added
stringdate
2025-04-01 04:05:38
2025-04-01 07:14:06
created
timestamp[us]date
2001-10-09 16:19:16
2025-01-01 03:51:31
id
stringlengths
4
10
metadata
dict
source
stringclasses
2 values
text
stringlengths
0
1.61M
2025-04-01T06:41:16.031433
2018-03-05T09:23:20
302224199
{ "authors": [ "KolushovAlexandr", "Railia" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12830", "repo": "it-projects-llc/pos-addons", "url": "https://github.com/it-projects-llc/pos-addons/pull/635" }
gharchive/pull-request
[REF] debts on paymentmethods rendering @KolushovAlexandr Displays negative values ​in debt history in POS after a purchase of Credit. merged in https://github.com/it-projects-llc/pos-addons/pull/632
2025-04-01T06:41:16.040214
2019-10-03T20:12:35
502273653
{ "authors": [ "jorgeorpinel" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12831", "repo": "iterative/dvc.org", "url": "https://github.com/iterative/dvc.org/pull/671" }
gharchive/pull-request
Even more regular updates (early Oct) towards new Use Case Note: I reverted be661b5 with 41c49ef instead of removing via force-push (since Ivan added a commit in between). Better to keep history linear (:
2025-04-01T06:41:16.041988
2016-04-21T09:45:08
150022544
{ "authors": [ "LeeSmet", "robvanmieghem" ], "license": "bsd-3-clause", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12832", "repo": "itsyouonline/identityserver", "url": "https://github.com/itsyouonline/identityserver/issues/48" }
gharchive/issue
Tree structure: show all organizations in the tree where someone has access to @robvanmieghem does this relate to the view on the user page, or the structure on the organization detail page? Either this is related to the user page, in which case its fixed, or it is related to the org page, which is more accurately tracked by #474
2025-04-01T06:41:16.085342
2020-07-20T09:33:09
661597223
{ "authors": [ "coveralls", "javrasya" ], "license": "bsd-3-clause", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12833", "repo": "javrasya/django-river", "url": "https://github.com/javrasya/django-river/pull/168" }
gharchive/pull-request
[#162] Don't cancel possible transitions even though it is the future of one of those that is cancelled Coverage decreased (-0.3%) to 80.217% when pulling 0c46393b32e9d1ba91f5523f12f75b1f2957cc8c on issue-162-dont-cancel-possible-transitions into f693cdb8fe1e0cfa56eea19a2d3bac89d4725504 on master. Coverage decreased (-0.3%) to 80.217% when pulling 0c46393b32e9d1ba91f5523f12f75b1f2957cc8c on issue-162-dont-cancel-possible-transitions into f693cdb8fe1e0cfa56eea19a2d3bac89d4725504 on master.
2025-04-01T06:41:16.123701
2021-06-21T16:15:59
926385308
{ "authors": [ "jan-janssen", "tjelvar-olsson" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12834", "repo": "jic-dtool/dtool-create", "url": "https://github.com/jic-dtool/dtool-create/pull/25" }
gharchive/pull-request
Include LICENSE.rst Thank you. Merged and released in: https://pypi.org/project/dtool-create/0.23.4/
2025-04-01T06:41:16.169821
2019-04-11T02:13:38
431798122
{ "authors": [ "jturner314" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12835", "repo": "jturner314/ndarray-stats", "url": "https://github.com/jturner314/ndarray-stats/pull/40" }
gharchive/pull-request
Update to Rust 2018 and replace quickcheck! with #[quickcheck] I've also bumped the Rust version to 1.34.
2025-04-01T06:41:16.244113
2017-10-16T02:47:42
265632127
{ "authors": [ "AyeshaShafique", "kailash09dabhi" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12836", "repo": "kailash09dabhi/OmRecorder", "url": "https://github.com/kailash09dabhi/OmRecorder/issues/30" }
gharchive/issue
Hi. I am new to Android. I just want to use mobile microphone to record wav audio and just find your repository. I ran it & it compiled successfully. But the problem is that the path you have used to store wav file is not actually having this file. Means i could not get the file, if i want to use that file for further processing. Note that i want to use this wav file for speech recognition purpose done in another device running on python. Any help would be more than welcome, as i am new bae. Sorry but please use stackoverflow to ask question.
2025-04-01T06:41:16.251586
2019-04-03T14:49:54
428812596
{ "authors": [ "claustres" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12837", "repo": "kalisio/kdk", "url": "https://github.com/kalisio/kdk/issues/34" }
gharchive/issue
Allow CLI to manage repositories in different organizations https://github.com/kalisio/kdk/commit/c46ca6ca8116ec6b8aae4fc925396c29461cbea9 allowed to provide organization as an argument for workspaces completely under the organization. Now we need to be able to also specify organization for specific modules in workspaces for projects using repositories from different organizations (e.g. KDK + Weacast).
2025-04-01T06:41:16.280002
2016-08-10T08:42:34
170363197
{ "authors": [ "coveralls", "wahwynn" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12838", "repo": "kayak/fireant", "url": "https://github.com/kayak/fireant/pull/1" }
gharchive/pull-request
Changed hostage references to fireant Coverage remained the same at 96.034% when pulling c1530b091934b9c72f6756d110ea5c6bf93d7cff on wahwynn:master into 488e483469a63161c744ce5d167de5470cc113c7 on kayak:master.
2025-04-01T06:41:16.294589
2018-02-13T07:11:27
296634985
{ "authors": [ "gunthercox", "torakiki", "wirecamp" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12839", "repo": "kelseyhightower/nocode", "url": "https://github.com/kelseyhightower/nocode/issues/1667" }
gharchive/issue
Not working on Internet Explorer I believe you can set the following in your configuration as a workaround. I tried that but I am getting this error ` ