
Change anything in the source code file, for example the string inside the print function
Save the script
Navigate back to the node and notice that the variable has been reset to true
Minimal reproduction project
ExportBoolReset.zip
Some additional information:
This doesn't happen for any other type as far as I can see. It also does not happen when the variable's default value is set to false. It makes no difference whether the variable is typed or not, @export var my_var:bool = true and @export var my_var = true lead to the same outcome.
Update: Apparently, saving the script is not what triggers the behavior, but parsing is. I didn't notice before, but when you make a change to the script and then sit back and wait for the automatic parse that gets called every idle_parse_delay seconds, you can see that the variable is reset right then and there.
| gharchive/issue | 2021-11-25T20:29:59 | 2025-04-01T06:44:20.538772 | {
"authors": [
"and-rad"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/55321",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1066686306 | Physics Error 3D (StaticBody): Condition "Math::is_nan(v.x)" is true/"Math::is_inf(v.x)" is true.
Godot version
3.4
System information
Windows 10, does not appear graphics related but NVIDEA GeForce GTX 1060 6GB, (GLES2, and GLES3 as far as I know)
Issue description
Note: Error appears similar to: #50450, #32611 (though the circumstances getting to the error seem to differ, so i'm unsure if it's a duplicate)
For larger projects run from the Godot editor this problem has causes me a crash of the Godot project itself (but NOT the editor). The message spams in console for a second, followed by windows problem reporting popping up (followed by windows problem reporting itself stopping working).
Appears to occur when static bodies are spawned already intersecting with each other (this could just be terrible practice on my part).
For smaller projects, as discovered when making the reproduction project, it never causes a crash, but still spams the console with the Math::is_nan error until the offending collision object (static body) is removed from the scene. I say static body only because thats the one I used, don't know about other physics bodies.
When running the project through the 3.4 editor, the error is a Math::is_nan(v.x) error, but when debugged via visual studio, the error is a Math::is_inf(v.x) error.
Another difference with visual studio and the Godot editor is when run via the editor, larger projects crash, while larger projects run via visual studio exhibit the same behaviour as smaller projects, spamming the error in console until the offending object is removed (but without crashing).
Here is the error in full:
ERROR: Condition "Math::is_nan(v.x)" is true.
at: instance_set_transform (servers/visual/visual_server_scene.cpp:691)
(and for visual studio with the code from https://github.com/godotengine/godot/tree/3.4-stable ):
ERROR: Condition "Math::is_inf(v.x)" is true.
at: instance_set_transform (servers/visual/visual_server_scene.cpp:690)
Not sure if it is helpful, but here is the call stack from visual studio when the error is thrown (using Godot 3.4 source code taken from
https://github.com/godotengine/godot/tree/3.4-stable
with scons vsproj):
godot.windows.tools.32.exe!VisualServerScene::instance_set_transform(RID p_instance, const Transform & p_transform) Line 690 C++
godot.windows.tools.32.exe!VisualServerRaster::instance_set_transform(RID arg1, const Transform & arg2) Line 548 C++
godot.windows.tools.32.exe!VisualServerWrapMT::instance_set_transform(RID p1, const Transform & p2) Line 471 C++
godot.windows.tools.32.exe!VisualInstance::_notification(int p_what) Line 87 C++
godot.windows.tools.32.exe!VisualInstance::_notificationv(int p_notification, bool p_reversed) Line 40 C++
godot.windows.tools.32.exe!GeometryInstance::_notificationv(int p_notification, bool p_reversed) Line 83 C++
godot.windows.tools.32.exe!MeshInstance::_notificationv(int p_notification, bool p_reversed) Line 42 C++
godot.windows.tools.32.exe!Object::notification(int p_notification, bool p_reversed) Line 929 C++
godot.windows.tools.32.exe!SceneTree::flush_transform_notifications() Line 159 C++
godot.windows.tools.32.exe!SceneTree::idle(float p_time) Line 541 C++
godot.windows.tools.32.exe!Main::iteration() Line 2188 C++
godot.windows.tools.32.exe!OS_Windows::run() Line 3342 C++
godot.windows.tools.32.exe!widechar_main(int argc, wchar_t * * argv) Line 161 C++
godot.windows.tools.32.exe!_main() Line 183 C++
godot.windows.tools.32.exe!main(int _argc, char * * _argv) Line 195 C++
[External Code]
kernel32.dll![Frames below may be incorrect and/or missing, no symbols loaded for kernel32.dll] Unknown
Steps to reproduce
I have found this 22.9KB project is reliable at getting the error quickly, but ultimately I have found this error is hard to predict, showing up from a few seconds to a few minutes after starting (and this project does not crash. I'm unsure what difference causes the crash in larger projects).
The project is a simple FPS, pan the camera around with mouse movement and click and hold left click to fire projectiles. Keep doing this, with the camera preferably facing upwards (shifting the camera 180 degrees up and down can help).
within the span of a few seconds to a few minutes you should get the error.
ERROR: Condition "Math::is_nan(v.x)" is true.
or
ERROR: Condition "Math::is_inf(v.x)" is true.
if debugging via visual studio.
A visual indication of the bug in game will be the 2 or more colliding projectiles that I have to assume cause the bug floating in the air, unresponsive to collisions.
Press the 'R' key and the scene will reset, removing the offending static bodies.
Minimal reproduction project
NANbowminireproducedproject.zip
Edited the original post, I said static body, but I meant rigid body.
from what I can gather from these values (from arrowProjectile.gd), it seems to me the static bodies have launched themselves away from each other so fast that their velocity/translation values could have gotten too big, causing a crash. They rotate alot, and I know other game engines don't appear to handle large rotational velocities well (e.g. Skyrim https://www.youtube.com/watch?v=8aiOR_KuRe4), so it could be the rotation that the projectiles experience.
Thus catching high velocities of rigid bodies early should prevent this crash. Honestly i'm not sure how to control the linear velocity of rigid bodies, but I tested this piece of code inserted in arrowProjectile.gd shown below (horrendously inefficient I know)...
Unfortunately, the problem persists, so I could be misinterpreting it.
While it won't be experienced in the minimal reproduction project the cause of the crash in the larger project appears to be caused by an access violation, though the position in the code where it breaks is not consistent. I'm unsure if this error is related to the crash at all, though I suspect it is, as they tend to appear together (though not always, the error in the title does not always cause a crash in the larger projects).
Also, in the latest crash at least, an access violation is caused because an array overruns its length trying to find a set of infinite values in a Vector3 (or at least that's what I interpreted from below.
The area in the code where the program crashes and the way it crashes has thus far varied, with the above being 1 of 2 different places (so far), though it has always been an access violation.
| gharchive/issue | 2021-11-30T01:35:26 | 2025-04-01T06:44:20.557030 | {
"authors": [
"Thaniel94"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/55458",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1548263892 | Builtin control nodes don't respond to touch if mouse event emulation is disabled
Godot version
d93b66ad4d6b1dae42cc318c16224d03bd1a1472
System information
arch linux, asus zenbook duo ux481 (dual touchscreen laptop)
goggle pixel 6
Issue description
Builtin control nodes don't respond to touch if emulate_mouse_from_touch is disabled in the project settings. This makes them unusable if the project needs this disabled.
Steps to reproduce
Place a button or text edit control into a scene. Attach a pressed event to the button and print to see if event is received.
Disable emulate_mouse_from_touch.
Attempt to interact with the controls on a touch capable device.
Minimal reproduction project
none
Duplicate of #24589
| gharchive/issue | 2023-01-18T23:14:49 | 2025-04-01T06:44:20.560294 | {
"authors": [
"akien-mga",
"maiself"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/71651",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1784015361 | Moving RichTextLabel into a CenterContainer node causes editor to crash
Godot version
v4.1.rc2.mono.official [46424488e]
System information
Windows 11 - v4.1.rc2 - Vulkan
Issue description
I tried to move a RichTextLabel into a CenterContainer node and the editor crashed.
It seems the latest working version was v4.1.beta2.mono.official [a2575cba4]
I have a crash with all the snapshots that came after that one.
ERROR: Failed method: CanvasItem::_redraw_callback. Message queue out of memory. Message queue out of memory. Try increasing 'memory. at: push_callablep (core/object/message_queue.cpp:96) ERROR: Failed method: CanvasItem::_redraw_callback. Message queue out of memory. Message queue out of memory. Try increasing 'memory. at: push_callablep (core/object/message_queue.cpp:96) ERROR: Failed method: CanvasItem::_redraw_callback. Message queue out of memory. Message queue out of memory. Try increasing 'memory. at: push_callablep (core/object/message_queue.cpp:96) ERROR: Failed method: CanvasItem::_redraw_callback. Message queue out of memory. Message queue out of memory. Try increasing 'memory. at: push_callablep (core/object/message_queue.cpp:96) ERROR: Failed method: EditorToaster::_error_handler_impl. Message queue out of memory. Message queue out of memory. Try increasing '. at: push_callablep (core/object/message_queue.cpp:96) ERROR: Failed method: EditorToaster::_error_handler_impl. Message queue out of memory. Message queue out of memory. Try increasing '.
Steps to reproduce
Open the project and the main scene
Move the RichTextLabel into the CenterContainer
Editor should crash
Minimal reproduction project
bug-ui.zip
backtrace leading to the failure:
* thread #1, name = 'godot.linuxbsd.', stop reason = signal SIGSEGV: invalid address (fault address: 0x722d00576e)
* frame #0: 0x000055555ca15185 godot.linuxbsd.editor.dev.x86_64.llvm`Callable::get_object(this=0x000055555f5d9320) const at callable.cpp:136:41
frame #1: 0x000055555ccb83e1 godot.linuxbsd.editor.dev.x86_64.llvm`CallQueue::statistics(this=0x000055555dc6f9b0) at message_queue.cpp:422:39
frame #2: 0x000055555ccb70d9 godot.linuxbsd.editor.dev.x86_64.llvm`CallQueue::push_callablep(this=0x000055555dc6f9b0, p_callable=<unavailable>, p_args=0x0000000000000000, p_argcount=0, p_show_error=false) at message_queue.cpp:97:4
frame #3: 0x000055555b02af4b godot.linuxbsd.editor.dev.x86_64.llvm`Control::update_minimum_size() [inlined] Error CallQueue::push_callable<>(this=0x000055555dc6f9b0, p_callable=0x00007fffffffd5e0) at message_queue.h:130:10
frame #4: 0x000055555b02af29 godot.linuxbsd.editor.dev.x86_64.llvm`Control::update_minimum_size(this=<unavailable>) at control.cpp:1628:33
frame #5: 0x000055555b096379 godot.linuxbsd.editor.dev.x86_64.llvm`Container::_child_minsize_changed(this=<unavailable>) at container.cpp:37:2
frame #6: 0x000055555ccc0a48 godot.linuxbsd.editor.dev.x86_64.llvm`Object::emit_signalp(this=0x0000555574ac6750, p_name=0x000055555edb7228, p_args=0x0000000000000000, p_argcount=0) at object.cpp:1068:15
frame #7: 0x000055555b0a47d2 godot.linuxbsd.editor.dev.x86_64.llvm`Control::_update_minimum_size() [inlined] Error Object::emit_signal<>(this=0x0000555574ac6750, p_name=<unavailable>) at object.h:881:10
frame #8: 0x000055555b0a47b3 godot.linuxbsd.editor.dev.x86_64.llvm`Control::_update_minimum_size(this=0x0000555574ac6750) at control.cpp:1592:3
frame #9: 0x000055555ccb9817 godot.linuxbsd.editor.dev.x86_64.llvm`CallQueue::_call_function(this=<unavailable>, p_callable=0x0000555577a6f098, p_args=0x0000555577a6f0b0, p_argcount=0, p_show_error=false) at message_queue.cpp:219:13
frame #10: 0x000055555ccb9c9b godot.linuxbsd.editor.dev.x86_64.llvm`CallQueue::flush(this=0x000055555dc6f9b0) at message_queue.cpp:320:6
frame #11: 0x000055555af36ab5 godot.linuxbsd.editor.dev.x86_64.llvm`SceneTree::physics_process(this=0x000055555f67f8d0, p_time=0) at scene_tree.cpp:471:33
frame #12: 0x0000555558fa153e godot.linuxbsd.editor.dev.x86_64.llvm`Main::iteration() at main.cpp:3383:45
The crash happens before MessageQueue runs out of memory, as usual.
The infinite loop's stacktrace seems to be:
0x0000000006cb4299 in TextServerAdvanced::_shaped_text_substr (this=0xbeaf7d0, p_shaped=..., p_start=0, p_length=8) at modules/text_server_adv/text_server_adv.cpp:4258
4258 new_sd->extra_spacing[i] = sd->extra_spacing[i];
(gdb) bt
#0 0x0000000006cb4299 in TextServerAdvanced::_shaped_text_substr (this=0xbeaf7d0, p_shaped=..., p_start=0, p_length=8) at modules/text_server_adv/text_server_adv.cpp:4258
#1 0x0000000006cc9a36 in TextServerAdvanced::shaped_text_substr (this=0xbeaf7d0, arg1=..., arg2=0, arg3=8) at modules/text_server_adv/text_server_adv.h:888
#2 0x0000000008fd0591 in TextParagraph::_shape_lines (this=0x23b73f10) at ./scene/resources/text_paragraph.cpp:194
#3 0x0000000008fd2eb3 in TextParagraph::get_size (this=0x23b73f10) at ./scene/resources/text_paragraph.cpp:534
#4 0x00000000086fda23 in RichTextLabel::Line::get_height (this=0x1f9fd230, line_separation=0) at ./scene/gui/rich_text_label.h:119
#5 0x00000000086fec2a in RichTextLabel::_calculate_line_vertical_offset (this=0x22d5b580, line=...) at ./scene/gui/rich_text_label.cpp:1724
#6 0x00000000085ef920 in RichTextLabel::_resize_line (this=0x22d5b580, p_frame=0x1f988c50, p_line=0, p_base_font=..., p_base_font_size=16, p_width=-7, p_h=0) at ./scene/gui/rich_text_label.cpp:448
#7 0x000000000860220e in RichTextLabel::_validate_line_caches (this=0x22d5b580) at ./scene/gui/rich_text_label.cpp:2854
#8 0x00000000085fd477 in RichTextLabel::_notification (this=0x22d5b580, p_what=30) at ./scene/gui/rich_text_label.cpp:1832
#9 0x00000000086fd916 in RichTextLabel::_notificationv (this=0x22d5b580, p_notification=30, p_reversed=false) at ./scene/gui/rich_text_label.h:41
#10 0x000000000a5e0423 in Object::notification (this=0x22d5b580, p_notification=30, p_reversed=false) at ./core/object/object.cpp:796
#11 0x0000000008335add in CanvasItem::_redraw_callback (this=0x22d5b580) at ./scene/main/canvas_item.cpp:140
#12 0x0000000007477bba in call_with_variant_args_helper<CanvasItem>(CanvasItem*, void (CanvasItem::*)(), Variant const**, Callable::CallError&, IndexSequence<>) (p_instance=0x22d5b580,
p_method=(void (CanvasItem::*)(CanvasItem * const)) 0x8335a46 <CanvasItem::_redraw_callback()>, p_args=0x0, r_error=...) at ./core/variant/binder_common.h:303
#13 0x0000000007468d65 in call_with_variant_args<CanvasItem> (p_instance=0x22d5b580, p_method=(void (CanvasItem::*)(CanvasItem * const)) 0x8335a46 <CanvasItem::_redraw_callback()>, p_args=0x0, p_argcount=0, r_error=...)
at ./core/variant/binder_common.h:417
#14 0x000000000742b894 in CallableCustomMethodPointer<CanvasItem>::call (this=0x22d51860, p_arguments=0x0, p_argcount=0, r_return_value=..., r_call_error=...) at ./core/object/callable_method_pointer.h:104
#15 0x000000000a2c5da9 in Callable::callp (this=0x24a211e8, p_arguments=0x0, p_argcount=0, r_return_value=..., r_call_error=...) at ./core/variant/callable.cpp:50
#16 0x000000000a5d95ff in CallQueue::_call_function (this=0xbbb16b0, p_callable=..., p_args=0x24a21200, p_argcount=0, p_show_error=false) at ./core/object/message_queue.cpp:219
#17 0x000000000a5d9dac in CallQueue::flush (this=0xbbb16b0) at ./core/object/message_queue.cpp:320
#18 0x0000000008383b11 in SceneTree::physics_process (this=0xda088b0, p_time=0.016666666666666666) at ./scene/main/scene_tree.cpp:471
#19 0x00000000057194cd in Main::iteration () at main/main.cpp:3383
#20 0x000000000569385a in OS_LinuxBSD::run (this=0x7fff0784cc30) at platform/linuxbsd/os_linuxbsd.cpp:912
#21 0x000000000568ab00 in main (argc=2, argv=0x7fff0784d218) at platform/linuxbsd/godot_linuxbsd.cpp:74
I was able to bisect it to #78241
Seems like it's caused by vscroll->get_combined_minimum_size() call.
Visually CenterContainer squeezes (fit content) to zero size
and when you put label inside it produces horizontal artefacts...
Seems like it's caused by vscroll->get_combined_minimum_size() call.
Not, this was wrong. It's something wrong with the sizing. Infinite loop is container sort children → resize rtl → rtl update_min_size → container sort children.
I guess the issue is update_min_size is called when processing is done, instead it should be called when control properties are changed and shaped data is invalidated. The same is true for Label, which had the same issue (not use why it's not crashing as well).
See also #78932 (potential duplicate).
| gharchive/issue | 2023-07-01T15:26:31 | 2025-04-01T06:44:20.570212 | {
"authors": [
"Sauermann",
"akien-mga",
"aldocd4",
"bruvzg",
"capnm"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/78912",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1806673371 | Linux - Editor Loses Keyboard Input after pressing F2 to rename nodes
Godot version
4.1.1-rc1
System information
Ubuntu MATE 20.04 LTS
Issue description
Sometimes the entire editor loses the ability to receive keyboard input and must be restarted.
This has been narrowed down to pressing F2 to rename nodes.
Steps to reproduce
This happens after pressing F2 to rename things. It doesn't happen every time. But almost always when hitting F2 with more than one node selected in the Scene Tree tab.
Possibly relevant: script editor is detached.
Minimal reproduction project
Shouldn't need an MRP for this
Possibly relevant: script editor is detached.
This is likely related to https://github.com/godotengine/godot/issues/78672.
This happens for me too on 4.2.2, keyboard inputs are not registered, only comes back after a project reload, not only when using F2, but also double clicking the node or anything that happens to show a little text input popup, such as editing values of a position. This only happens on KDE Plasma (Xorg), was using Gnome (Xorg) previously and never seem this issue.
As a workaround, turning on "Single Window Mode" in editor settings helped.
This is resolved for me now, not sure when it became resolved
| gharchive/issue | 2023-07-16T18:16:56 | 2025-04-01T06:44:20.575159 | {
"authors": [
"Calinou",
"mauriciomarinho2",
"naturally-intelligent"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/79556",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1876519019 | "Invalid include exit hint @@< without matching enter hint." error setting built-ins after shader #include
Godot version
4.1.1.stable.official
System information
Windows 10, Vulkan forward +
Issue description
When trying to use shader includes inside the body of the fragment() function, an error occurs when setting EMISSION after the include: Invalid include exit hint @@< without matching enter hint.
Edit: Seems this also happens setting ALPHA, ROUGHNESS, etc. Likely any built-in that can be set.
Steps to reproduce
Create a material.
Convert it to a shader material.
Copy the contents in the fragment() function into a new shader include.
Include the shader include inside the fragment() function.
Set EMISSION to something after the include. Note error. Setting before the include works fine.
Minimal reproduction project
(upload failed, will post in reply).
Looks like the shader parser does some backtracking which breaks the include end marker as it is processed multiple times. Basically any assignment after an #include inside a function body will cause this. Workaround is trivial, for example add a dummy variable declaration or re-order the statements as you noticed:
#include "test_include.gdshaderinc"
float dummy;
EMISSION = vec3(1.0, 0.0, 0.0); // Works again.
Still, it's a bug and should be fixed.
Ah, I thought I tried that.
I ended up scrapping this method of including shaders mid function due to the errors and also I started running into a problem where every time I edited a shader, it would get stuck in a near-infinite loop of triggering shader recompiling causing the editor to freeze up for sometimes minutes at a time, making it basically impossible to edit shaders. I couldn't figure out how to repro it in a minimal case, though, and ended up just reverting everything in frustration. Should have saved the problematic shaders. Ah, well.
Decided to go with more of an uber shader with #define's instead.
| gharchive/issue | 2023-09-01T01:05:16 | 2025-04-01T06:44:20.580452 | {
"authors": [
"bitsawer",
"jitspoe"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/81215",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1897611664 | Drag and drop files from Windows explorer to Godot not working
Godot version
v4.1.1.stable.official [bd6af8e0e]
System information
Godot v4.1.1.stable - Windows 10.0.22621 - Vulkan (Forward+) - dedicated NVIDIA GeForce RTX 2060 SUPER (NVIDIA; 31.0.15.3713) - Intel(R) Core(TM) i5-10400 CPU @ 2.90GHz (12 Threads)
Issue description
I'm trying to import external assets into my project and I'm not succeeding. I tried to drag and drop the files from a folder straight to the Godot editor and nothing happened.
I tried several file formats: .jpg, .png, .txt, .shader, .mp3, .mp4,
I tried navigating to other folders through the file manager, but I get stuck in the res:// folder
I also tried throwing the files directly into the project directory through Windows Explorer, but nothing changes either. I've closed and opened the editor several times and I simply can't import anything from outside of Godot.
I don't know if I'm doing something wrong but I believe that an import window should appear when I drop external files into the editor. I tried to download the same version from the official web page on another computer and that behaviour persists.
Sorry guys, I'm just new here and I don't know if this is the normal behaviour of Godot, but seems a little bit strange for me.
Steps to reproduce
Download the 4.1.1 version from official web page, run Godot and try to import anything to the files viewport through drag and drop.
Minimal reproduction project
Pages.zip
I tried in Godot 3.5.1 and I got it. Actually dragging and dropping the file directly into the editor doesn't work, but placing files directly through Windows Explorer the Godot editor 'updates' and shows me the file that I imported normally.
In Godot 4.1.1, even placing the files directly in the folder through the explorer, the editor does not update and does not show me the files, even after restarting.
I don't think drag-and-drop of arbitrary files onto the FileSystem dock is implemented.
If you're not seeing files in the FileSystem dock in 4.x, double-check that res:// is unfolded in the FileSystem dock (the arrow next to it should point downwards). By default, it may be folded, which means you won't see any files/folders contained within.
I put the file in the folder like I did in 3.x but in the FileSystem dock the folder looks empty
The .shader extension needs to be renamed in .gdshader in Godot 4.
That worked! Sorry about this. I just couldn't find it anywhere. Really lame by me. Thank you very much!
Closing per the above comment.
| gharchive/issue | 2023-09-15T02:52:02 | 2025-04-01T06:44:20.587964 | {
"authors": [
"Calinou",
"Kelvym115"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/81678",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1906219704 | Runtime loading GLTF models crashes in exported projects when using HANDLE_BINARY_EMBED_AS_BASISU
Godot version
4.1.1
System information
Godot v4.1.1.stable - Windows 10.0.19045 - Vulkan (Mobile) - dedicated NVIDIA GeForce RTX 3080 (NVIDIA; 31.0.15.3623) - AMD Ryzen 7 5800X 8-Core Processor (16 Threads)
Issue description
When loading an arbitrary model at runtime in exported projects, Godot will crash when
GLTFState.handle_binary_image is set to GLTFState.HANDLE_BINARY_EMBED_AS_BASISU.
When running in the editor, setting the flag will not lead to a crash.
Steps to reproduce
The code below will reproduce the crash in exported projects. When running in the editor, loading will work as expected with or without the flag.
var gltf := GLTFDocument.new()
var state := GLTFState.new()
# This causes crashes in godot 4.1.1 exported projects
# Commenting out the line below will avoid the crash
state.handle_binary_image = GLTFState.HANDLE_BINARY_EMBED_AS_BASISU
# Substitute in any gltf model here
gltf.append_from_file("./duck/duck.glb", state)
add_child(gltf.generate_scene(state))
Minimal reproduction project
repo with MRP
Zipped repo: runtime-gltf-import-crash.zip
The crash was also reproduced in the following environment:
System Information
Godot v4.2.dev (0ca854232) - Windows 10.0.22621 - Vulkan (Mobile) - dedicated NVIDIA GeForce RTX 4070 Ti (NVIDIA; 31.0.15.3598) - 13th Gen Intel(R) Core(TM) i7-13700KF (24 Threads)
Upon checking with the debugger, it appears that Image::basis_universal_packer is a nullptr.
https://github.com/godotengine/godot/blob/master/scene/resources/portable_compressed_texture.cpp#L147-L150
This is a guess, but if TOOL_ENABLED is not defined, it may be that Image::basis_universal_packer is not initialized. This may be the cause of the crash in the exported project.
https://github.com/godotengine/godot/blob/master/modules/basis_universal/register_types.cpp#L285-L286
| gharchive/issue | 2023-09-21T06:35:39 | 2025-04-01T06:44:20.594501 | {
"authors": [
"kmqwerty",
"you-win"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/82029",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
2255063596 | project.godot converted to project.binary on export when ProjectSettings.editor/export/convert_text_resources_to_binary is set to false
Tested versions
Reproducible in: 4.0-stable, 4.2.1-stable, 4.3-dev5
System information
Godot v4.2.1 - Windows 10.0.22631 - Vulkan (Mobile)
Issue description
The documentation for ProjectSettings.editor/export/convert_text_resources_to_binary states the following:
Note: If Editor > Export > Convert Text Resources to Binary is true, @GDScript.load() will not be able to return the converted files in an exported project. Some file paths within the exported PCK will also change, such as project.godot becoming project.binary. If you rely on run-time loading of files present within the PCK, set Editor > Export > Convert Text Resources to Binary to false.
This setting works as expected for all other text-format files; however, project.godot is always converted to project.binary on export, regardless of this setting, which does not match the documentation.
Note that it is still possible to export project.godot to the .pck using Export Filter options during the export process; however, project.binary will always be included, regardless of filters, and takes priority over project.godot during initialization.
Steps to reproduce
Create a project
In Project Settings, turn on Advanced Options
Under Editor>Export, uncheck the box for "Convert Text Resources to Binary"
Export the project
Inspect the exported .pck using debug code or with any available .pck unpacking tool
Minimal reproduction project (MRP)
For convenience, this project includes a script to print all files in "res://" to the debug console.
PCK_Test.zip
Seems like a bug to me:
https://github.com/godotengine/godot/blob/4a0160241fd0c1e874e297f6b08676cf0761e5e8/editor/export/editor_export_platform.cpp#L1406-L1410
Project settings is technically not a "resource" :P
I'd say it's only an issue because it's mentioned in the docs explicitly, the intention seems to not have been to respect it but always convert it, but wouldn't hurt to make it respect the option as long as the rest of the engine handles it
| gharchive/issue | 2024-04-21T13:02:40 | 2025-04-01T06:44:20.600694 | {
"authors": [
"AThousandShips",
"ShroudedMuridian",
"jsjtxietian",
"timothyqiu"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/90981",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
2491914867 | script_editor_cache.cfg stores old version of script filename resulting in "Case mismatch opening requested file" console spam.
Tested versions
4.3
System information
Windows 10, Vulkan forward +, Nvidia 3070
Issue description
I renamed a script file outside of the editor that had an upper case letter to match the standard snake_case filenames and, after restarting the project, got a bunch of spam in the console, even though nothing referenced the original filename:
drivers/windows/file_access_windows.cpp:181 - Case mismatch opening requested file 'res://addons/console/Console.gd', stored as 'res://addons/console/console.gd' in the filesystem. This file will not open when exported to other case-sensitive platforms.
Steps to reproduce
Not sure if this is exclusive to addons, but I have a simple developer console, and it's had a non-standard filename, so I decided to fix this. In my active project, it wasn't a problem, as I renamed it within Godot, but when I replaced the addon with the updated version in a test project, I started getting errors, even though nothing referenced the original "Console.gd" filename. I tried enabling/disabling the addon and restarting the project several times and ensured nothing was referencing it in the autoload/plugin/scripts/etc.
Before:
func _enter_tree():
print("Console plugin activated.")
add_autoload_singleton("Console", "res://addons/console/Console.gd")
After:
func _enter_tree():
print("Console plugin activated.")
add_autoload_singleton("Console", "res://addons/console/console.gd")
Deleting the ".godot" directory fixes the issue, and upon further inspection, it seems the project_metadata.cfg was the culprit:
scripts=["res://addons/console/console.gd", "res://addons/console/plugin.cfg", "res://addons/console/console_plugin.gd", "res://addons/console/Console.gd", "res://main_scene.gd", "res://spinning_cube.gd"]
The script_editor_cache.cfg and editor_layout.cfg might also cause the problem as they also contain references to the file.
Edit: Seems the project_metadata.cfg does not cause it. The script_editor_cache definitely does, though. Editor_layout likely does as well. As a side note, it's kind of interesting how many times the error message pops up suggesting that Godot might be reading these files more often than it needs to and reducing these reads could speed up project reload times.
Minimal reproduction project (MRP)
test-console_case_error.zip
Note: .godot directory is necessary to reproduce the issue.
I deleted the .godot folder as suggested in the post, and that did not fix the issue in my own project when this happened. Haven't determined a user fix yet.
I'd do a full search through your whole project for anything referencing the original filename.
It might be worth noting that in addition to the multiple *cache.cfg files that hold references to the there's also filesystem_cache files that had references to it, but those are also in the .godot subdirectory.
If you're specifying specific export files, it's possible the file could be referenced in the export_presets.cfg outside of the .godot directory.
User error. I found 2 preloads with the old path names in my script that prevented the fix. No clue how I didn't catch that originally, but ain't that just the way?
So, to amend my claim: Deleting the .godot folder as suggested seems to work; remaining errors were my user error.
| gharchive/issue | 2024-08-28T12:14:53 | 2025-04-01T06:44:20.609070 | {
"authors": [
"Kylemcarthur",
"jitspoe"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/issues/96221",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
477135000 | Implemented check_vadjust in CheckBox. It was exposed to the editor, …
…but wasn't used.
It's also defined and unused in CheckButton.
Do you need it for something specific? If it never worked in two nodes but was never reported by any user, it might be better to simply drop this constant.
Yes, the reason I realized it wasn't implemented was that I was trying to use it and it didn't do anything. Without this constant it can be challenging to get the icons to nicely line up with it's text.
I can look at the code for CheckButton too.
Thanks!
| gharchive/pull-request | 2019-08-06T02:19:18 | 2025-04-01T06:44:20.611312 | {
"authors": [
"MuffinManKen",
"akien-mga"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/31127",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
619999115 | Remove the custom_defines section from the editor
Fixes #38822
Thanks!
| gharchive/pull-request | 2020-05-18T08:09:33 | 2025-04-01T06:44:20.612245 | {
"authors": [
"akien-mga",
"m4gr3d"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/38824",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
781494076 | Make links on the HTML5 editor more readable
This also tweaks the focus style to apply to all elements for better keyboard navigation.
Preview
Before
After
Thanks!
Thanks!
Cherry-picked for 3.2.4.
Cherry-picked for 3.2.4.
| gharchive/pull-request | 2021-01-07T17:47:38 | 2025-04-01T06:44:20.614935 | {
"authors": [
"Calinou",
"akien-mga"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/44995",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
855115708 | Improve the Engine.editor_hint property documentation
Note when cherry-picking to 3.x: Replace @tool with tool and replace the documentation URL with https://docs.godotengine.org/en/stable/tutorials/misc/running_code_in_the_editor.html.
This also includes a typo fix in doc/classes/AStar.xml.
Thanks!
Cherry-picked for 3.3.
| gharchive/pull-request | 2021-04-10T17:21:00 | 2025-04-01T06:44:20.616999 | {
"authors": [
"Calinou",
"akien-mga"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/47772",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
994171340 | Make restore_scenes_on_load true by default
It's a very useful option that many people don't know about. This changes only the default, if you have a "big project" and this affects your startup time (which is unlikely; I have such project myself and scenes add ~5 seconds to the regular >1 minute startup) you will also know how to disable it.
Closes #16317
Thanks!
Hmm, I don't think @reduz will be happy about it. This was never made the default because huge 3D scenes did take a lot of time to launch, like the third person shooter demo. It's not a matter of how big your project is, more like what your currently loaded scenes contain.
I think we should not have made this the default until we solve this problem (that could be by having away to disable the scene loading for example, or a popup asking you if you want to load the previously opened scenes).
Much more people benefit from it being true by default than there are people affected negatively. And as already mentioned, if someone doesn't like their load times, they can disable this option.
This was never made the default because huge 3D scenes did take a lot of time to launch, like the third person shooter demo.
Isn't the biggest scene in 3rd person shooter the main one? If yes, then it will load automatically anyways. Also most people don't open this project more than once, in which case they are not affected.
In any case, I wouldn't worry about the "problem" until someone actually reports it. And there is a chance that it will be solved even before such report appears (although this is unlikely, given the issue has been around for 3 years).
Much more people benefit from it being true by default than there are people affected negatively. And as already mentioned, if someone doesn't like their load times, they can disable this option.
Well, that's not how things work. If an option benefits most people, but makes the engine unusable is other cases, then it should not be made the default. That's exactly why we make the engine "just work" with small screens, even if most people have wide screens.
This was discussed already several time. There https://github.com/godotengine/godot/issues/16317 or there https://github.com/godotengine/godot/issues/2385, and there was no clear consensus on the fact that was a UX improvement. I would have refrained merging this before we reach a consensus. I also suggested there to add a way to cancel the loading would be a good compromise.
Don't get me wrong, even for me I prefer having my scenes back at startup, but we are not sure this won't have a significant impact on other users situations.
If someone had this option disabled, their first load will still be fast. The most affected users are the ones that didn't know the default changed or don't know about this option and just started their project. And even if someone faces this problem, a couple of minutes loading time doesn't make the engine unusable (especially when, as I said, it can be disabled, which many users will do when this happens).
I've seen numerous users asking about the "Restore Scenes on Load" feature or even opening proposals about it, even though it already exists. IMO it's ok to solve the potential problems after enabling it by default; there is a chance that we can come up with a solution faster, now that it was brought to attention.
An idea I came up with just now is that when editor detects that it loaded longer than X minutes, it could show a dialog asking if you want to disable reopening scenes. Should be easy to do.
I chose to merge so that we can have users actually test it and provide feedback. Consensus is very hard to reach on this kind of option when most users don't even know it exists, so we can only speculate on possible problems. It's simpler to just force it on during alpha, see if users like or hate it, and revisit accordingly.
| gharchive/pull-request | 2021-09-12T14:18:17 | 2025-04-01T06:44:20.624153 | {
"authors": [
"KoBeWi",
"akien-mga",
"groud"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/52601",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1054500944 | [macOS] Fix crash handler not printing function names on M1 Macs.
Fixes incorrect architecture name used to call atos.
Thanks!
This fix is relevant for 3.x as well.
Cherry-picked for 3.5.
Cherry-picked for 3.4.1.
| gharchive/pull-request | 2021-11-16T06:34:55 | 2025-04-01T06:44:20.626155 | {
"authors": [
"akien-mga",
"bruvzg"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/55019",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1086490625 | Update CharFXTransform: explain "relative index"/"absolute index"
Confusingly, these two properties had identical descriptions even though they measure different things.
"relative_index" measures character count from the custom effect's bbcode opening tag.
"absolute_index" measures character count from the start of the bbcode text that includes the custom effect.
See the code author's own explanation here: https://github.com/godotengine/godot/pull/23658
NOTE: Doco for CharFXTransform.xml has changed significantly in 4.0, where terminology has changed to "glyph". Therefore, proposing this change for 3.x branch only.
Thank you! And congrats on your first contribution to Godot 🎉
Cherry-picked for 3.4.3.
| gharchive/pull-request | 2021-12-22T06:19:37 | 2025-04-01T06:44:20.628799 | {
"authors": [
"akien-mga",
"belzecue",
"mhilbrunner"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/56152",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1426472864 | update default cursor shape for javascript export
this looks to already be fixed in master: https://github.com/godotengine/godot/blob/master/platform/web/js/libs/library_godot_display.js#L177
It seems I needed to update the default shape in $GodotDisplayCursor to match the changes made in 3f8830ec50ba349bed473fd55f7cbe77d266e0fa
This MR should fix #67925.
Also requires changes in clear and set_shape. See: https://github.com/godotengine/godot/pull/67045/files
I've applied the changes
I've applied the changes
@tehsquidge A single commit is preffered, please squash your commits (see: PR workflow).
Thanks @kleonc
The docs are really good, I should have read them better. :stuck_out_tongue:
I believe I've squashed my commits correctly.
Thanks for opening a pull request :slightly_smiling_face:
Pull requests targeting the stable branch should be opened against the 3.x branch instead of 3.5.We can then cherry-pick specific bug fix PRs to the 3.5 branch.
Shouldn't we cherry-pick #67044 to 3.5 instead? Cherry-picking makes it easier to track commits across branches, instead of merging separate pull requests.
Ah, yep. I didn't realize this was already fixed in 3.x
One thought. My original PR that caused this regression was not merged without a good reason, too.
So if this is reverted, the other problem comes back.
Wouldn't it make sense to check if a custom cursor is used before using this behaviour? Otherwise use the behaviour I added in my PR.
That would solve both problems.
It'd make more sense to cherrypick #67044. I hadn't realised this was already fixed in 3.x/4 when I put this PR in.
My intention wasn't to revert the changes made by @RPicster - is it definitely the case that has happened? Because my PR is identical to the PR already in 3.x
I've opened a new MR to cherry-pick as suggested here #68214
I'll close this one
| gharchive/pull-request | 2022-10-28T01:36:09 | 2025-04-01T06:44:20.636062 | {
"authors": [
"Calinou",
"RPicster",
"clayjohn",
"kleonc",
"tehsquidge"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/67964",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1660771387 | Broken link removed
There is no translation readme available on editor/translations/README.md
This file was removed in #70623, the changes there should be reflected instead of just removing the link
| gharchive/pull-request | 2023-04-10T13:13:18 | 2025-04-01T06:44:20.637600 | {
"authors": [
"AThousandShips",
"pablogila"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/75887",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1751507219 | MSVC: Disable ICF (Identical COMDAT Folding) for optimize=speed_trace
ICF means that methods (and readonly globals) with identical code get folded into a single version, which means that in stack traces the various methods cannot be distinguished and show up as any one of them, making the stack trace less useful.
Notable methods where this can be an issue are memdelete<T> and CallableCustomMethodPointer<T>::get_object (#77540) where the template argument(s) can be very important information that is not available with ICF enabled.
Changing this for optimize=speed_trace only as that optimization level is supposed to be better than speed for debugging, but MSVC generally doesn't have such an optimization level.
Docs: https://learn.microsoft.com/en-us/cpp/build/reference/opt-optimizations
Thanks!
| gharchive/pull-request | 2023-06-11T16:34:08 | 2025-04-01T06:44:20.639970 | {
"authors": [
"RedworkDE",
"akien-mga"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/78120",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1889297479 | Allow contextual plugins to persist temporarily
Fixes #72271
This makes the theme editor semi-persistent. If you edit a theme from the FileSystem Dock or directly in the inspector (via Edit), the editor will not be closed when something else takes over its editing context. A close button was added to close the editor manually.
The editor can still close automatically if its editing context vanishes (EditorProperty).
https://github.com/godotengine/godot/assets/2223172/7c23c1c8-be56-4085-83f3-b60316eab9d4
The bug is fixed.
But the case of editing a theme attached to a control is still important, I believe, and this trick doesn't cover that case.
The Theme Editor is closed, because the EditorProperty that was editing the theme is gone. I could make it persist, but then deselecting the node would keep the theme edited.
Done, but I had to do a questionable change. Inspector sets new object before clearing, so that plugins can know it before being cleared.
Going back in history is a font of all sorts of bugs 🙁 Closing the theme normally seems to work fine. But once you close it under the conditions demonstrated in the clip, you can no longer open the theme edit:
https://github.com/godotengine/godot/assets/11782833/02f4b251-d7b1-47d3-9304-7e0a5586874e
Regarding the unnatural thing that you had to do, it's probably fine. But to be absolutely safe you could add get_next_edited_object() or something like that? That's all you need from the hack, to be able to get a reference through the inspector dock to the upcoming edited object.
@KoBeWi Any news on fixing the bug (and maybe doing the pre-edit step in a less nasty manner)?
I will finish this one day.
I'll put it into draft for now, feel free to undraft once it's ready
Ok I fixed the history bug. I just forgot to unref Theme when closing the editor.
Also added get_next_edited_object() like you suggested.
As described on RC, there is currently an issue because behavior is different between the main inspector and sub inspectors. For subinspectors get_next_edited_object() doesn't work, because it gets cleared before we get to the theme editor, so it has nothing to work with. For subinspectors using the old get_edited_object() method is sufficient. But it works the other way around for the main inspector.
The solution can be as simple as moving the clearing step to the last possible moment:
diff --git a/editor/editor_inspector.cpp b/editor/editor_inspector.cpp
index 5985bdf525..233fcd10ce 100644
--- a/editor/editor_inspector.cpp
+++ b/editor/editor_inspector.cpp
@@ -3408,7 +3408,6 @@ void EditorInspector::edit(Object *p_object) {
per_array_page.clear();
object = p_object;
- next_object = nullptr;
if (object) {
update_scroll_request = 0; //reset
@@ -3418,6 +3417,10 @@ void EditorInspector::edit(Object *p_object) {
object->connect("property_list_changed", callable_mp(this, &EditorInspector::_changed_callback));
update_tree();
}
+
+ // Keep it available until the end so it works with both main and sub inspectors.
+ next_object = nullptr;
+
emit_signal(SNAME("edited_object_changed"));
}
However, you can still manage to permanently disable the theme editor with this PR. Open a theme resource, then edit its subresource. Use the inspector history to navigate back to the theme and close the editor. You will no longer be able to open it by any means.
https://github.com/godotengine/godot/assets/11782833/4a12faad-8f8f-4392-a2ba-ab7947f67daa
This does NOT happen if you open a control that has a theme as a subresource. Doing similar steps still restores the theme editor.
Fixed. Theme editor was clearing the inspector to remove itself, but the code for hiding editors did not handle self-owning plugins.
Thanks!
| gharchive/pull-request | 2023-09-10T19:33:03 | 2025-04-01T06:44:20.647953 | {
"authors": [
"KoBeWi",
"YuriSizov",
"akien-mga"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/81523",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1950656715 | Allow coexistence of GDScript and GDExtension virtual methods in the same object
…same object
Resolves https://github.com/godotengine/godot-cpp/issues/1224
So, I did some testing with a rather simple gdext project (thus, using Rust bindings - things should probably behave the same as with other bindings though, since very similar, if not identical, bugs have been observed compared to godot-cpp), and I wanted to share some of the results.
The source of the project is here: https://github.com/PgBiel/hello-gdext-wasm/tree/56916dabe21cc45eb721889a17d059d729384ccf (simply cd rust && cargo build to compile the extension).
Basically, the project contains a Player class created by the extension, subclassing Sprite2D. It overrides init, ready (prints "Rust is ready") and physics_process (rotates the node forever). Here I attempt to add a script to it and see the results.
Before this PR (4.2-beta1):
Without a script, everything works (prints "Rust is ready" and the node rotates, i.e. physics_process works).
By attaching a newly-created script (even when removing all of the default functions in it), nothing proceeds to happen ("Rust is ready" isn't printed and the node isn't rotated, so none of the extension's overrides are recognized anymore).
Explicitly overriding the ready function in GDScript, making it print Godot ready, results in only that being printed - Rust ready isn't printed and the node still doesn't rotate, indicating GDScript has indeed not only fully "disabled" but also overridden (the latter is, of course, to be expected) the extension's virtual method overrides.
Attempting to call super() from _ready in the script (in an attempt to recover the extension's behavior) results in Parser error: Cannot call the parent class' virtual function "_ready()" because it hasn't been defined.
With this PR:
Without a script, everything still works, like before.
By attaching a newly-created script, without overriding any functions (thus removing the default overrides), everything works ("Rust is ready" is printed and the node rotates), indicating this PR does make it so GDScript doesn't fully "disable" the extension's virtual methods anymore.
Explicitly overriding the ready function like before still makes it print Godot ready but not Rust is ready. Regardless, the node still rotates, so physics_process from the extension still works.
Attempting to call super() from GDScript's _ready results in the same parser error from before.
So I believe this PR fixes the very obvious bug of GDScript fully 'disabling' extensions' virtual method overrides even when the script does not override those particular methods at all, which is great!
Based on my debugging above, however, I don't think this will fully fix issues such as https://github.com/godotengine/godot-cpp/issues/1022 just yet (since e.g. you can't even call super to opt into preserving the extension's behavior as supposedly the method "isn't defined"). But this PR's fix would already be a very important step forward in this regard by fixing (issues such as) https://github.com/godotengine/godot-cpp/issues/1224 .
I've tested this PR and I'm also seeing the same results as @PgBiel. Thanks for the detailed report
[snip]
3. Explicitly overriding the `ready` function like before still makes it print `Godot ready` but not `Rust is ready`. Regardless, the node still rotates, so `physics_process` from the extension still works.
4. Attempting to call `super()` from GDScript's `_ready` results in the same parser error from before.
Based on what I see in core/os/main_loop.cpp void MainLoop::_bind_methods() I think _ready (and some others like _enter_tree) are not actually virtual functions so that is a correct error? (I've never done anything like that)
[snip]
3. Explicitly overriding the `ready` function like before still makes it print `Godot ready` but not `Rust is ready`. Regardless, the node still rotates, so `physics_process` from the extension still works.
4. Attempting to call `super()` from GDScript's `_ready` results in the same parser error from before.
Based on what I see in core/os/main_loop.cpp void MainLoop::_bind_methods() I think _ready (and some others like _enter_tree) are not actually virtual functions so that is a correct error? (I've never done anything like that)
Unfortunately, the exact same error (bar the method name) occurs with physics_process (apparently bound in the function you refer to). I believe ready is bound here:
https://github.com/godotengine/godot/blob/f8818f85e6c43cdf1277e8ae85eba19ca0a003b0/scene/main/node.cpp#L3527
The error I show seems to point to the code below - I don't know the full implications yet, but this seems to suggest to me that whatever it's trying to access with super() is a non-overridden virtual method (perhaps it's skipping the GDExtension's virtual method override and going straight for the native type's non-overridden impl?):
https://github.com/godotengine/godot/blob/f8818f85e6c43cdf1277e8ae85eba19ca0a003b0/modules/gdscript/gdscript_analyzer.cpp#L3298-L3303
But this is getting a bit off-topic for the PR now; I believe we should move this discussion to a separate issue (Godot maintainers, feel free to guide us!). The following issue seems the most appropriate: https://github.com/godotengine/godot/issues/72034
The following issue also seems to be at least tangentially related, but doesn't mention native types: https://github.com/godotengine/godot/issues/82267
Thanks! And congrats for your first merged Godot contribution :tada:
| gharchive/pull-request | 2023-10-18T21:06:05 | 2025-04-01T06:44:20.662651 | {
"authors": [
"PgBiel",
"akien-mga",
"anrp",
"ducklin5"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/83583",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
2034471172 | Fix script of a scene dependency not freed when closing a scene tab
Fixes #85983
EditorNode::_remove_scene from editor/editor_node.cpp calls editor_data.clear_script_icon_cache() for cleaning all script icon cache, previously, the implementation inside _remove_edited_scene(p_change_tab) loops through all scene tabs including the one we are closing, results in a call to EditorData::get_script_icon() on the closing scene tab, and create Ref<Script>s to that unloading scene lives inside _script_icon_cache.
This issue causes, primarily, the c-sharp scripts not to get freed when closing scene tabs, and might be a cause of #78513.
These wild script references may live inside the ScriptManagerBridge.ScriptTypeBiMap, resulting in the duplicate key issue when unloading the assebly.
modules/mono/glue/runtime_interop.cpp:1324 - System.ArgumentException: An item with the same key has already been added. Key: Game.WeaponAbilityData
at System.Collections.Generic.Dictionary`2.TryInsert(TKey key, TValue value, InsertionBehavior behavior)
at System.Collections.Generic.Dictionary`2.Add(TKey key, TValue value)
at Godot.Bridge.ScriptManagerBridge.ScriptTypeBiMap.Add(IntPtr scriptPtr, Type scriptType) in /root/godot/modules/mono/glue/GodotSharp/GodotSharp/Core/Bridge/ScriptManagerBridge.types.cs:line 23
at Godot.Bridge.ScriptManagerBridge.AddScriptBridge(IntPtr scriptPtr, godot_string* scriptPath) in /root/godot/modules/mono/glue/GodotSharp/GodotSharp/Core/Bridge/ScriptManagerBridge.cs:line 419
Please give your PR a name that describes what the PR does
You should target master not 4.2, changes can then be cherry picked.
If you need more information about the update process please see: here. If wanted I can fix the branch for you.
I don't think adding this removing_scene field inside editor_data is healthy, a better approach would directly modify the logic inside EditorNode::_remove_edited_scene(bool p_change_tab) since we just need to find a way for calling editor_data.remove_scene(old_index) before _set_current_scene(new_index). However, the program started crashing after modifying the logic there.
Original Logic
void EditorNode::_remove_edited_scene(bool p_change_tab) {
// When scene gets closed no node is edited anymore, so make sure the editors are notified before nodes are freed.
hide_unused_editors(SceneTreeDock::get_singleton());
int new_index = editor_data.get_edited_scene();
int old_index = new_index;
if (new_index > 0) {
new_index = new_index - 1;
} else if (editor_data.get_edited_scene_count() > 1) {
new_index = 1;
} else {
editor_data.add_edited_scene(-1);
new_index = 1;
}
if (p_change_tab) {
_set_current_scene(new_index);
}
editor_data.remove_scene(old_index);
_update_title();
scene_tabs->update_scene_tabs();
}
New Logic that Causes Crashing
void EditorNode::_remove_edited_scene(bool p_change_tab) {
// When scene gets closed no node is edited anymore, so make sure the editors are notified before nodes are freed.
hide_unused_editors(SceneTreeDock::get_singleton());
int old_index = editor_data.get_edited_scene();
editor_data.remove_scene(old_index);
if (p_change_tab) {
int new_index = editor_data.get_edited_scene();
if (new_index == 0) {
if (editor_data.get_edited_scene_count() == 0) {
editor_data.add_edited_scene(-1);
}
} else {
new_index = new_index - 1;
}
_set_current_scene(new_index);
}
_update_title();
scene_tabs->update_scene_tabs();
}
The Crash Message
CrashHandlerException: Program crashed
Engine version: Godot Engine v4.2.1.rc.mono.custom_build (daeb1c7292cbb426fd45c5ca98b1c7da40b390ba)
Dumping the backtrace. Please include this when reporting the bug to the project developer.
[0] Node::get_configuration_warnings_as_string (D:\UnityProject\godot\scene\main\node.cpp:3067)
[1] SceneTreeEditor::_add_nodes (D:\UnityProject\godot\editor\gui\scene_tree_editor.cpp:287)
[2] SceneTreeEditor::_update_tree (D:\UnityProject\godot\editor\gui\scene_tree_editor.cpp:617)
[3] call_with_variant_args_helper<SceneTreeEditor,bool,0> (D:\UnityProject\godot\core\variant\binder_common.h:308)
[4] call_with_variant_args_dv<SceneTreeEditor,bool> (D:\UnityProject\godot\core\variant\binder_common.h:451)
[5] MethodBindT<SceneTreeEditor,bool>::call (D:\UnityProject\godot\core\object\method_bind.h:337)
[6] Object::callp (D:\UnityProject\godot\core\object\object.cpp:775)
[7] Callable::callp (D:\UnityProject\godot\core\variant\callable.cpp:69)
[8] CallQueue::_call_function (D:\UnityProject\godot\core\object\message_queue.cpp:222)
[9] CallQueue::flush (D:\UnityProject\godot\core\object\message_queue.cpp:328)
[10] SceneTree::physics_process (D:\UnityProject\godot\scene\main\scene_tree.cpp:473)
[11] Main::iteration (D:\UnityProject\godot\main\main.cpp:3596)
[12] OS_Windows::run (D:\UnityProject\godot\platform\windows\os_windows.cpp:1474)
[13] widechar_main (D:\UnityProject\godot\platform\windows\godot_windows.cpp:182)
[14] _main (D:\UnityProject\godot\platform\windows\godot_windows.cpp:204)
[15] main (D:\UnityProject\godot\platform\windows\godot_windows.cpp:218)
[16] WinMain (D:\UnityProject\godot\platform\windows\godot_windows.cpp:232)
[17] __scrt_common_main_seh (D:\a\_work\1\s\src\vctools\crt\vcstartup\src\startup\exe_common.inl:288)
[18] <couldn't map PC to fn name>
-- END OF BACKTRACE --
I checked why the tab updates while deleting and seems like when closing the tab, update_scene_tabs() is called at least 3 times, for no reason. I can't reproduce the issue, so I can't test, but #83957 might fix it too.
The fix here looks like a workaround for bad tab management.
I checked why the tab updates while deleting and seems like when closing the tab, update_scene_tabs() is called at least 3 times, for no reason. I can't reproduce the issue, so I can't test, but #83957 might fix it too.
The fix here looks like a workaround for bad tab management.
Detailed reproduction steps are written in issue #85983.
I just want to emphasize that this specific fix is targeting the CSharp script system (un-freed (re-referenced) script reference causing issue & module crash on assembly reload), and I agree that #83957 is a better way as long as that resolves the CSharp scripting issue as well.
Detailed reproduction steps are written in issue
I know, I just don't have a C# setup to test. I tried with GDScript and other resources and it doesn't seem to happen.
I agree that https://github.com/godotengine/godot/pull/83957 is a better way as long as that resolves the CSharp scripting issue as well.
It's already merged, please give it a test.
It's already merged, please give it a test.
The test indicates the issue persists to exist.
It looks like the core of this issue is not calling update_scene_tabs() multiple times, but calling update_scene_tabs() in the middle of deleting a tab will cause a re-reference of script dependencies.
This is a flowchart for explaining this issue.
Well the issue is still updating the tabs before the old tab was completely removed. So I think a better solution would be delaying tab update.
Well the issue is still updating the tabs before the old tab was completely removed. So I think a better solution would be delaying tab update.
I will give that a try
Could you please squash your commits into one? Make sure that the final commit has a short but descriptive message (the title of this PR is a good option). See this documentation, if you need help with squashing.
Thanks, and congrats on your first merged Godot contribution!
| gharchive/pull-request | 2023-12-10T17:58:08 | 2025-04-01T06:44:20.675928 | {
"authors": [
"AThousandShips",
"Delsin-Yu",
"KoBeWi",
"YuriSizov"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/86008",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
2163682218 | Add Codium support to C# external editors
Selecting the VS Code option will now work if Codium is installed, picking it as the last option after attempting to run VS Code.
Not tested on any OS, so please test on macOS and Windows or Linux before merging.
This closes https://github.com/godotengine/godot-proposals/issues/6518.
Thanks!
| gharchive/pull-request | 2024-03-01T15:51:42 | 2025-04-01T06:44:20.678403 | {
"authors": [
"Calinou",
"Repiteo"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/89051",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
2188317549 | Editor: Customized Engine Build Configuration: auto detect usage of OpenXR
If OpenXR isn't enabled int the project settings, the auto detector will disable the OpenXR in profile
WARNING: Possible regression: Creates empty:
"disabled_build_options": {
},
How to test:
Enable disable the OpenXr by setting "xr/openxr/enabled" property. Go to the Project->Customize Engine Build Configuration... and click Detect from Project with disabled and enabled OpenXR
I've just thought it isn't very perfect... There could be added different xr/openxr/enabled.* with different values... When the build profile creator supports creation of one profile without platform dependency....
@reduz Hello
I've just was pointed about https://github.com/godotengine/godot/pull/62996 and I've just though as you are one of those who wrote this part of the editor and the one who cares about it. Could you look at this PR?
Anyway won't be merged alone. Should be modified and merged with auto-detection of other features...
| gharchive/pull-request | 2024-03-15T11:37:38 | 2025-04-01T06:44:20.682098 | {
"authors": [
"dustdfg"
],
"repo": "godotengine/godot",
"url": "https://github.com/godotengine/godot/pull/89512",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
1403444210 | add transaction hook for package gdb
增加transaction的commit和rollback两种hook
支持添加多个hook
允许通过条件判断，延迟执行。例如子事务中的hook在主事务commit或者rollback时执行
#1736
Codecov Report
Base: 76.49% // Head: 76.54% // Increases project coverage by +0.04% :tada:
Coverage data is based on head (8887cf7) compared to base (847f016).
Patch coverage: 96.82% of modified lines in pull request are covered.
:exclamation: Current head 8887cf7 differs from pull request most recent head d6974b0. Consider uploading reports for the commit d6974b0 to get more accurate results
Additional details and impacted files
@@ Coverage Diff @@
## master #2189 +/- ##
==========================================
+ Coverage 76.49% 76.54% +0.04%
==========================================
Files 580 581 +1
Lines 48176 48274 +98
==========================================
+ Hits 36853 36951 +98
+ Misses 9304 9301 -3
- Partials 2019 2022 +3
Flag
Coverage Δ
go-1.15-386
76.53% <96.82%> (+0.04%)
:arrow_up:
go-1.15-amd64
76.56% <96.82%> (+0.05%)
:arrow_up:
go-1.16-386
76.54% <96.82%> (+0.07%)
:arrow_up:
go-1.16-amd64
76.54% <96.82%> (+0.05%)
:arrow_up:
go-1.17-386
76.54% <96.82%> (+0.01%)
:arrow_up:
go-1.17-amd64
76.56% <96.82%> (+0.04%)
:arrow_up:
go-1.18-386
76.47% <96.82%> (+0.03%)
:arrow_up:
go-1.18-amd64
76.47% <96.82%> (+0.03%)
:arrow_up:
Flags with carried forward coverage won't be shown. Click here to find out more.
Impacted Files
Coverage Δ
database/gdb/gdb_core_transaction_hook.go
95.12% <95.12%> (ø)
database/gdb/gdb_core_transaction.go
65.53% <100.00%> (+0.80%)
:arrow_up:
database/gdb/gdb_model_hook.go
90.38% <100.00%> (+2.28%)
:arrow_up:
os/gfsnotify/gfsnotify_watcher_loop.go
83.19% <0.00%> (ø)
test/gtest/gtest_util.go
54.34% <0.00%> (+0.86%)
:arrow_up:
database/gdb/gdb_model_transaction.go
25.00% <0.00%> (+25.00%)
:arrow_up:
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.
:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.
@qinyuguang 我没想到有具体应用的业务场景，我想的是增加了ORM的复杂度，带来的收益有多大？
@qinyuguang 另外，我建议可以提供Commit的HOOK，这样可能更有价值一些，而不是单独给TX加个HOOK功能。
@qinyuguang 我没想到有具体应用的业务场景，我想的是增加了ORM的复杂度，带来的收益有多大？
写后读的场景会用到
比如事务消息的场景，可能会在多个子事务中insert数据，并会在主事务commit后，起个goroutine去查询插入的数据并发送mq
改为DoCommit实现
https://github.com/gogf/gf/pull/2221
| gharchive/pull-request | 2022-10-10T16:37:32 | 2025-04-01T06:44:20.713773 | {
"authors": [
"codecov-commenter",
"gqcn",
"qinyuguang"
],
"repo": "gogf/gf",
"url": "https://github.com/gogf/gf/pull/2189",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
119186169 | Team repo user access
In 0.7.20.1121 Beta on x64 windows server team members (not owners) of organization can't access (404) to private repository page.
Because the user is not in any team has access to that private repository.
Close now but open for continuing discussion.
Hmm maybe more meaningful pages explaining they don't have access?
Members in the same organization are able to view activities in the dashboard on repos that are not accessible to them. Clicking on it will return 404 which might be mistaken for a bug.
Or a better suggestion is to hide all the activities for repos that they do not have access.
@bhchew
Or a better suggestion is to hide all the activities for repos that they do not have access.
Good point, please file an issue for this.
yep will do!
| gharchive/issue | 2015-11-27T12:10:39 | 2025-04-01T06:44:20.720289 | {
"authors": [
"ChkonCSD",
"Unknwon",
"bhchew"
],
"repo": "gogits/gogs",
"url": "https://github.com/gogits/gogs/issues/2043",
"license": "MIT",
"license_type": "permissive",
"license_source": "github-api"
} |
942079196 | Shortcodes not behaving the same with --minify flag than without
What version of Hugo are you using (hugo version)?
$ hugo version v0.83.1
Does this issue reproduce with the latest release?
Yes, my website has been deployed with GitHub Actions using 'latest' for the Hugo version, so I assume it used v0.85.0, which is the latest version right now and the problem occured there too.
The problem briefly explained
Building my site with the --minify flag makes my shortcodes behave slightly different. Precisely, there is no space between the word before the content wrapped into the shortcode layout and this content.
Without using --minify, the same Markdown yields different result (with a normal space, as expected).
See The problem in detail, Example 1 and Example 2 for further information and examples to reproduce the issue.
The problem in detail
I noticed that my shortcodes don’t behave the same in the deployed site (hosted with GitHub Pages, deployed with GitHub Actions as described in the official documentation) as in a local build (built locally with hugo server or hugo server -D, localhost).
In my locally built site, the shortcode behaves as expected. It’s a simple shortcode formatting the content of it. The only problem is, that even though I actually added a space between the text and the declaration of the shortcode (see examples below), it doesn’t add a space in the deployed site on GitHub Pages, whereas it behaves as expected in a locally built site.
After a conversation with @bep on the Hugo forum, we came to the conclusion that it's most likely an issue with the built-in minifier in Hugo, because I never used --minify in my local builds, whereas the action deployed the page with GitHub pages using --minify. That's why I'm creating this issue. Going even further, I tried to use --minify for my local build (hugo server --minify in the command line) and I managed to reproduce the issue on a local build too.
Example 1
Markdown content (inside a simple block of text):
Apart from the {{<link `index.html` `_blank`>}}official API reference{{</link>}}
Shortcode link.html:
<head>
<style>
.shortcode-link {
color:#fc3f65;
}
.shortcode-link:visited {
color:#fc3f65;
}
.shortcode-link:active {
color: #c92d4c;
}
</style>
</head>
{{ $p0 := .Get 0 }}
{{ $p1 := .Get 1 }}
{{ $inner := .Inner }}
<a href="{{ $p0 }}" target="{{ $p1 }}" class="shortcode-link">{{$inner}}</a>
Without --minify (expected result, with space):
With --minify (unexpected result, without a space between the and official API reference):
Example 2
Markdown content (inside a simple block of text):
The {{<codeline>}}[SerializeField]{{</codeline>}} attribute makes sure that the field is exposed to the inspector
Shortcode codeline.html:
<head>
<style>
.single-code-line {
font-family: CodeMono;
background-color: #202020;
color:white;
padding-left: 7px;
padding-right: 7px;
text-align: justify;
word-wrap: break-word;
}
</style>
</head>
<code class="single-code-line">{{ .Inner }}</code>
Without --minify (expected result, with space):
With --minify (unexpected result, no space between The and [SerializeField]):
First, <head> does not belong here. It must be the first child of the <html> element.
See https://developer.mozilla.org/en-US/docs/Web/HTML/Element/head.
Second, <style> does not belong here.
See https://developer.mozilla.org/en-US/docs/Web/HTML/Element/style#technical_summary.
I don't think that is the reason for the issue, mostly because the link is still red in the first example and in the second example nothing is removed. Most modern browsers can work around wrong markup. The issue at hand seems to be that hugo minify removes spaces around the shortcode in certain circumstances. Might be an upstream issue.
Yes, excuse me for these examples. It was a quick example I made based on my code, which actually does look like that, just not with the . I included it, because I had previous positive experiences by adding the . But with or without that, the issue is still the same.
The issue at hand seems to be that hugo minify removes spaces around the shortcode in certain circumstances. Might be an upstream issue.
So, --minify is still the issue, right? Can I somehow workaround that (without having to abandon --minify, which isn't recommended anyway or is it really an internal issue, which would need to be fixed on your side?
{{- removes all spaces left of the golang-template-tag. -}} removes all whitespaces right of the tag.
@duck-dev
This is an example of the HTML produced by your markdown+shortcode
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, minimum-scale=1.0, initial-scale=1.0">
<title>Test</title>
</head>
<body>
<h1>Test</h1>
<p>Apart from the <style>
.shortcode-link {
color: #fc3f65;
}
.shortcode-link:visited {
color: #fc3f65;
}
.shortcode-link:active {
color: #c92d4c;
}
</style>
<a href="path" target="_blank" class="shortcode-link">official API reference</a>
</p>
</body>
</html>
You have a <style> element within a <p> element. Now copy and paste this HTML into a validator:
Either move the <style> element, or place the CSS in a CSS file.
Yes, the browsers that I tested did not complain about the invalid HTML, but that doesn't mean that the minifier should be modified to accommodate HTML that doesn't follow the specification.
A closed issue is ... closed. If this issue is still relevant, open another issue.
| gharchive/issue | 2021-07-12T13:43:21 | 2025-04-01T06:44:20.767099 | {
"authors": [
"bep",
"davidsneighbour",
"duck-dev",
"jmooring"
],
"repo": "gohugoio/hugo",
"url": "https://github.com/gohugoio/hugo/issues/8748",
"license": "Apache-2.0",
"license_type": "permissive",
"license_source": "github-api"
} |
1698542007 | Add missing space in documentation
This is a minor typo fix.
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.
| gharchive/pull-request | 2023-05-06T08:51:24 | 2025-04-01T06:44:20.769773 | {
"authors": [
"CLAassistant",
"kirillbobyrev"
],
"repo": "gohugoio/hugo",
"url": "https://github.com/gohugoio/hugo/pull/10929",
"license": "Apache-2.0",
"license_type": "permissive",
"license_source": "github-api"
} |
487789124 | Escape RSS description as XML
This is to avoid including characters invalid for XML.
Fixes #3268
I just found an Hugo built rss feed failing at https://pawelgrzybek.com/feed.xml. I think it is due to this old issue. I tested the latest version of Hugo and it still fails as described in #3286. The fix written there by @horgh still seems to work as well, so I've rebased his branch here for your consideration. Thanks.
I don't mind adding this if we need it, but Is wrapping it in CDATA not an option for this partiular case?
@bep I would've said yes CDATA would be enough. But looking through the issue comments again it looks like that was tried and didn't work (I haven't tried to reproduce this.):
@moorereason commented on 8 Oct 2018
I attempted to use <!CDATA[ ... ]]> around the contents, but that didn't fix the issue. The illegal character within the CDATA block still violates the XML spec. (Additionally, we use html/template for the RSS feed, so the <!CDATA gets escaped if we try to use that in the RSS template, anyway.)
There are ways to prevent that. @moorereason may remember ...?
| gharchive/pull-request | 2019-08-31T16:50:52 | 2025-04-01T06:44:20.773314 | {
"authors": [
"bep",
"pwindle"
],
"repo": "gohugoio/hugo",
"url": "https://github.com/gohugoio/hugo/pull/6292",
"license": "Apache-2.0",
"license_type": "permissive",
"license_source": "github-api"
} |
198916499 | To integrate Go's unit test report
I recommended to integrate Go's unit test report (GoConvey: https://github.com/smartystreets/goconvey), the Go's unit test report is very important and useful.
@penhauer-xiao this looks like a library that builds on top of go test? I'm not sure how we'd use this, it seems like it would depend on whether our users are running GoConvey or not. In any case I don't think we have plans to add unit test coverage at the moment, since many of our users have quite large repos and it would just be impossible with our current architecture to run tests efficiently. Thanks for filing an issue though.
| gharchive/issue | 2017-01-05T09:57:41 | 2025-04-01T06:44:20.778652 | {
"authors": [
"penhauer-xiao",
"shawnps"
],
"repo": "gojp/goreportcard",
"url": "https://github.com/gojp/goreportcard/issues/171",
"license": "apache-2.0",
"license_type": "permissive",
"license_source": "bigquery"
} |
578952017 | cc1.exe: out of memory allocating 65536 bytes
build example widget
winx64,
the newly GI
void __cgo_f_2043_5(void) { static const char __cgo_undefined__5[] = (glowProgramUniform1dEXT); }
#line 2044 "not-declared"
void __cgo_f_2044_1(void) { __typeof__(glowProgramUniform1dv) *__cgo_undefined__1; }
#line 2044 "not-type"
void __cgo_f_2044_2(void) { glowProgramUniform1dv *__cgo_undefined__2; }
#line 2044 "not-int-const"
void __cgo_f_2044_3(void) { enum { __cgo_undefined__3 = (glowProgramUniform1dv)*1 }; }
#line 2044 "not-num-const"
void __cgo_f_2044_4(void) { static const double __cgo_undefined__4 = (glowProgramUniform1dv); }
#line 2044 "not-str-lit"
void __cgo_f_2044_5(void) { static const char __cgo_undefined__5[] = (glowProgramUniform1dv); }
#line 2045 "not-declared"
.......
........
void __cgo_f_2542_5(void) { static const char __cgo_undefined__5[] = (glowWindowRectanglesEXT); }
#line 1 "completed"
int __cgo__1 = __cgo__2;
full error output:
cc1.exe: out of memory allocating 65536 bytes
did you use http://tdm-gcc.tdragon.net/ ?
closing for now -- reopoen if you have more data.
| gharchive/issue | 2020-03-11T01:26:34 | 2025-04-01T06:44:20.780716 | {
"authors": [
"gonboy",
"rcoreilly"
],
"repo": "goki/gi",
"url": "https://github.com/goki/gi/issues/379",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
94610874 | Godoc reports Internal Server Error
I have created a new version of a library with a breaking change from the previous version. I am using Gustavo Niemeyer's gopkg.in facility for versioning APIs. The godoc URL to the new version is http://godoc.org/gopkg.in/stomp.v2.
This URL reports one of two error messages:
Internal server error.
Another error about failing to retrieve code. I only see this very occasionally and cannot reproduce at the moment.
I have communicated this problem to Gustavo, and he thinks it looks like a problem with godoc.
Can anyone help me with diagnosing this problem?
I can see that this is a duplicate of issue #286.
| gharchive/issue | 2015-07-12T23:52:37 | 2025-04-01T06:44:20.787826 | {
"authors": [
"jjeffery"
],
"repo": "golang/gddo",
"url": "https://github.com/golang/gddo/issues/288",
"license": "bsd-3-clause",
"license_type": "permissive",
"license_source": "bigquery"
} |
90089834 | x/tools/go/types: overflow in int -> string
Gc rejects to compile the following program:
package a
var a = string(9223372036854775808)
saying:
overflow in int -> string
go/types compile it successfully.
Compilers must agree on whether it is a valid Go program or not.
on commit af817890aa7b628234075a6cb01a3a435fb8317d
Given the spec as is, I believe this should compile w/o errors. The spec says:
http://tip.golang.org/ref/spec#Conversions
A constant value x can be converted to type T in any of these cases:
...
x is an integer constant and T is a string type. The same rule as for non-constant x applies in this case.
and:
http://tip.golang.org/ref/spec#Conversions_to_and_from_a_string_type
Converting a signed or unsigned integer value to a string type yields a string containing the UTF-8 representation of the integer. Values outside the range of valid Unicode code points are converted to "\uFFFD".
There's no mention of a compile-time (or even run-time) error. The integer should be converted into "\uFFFD".
Closed by mistake.
Then this is a bug in gc, right?
@dvyukov yes, I believe this is a gc compiler bug.
We could make the spec more restrictive, but if the valid Unicode ranges changes, the compiler would have to be updated (that said, it would probably be a range extension and thus a backward-compatible change).
Too late for Go 1.5. The only possible visible effect of making changes to fix hypothetical bugs like this one is introducing real ones.
This was fixed via https://go-review.googlesource.com/#/c/21344/ (fix for #15039).
| gharchive/issue | 2015-06-22T12:51:17 | 2025-04-01T06:44:20.793358 | {
"authors": [
"dvyukov",
"griesemer",
"rsc"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/11330",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
94726354 | tour: no coding environment is available on the page
Context: /concurrency/7
The page is about an exercise. Meanwhile there's nothing on the left to write the code or run it.
The excercise is too big so it was split into two pages - /concurrency/7 and /concurrency/8
While this is working as intended, this might require some refactoring to make it more obvious.
CC @adg @campoy
Would adding a small clarification at the bottom of the page help?
https://go-review.googlesource.com/#/c/12179/1
| gharchive/issue | 2015-07-13T13:54:21 | 2025-04-01T06:44:20.795554 | {
"authors": [
"campoy",
"ianlancetaylor",
"severeone",
"snakeru"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/11692",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
175849776 | test/run.go: errcheck ineffective when error pattern matches test case file position information.
What version of Go are you using (go version)?
tip @ https://github.com/golang/go/commit/ed8f207940c8787d344664a43071b235e2ce5c68
What operating system and processor architecture are you using (go env)?
GOARCH="amd64"
GOBIN=""
GOEXE=""
GOHOSTARCH="amd64"
GOHOSTOS="linux"
GOOS="linux"
GOPATH="/home/jnml"
GORACE=""
GOROOT="/home/jnml/go"
GOTOOLDIR="/home/jnml/go/pkg/tool/linux_amd64"
CC="gcc"
GOGCCFLAGS="-fPIC -m64 -pthread -fmessage-length=0 -fdebug-prefix-map=/tmp/go-build952459844=/tmp/go-build -gno-record-gcc-switches"
CXX="g++"
CGO_ENABLED="1"
Discussion
The expected error pattern is checked against the actual error message at run.go:857. Invoking, for example, go run run.go - ./method2.go in $GOROOT/test, the performed regexp vs errmsg pairs are
`receiver.* pointer|invalid pointer or interface receiver` `method2.go:18: invalid receiver type P (P is a pointer type)`
`receiver.* pointer|invalid pointer or interface receiver` `method2.go:19: invalid receiver type *P1 (P1 is a pointer type)`
`receiver.*interface|invalid pointer or interface receiver` `method2.go:24: invalid receiver type I (I is an interface type)`
`receiver.*interface|invalid pointer or interface receiver` `method2.go:25: invalid receiver type *I1 (I1 is an interface type)`
`method` `method2.go:31: (*Val).val undefined (type *Val has no method val)`
`method` `method2.go:36: pv.val undefined (type *Val is pointer to interface, not interface)`
`method` `method2.go:37: pv.val undefined (type *Val is pointer to interface, not interface)`
Error patterns for lines {31,36,37} above are considered satisfied regardless of the compiler error message because the pattern matches the test case file position information itself, ie. errcheck fails to detect the actual mismatches at, in this example, lines 36 and 37.
FTR (the diff is a QND hack, not a fix proposal)
jnml@r550:~/go/test$ git diff -u
diff --git a/test/run.go b/test/run.go
index 19210b4..d7ea67b 100644
--- a/test/run.go
+++ b/test/run.go
@@ -854,7 +854,7 @@ func (t *test) errorCheck(outStr string, fullshort ...string) (err error) {
matched := false
n := len(out)
for _, errmsg := range errmsgs {
- if we.re.MatchString(errmsg) {
+ if we.re.MatchString(strings.SplitN(errmsg, " ", 2)[1]) {
matched = true
} else {
out = append(out, errmsg)
jnml@r550:~/go/test$ go run run.go
# go run run.go -- method2.go
method2.go:36: no match for `method` in:
method2.go:36: pv.val undefined (type *Val is pointer to interface, not interface)
method2.go:37: no match for `method` in:
method2.go:37: pv.val undefined (type *Val is pointer to interface, not interface)
Unmatched Errors:
method2.go:36: pv.val undefined (type *Val is pointer to interface, not interface)
method2.go:37: pv.val undefined (type *Val is pointer to interface, not interface)
FAIL method2.go 0.076s
# go run run.go -- fixedbugs/bug332.go
bug332.go:17: no match for `bug332` in:
bug332.go:17: undefined: T
Unmatched Errors:
bug332.go:17: undefined: T
FAIL fixedbugs/bug332.go 0.062s
exit status 1
jnml@r550:~/go/test$
| gharchive/issue | 2016-09-08T20:02:19 | 2025-04-01T06:44:20.799944 | {
"authors": [
"cznic"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/17030",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
209692704 | internal/poll: Changed error string breaks user application
Please answer these questions before submitting your issue. Thanks!
What version of Go are you using (go version)?
go version devel +3792db5 Mon Feb 13 18:36:28 2017 +0000 linux/amd64
What operating system and processor architecture are you using (go env)?
GOARCH="amd64"
GOBIN=""
GOEXE=""
GOHOSTARCH="amd64"
GOHOSTOS="linux"
GOOS="linux"
GOPATH="/home/mura/devel/go"
GORACE=""
GOROOT="/home/mura/tools/go"
GOTOOLDIR="/home/mura/tools/go/pkg/tool/linux_amd64"
GCCGO="gccgo"
CC="gcc"
GOGCCFLAGS="-fPIC -m64 -pthread -fmessage-length=0 -fdebug-prefix-map=/tmp/go-build174238897=/tmp/go-build -gno-record-gcc-switches"
CXX="g++"
CGO_ENABLED="1"
PKG_CONFIG="pkg-config"
CGO_CFLAGS="-g -O2"
CGO_CPPFLAGS=""
CGO_CXXFLAGS="-g -O2"
CGO_FFLAGS="-g -O2"
CGO_LDFLAGS="-g -O2"
What did you do?
make integration-test for the code I am working on.
https://github.com/go-gitea/gitea/pull/741#discussion_r102649390
But I think that doesn't matter after all.
If possible, provide a recipe for reproducing the error.
A complete runnable program is good.
A link on play.golang.org is best.
What did you expect to see?
All tests passed.
What did you see instead?
All tests failed.
It turned out that one of the dependent libraries ( (See https://github.com/facebookgo/httpdown/blob/master/httpdown.go#L338) uses the exact error string to make its decision and the string in question was changed at some point of time (https://github.com/golang/go/commit/3792db518327c685da17ca6c6faa4e1d2da4c33c#diff-18782360650b1a2ccfee593591ff45bbR17). This may not be the fault of the error string changing but I'd like to know what the recommended solution (an idimatic way for differentiating errors) is for the case. And maybe the possible impact can be prepared before the next production release.
As far as I know, error strings are not under the Go1 compatibility contract. If your program depends on them, it might break in the future or not be portable at all (e.g. some os errors change between operating systems, I believe).
The way to do this in a portable and backwards/forwards compatible way is to check for error types, not error messages. Although that won't be possible if the package just uses errors.New().
This is related to #4373.
Marking for Go 1.9.
This may not be the fault of the error string changing but I'd like to know what the recommended solution (an idimatic way for differentiating errors) is for the case
Indeed.
| gharchive/issue | 2017-02-23T08:16:55 | 2025-04-01T06:44:20.807113 | {
"authors": [
"bradfitz",
"ianlancetaylor",
"mvdan",
"typeless"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/19252",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
226148915 | cmd/compile, go/types: disagreement about whether 6e5518446744 is well-formed
https://play.golang.org/p/_x_9PkQc5F
package main
const x = 6e5518446744
go/types says yes.
cmd/compile says no:
tmp/sandbox552762073/main.go:3: malformed constant: 6e5518446744
@griesemer @mdempsky
The spec says:
Implementation restriction: Although numeric constants have arbitrary precision in the language, a compiler may implement them using an internal representation with limited precision. That said, every implementation must:
Represent integer constants with at least 256 bits.
Represent floating-point constants, including the parts of a complex constant, with a mantissa of at least 256 bits and a signed binary exponent of at least 16 bits.
Give an error if unable to represent an integer constant precisely.
Give an error if unable to represent a floating-point or complex constant due to overflow.
So it's okay for go/types to accept it; but since the exponent exceeds 16 bits, it's also okay for cmd/compile to reject it with an error.
However, we could definitely emit a better error message. This is the TODO in gc/mpfloat.go:
f, ok := a.Val.SetString(as)
if !ok {
// At the moment we lose precise error cause;
// the old code additionally distinguished between:
// - malformed hex constant
// - decimal point in hex constant
// - constant exponent out of range
// - decimal point and binary point in constant
// TODO(gri) use different conversion function or check separately
yyerror("malformed constant: %s", as)
| gharchive/issue | 2017-05-04T00:13:58 | 2025-04-01T06:44:20.811028 | {
"authors": [
"josharian",
"mdempsky"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/20232",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
250153627 | Bootstrapping fails on MIPS64
What version of Go are you using (go version)?
go version devel +c8e9fd5 Mon Aug 14 01:38:08 2017 +0000 linux/mips64
From go version go1.8.3 linux/amd64 GOOS=linux GOARCH=mips64 ./bootstrap.bash
What operating system and processor architecture are you using (go env)?
GOARCH="mips64"
GOBIN="/home/myusername/go/bin"
GOEXE=""
GOHOSTARCH="mips64"
GOHOSTOS="linux"
GOOS="linux"
GOPATH="/root/go"
GORACE=""
GOROOT="/home/myusername/go"
GOTOOLDIR="/home/myusername/go/pkg/tool/linux_mips64"
GCCGO="gccgo"
CC="gcc"
GOGCCFLAGS="-fPIC -mabi=64 -pthread -fmessage-length=0 -fdebug-prefix-map=/tmp/go-build798056280=/tmp/go-build"
CXX="g++"
CGO_ENABLED="1"
CGO_CFLAGS="-g -O2"
CGO_CPPFLAGS=""
CGO_CXXFLAGS="-g -O2"
CGO_FFLAGS="-g -O2"
CGO_LDFLAGS="-g -O2"
PKG_CONFIG="pkg-config"
"$GOTOOLDIR"/dist env -p
CC="gcc"
CC_FOR_TARGET="gcc"
GOROOT="/home/myusername/go"
GOBIN="/home/myusername/go/bin"
GOARCH="mips64"
GOOS="linux"
GOHOSTARCH="mips64"
GOHOSTOS="linux"
GOTOOLDIR="/home/myusername/go/pkg/tool/linux_mips64"
PATH="/home/myusername/go/bin:/home/myusername/go/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
What did you do?
I tried to run GOROOT_BOOTSTRAP=~/go-linux-mips64-bootstrap ./all.bash. It was failing. I ran ./all.bash --dist-tool and then started manually going through the commands in ./make.bash.
I ran CC=$CC_FOR_TARGET "$GOTOOLDIR"/go_bootstrap install $GO_FLAGS -gcflags "$GO_GCFLAGS" -ldflags "$GO_LDFLAGS" -v std cmd || echo "Oh noes! $?"
What did you expect to see?
A successful completion of go_bootstrap install std cmd.
What did you see instead?
go_bootstrap install std cmd failed with exit code 2 and the following output:
...
runtime/internal/sys
runtime/internal/atomic
runtime
errors
internal/race
sync/atomic
unicode
sync
io
unicode/utf8
math
bytes
bufio
syscall
strconv
reflect
time
internal/poll
os
strings
fmt
sort
path/filepath
flag
log
cmd/internal/src
encoding/binary
cmd/internal/objabi
cmd/internal/goobj
cmd/internal/sys
container/list
path
math/bits
debug/dwarf
compress/flate
hash
hash/adler32
compress/zlib
debug/gosym
debug/elf
debug/macho
debug/pe
debug/plan9obj
cmd/vendor/golang.org/x/arch/arm/armasm
cmd/vendor/golang.org/x/arch/ppc64/ppc64asm
cmd/vendor/golang.org/x/arch/x86/x86asm
io/ioutil
regexp/syntax
regexp
text/tabwriter
archive/tar
cmd/internal/objfile
hash/crc32
archive/zip
cmd/addr2line
compress/bzip2
compress/gzip
compress/lzw
container/heap
container/ring
context
crypto
crypto/subtle
crypto/cipher
math/rand
crypto/aes
crypto/des
math/big
crypto/sha512
crypto/hmac
crypto/internal/cipherhw
crypto/md5
internal/syscall/unix
crypto/rc4
crypto/sha1
crypto/sha256
encoding/hex
encoding/base64
encoding/pem
internal/nettrace
internal/singleflight
go/token
go/scanner
go/ast
crypto/dsa
crypto/elliptic
encoding/asn1
crypto/rand
crypto/rsa
crypto/ecdsa
crypto/x509/pkix
os/exec
go/parser
go/printer
vendor/golang_org/x/crypto/chacha20poly1305/internal/chacha20
vendor/golang_org/x/crypto/poly1305
vendor/golang_org/x/crypto/chacha20poly1305
vendor/golang_org/x/crypto/curve25519
database/sql/driver
cmd/cgo
database/sql
encoding
encoding/ascii85
encoding/base32
encoding/csv
encoding/gob
unicode/utf16
encoding/json
runtime/cgo
# runtime/cgo
_cgo_export.c:2:20: fatal error: stdlib.h: No such file or directory
compilation terminated.
encoding/xml
vendor/golang_org/x/net/http2/hpack
vendor/golang_org/x/text/transform
vendor/golang_org/x/text/unicode/bidi
vendor/golang_org/x/text/unicode/norm
vendor/golang_org/x/text/secure/bidirule
net/url
mime
mime/quotedprintable
net/http/internal
vendor/golang_org/x/net/idna
text/template/parse
go/constant
text/template
go/format
go/types
go/doc
go/build
text/scanner
hash/crc64
hash/fnv
html
html/template
go/internal/gccgoimporter
go/internal/gcimporter
go/internal/srcimporter
image/color
go/importer
image
image/color/palette
index/suffixarray
internal/cpu
internal/syscall/windows
internal/syscall/windows/registry
internal/syscall/windows/sysdll
runtime/debug
runtime/trace
testing
image/internal/imageutil
image/draw
image/gif
image/jpeg
image/png
internal/testenv
internal/trace
math/cmplx
runtime/pprof
net/internal/socktest
os/signal
runtime/pprof/internal/profile
runtime/race
testing/internal/testdeps
testing/iotest
testing/quick
vendor/golang_org/x/text/secure
vendor/golang_org/x/text/unicode
cmd/api
cmd/internal/dwarf
cmd/internal/obj
cmd/internal/obj/arm
cmd/internal/obj/arm64
cmd/internal/obj/mips
cmd/internal/obj/ppc64
cmd/internal/obj/s390x
cmd/internal/obj/x86
cmd/asm/internal/flags
cmd/asm/internal/lex
cmd/internal/bio
cmd/compile/internal/types
cmd/asm/internal/arch
cmd/asm/internal/asm
cmd/compile/internal/ssa
cmd/asm
cmd/compile/internal/syntax
cmd/internal/gcprog
cmd/compile/internal/test
cmd/internal/browser
cmd/cover
cmd/dist
cmd/doc
go build cmd/compile/internal/ssa: /home/myusername/go/pkg/tool/linux_mips64/compile: signal: killed
cmd/fix
cmd/go/internal/cfg
cmd/go/internal/str
cmd/go/internal/base
cmd/go/internal/buildid
cmd/go/internal/load
cmd/go/internal/doc
cmd/go/internal/help
cmd/go/internal/work
cmd/go/internal/fix
cmd/go/internal/fmtcmd
cmd/go/internal/cmdflag
cmd/go/internal/tool
cmd/go/internal/version
cmd/gofmt
cmd/go/internal/envcmd
cmd/go/internal/clean
cmd/go/internal/generate
cmd/go/internal/list
cmd/go/internal/run
cmd/go/internal/test
cmd/go/internal/vet
cmd/link/internal/ld
cmd/nm
cmd/objdump
cmd/pack
cmd/vendor/github.com/google/pprof/internal/elfexec
cmd/vendor/github.com/google/pprof/profile
cmd/link/internal/amd64
cmd/link/internal/arm
cmd/link/internal/arm64
cmd/link/internal/mips
cmd/link/internal/mips64
cmd/link/internal/ppc64
cmd/link/internal/s390x
cmd/link/internal/x86
cmd/vendor/github.com/google/pprof/internal/plugin
cmd/vendor/github.com/ianlancetaylor/demangle
cmd/link
cmd/vendor/github.com/google/pprof/internal/measurement
cmd/vendor/github.com/google/pprof/internal/graph
cmd/vendor/github.com/google/pprof/internal/binutils
cmd/vendor/github.com/google/pprof/internal/report
cmd/vendor/github.com/google/pprof/internal/symbolz
cmd/vendor/github.com/google/pprof/third_party/svg
cmd/vendor/github.com/google/pprof/internal/proftest
cmd/vet/internal/cfg
cmd/vet/internal/whitelist
cmd/vet
Oh noes! 2
# runtime/cgo
_cgo_export.c:2:20: fatal error: stdlib.h: No such file or directory
This seems there is no system headers installed. Either install them, or try setting CGO_ENABLED=0.
go build cmd/compile/internal/ssa: /home/myusername/go/pkg/tool/linux_mips64/compile: signal: killed
This is probably OOM-killed. How much memory does the machine have? Try set up some swap, or set GOGC=50 or some smaller number (and expect longer time to finish).
I installed libc6-dev and got the following result (partial build?):
cmd/addr2line
runtime/cgo
# runtime/cgo
In file included from /usr/include/stdlib.h:25:0,
from _cgo_export.c:2:
/usr/include/features.h:323:26: fatal error: bits/predefs.h: No such file or directory
compilation terminated.
cmd/api
cmd/asm
cmd/compile/internal/ssa
cmd/cover
cmd/dist
cmd/doc
cmd/fix
cmd/link
go build cmd/compile/internal/ssa: /home/firelizzard/go/pkg/tool/linux_mips64/compile: signal: killed
cmd/nm
cmd/objdump
cmd/pack
cmd/vet
Oh noes! 2
I'm on an EdgeRouter Lite, which has <0.5GB of RAM. I'll try GOGC=50
You need gcc-multilib if you are crosscompiling
That's unlikely to be enough, IME on a 64 bit system you need 768mb of ram
per core.
On Tue, 15 Aug 2017, 08:37 Ethan Reesor notifications@github.com wrote:
I'm on an EdgeRouter Lite, which has <0.5GB of RAM. I'll try GOGC=50
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
https://github.com/golang/go/issues/21444#issuecomment-322328992, or mute
the thread
https://github.com/notifications/unsubscribe-auth/AAAcAzl-Kr2F5bGQRbjdeUg-z4YUgym2ks5sYMwegaJpZM4O25bo
.
Oh joy. Is there any way I can cross-compile the entire toolchain from a different machine? Would it be safe to use go_bootstrap as if it were just go?
I'm trying to do this because I want to build github.com/elastic/beats/tree/master/packetbeat for the ERL but that requires some CGO stuff and I've had a hell of a time trying to get my gentoo box's GCC to cross-compile for MIPS.
I tried installing gcc-multilib, but installing things isn't working well because of some weird BS with aptitude and the fact that my system has an ever so slightly newer version of libc and libc-bin than the newest available MIPS version of libc6-dev (lots of complaints about wanting some-pkg ...deb7u10 but having some-pkg ...deb7u11).
# runtime/cgo
In file included from /usr/include/stdlib.h:25:0,
from _cgo_export.c:2:
/usr/include/features.h:323:26: fatal error: bits/predefs.h: No such file or directory
compilation terminated.
I think you also need linux-headers or like. (Or, set CGO_ENABLED=0)
I'm on an EdgeRouter Lite, which has <0.5GB of RAM. I'll try GOGC=50
I have a Loongson 2E with 512 MB RAM. I set up ~1 GB swap, and it works even with default GOGC=100.
@firelizzard18
https://golang.org/doc/install/source
To cross-compile a bootstrap tool chain from source, which is necessary on systems Go 1.4 did not target (for example, linux/ppc64le), install Go on a different system and run bootstrap.bash.
On Tue, Aug 15, 2017 at 9:01 AM, Ethan Reesor notifications@github.com
wrote:
Oh joy. Is there any way I can cross-compile the entire toolchain from a
different machine? Would it be safe to use go_bootstrap as if it were
just go?
That probably won't have the right cgo parts enabled. I don't know for
sure, I've never tried.
I'm trying to do this because I want to build
github.com/elastic/beats/tree/master/packetbeat for the ERL but that
requires some CGO stuff and I've had a hell of a time trying to get my
gentoo box's GCC to cross-compile for MIPS.
I tried installing gcc-multilib, but installing things isn't working well
because of some weird BS with aptitude and the fact that my system has an
ever so slightly newer version of libc and libc-bin than the newest
available MIPS version of libc6-dev (lots of complaints about wanting some-pkg
...deb7u10 but having some-pkg ...deb7u11).
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/golang/go/issues/21444#issuecomment-322333027, or mute
the thread
https://github.com/notifications/unsubscribe-auth/AAAcA0_3eyV6AJlUIQoWTRoELlJRWGztks5sYNHHgaJpZM4O25bo
.
If there are still issues here, they are better discussed on a forum, not on the issue tracker. See https://golang.org/wiki/Questions .
| gharchive/issue | 2017-08-14T21:03:54 | 2025-04-01T06:44:20.829679 | {
"authors": [
"AlexRouSg",
"cherrymui",
"davecheney",
"firelizzard18",
"ianlancetaylor"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/21444",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
273226438 | net/http2: data race in Server
What version of Go are you using (go version)?
go version devel +36ef06cc10 Fri Nov 10 18:17:38 2017 +0000 linux/amd64
Does this issue reproduce with the latest release?
No, this only happens on Go devel +36ef06cc10.
What operating system and processor architecture are you using (go env)?
Travis env :
Operating System Details
Distributor ID: Ubuntu
Description: Ubuntu 14.04.5 LTS
Release: 14.04
Codename: trusty
Cookbooks Version
4642454 https://github.com/travis-ci/travis-cookbooks/tree/4642454
git version
git version 2.14.1
bash version
GNU bash, version 4.3.11(1)-release (x86_64-pc-linux-gnu)
gcc version
gcc (Ubuntu 4.8.4-2ubuntu1~14.04.3) 4.8.4
Copyright (C) 2013 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
docker version
Client:
Version: 17.03.1-ce
API version: 1.27
Go version: go1.7.5
Git commit: c6d412e
Built: Mon Mar 27 16:58:30 2017
OS/Arch: linux/amd64
clang version
clang version 3.9.0 (tags/RELEASE_390/final)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /usr/local/clang-3.9.0/bin
jq version
jq-1.5
bats version
Bats 0.4.0
shellcheck version
0.4.6
shfmt version
v1.3.1
ccache version
ccache version 3.1.9
What did you do?
pkg src on github
test log on travis
The pkg does things like send H2 Pushes based on Link headers. Since there isn't a Go client that can receive H2 Pushes, I pull in the http2 package to be able to write tests that have access to http2 internals. Tests are run daily against Go master to ensure everything keeps working.
example H2 Push test
Since Friday I noticed there where intermittent data races. As far as I can tell these are not caused by the tests I added.
I first pulled in the latest http2 code and retried but data races still happen from time to time.
What did you expect to see?
No data races.
What did you see instead?
Data race:
Seems related to #20704
WARNING: DATA RACE
Read at 0x00c4213a2000 by goroutine 212:
runtime.slicecopy()
/home/travis/.gimme/versions/go/src/runtime/slice.go:192 +0x0
github.com/romainmenke/pusher/http2.(*Framer).WriteDataPadded()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/frame.go:683 +0x348
github.com/romainmenke/pusher/http2.(*Framer).WriteData()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/frame.go:643 +0x85
github.com/romainmenke/pusher/http2.(*writeData).writeFrame()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/write.go:109 +0xb1
github.com/romainmenke/pusher/http2.(*serverConn).writeFrameAsync()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:724 +0x58
Previous write at 0x00c4213a2000 by goroutine 202:
runtime.slicecopy()
/home/travis/.gimme/versions/go/src/runtime/slice.go:192 +0x0
bufio.(*Writer).Write()
/home/travis/.gimme/versions/go/src/bufio/bufio.go:617 +0x3f4
github.com/romainmenke/pusher/http2.(*responseWriter).write()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:2550 +0x1d9
github.com/romainmenke/pusher/http2.(*responseWriter).Write()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:2524 +0x70
github.com/romainmenke/pusher/http2.TestIssue20704Race.func1()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server_test.go:3701 +0x90
net/http.HandlerFunc.ServeHTTP()
/home/travis/.gimme/versions/go/src/net/http/server.go:1921 +0x51
net/http.serverHandler.ServeHTTP()
/home/travis/.gimme/versions/go/src/net/http/server.go:2646 +0xb9
net/http.initNPNRequest.ServeHTTP()
/home/travis/.gimme/versions/go/src/net/http/server.go:3209 +0x100
net/http.(*initNPNRequest).ServeHTTP()
<autogenerated>:1 +0x82
net/http.(Handler).ServeHTTP-fm()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:1774 +0x64
github.com/romainmenke/pusher/http2.(*serverConn).runHandler()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:2059 +0x96
Goroutine 212 (running) created at:
github.com/romainmenke/pusher/http2.(*serverConn).startFrameWrite()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:1099 +0x2fd
github.com/romainmenke/pusher/http2.(*serverConn).scheduleFrameWrite()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:1200 +0x353
github.com/romainmenke/pusher/http2.(*serverConn).writeFrame()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:1054 +0xfc
github.com/romainmenke/pusher/http2.(*serverConn).serve()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:815 +0xaf3
github.com/romainmenke/pusher/http2.(*Server).ServeConn()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:431 +0xcee
github.com/romainmenke/pusher/http2.ConfigureServer.func1()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:269 +0xda
net/http.(*conn).serve()
/home/travis/.gimme/versions/go/src/net/http/server.go:1725 +0x1845
Goroutine 202 (running) created at:
github.com/romainmenke/pusher/http2.(*serverConn).processHeaders()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:1793 +0x7e7
github.com/romainmenke/pusher/http2.(*serverConn).processFrame()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:1340 +0x419
github.com/romainmenke/pusher/http2.(*serverConn).processFrameFromReader()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:1298 +0x7b3
github.com/romainmenke/pusher/http2.(*serverConn).serve()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:819 +0x1018
github.com/romainmenke/pusher/http2.(*Server).ServeConn()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:431 +0xcee
github.com/romainmenke/pusher/http2.ConfigureServer.func1()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:269 +0xda
net/http.(*conn).serve()
/home/travis/.gimme/versions/go/src/net/http/server.go:1725 +0x1845
==================
--- FAIL: TestIssue20704Race (6.48s)
testing.go:722: race detected during execution of test
2017/11/10 19:22:08 protocol error: received DATA on a HEAD request
2017/11/10 19:23:33 protocol error: received DATA before a HEADERS frame
FAIL
FAIL github.com/romainmenke/pusher/http2 104.759s
/cc @bradfitz @tombergan
@romainmenke you mention in your bug report that it only happens with 36ef06cc10 (which is very recent) but in the details on Travis that it runs on Go1.7.5(which I don't think we support anymore). Could you please confirm if you can actually reproduce this on tip? Also what's the output of
cd $GOPATH/src/golang.org/x/net && git rev-parse HEAD
@odeke-em The Go version listed in the Travis details is most likely the Go version used to compile the latest Go from source. In the Travis log on line 438 you can see that it uses 36ef06c for the pkg build and tests.
cd $GOPATH/src/golang.org/x/net && git rev-parse HEAD
a337091b0525af65de94df2eb7e98bd9962dcbe2
@romainmenke Can you verify that you are using the latest version of github.com/romainmenke/pusher? Or, better, can you use the latest version of golang.org/x/net/http2 instead of github.com/romainmenke/pusher/http? The stack trace has:
github.com/romainmenke/pusher/http2.(*responseWriter).Write()
/home/travis/gopath/src/github.com/romainmenke/pusher/http2/server.go:2524 +0x70
However, https://github.com/romainmenke/pusher/blob/master/http2/server.go#L2524 is outside of responseWriter.Write.
Is this still a problem in 1.10.1?
I checked the test logs on Travis and this hasn't happened on master in a while. Seems to be fixed to me.
| gharchive/issue | 2017-11-12T12:47:17 | 2025-04-01T06:44:20.840079 | {
"authors": [
"ianlancetaylor",
"odeke-em",
"romainmenke",
"tombergan"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/22678",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
340374678 | x/review/git-codereview: use the gofmt from $GOROOT/bin/gofmt
gofmt changed in Go 1.11, and git-codereview uses the gofmt from $PATH, so when that's the Go 1.10 version, it complains as git-codereview: gofmt needs to format these files (run 'git gofmt').
It should instead use the gofmt in GOROOT.
It should instead use the gofmt in GOROOT.
This is not obvious to me. Why is GOROOT better than PATH in this context?
We go to pains to tell people not to set the GOROOT env var, as it is a source of significant confusion. (This is one of @davecheney's favorites.) And tools that memorize the toolchain that they were built with are really frustrating.
It seems to me that using the gofmt in your PATH is the best available solution. Or at most fixing https://github.com/golang/go/issues/27166 and then using the results of that.
| gharchive/issue | 2018-07-11T19:27:19 | 2025-04-01T06:44:20.843539 | {
"authors": [
"FiloSottile",
"josharian"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/26336",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
342584654 | cmd/go: TestScript use of localhost GOPROXY failing on Windows, now skipped
cmd/go's tests start an HTTP server by doing net.Listen("tcp", "localhost:0") and then try to make URLs using l.Addr().String(), which ends up being localhost:<port>.
This worked for some tests but somehow is failing when run as part of TestScript, as of CL 124697.
I can't see what is different between the old hand-written-in-Go tests and the new script-based tests.
One possibility was the use of t.Parallel, but I commented that out in TestScript and still no luck.
The trybots failed on that CL and the gomote failed too when run by hand. I submitted it anyway to see if the builders were any different, since it is so mysterious. I expect them to fail too. If they do I will skip those scripts on Windows.
--- FAIL: TestScript/mod_require_exclude (0.04s)
script_test.go:134:
# build with no newer version to satisfy exclude (0.036s)
> env GO111MODULE=on
> ! go list -m all
[stderr]
go: errors parsing go.mod:
$WORK\gopath\src\go.mod:2: invalid module version "latest": Get http://127.0.0.1:49713/mod/rsc.io/sampler/@v/list: dial tcp 127.0.0.1:49713: socket: The requested service provider could not be loaded or initialized.
$WORK\gopath\src\go.mod:3: invalid module version "latest": Get http://127.0.0.1:49713/mod/rsc.io/sampler/@v/list: dial tcp 127.0.0.1:49713: socket: The requested service provider could not be loaded or initialized.
[exit status 1]
> stderr 'no newer version available'
FAIL: testdata\script\mod_require_exclude.txt:4: no match for `no newer version available` found in stderr
We need to turn the script tests back on for Windows before the release.
I also tried using [::1]:0 instead of localhost:0 and it did not help.
It looks like this might happen when some environment variables are missing: https://github.com/gevent/gevent/issues/284 I don’t have Windows so I can’t check further.
It looks like this might happen when some environment variables are missing: gevent/gevent#284 I don’t have Windows so I can’t check further.
I agree. Russ's error looks similar to #25210. I did not investigate this bug (I do not have Windows PC near by), but Brad's problem was that he created Windows process with no environment variables. Generally you cannot do that. You should copy your current process variables and add / delete / modify whatever variables you want.
Alex
That makes sense, thanks - that comment saves me a ton of time. I am trying to make sure extraneous env vars don't pollute the tests but I'll copy whatever Windows needs.
| gharchive/issue | 2018-07-19T05:34:10 | 2025-04-01T06:44:20.848434 | {
"authors": [
"alexbrainman",
"chlunde",
"rsc"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/26457",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
370761726 | fmt: Provide more examples in godocs
Most fmt functions do not have examples in godocs (currently, only Stringer)
As I've not contributed to go before, I thought providing examples would be a good starter.
A few people recently added several examples; see the tip version of the website:
https://tip.golang.org/pkg/fmt/
These new example may also need some clean-up, as Rob said in #27554.
Duplicate of #27554, in effect.
| gharchive/issue | 2018-10-16T19:22:41 | 2025-04-01T06:44:20.850432 | {
"authors": [
"ALTree",
"robpike",
"rwhittington"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/28238",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
408463017 | encoding/json: panic on embedded struct pointer that implements Unmarshaler
What version of Go are you using (go version)?
$ go version
go version go1.11.5 linux/amd64
Does this issue reproduce with the latest release?
Yes.
What operating system and processor architecture are you using (go env)?
go env Output
$ go env
GOARCH="amd64"
GOBIN=""
GOCACHE="<snip>/.cache/go-build"
GOEXE=""
GOFLAGS=""
GOHOSTARCH="amd64"
GOHOSTOS="linux"
GOOS="linux"
GOPATH="<snip>/go"
GOPROXY=""
GORACE=""
GOROOT="/usr/local/go"
GOTMPDIR=""
GOTOOLDIR="/usr/local/go/pkg/tool/linux_amd64"
GCCGO="gccgo"
CC="gcc"
CXX="g++"
CGO_ENABLED="1"
GOMOD=""
CGO_CFLAGS="-g -O2"
CGO_CPPFLAGS=""
CGO_CXXFLAGS="-g -O2"
CGO_FFLAGS="-g -O2"
CGO_LDFLAGS="-g -O2"
PKG_CONFIG="pkg-config"
GOGCCFLAGS="-fPIC -m64 -pthread -fmessage-length=0 -fdebug-prefix-map=/tmp/go-build174238184=/tmp/go-build -gno-record-gcc-switches"
What did you do?
https://play.golang.org/p/c6Q_ehlEHMQ
What did you expect to see?
Unmarshaling works.
What did you see instead?
panic: json: Unmarshal(nil *main.A)
Analysis
This is caused by encoding/json not initializing (new) the embedded struct pointer field. Type asserting the embedding struct to Unmarshaler works, with the interface value holding a pointer to the embedding struct, but on calling the UnmarshalJSON method it gets called with the embedded struct as the receiver, which is nil.
Also see https://github.com/go-yaml/yaml/issues/436
I don't think there's anything that we can do without changing the reflect package (and since the reflect package is a dynamic representation of the Go language, this may mean changing the Go language itself).
The fundamental problem is that embedded types have their methods forwarded to the parent. Thus, type B has an UnmarshalJSON method. From the perspective of the json package, it has no idea that it needs to also allocate a value of type A since it sees the UnmarshalJSON method on B first and calls it. You could argue that json should investigate whether the UnmarshalJSON was forwarded from embedding, and if so, allocate the necessary receivers. However, reflect doesn't seem to provide an API to distinguish between methods directly declared on the parent, versus those from embedded children.
Also, the fact that you can't directly call UnmarshalJSON on an allocated B suggest that the user code itself is problematic.
var b B
err := b.UnmarshalJSON([]byte(`{"hello": "world"}`))
fmt.Println(err) // json: Unmarshal(nil *main.A)
I guess fixing this would require extending reflect with the ability to give you information about the chain of embedded structs from where a method implementation comes from. But maybe it also takes additional special handling so that an embedded UnmarshalJSON doesn't interfere with serializing the struct itself.
Any sort of input on the zero value of B will trigger this since encoding/json will try to call it's embedded UnmarshalJSON method.
There is no way in Go in embed a type and not have it's method set be forwarded. Embedding forwards both fields and methods and is a common gotcha in Go and a source of bugs.
That said, you can destructively cancel out the forwarded method if another embedded type has the same method. In such a case, the parent type will not have forwarded methods that have naming conflicts.
Consider this modified example of yours: https://play.golang.org/p/PTTh6Jqf7El
However, I should note that this approach is a hack. In general I recommend against the use of embedding. It tends to cause more problems than it solves.
@segevfiner is this still a bug that you want addressed, or do you feel that your question has been answered? If you think this is more of a question, and less of a bug at this point, you can continue the conversation here: https://golang.org/wiki/Questions
Definitely a bug. If an embedded value works, why shouldn't an embedded pointer? Also considering that this is not documented anywhere, and calls the user UnmarshalJSON with a nil pointer receiver which is completely unexpected here and will likely cause a random panic suprising the user/developer.
@segevfiner thanks for following up.
/cc @rsc
Definitely a bug.
I disagree.
If an embedded value works, why shouldn't an embedded pointer?
Embedded pointers already have asymmetric behavior compared to the non-pointer versions.
Examples:
b.A = nil // possible for embedded pointer, obviously not possible for embedded value
b.A = 5 // may panic for embedded pointer, never panics for embedded value
b.a = new(int) // impossible to initialize an embedded unexported pointer (see #21357)
I don't believe asymmetry here is an argument that this is a bug.
calls the user UnmarshalJSON with a nil pointer receiver which is completely unexpected here
The json did call UnmarshalJSON with an allocated pointer to (*B).UnmarshalJSON. It is the Go language semantics that then forwarded the call to the underlying (*A).UnmarshalJSON with a nil pointer. Again, this is not the json package's fault but rather surprising behavior around embedding in Go.
cause a random panic surprising the user/developer.
The json package is not panicking. Your example code is calling panic on L30.
I don't think there is anything actionable here for json other than a proposal to change reflect or the Go language.
Whether you call this a bug or not. It's still a behavior that surprised me and I didn't expect.
Of course they have different/asymmetric behavior, but Go does try to hide some of the differences where it can between pointers and values. Not all cases of course since they arw different after all.
Do consider that this can easily happen in some deeply nested struct inside some other struct and maybe even on some slice of such structs which allocates items dynamically.
Yeah json isn't panicking. But the user UnmarshalJSON called with an unexpected nil receiver can easily do.
Except documenting this, if there is any convention of documenting such gotchas... It's quite possible that fixing this would require extending/modifying reflect to be able to fix this.
Embedded are in general difficult to use properly and naturally lead to surprisingly spooky action at a distance. This is a general gotcha that isn't specific to json. As such, I don't believe we need to complicate the documentation (which is already complicated) with gotchas from the language itself.
I'm going to close this issue, then, and we can re-open if this needs to be evaluated again later on.
| gharchive/issue | 2019-02-09T19:33:18 | 2025-04-01T06:44:20.866518 | {
"authors": [
"dsnet",
"katiehockman",
"segevfiner"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/30148",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
489034160 | net/http: error message in case of bad certificate leaks implementation details in Go 1.13
What version of Go are you using (go version)?
$ go version
go version go1.13 darwin/amd64
Does this issue reproduce with the latest release?
Yes
What operating system and processor architecture are you using (go env)?
go env Output
$ go env
GO111MODULE=""
GOARCH="amd64"
GOBIN=""
GOCACHE="/Users/cuonglm/Library/Caches/go-build"
GOENV="/Users/cuonglm/Library/Application Support/go/env"
GOEXE=""
GOFLAGS=""
GOHOSTARCH="amd64"
GOHOSTOS="darwin"
GONOPROXY=""
GONOSUMDB=""
GOOS="darwin"
GOPATH="/Users/cuonglm/go"
GOPRIVATE=""
GOPROXY="https://proxy.golang.org,direct"
GOROOT="/Users/cuonglm/sdk/go1.13"
GOSUMDB="sum.golang.org"
GOTMPDIR=""
GOTOOLDIR="/Users/cuonglm/sdk/go1.13/pkg/tool/darwin_amd64"
GCCGO="gccgo"
AR="ar"
CC="clang"
CXX="clang++"
CGO_ENABLED="1"
GOMOD=""
CGO_CFLAGS="-g -O2"
CGO_CPPFLAGS=""
CGO_CXXFLAGS="-g -O2"
CGO_FFLAGS="-g -O2"
CGO_LDFLAGS="-g -O2"
PKG_CONFIG="pkg-config"
GOGCCFLAGS="-fPIC -m64 -pthread -fno-caret-diagnostics -Qunused-arguments -fmessage-length=0 -fdebug-prefix-map=/var/folders/y4/hs76ltbn7sb66lw_6934kq4m0000gn/T/go-build839999394=/tmp/go-build -gno-record-gcc-switches -fno-common"
What did you do?
$ export GO111MODULE=off
$ go1.13 get -d github.com/loadimpact/k6
$ cd $GOPATH/src/github.com/loadimpact/k6/js
$ go1.13 test -count=1 -race -run=TestVUIntegrationClientCerts/Unauthenticated
--- FAIL: TestVUIntegrationClientCerts (5.16s)
--- FAIL: TestVUIntegrationClientCerts/Unauthenticated (0.56s)
--- FAIL: TestVUIntegrationClientCerts/Unauthenticated/Source (0.11s)
Error Trace: runner_test.go:1168
Error: Error message not equal:
expected: "GoError: Get https://127.0.0.1:51407: remote error: tls: bad certificate"
actual : "GoError: Get https://127.0.0.1:51407: readLoopPeekFailLocked: remote error: tls: bad certificate"
Test: TestVUIntegrationClientCerts/Unauthenticated/Source
--- FAIL: TestVUIntegrationClientCerts/Unauthenticated/Archive (0.11s)
Error Trace: runner_test.go:1168
Error: Error message not equal:
expected: "GoError: Get https://127.0.0.1:51407: remote error: tls: bad certificate"
actual : "GoError: Get https://127.0.0.1:51407: readLoopPeekFailLocked: remote error: tls: bad certificate"
Test: TestVUIntegrationClientCerts/Unauthenticated/Archive
FAIL
exit status 1
FAIL github.com/loadimpact/k6/js 5.344s
What did you expect to see?
Test passed.
What did you see instead?
Test failed.
Note
The test will pass with go1.11 to go1.12.9 or go1.13 without -race.
The problem also occurs in linux/amd64.
If you write tests that compare error string values for exact matches, don't you already run into failures every other Go release?
@mvdan No, it works consistently at least from go1.11.x. Also, go1.13 does not have the issue without -race, the problem happens when run test under -race.
I chased down the code and point to https://github.com/golang/go/blob/release-branch.go1.13/src/net/http/transport.go#L2060
@mvdan those lines are there since 7fa984674971e801c549e4e2f1715f6f39c962e5, so it's high chance that something changes in go1.13 that break the test.
The line in question hasn't changed in four years, but this may be related to CL 179457 (#32310), which changed some nearby code.
I tend to agree with @mvdan that tests should not rely on the error strings.
On the other hand, the fact that the string now contains readLoopPeekFailLocked (an implementation detail of the net/http package) seems like a broken abstraction boundary, at the very least.
CC @bradfitz @dmitshur @ianlancetaylor
I think what changed is that the TLS error used to be returned by conn.Handshake() but is now returned by a conn.Read() because of the structure of TLS 1.3. See the last paragraph of https://golang.org/doc/go1.12#tls_1_3. Not sure why -race makes a difference though, maybe scheduler timing?
This is not a regression because we don't guarantee error messages across versions, but we should figure out the -race difference and make the error prettier.
@FiloSottile From I observed, the error is still the same, but the condition peekErr == io.EOF false causing readLoopPeekFailLocked: was appended to current error https://github.com/golang/go/blob/release-branch.go1.13/src/net/http/transport.go#L2060
I agree that an error message reported back to the user for a failure like this shouldn't say readLoopPeekFailLocked. @bradfitz this was added in https://golang.org/cl/18282 back in 2016. Can we just remove it?
I also agree that this test should be fixed to not expect a specific error message. The Go 1 compatibility guarantee does not extend to the exact text of error messages. Especially not from the net/http package.
In that CL, the new errServerClosedIdle variable was documented with this comment:
// error values for debugging and testing, not seen by users.
If errServerClosedIdle wasn't meant to be seen by users, then it's likely fmt.Errorf("readLoopPeekFailLocked: %v", peekErr) wasn't intended to be seen by users either.
I agree that an error message reported back to the user for a failure like this shouldn't say readLoopPeekFailLocked. @bradfitz this was added in https://golang.org/cl/18282 back in 2016. Can we just remove it?
I also agree that this test should be fixed to not expect a specific error message. The Go 1 compatibility guarantee does not extend to the exact text of error messages. Especially not from the net/http package.
I fixed the test already.
I just want to insist that something internal was changed, because the test is still passed without -race.
I'm trying digging more into this problem. As I can see, with -race, this condition:
https://github.com/golang/go/blob/master/src/net/http/transport.go#L1920
becomes true. The only place that increase numExpectedResponse is:
https://github.com/golang/go/blob/master/src/net/http/transport.go#L2363
Without race, roundTrip always happens before readLoop. What's the right logic for numExpectedResponse? Must it be increased before readLoop run?
| gharchive/issue | 2019-09-04T09:04:57 | 2025-04-01T06:44:20.883015 | {
"authors": [
"FiloSottile",
"bcmills",
"cuonglm",
"dmitshur",
"ianlancetaylor",
"mvdan"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/34066",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
503106834 | text/template: generated slice not initialized to zero
What version of Go are you using (go version)?
$ go version
go version go1.13.1 windows/amd64
and same on:
go version go1.13.1 linux/amd64
Does this issue reproduce with the latest release?
Yes.
What operating system and processor architecture are you using (go env)?
windows/amd64
and
linux/amd64
GOARCH=amd64
go env Output
$ go env
GO111MODULE=""
GOARCH="amd64"
GOBIN="/home/user/go/bin"
GOCACHE="/home/user/.cache/go-build"
GOENV="/home/user/.config/go/env"
GOEXE=""
GOFLAGS=""
GOHOSTARCH="amd64"
GOHOSTOS="linux"
GONOPROXY=""
GONOSUMDB=""
GOOS="linux"
GOPATH="/home/user/go"
GOPRIVATE=""
GOPROXY="https://proxy.golang.org,direct"
GOROOT="/usr/local/go"
GOSUMDB="sum.golang.org"
GOTMPDIR="/dev/shm"
GOTOOLDIR="/usr/local/go/pkg/tool/linux_amd64"
GCCGO="/usr/bin/gccgo"
CC="gcc"
CXX="g++"
CGO_ENABLED="1"
GOMOD=""
CGO_CFLAGS="-g -O2"
CGO_CPPFLAGS=""
CGO_CXXFLAGS="-g -O2"
CGO_FFLAGS="-g -O2"
CGO_LDFLAGS="-g -O2"
PKG_CONFIG="pkg-config"
GOGCCFLAGS="-fPIC -m64 -pthread -fmessage-length=0 -fdebug-prefix-map=/dev/shm/go-build949846437=/tmp/go-build -gno-record-gcc-switches"
What did you do?
Try this
Then try this
What did you expect to see?
Initialize the new underlying array of the template generated slice to zero.
What did you see instead?
Not zero.
Code one:
package main
import (
"bytes"
"fmt"
"io"
"text/template"
)
func main() {
buf := &bytes.Buffer{}
my := &myWriter{"You", buf}
template.Must(template.New("my").Parse("Hi{{.Name}}Bye.")).Execute(my, my)
fmt.Printf("<<%q>>\n", buf.String())
}
func (m *myWriter) Write(p []byte) (n int, err error) {
fmt.Printf("len=%v cap=%v\t%v %v\n", len(p), cap(p), string(p), p[:cap(p)])
no++
fmt.Println("gen:", no, gen())
m.Writer.Write(p)
// m.Writer.Write(p[:8])
return 8, nil
}
type myWriter struct {
Name string
io.Writer
}
const genLen = 8
func gen() string {
b := [genLen]byte{}
for i := range b {
b[i] = no
}
return string(b[:])
}
var no = byte(49) //'1'
Output:
len=2 cap=8 Hi [72 105 0 0 0 0 0 0]
gen: 50 22222222
len=3 cap=64 You [89 111 117 58 32 53 48 32 50 50 50 50 50 50 50 50 10 50 32 49 48 53 32 48 32 48 32 48 32 48 32 48 32 48 93 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
gen: 51 33333333
len=4 cap=8 Bye. [66 121 101 46 0 0 0 0]
gen: 52 44444444
<<"HiYouBye.">>
Code two:
package main
import (
"bytes"
"fmt"
"io"
"text/template"
)
func main() {
buf := &bytes.Buffer{}
my := &myWriter{"You", buf}
template.Must(template.New("my").Parse("Hi{{.Name}}Bye.")).Execute(my, my)
fmt.Printf("<<%q>>\n", buf.String())
}
func (m *myWriter) Write(p []byte) (n int, err error) {
fmt.Printf("len=%v cap=%v\t%v %v\n", len(p), cap(p), string(p), p[:cap(p)])
no++
fmt.Println("gen:", no, gen())
m.Writer.Write(p)
// m.Writer.Write(p[:8])
return 8, nil
}
type myWriter struct {
Name string
io.Writer
}
const genLen = 64
func gen() string {
b := [genLen]byte{}
for i := range b {
b[i] = no
}
return string(b[:])
}
var no = byte(49) //'1'
Output:
len=2 cap=8 Hi [72 105 0 0 0 0 0 0]
gen: 50 2222222222222222222222222222222222222222222222222222222222222222
len=3 cap=128 You [89 111 117 58 32 53 48 32 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
gen: 51 3333333333333333333333333333333333333333333333333333333333333333
len=4 cap=8 Bye. [66 121 101 46 0 0 0 0]
gen: 52 4444444444444444444444444444444444444444444444444444444444444444
<<"HiYouBye.">>
This should be all zeros except the first three:
[89 111 117 58 32 53 48 32 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
The t.Execute(my, my) calls func (m *myWriter) Write(p []byte) so the p with len=3 and cap=128 generated by the tamplate engine.
After debugging the 2nd code inside /usr/local/go/src/fmt/print.go file at line 230, it seems it is fmt.buffer with length=3, and cap=128, here:
func Fprint(w io.Writer, a ...interface{}) (n int, err error) {
p := newPrinter()
p.doPrint(a)
n, err = w.Write(p.buf)
p.free()
return
}
The call to p := newPrinter() initialized here p.fmt.init(&p.buf):
// newPrinter allocates a new pp struct or grabs a cached one.
func newPrinter() *pp {
p := ppFree.Get().(*pp)
p.panicking = false
p.erroring = false
p.wrapErrs = false
p.fmt.init(&p.buf)
return p
}
Gets and returns the free memory without initializing it to zero.
Solution:
In /usr/local/go/src/fmt/format.go file at line 58 set the buffer to all zero:
func (f *fmt) init(buf *buffer) {
b := (*buf)[:cap(*buf)]
for i := range b {
b[i] = 0
}
f.buf = buf
f.clearflags()
}
Then Output of the 2nd code is all zero:
len=2 cap=8 Hi [72 105 0 0 0 0 0 0]
gen: 50 2222222222222222222222222222222222222222222222222222222222222222
len=3 cap=128 You [89 111 117 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
gen: 51 3333333333333333333333333333333333333333333333333333333333333333
len=4 cap=8 Bye. [66 121 101 46 0 0 0 0]
gen: 52 4444444444444444444444444444444444444444444444444444444444444444
<<"HiYouBye.">>
/cc @robpike
/cc @robpike
This doesn't look like an actual issue, the slices passed have the correct length, and contents in the extra capacity aren't really relevant.
| gharchive/issue | 2019-10-06T14:19:33 | 2025-04-01T06:44:20.898069 | {
"authors": [
"katiehockman",
"seankhliao",
"wasmup"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/34720",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
509959679 | cmd/go: files in module cache may be writable on Windows
In the course of testing #31481, I found the following failure on the windows builders at patch set 3 (https://storage.googleapis.com/go-build-log/096e97bd/windows-386-2008_7b478c28.log):
--- FAIL: TestScript (0.00s)
--- FAIL: TestScript/mod_cache_rw (0.29s)
script_test.go:191:
# Regression test for golang.org/issue/31481. (0.000s)
# golang.org/issue/31481: an explicit flag should make directories in the module
# cache writable in order to work around the historical inability of 'rm -rf' to
# forcibly remove files in unwritable directories. (0.210s)
# After adding an extraneous file, 'go mod verify' should fail. (0.046s)
# However, files within those directories should still be read-only to avoid
# accidental mutations. (0.001s)
> [!root] ! cp $WORK/extraneous.txt $GOPATH/pkg/mod/rsc.io/quote@v1.5.2/go.mod
FAIL: testdata\script\mod_cache_rw.txt:16: unexpected command success
FAIL
FAIL cmd/go 75.878s
This suggests that either the root condition in TestScript is failing to detect a user acting in some sort of root-like role on Windows, or else we are failing to mark the contents of the module cache read-only on Windows.
CC @jayconrod @alexbrainman @zx2c4
I'll have something on gerrit in 5 minutes.
Suggested issue title change: "syscall: Windows does not respect permission bits passed to Open"
| gharchive/issue | 2019-10-21T13:23:11 | 2025-04-01T06:44:20.901726 | {
"authors": [
"bcmills",
"zx2c4"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/35033",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
614525523 | Panic occurred when handle Segmentation fault signal
What version of Go are you using (go version)?
$ go version
go version go1.11.13 linux/amd64
Does this issue reproduce with the latest release?
Don't konw, we did not tried.
What operating system and processor architecture are you using (go env)?
go env Output
$ go env
GOARCH="amd64"
GOBIN=""
GOCACHE="/root/.cache/go-build"
GOEXE=""
GOFLAGS=""
GOHOSTARCH="amd64"
GOHOSTOS="linux"
GOOS="linux"
GOPATH="/data/jenkins/workspace/gohere"
GOPROXY=""
GORACE=""
GOROOT="/root/go"
GOTMPDIR=""
GOTOOLDIR="/root/go/pkg/tool/linux_amd64"
GCCGO="gccgo"
CC="gcc"
CXX="g++"
CGO_ENABLED="1"
GOMOD=""
CGO_CFLAGS="-g -O2"
CGO_CPPFLAGS=""
CGO_CXXFLAGS="-g -O2"
CGO_FFLAGS="-g -O2"
CGO_LDFLAGS="-g -O2"
PKG_CONFIG="pkg-config"
GOGCCFLAGS="-fPIC -m64 -pthread -fmessage-length=0 -fdebug-prefix-map=/tmp/go-build208195420=/tmp/go-build -gno-record-gcc-switches"
What did you do?
This is our online service, and we use cgo to invoke our C++ code, recently we find our service crashed something. So,
We setted GOTRACEBACK=crash
# ulimit -c unlimited
Started the service again to expect another crash
What did you expect to see?
We want a core file and use gdb to see what happend.
What did you see instead?
We seen this with gdb and the core file:
Core was generated by `xxxxx'.
Program terminated with signal 11, Segmentation fault.
#0 0x0000000000603e3e in runtime.(*sigctxt).preparePanic (c=0xc001389258, gp=<optimized out>, sig=<optimized out>) at /root/go/src/runtime/signal_linux_amd64.go:20
20 return (*sigcontext)(unsafe.Pointer(&(*ucontext)(c.ctxt).uc_mcontext))
Loading Go Runtime support.
I want to know is this normal?
Yes, this is the result of using unsafe in your code. Unless you are sure there are no bugs, your code can crash.
If you think there's a bug in Go rather than in your code, please post a short reproducer containing the code that triggers this. But given the fact that there is cgo and unsafe involved, it is unlikely that there is any bug here.
If you use Go 1.14 with -race, you can get the benefits of the -d=checkptr flag which adds instrumentation to ensure that unsafe rules are followed properly. https://tip.golang.org/doc/go1.14#compiler.
Thanks for your reply, but I want to know how to get the real reason of the crash?
It's a bug in your code, due to which a segmentation fault occurs. That's the real reason. I'm afraid I cannot explain any better than this.
OK, I expect an crash detail of my code in core dump, but currently I only got a signal handle panic.
| gharchive/issue | 2020-05-08T06:02:49 | 2025-04-01T06:44:20.907924 | {
"authors": [
"agnivade",
"coldnight"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/38944",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
749844710 | Go net/http crash in Transport.queueForIdleConn(...)
What version of Go are you using (go version)?
go1.15.3
Does this issue reproduce with the latest release?
Not sure. Only happened once in our PROD env.
What operating system and processor architecture are you using (go env)?
PROD runtime env, there is no go.
$ uname -a
Linux ashads097.fwmrm.net 3.10.0-514.el7.x86_64 #1 SMP Tue Nov 22 16:42:41 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux
What did you do?
Sending an HTTPS request, and Got crashed.
fatal error: fault
[signal SIGSEGV: segmentation violation code=0x80 addr=0x0 pc=0x74f58d]
goroutine 40123707 [running]:
runtime.throw(0xe8844d, 0x5)
/usr/local/go/src/runtime/panic.go:1116 +0x72 fp=0xc00deb3330 sp=0xc00deb3300 pc=0x43a432
runtime.sigpanic()
/usr/local/go/src/runtime/signal_unix.go:727 +0x405 fp=0xc00deb3360 sp=0xc00deb3330 pc=0x450be5
net/http.(*persistConn).isBroken(0x7c736d3838313537, 0xc00616f650)
/usr/local/go/src/net/http/transport.go:1897 +0x2d fp=0xc00deb3390 sp=0xc00deb3360 pc=0x74f58d
net/http.(*Transport).queueForIdleConn(0xc00028c000, 0xc0131f9080, 0xc00cb96a00)
/usr/local/go/src/net/http/transport.go:1021 +0x1e5 fp=0xc00deb3478 sp=0xc00deb3390 pc=0x749845
net/http.(*Transport).getConn(0xc00028c000, 0xc00f51dbc0, 0x0, 0xc006086500, 0x5, 0xc00e9184c0, 0x16, 0x0, 0x0, 0x0, ...)
/usr/local/go/src/net/http/transport.go:1326 +0x297 fp=0xc00deb3718 sp=0xc00deb3478 pc=0x74b257
net/http.(*Transport).roundTrip(0xc00028c000, 0xc004784200, 0x30, 0x7f7f604ade90, 0x150)
/usr/local/go/src/net/http/transport.go:569 +0x77c fp=0xc00deb3960 sp=0xc00deb3718 pc=0x746c7c
net/http.(*Transport).RoundTrip(0xc00028c000, 0xc004784200, 0xc00028c000, 0x0, 0x0)
/usr/local/go/src/net/http/roundtrip.go:17 +0x35 fp=0xc00deb3998 sp=0xc00deb3960 pc=0x72ccb5
net/http.send(0xc004784200, 0xfb5860, 0xc00028c000, 0x0, 0x0, 0x0, 0xc012b706a0, 0x203001, 0x1, 0x0)
/usr/local/go/src/net/http/client.go:252 +0x453 fp=0xc00deb3b40 sp=0xc00deb3998 pc=0x6e8b93
net/http.(*Client).send(0xc000a36750, 0xc004784200, 0x0, 0x0, 0x0, 0xc012b706a0, 0x0, 0x1, 0x203003)
/usr/local/go/src/net/http/client.go:176 +0xff fp=0xc00deb3bc0 sp=0xc00deb3b40 pc=0x6e857f
net/http.(*Client).do(0xc000a36750, 0xc004784200, 0x0, 0x0, 0x0)
/usr/local/go/src/net/http/client.go:718 +0x45f fp=0xc00deb3d98 sp=0xc00deb3bc0 pc=0x6ea59f
net/http.(*Client).Do(...)
/usr/local/go/src/net/http/client.go:586
(dlv) frame 6
> runtime.raise() /usr/local/go/src/runtime/sys_linux_amd64.s:165 (PC: 0x4749e1)
Warning: debugging optimized function
Frame 6: /usr/local/go/src/net/http/transport.go:1021 (PC: 749845)
(dlv) locals
oldTime = time.Time {wall: 0, ext: 0, loc: *time.Location nil}
list = []*net/http.persistConn len: 7, cap: 64, [...]
stop = false
delivered = false
pconn = ("*net/http.persistConn")(0x7c736d3838313537)
tooOld = false
(dlv) p list
[]*net/http.persistConn len: 7, cap: 64, [
*{
alt: net/http.RoundTripper nil,
t: *(unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
cacheKey: (*"net/http.connectMethodKey")(0x71655265746f726f),
conn: net.Conn nil,
tlsState: *(unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
br: *(unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
bw: *(unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
nwrite: (unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
reqch: chan net/http.requestAndChan {},
writech: chan net/http.writeRequest {},
closech: chan struct {} {},
isProxy: (unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
sawEOF: (unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
readLimit: (unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
writeErrCh: chan error {},
writeLoopDone: chan struct {} {},
idleAt: (*time.Time)(0x71655265746f730f),
idleTimer: *(unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
mu: (*sync.Mutex)(0x71655265746f732f),
numExpectedResponses: (unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
closed: error nil,
canceledErr: error nil,
broken: (unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
reused: (unreadable error while reading spliced memory at 0x71655265746f7257: EOF),
mutateHeaderFunc: (unreadable error while reading spliced memory at 0x71655265746f7257: EOF),},
*{
alt: net/http.RoundTripper nil,
t: *(unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
cacheKey: (*"net/http.connectMethodKey")(0x2e3332207473658d),
conn: net.Conn nil,
tlsState: *(unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
br: *(unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
bw: *(unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
nwrite: (unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
reqch: chan net/http.requestAndChan {},
writech: chan net/http.writeRequest {},
closech: chan struct {} {},
isProxy: (unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
sawEOF: (unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
readLimit: (unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
writeErrCh: chan error {},
writeLoopDone: chan struct {} {},
idleAt: (*time.Time)(0x2e3332207473662d),
idleTimer: *(unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
mu: (*sync.Mutex)(0x2e3332207473664d),
numExpectedResponses: (unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
closed: error nil,
canceledErr: error nil,
broken: (unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
reused: (unreadable error while reading spliced memory at 0x2e33322074736575: EOF),
mutateHeaderFunc: (unreadable error while reading spliced memory at 0x2e33322074736575: EOF),},
*{
alt: net/http.RoundTripper nil,
t: *(unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
cacheKey: (*"net/http.connectMethodKey")(0x6f477c73b5c2374f),
conn: net.Conn nil,
tlsState: *(unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
br: *(unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
bw: *(unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
nwrite: (unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
reqch: chan net/http.requestAndChan {},
writech: chan net/http.writeRequest {},
closech: chan struct {} {},
isProxy: (unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
sawEOF: (unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
readLimit: (unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
writeErrCh: chan error {},
writeLoopDone: chan struct {} {},
idleAt: (*time.Time)(0x6f477c73b5c237ef),
idleTimer: *(unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
mu: (*sync.Mutex)(0x6f477c73b5c2380f),
numExpectedResponses: (unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
closed: error nil,
canceledErr: error nil,
broken: (unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
reused: (unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),
mutateHeaderFunc: (unreadable error while reading spliced memory at 0x6f477c73b5c23737: EOF),},
*{
alt: net/http.RoundTripper nil,
t: *(unreadable error while reading spliced memory at 0x6552747372694674: EOF),
cacheKey: (*"net/http.connectMethodKey")(0x655274737269468c),
conn: net.Conn nil,
tlsState: *(unreadable error while reading spliced memory at 0x6552747372694674: EOF),
br: *(unreadable error while reading spliced memory at 0x6552747372694674: EOF),
bw: *(unreadable error while reading spliced memory at 0x6552747372694674: EOF),
nwrite: (unreadable error while reading spliced memory at 0x6552747372694674: EOF),
reqch: chan net/http.requestAndChan {},
writech: chan net/http.writeRequest {},
closech: chan struct {} {},
isProxy: (unreadable error while reading spliced memory at 0x6552747372694674: EOF),
sawEOF: (unreadable error while reading spliced memory at 0x6552747372694674: EOF),
readLimit: (unreadable error while reading spliced memory at 0x6552747372694674: EOF),
writeErrCh: chan error {},
writeLoopDone: chan struct {} {},
idleAt: (*time.Time)(0x655274737269472c),
idleTimer: *(unreadable error while reading spliced memory at 0x6552747372694674: EOF),
mu: (*sync.Mutex)(0x655274737269474c),
numExpectedResponses: (unreadable error while reading spliced memory at 0x6552747372694674: EOF),
closed: error nil,
canceledErr: error nil,
broken: (unreadable error while reading spliced memory at 0x6552747372694674: EOF),
reused: (unreadable error while reading spliced memory at 0x6552747372694674: EOF),
mutateHeaderFunc: (unreadable error while reading spliced memory at 0x6552747372694674: EOF),},
*{
alt: net/http.RoundTripper nil,
t: *(unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
cacheKey: (*"net/http.connectMethodKey")(0x794265736e6f708b),
conn: net.Conn nil,
tlsState: *(unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
br: *(unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
bw: *(unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
nwrite: (unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
reqch: chan net/http.requestAndChan {},
writech: chan net/http.writeRequest {},
closech: chan struct {} {},
isProxy: (unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
sawEOF: (unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
readLimit: (unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
writeErrCh: chan error {},
writeLoopDone: chan struct {} {},
idleAt: (*time.Time)(0x794265736e6f712b),
idleTimer: *(unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
mu: (*sync.Mutex)(0x794265736e6f714b),
numExpectedResponses: (unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
closed: error nil,
canceledErr: error nil,
broken: (unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
reused: (unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),
mutateHeaderFunc: (unreadable error while reading spliced memory at 0x794265736e6f7073: EOF),},
*{
alt: net/http.RoundTripper nil,
t: *(unreadable error while reading spliced memory at 0x382e353031206574: EOF),
cacheKey: (*"net/http.connectMethodKey")(0x382e35303120658c),
conn: net.Conn nil,
tlsState: *(unreadable error while reading spliced memory at 0x382e353031206574: EOF),
br: *(unreadable error while reading spliced memory at 0x382e353031206574: EOF),
bw: *(unreadable error while reading spliced memory at 0x382e353031206574: EOF),
nwrite: (unreadable error while reading spliced memory at 0x382e353031206574: EOF),
reqch: chan net/http.requestAndChan {},
writech: chan net/http.writeRequest {},
closech: chan struct {} {},
isProxy: (unreadable error while reading spliced memory at 0x382e353031206574: EOF),
sawEOF: (unreadable error while reading spliced memory at 0x382e353031206574: EOF),
readLimit: (unreadable error while reading spliced memory at 0x382e353031206574: EOF),
writeErrCh: chan error {},
writeLoopDone: chan struct {} {},
idleAt: (*time.Time)(0x382e35303120662c),
idleTimer: *(unreadable error while reading spliced memory at 0x382e353031206574: EOF),
mu: (*sync.Mutex)(0x382e35303120664c),
numExpectedResponses: (unreadable error while reading spliced memory at 0x382e353031206574: EOF),
closed: error nil,
canceledErr: error nil,
broken: (unreadable error while reading spliced memory at 0x382e353031206574: EOF),
reused: (unreadable error while reading spliced memory at 0x382e353031206574: EOF),
mutateHeaderFunc: (unreadable error while reading spliced memory at 0x382e353031206574: EOF),},
*{
alt: net/http.RoundTripper nil,
t: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
cacheKey: (*"net/http.connectMethodKey")(0x7c736d383831354f),
conn: net.Conn nil,
tlsState: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
br: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
bw: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
nwrite: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
reqch: chan net/http.requestAndChan {},
writech: chan net/http.writeRequest {},
closech: chan struct {} {},
isProxy: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
sawEOF: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
readLimit: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
writeErrCh: chan error {},
writeLoopDone: chan struct {} {},
idleAt: (*time.Time)(0x7c736d38383135ef),
idleTimer: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
mu: (*sync.Mutex)(0x7c736d383831360f),
numExpectedResponses: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
closed: error nil,
canceledErr: error nil,
broken: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
reused: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
mutateHeaderFunc: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),},
]
(dlv) p pconn
*net/http.persistConn {
alt: net/http.RoundTripper nil,
t: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
cacheKey: net/http.connectMethodKey {proxy: (unreadable could not read string len error while reading spliced memory at 0x7c736d3838313537: EOF), scheme: (unreadable could not read string len error while reading spliced memory at 0x7c736d3838313537: EOF), addr: (unreadable could not read string len error while reading spliced memory at 0x7c736d3838313537: EOF), onlyH1: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF)},
conn: net.Conn nil,
tlsState: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
br: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
bw: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
nwrite: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
reqch: chan net/http.requestAndChan {},
writech: chan net/http.writeRequest {},
closech: chan struct {} {},
isProxy: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
sawEOF: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
readLimit: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
writeErrCh: chan error {},
writeLoopDone: chan struct {} {},
idleAt: time.Time {
wall: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
ext: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
loc: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),},
idleTimer: *(unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
mu: sync.Mutex {state: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF), sema: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF)},
numExpectedResponses: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
closed: error nil,
canceledErr: error nil,
broken: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
reused: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),
mutateHeaderFunc: (unreadable error while reading spliced memory at 0x7c736d3838313537: EOF),}
What did you expect to see?
Successfully sending out HTTPS request
What did you see instead?
The memory of idle connections list seems corrupted
func (t *Transport) queueForIdleConn(w *wantConn) (delivered bool) {
// ...
// Look for most recently-used idle connection.
if list, ok := t.idleConn[w.key]; ok {
stop := false
delivered := false
for len(list) > 0 && !stop {
pconn := list[len(list)-1]
@davecheney thanks for the info. Will try to test it.
BTW, we get another net/http related crash stack trace.
panic: close of closed channel
goroutine 27732906 [running]:
panic(0xd52840, 0xfa4110)
/usr/local/go/src/runtime/panic.go:1064 +0x545 fp=0xc00ecd33c0 sp=0xc00ecd32f8 pc=0x43a025
runtime.closechan(0xc00f5ebe00)
/usr/local/go/src/runtime/chan.go:365 +0x25a fp=0xc00ecd3400 sp=0xc00ecd33c0 pc=0x408eda
net/http.(*persistConn).roundTrip(0xc00ea685a0, 0xc01a7c3800, 0xc010707710, 0x0, 0x0)
/usr/local/go/src/net/http/transport.go:2603 +0xb3a fp=0xc00ecd3718 sp=0xc00ecd3400 pc=0x752dda
net/http.(*Transport).roundTrip(0xc000402280, 0xc00e266100, 0x30, 0x7f5be55f9188, 0x150)
/usr/local/go/src/net/http/transport.go:582 +0xa65 fp=0xc00ecd3960 sp=0xc00ecd3718 pc=0x746f65
net/http.(*Transport).RoundTrip(0xc000402280, 0xc00e266100, 0xc000402280, 0x0, 0x0)
/usr/local/go/src/net/http/roundtrip.go:17 +0x35 fp=0xc00ecd3998 sp=0xc00ecd3960 pc=0x72ccb5
net/http.send(0xc00e266100, 0xfb5940, 0xc000402280, 0x0, 0x0, 0x0, 0xc006cb7230, 0x203005, 0x1, 0x0)
/usr/local/go/src/net/http/client.go:252 +0x453 fp=0xc00ecd3b40 sp=0xc00ecd3998 pc=0x6e8b93
net/http.(*Client).send(0xc0000d5230, 0xc00e266100, 0x0, 0x0, 0x0, 0xc006cb7230, 0x0, 0x1, 0x203005)
/usr/local/go/src/net/http/client.go:176 +0xff fp=0xc00ecd3bc0 sp=0xc00ecd3b40 pc=0x6e857f
net/http.(*Client).do(0xc0000d5230, 0xc00e266100, 0x0, 0x0, 0x0)
/usr/local/go/src/net/http/client.go:718 +0x45f fp=0xc00ecd3d98 sp=0xc00ecd3bc0 pc=0x6ea59f
net/http.(*Client).Do(...)
/usr/local/go/src/net/http/client.go:586
Found one buz code change: 1st time to fully use
httptrace strings.Builder
currentTraceTime := time.Now()
trace := &httptrace.ClientTrace{
GetConn: func(hostPort string) {
fmt.Fprintf(&httptrace, "\n\tHttpTrace: |%s %v|", "GetConn", time.Since(currentTraceTime))
currentTraceTime = time.Now()
},
DNSStart: func(dnsInfo httptrace.DNSStartInfo) {
fmt.Fprintf(&httptrace, "%s %v|", "DNSStart", time.Since(currentTraceTime))
currentTraceTime = time.Now()
},
DNSDone: func(dnsInfo httptrace.DNSDoneInfo) {
fmt.Fprintf(&httptrace, "%s %v|", "DNSDone", time.Since(currentTraceTime))
currentTraceTime = time.Now()
},
ConnectStart: func(network, addr string) {
fmt.Fprintf(&httptrace, "%s %v|", "ConnectStart", time.Since(currentTraceTime))
currentTraceTime = time.Now()
},
ConnectDone: func(network, addr string, err error) {
fmt.Fprintf(&httptrace, "%s %v|", "ConnectDone", time.Since(currentTraceTime))
currentTraceTime = time.Now()
},
TLSHandshakeStart: func() {
fmt.Fprintf(&httptrace, "%s %v|", "TLSHandshakeStart", time.Since(currentTraceTime))
currentTraceTime = time.Now()
},
TLSHandshakeDone: func(connState tls.ConnectionState, err error) {
fmt.Fprintf(&httptrace, "%s %v|", "TLSHandshakeDone", time.Since(currentTraceTime))
currentTraceTime = time.Now()
},
GotConn: func(connInfo httptrace.GotConnInfo) {
fmt.Fprintf(&httptrace, "%s %v|", "GotConn", time.Since(currentTraceTime))
r.remoteAddr = util.ExtractIP(connInfo.Conn.RemoteAddr().String())
currentTraceTime = time.Now()
},
WroteRequest: func(wrInfo httptrace.WroteRequestInfo) {
fmt.Fprintf(&httptrace, "%s %v|", "WroteRequest", time.Since(currentTraceTime))
currentTraceTime = time.Now()
},
GotFirstResponseByte: func() {
fmt.Fprintf(&httptrace, "%s %v|", "GotFirstResponseByte", time.Since(currentTraceTime))
},
}
Hi @davecheney
Actually, there is racing for shared variables(like currentTraceTime, strings.Builder) among different HTTP trace hooks(or callbacks).
Do you have any proposal for the HTTP trace usage? One idea is to lock in each hook.
Thanks.
Have a look at how we do it in github.com/davecheney/httpstat
Hi @davecheney, Thanks for the help so much.
httpstat seems mainly for a single HTTP request.
For PRD usage, the net/http client requests concurrency may be quite high. I noticed one comment "some may be called after the request has completed or failed" for ClientTrace.
I have a concern if there is data access racing for the shared variable between ClientTrace hook and the usage after getting the response, like the connInfo.Conn.RemoteAddr() from httptrace.GotConnInfo.
// ClientTrace is a set of hooks to run at various stages of an outgoing
// HTTP request. Any particular hook may be nil. Functions may be
// called concurrently from different goroutines and some may be called
// after the request has completed or failed.
//
// ClientTrace currently traces a single HTTP request & response
// during a single round trip and has no hooks that span a series
// of redirected requests.
//
// See https://blog.golang.org/http-tracing for more.
type ClientTrace struct {
There could be a possibility of data races during the different phases of ClientTrace as the memory ownership and happens before relationships are not (IMO) well defined for that use case. I would proceed cautiously and make heavy use of the race detector to check your work.
Also, I should remind you that unlike many projects, the Go project does not use GitHub Issues for general discussion or asking questions. GitHub Issues are used for tracking bugs and proposals only.
For asking questions, see:
Stack Overflow with questions tagged "go"
The Go Forum, a web-based forum
Gophers Slack, use the invite app for access. The #general channel is a good starting point.
Go Community on Hashnode with questions and posts tagged with "go"
The golang-nuts mailing list
IRC channel #go-nuts on Freenode
Thanks for the reminder and all the help, which is truly impressive.
I got such log: TLSHandshakeStart is after GotConn/WroteRequest, which means the variable shared between the trace hook and the usage after got response body may encounter race condition. One approach is to use atomic/mutex for the shared variables.
HttpTrace: |GetConn 37.462µs|GotConn 136.729174ms|WroteRequest 44.956µs|TLSHandshakeStart 102.646256ms|TLSHandshakeDone 1.03µs|
| gharchive/issue | 2020-11-24T16:09:19 | 2025-04-01T06:44:20.923679 | {
"authors": [
"davecheney",
"xqzhang2015"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/42810",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
755340893 | x/perf: build failures due to C compiler warnings in github.com/mattn/go-sqlite3
2020-09-18T15:55:09-d949658/solaris-amd64-oraclerel
# github.com/mattn/go-sqlite3
sqlite3-binding.c: In function ‘sqlite3SelectNew.isra’:
sqlite3-binding.c:114528:10: warning: function may return address of local variable [-Wreturn-local-addr]
114528 | return pNew;
| ^~~~
sqlite3-binding.c:114488:10: note: declared here
114488 | Select standin;
| ^~~~~~~
go build github.com/mattn/go-sqlite3: C compiler warning promoted to error on Go builders
2020-09-18T15:55:09-d949658/openbsd-386-68
2020-09-18T15:55:09-d949658/openbsd-amd64-68
# github.com/mattn/go-sqlite3
sqlite3-binding.c:104476:38: warning: implicit conversion from 'long long' to 'double' changes value from 9223372036854775806 to 9223372036854775808 [-Wimplicit-int-float-conversion]
sqlite3-binding.c:104478:46: warning: implicit conversion from 'long long' to 'double' changes value from 9223372036854775806 to 9223372036854775808 [-Wimplicit-int-float-conversion]
go build github.com/mattn/go-sqlite3: C compiler warning promoted to error on Go builders
CC @golang/release via #11811
These failures appear to have started as of CL 222540, which suggests that those builders were previously using the vendor directory, although I'm not sure how. (The default -mod=vendor trigger is go 1.14 or higher in the go.mod file, but the go.mod file for this module is only go 1.11.)
The implicit conversion seems to be mattn/go-sqlite3#867, which AFAICT is fixed upstream at head.
The function may return address of local variable warning seems to be mattn/go-sqlite3#803 / mattn/go-sqlite3#822.
Fixed by CL 275192.
| gharchive/issue | 2020-12-02T14:51:48 | 2025-04-01T06:44:20.930258 | {
"authors": [
"bcmills"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/42941",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
772524232 | x/build: trybots are down
Go Bot hasn't acked any Run-TryBot+1 requests since about 1:15pm.
https://go-review.googlesource.com/c/go/+/276653 (patch set 10) is the most recent CL I've seen to get acknowledged by Go Bot, after Run-TryBot+1 at 1:00pm.
https://go-review.googlesource.com/c/go/+/277934 (patch set 7) is the earliest CL I've seen to not get acknowledged by Go Bot, after Run-TryBot+1 at 1:17pm.
@aclements mentions having uploaded duplicate CLs to Gerrit around this time window, which they said seemed to have broken Gerrit. Perhaps the CLs broke Go Bot too?
/cc @golang/release
Thanks for reporting this.
Coordinator finds work by using Gerrit's API, so if Gerrit is having issues, that would explain trybots not starting. We should figure out if the problem is on the side of Gerrit or not.
I see at https://farmer.golang.org/#trybots there are active trybot runs right now, so maybe this is an issue only affecting a subset of CLs?
I see at https://farmer.golang.org/#trybots there are active trybot runs right now,
They all seem to be stuck though. All builds have completed, yet the trybot run isn't completed.
From coordinator logs, it seems clear that the problem is coordinator is failing to find work:
$ kubectl logs coordinator-deployment-6cff4d4d8d-5j6f5 | grep "failed to find trybot work:"
[...]
2020/12/21 23:37:21 failed to find trybot work: rpc error: code = 2 desc = grpc: HTTP status 404 Not Found; Multiple changes found for build%7Emaster%7EIb1c7ae1e914116dd8a4440db8ee46d6af3ed1ad5%0A
2020/12/21 23:37:36 failed to find trybot work: rpc error: code = 2 desc = grpc: HTTP status 404 Not Found; Multiple changes found for build%7Emaster%7EIb1c7ae1e914116dd8a4440db8ee46d6af3ed1ad5%0A
2020/12/21 23:37:52 failed to find trybot work: rpc error: code = 2 desc = grpc: HTTP status 404 Not Found; Multiple changes found for build%7Emaster%7EIb1c7ae1e914116dd8a4440db8ee46d6af3ed1ad5%0A
2020/12/21 23:38:06 failed to find trybot work: rpc error: code = 2 desc = grpc: HTTP status 404 Not Found; Multiple changes found for build%7Emaster%7EIb1c7ae1e914116dd8a4440db8ee46d6af3ed1ad5%0A
2020/12/21 23:38:22 failed to find trybot work: rpc error: code = 2 desc = grpc: HTTP status 404 Not Found; Multiple changes found for build%7Emaster%7EIb1c7ae1e914116dd8a4440db8ee46d6af3ed1ad5%0A
Ib1c7ae1e914116dd8a4440db8ee46d6af3ed1ad5 is the Change-Id of the duplicate CL that was mentioned.
Coordinator will need to be updated to handle it without an error.
Is it possible to Gerrit super-user force delete https://go-review.googlesource.com/c/build/+/279515 or something? Maybe that would help unstick the situation?
Yes, deleting the CL may work.
We should also make coordinator more resilient so this won't need to be dealt with manually if a duplicate Gerrit CL happens again in the future.
I suspect the problem could be how we're specifying the ID in the call here:
comments, err := gerritc.ListChangeComments(ctx, ci.ID)
if err != nil {
return nil, err
}
But I can't reproduce it locally with maintq try-work. Not sure why yet.
We should also make coordinator more resilient so this won't need to be dealt with manually if a duplicate Gerrit CL happens again in the future.
Probably. I think it's a bug that Gerrit was able to get into this state in the first place though. I thought that Gerrit is supposed to ensure a 1:1 correspondence between Branch+Change-Id and CL. I'm not sure how @aclements was able to create those duplicate CLs in the first place.
Ok, I could reproduce it by hitting the maintapi server directly, after some tries:
$ maintq try-work
rpc error: code = 2 desc = grpc: HTTP status 404 Not Found; Multiple changes found for build%7Emaster%7EIb1c7ae1e914116dd8a4440db8ee46d6af3ed1ad5%0A
exit status 1
It succeeds more often than fails, so maybe this isn't what's causing TryBots not to start. (Although it's possible it is what caused coordinator to get into a bad/stuck state, at least in its ability to start/complete TryBot runs.)
I think a good next step will be to restart coordinator and see how behaves, before we try to do more. I'll do that now.
Restarting alone doesn't seem to make a big difference, TryBot runs are still reaching "Builds remaining: 0" and not completing (e.g., https://farmer.golang.org/try?commit=0ca67663 has reached that state some time ago and is still "Active"). This seems to be a problem that needs to be fixed in addition to the "findTryWork" one.
This needs some more investigation, and I'll resume working on it tomorrow morning.
I've deployed CL 279672 for the problem spotted in https://github.com/golang/go/issues/43312#issuecomment-749263814, and I'm seeing much TryBot activity has resumed.
This problem is resolved now.
I've filed 3 issues for some follow up tasks:
#43326—x/build/cmd/coordinator: add health item for work queue size and errors in finding work
#43323—x/build/cmd/coordinator: understand why TryBot runs sometimes fail to complete after reaching 0 builds remaining
#43324—x/build/cmd/coordinator: failing to complete TryBot run for change with duplicate Change-Id line
| gharchive/issue | 2020-12-21T23:18:31 | 2025-04-01T06:44:20.942859 | {
"authors": [
"dmitshur",
"mdempsky"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/43312",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
890354284 | crypto/ed25519: TestAllocations fails with BoringCrypto or without optimizations
--- FAIL: TestAllocations (0.02s)
ed25519_test.go:202: expected zero allocations, got 3e+01
Testing performance profiles is brittle: it currently fails on the noopt builder, and on the dev.boringcrypto branch. Even if we skipped it on both, it would fail on go test all if run without optimizations, like Delve does.
We can make it run only on a specific builder, but I am worried that will eventually drift out of sync with builder naming and end up never running. Better than nothing, I guess.
@FiloSottile thank you for filing this issue. We haven't done anything related for this cycle Go1.18, hence I am punting the issue to Go1.19.
CC @golang/security
Looks like this didn't make 1.19. Moving to backlog. Please recategorize as appropriate.
| gharchive/issue | 2021-05-12T18:06:04 | 2025-04-01T06:44:20.945914 | {
"authors": [
"FiloSottile",
"ianlancetaylor",
"odeke-em"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/46134",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
909087711 | x/tour: not rendering with file corrupted error message
The tour.golang.org website appears to be broken in Firefox.
I see the following errors in the console:
Exception { name: "NS_ERROR_FILE_CORRUPTED", message: "", result: 2152857611, filename: "https://tour.golang.org/script.js", lineNumber: 10805, columnNumber: 0, data: null, stack: "get@https://tour.golang.org/script.js:10805:45\n@https://tour.golang.org/script.js:10826:30\nd@https://tour.golang.org/script.js:633:127\nsb/l.$injector<@https://tour.golang.org/script.js:634:298\nc@https://tour.golang.org/script.js:631:479\nd@https://tour.golang.org/script.js:632:111\nj/</<@https://tour.golang.org/script.js:641:469\nn@https://tour.golang.org/script.js:611:192\nj/<@https://tour.golang.org/script.js:641:439\nd@https://tour.golang.org/script.js:633:127\nsb/l.$injector<@https://tour.golang.org/script.js:634:298\nc@https://tour.golang.org/script.js:631:479\nr@https://tour.golang.org/script.js:651:142\nW@https://tour.golang.org/script.js:645:344\nz@https://tour.golang.org/script.js:644:336\nz@https://tour.golang.org/script.js:644:447\nz@https://tour.golang.org/script.js:644:447\nP@https://tour.golang.org/script.js:643:115\nk@https://tour.golang.org/script.js:755:16\n$broadcast@https://tour.golang.org/script.js:695:240\ni/<@https://tour.golang.org/script.js:687:191\nh@https://tour.golang.org/script.js:682:437\nh@https://tour.golang.org/script.js:682:437\nthen/<@https://tour.golang.org/script.js:683:169\n$eval@https://tour.golang.org/script.js:693:272\n$digest@https://tour.golang.org/script.js:691:126\n$apply@https://tour.golang.org/script.js:693:431\ne@https://tour.golang.org/script.js:699:424\no@https://tour.golang.org/script.js:703:4\nXc/</s.onreadystatechange@https://tour.golang.org/script.js:704:132\n" }
Error: Circular dependency: syntaxCheckboxDirective
c@https://tour.golang.org/script.js:631:380
d@https://tour.golang.org/script.js:632:111
j/</<@https://tour.golang.org/script.js:641:469
n@https://tour.golang.org/script.js:611:192
j/<@https://tour.golang.org/script.js:641:439
d@https://tour.golang.org/script.js:633:127
sb/l.$injector<@https://tour.golang.org/script.js:634:298
c@https://tour.golang.org/script.js:631:479
r@https://tour.golang.org/script.js:651:142
W@https://tour.golang.org/script.js:645:344
z@https://tour.golang.org/script.js:644:336
z@https://tour.golang.org/script.js:644:447
z@https://tour.golang.org/script.js:644:447
z@https://tour.golang.org/script.js:644:447
z@https://tour.golang.org/script.js:644:447
P@https://tour.golang.org/script.js:643:115
k@https://tour.golang.org/script.js:755:16
$broadcast@https://tour.golang.org/script.js:695:240
i/<@https://tour.golang.org/script.js:687:191
h@https://tour.golang.org/script.js:682:437
h@https://tour.golang.org/script.js:682:437
then/<@https://tour.golang.org/script.js:683:169
$eval@https://tour.golang.org/script.js:693:272
$digest@https://tour.golang.org/script.js:691:126
$apply@https://tour.golang.org/script.js:693:431
e@https://tour.golang.org/script.js:699:424
o@https://tour.golang.org/script.js:703:4
Xc/</s.onreadystatechange@https://tour.golang.org/script.js:704:132
Clearing local storage fixed the main problem, the page renders, but sometimes I still see the file corrupted message intermittently.
I'm not sure what data or how critical the data is the tour stores in local storage, but possibly there was a change made to the tour that has caused local storage to be interpreted in a way that it doesn't expect?
The last time I visited the tour was over 7 days ago, but less than 21 days ago, and in the last 7 days the following two commits were made, neither seem meaningful, except the second upgrading dependencies could suggest a change in dependency has subtly changed something.
https://github.com/golang/tour/commit/3891a3eb15c053bf644848461c84facbfe1cfb34
https://github.com/golang/tour/commit/fa9f4b9b72b09cbc55464f936d32179709b937cd
It's also possible this is an isolated issue and for some reason my local storage was corrupt in some way.
Clearing local storage fixed the main problem, the page renders
I spoke too soon about this. Visiting the tour a second time after clearing local storage causes the problem to recur.
This turned out to be a Firefox local storage corruption issue, unrelated to this specific website.
| gharchive/issue | 2021-06-02T04:30:16 | 2025-04-01T06:44:20.950998 | {
"authors": [
"leighmcculloch"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/46512",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
1116589040 | x/build/cmd/relui: compute Google Groups URL of sent announcement email
A part of completing the task that sends a release announcement email (#47405) is being able to tell that the email has been received by Google Groups (and moderated as needed), and determining its URL. This URL serves as the canonical announcement URL, and is currently included in the release tweet (see ReleaseTweet.Announcement).
For example, the email announcing the release of Go 1.17.6 and Go 1.16.13 was sent on Jan 6. It was successfully received, moderated, and its URL became https://groups.google.com/g/golang-announce/c/95ZD3rKn4DI/m/93cyN8F1BAAJ.
We don't find out what the final URL will be on google groups as part of sending the email, this needs to be computed afterwards, so breaking it out as a smaller issue.
Since the announcement group is highly moderated, it's likely sufficient for us to match based on subject. If the subject matches, it must be the right email:
// Input:
"Go 1.17.6 and Go 1.16.13 are released"
// Output:
"https://groups.google.com/g/golang-announce/c/95ZD3rKn4DI/m/93cyN8F1BAAJ"
"https://groups.google.com/g/golang-announce/c/95ZD3rKn4DI" // Alternative, also okay.
// Input:
"Go 1.17 Release Candidate 3 is released"
// Output:
os.ErrNotExist or context.DeadlineExceeded // Accurate on 2022-01-27.
It's expected that emails don't travel instantaneously, so the code would need to run in a loop that tries (with a reasonable deadline) multiple times, and once the email is received by google groups, eventually succeeds.
Options:
Extract the subject and URL from the HTML of Google Groups (https://groups.google.com/g/golang-announce).
Use an API to find a matching thread, if such an API is available and using it is viable.
CC @golang/release.
I've explored the aforementioned options.
There may be an API we can use, but using it has some overhead in initial setup and maintenance. Based on what I've learned, I think we can defer considing/exploring it more until a later milestone.
Parsing the HTML appears to be a more attractive immediate solution to this (to unblock #47405). I've prototyped it locally and it seems to meet the needs of this task. Unless we learn of a reason it can't work well, I'll clean it up and send a CL.
| gharchive/issue | 2022-01-27T18:07:50 | 2025-04-01T06:44:20.956217 | {
"authors": [
"dmitshur"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/50864",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
51283541 | proposal: spec: byte view: type that can represent a []byte or string
Go has no built-in type which can represent a read-only view of possibly-changing bytes.
We have two current types which internally are a byte pointer and a length: (the first
also has a capacity, irrelevant to this issue)
[]byte
Allowance: "you can mess with this memory"
Restriction: "this memory isn't necessarily forever; anybody else might be changing it now (race) or later (iterator becoming invalidated after this call, etc)"
string
Allowance: "if you have a reference to this, it's good forever, and nobody will ever mess with it".
Restriction: "you can NOT change"
And because of their conflicting allowances and restrictions, any conversion from
string->[]byte or []byte->string necessarily has to make a copy and often
generates garbage.
Both are great, but there's a missing piece:
Many places in Go's API want a type with *neither* of those allowances, and are happy
with both of those restrictions:
-- much of pkg strings
-- much of pkg bytes
-- all the strconv.Parse* functions (issue #2632)
-- most of the string arguments to system calls
-- the io.Writer interface's argument (no need then for io.WriteString)
-- leveldb's Key/Value accessors.
Look at leveldb's Iterator:
http://godoc.org/code.google.com/p/leveldb-go/leveldb/db#Iterator
It has to go out of its way to say "please do not corrupt my memory". If
somebody uses memdb directly
(http://godoc.org/code.google.com/p/leveldb-go/leveldb/memdb) and misuses the Iterator
type, all the sudden the memory corruption and stack traces implicate the memdb package,
even though it's a caller of memdb who violated the db.Iterator contract.
All leveldb really wants is to give callers a view of the bytes.
That is, in addition to "string" with its promise A (handle for life) and
"[]byte" with its promise B (you can mutate), we need a a common type to both
of those with neither promise, what is constant-time assignable from a string or a
[]byte.
s := "string"
b := []byte(s)
var v byteview = s // constant time assignment
var v byteview = b // constant time assignment
A byteview would be represented in memory just like a string (*byte, int), but the
compiler would prevent mutations or addressing (like string), and callers would always
treat its backing data as ephemeral, like a []byte, unless negotiated otherwise.
Obviously this isn't a Go 1.n item, considering the impact it would have on the standard
library.
A good name for byteview would be "bytes", but we have a package
"bytes" already. Maybe we can get rid of that package.
The discussion of this is at https://groups.google.com/d/msg/golang-dev/Y7j4B2r_eDw/Ovu7WbSzlOUJ.
Certain implementations of generics (#15292) would provide this. But not all--it's possible to imagine generics proposals that do not support this.
Putting on hold until we have some more experience with generics.
Update: I wrote go4.org/mem which I'm now happily using in a number of places.
Update: I wrote go4.org/mem which I'm now happily using in a number of places.
Everyone was forced to do something like this for ages...
https://godoc.org/lab.nexedi.com/kirr/go123/mem#String
https://godoc.org/lab.nexedi.com/kirr/go123/mem#Bytes
@navytux, the difference is that go4.org/mem doesn't let callers do unsafe things.
@bradfitz, indeed, thanks for feedback.
| gharchive/issue | 2013-04-30T03:40:15 | 2025-04-01T06:44:20.961835 | {
"authors": [
"bradfitz",
"cespare",
"ianlancetaylor",
"navytux"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/5376",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
1473375450 | Module's local transitive dependencies work only with explicit replace directive
What version of Go are you using (go version)?
$ go version
go version go1.20-pre3 cl/474093167 +a813be86df linux/amd64
Does this issue reproduce with the latest release?
Yes
What operating system and processor architecture are you using (go env)?
go env Output
$ go env
GO111MODULE=""
GOARCH="amd64"
GOBIN=""
GOCACHE="$HOME/.cache/go-build"
GOENV="$HOME/.config/go/env"
GOEXE=""
GOEXPERIMENT=""
GOFLAGS=""
GOHOSTARCH="amd64"
GOHOSTOS="linux"
GOINSECURE=""
GOMODCACHE="$HOME/go/pkg/mod"
GONOPROXY=""
GONOSUMDB=""
GOOS="linux"
GOPATH="$HOME/go"
GOPRIVATE=""
GOPROXY="https://proxy.golang.org,direct"
GOROOT="/usr/lib/google-golang"
GOSUMDB="sum.golang.org"
GOTMPDIR=""
GOTOOLDIR="/usr/lib/google-golang/pkg/tool/linux_amd64"
GOVCS=""
GOVERSION="go1.20-pre3 cl/474093167 +a813be86df"
GCCGO="gccgo"
GOAMD64="v1"
AR="ar"
CC="clang"
CXX="clang++"
CGO_ENABLED="1"
GOMOD="/dev/null"
GOWORK=""
CGO_CFLAGS="-g -O2"
CGO_CPPFLAGS=""
CGO_CXXFLAGS="-g -O2"
CGO_FFLAGS="-g -O2"
CGO_LDFLAGS="-g -O2"
PKG_CONFIG="pkg-config"
GOGCCFLAGS="-fPIC -m64 -pthread -fno-caret-diagnostics -Qunused-arguments -Wl,--no-gc-sections -fmessage-length=0 -fdebug-prefix-map=/tmp/go-build2517995585=/tmp/go-build -gno-record-gcc-switches"
GOROOT/bin/go version: go version go1.20-pre3 cl/474093167 +a813be86df linux/amd64
GOROOT/bin/go tool compile -V: compile version go1.20-pre3 cl/474093167 +a813be86df
uname -sr: Linux 5.18.16-1rodete4-amd64
Distributor ID: Debian
Description: Debian GNU/Linux rodete
Release: rodete
Codename: rodete
/lib/x86_64-linux-gnu/libc.so.6: GNU C Library (Debian GLIBC 2.35-1) stable release version 2.35.
gdb --version: GNU gdb (GDB) 10.0-gg5
What did you do?
There are three local modules, with top depending on middle, and middle depending on bottom
$ find . -type f -printf "\n---%P---\n" -exec cat {} \;
---middle/go.mod---
module middle
require _/bottom v0.0.0-00010101000000-000000000000
replace _/bottom => ../bottom
go 1.20
---middle/middle.go---
package middle
import "_/bottom"
var X = bottom.X
---top/go.mod---
module main
require _/middle v0.0.0-00010101000000-000000000000
replace _/middle => ../middle
go 1.20
---top/top.go---
package main
import "_/middle"
func main() {
println(middle.X)
}
---bottom/go.mod---
module bottom
go 1.20
---bottom/bottom.go---
package bottom
var X int = 1
What did you expect to see?
cd top && go run . should succeed
What did you see instead?
$ (cd top && go run .)
../middle/middle.go:2:8: missing go.sum entry for module providing package _/bottom (imported by _/middle); to add:
go get _/middle@v0.0.0-00010101000000-000000000000
Doing what is suggested doesn't help:
$ (cd top && go get _/middle@v0.0.0-00010101000000-000000000000)
_/middle imports
_/bottom: malformed module path "_/bottom": missing dot in first path element
What does help is explicitly adding replace directive to top/go.mod:
(cd top && go mod edit -replace _/bottom=../bottom && go mod tidy)
$ (cd top && go run .)
1
Requesting that a module keeps track of all its transitive local dependencies is IMHO a maintenance problem.
Working as intended, replace directives only apply at the main module, if you need to work with multiple modules, consider using workspaces.
Unlike many projects, the Go project does not use GitHub Issues for general discussion or asking questions. GitHub Issues are used for tracking bugs and proposals only.
For questions please refer to https://github.com/golang/go/wiki/Questions
Hmm, I don't think this is working as intended. go get _/middle@v0.0.0-00010101000000-000000000000 within module top should apply the replace directive from top/go.mod, which should arguably avoid the missing dot in first path element error.
I suspect we're failing to apply the replacement here:
https://cs.opensource.google/go/go/+/master:src/cmd/go/internal/modload/query.go;l=184-187;drc=c85848a4a62f839427bb56cb2772e008e405cf1f
doesn't top have no require / replace for _/bottom, and the error comes from _/bottom being an unretrievable module?
Ah, you're right! The error message _/bottom: malformed module path "_/bottom": missing dot in first path element is indeed for that module.
| gharchive/issue | 2022-12-02T19:56:12 | 2025-04-01T06:44:20.970512 | {
"authors": [
"asmundak",
"bcmills",
"seankhliao"
],
"repo": "golang/go",
"url": "https://github.com/golang/go/issues/57056",
"license": "BSD-3-Clause",
"license_type": "permissive",
"license_source": "github-api"
} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.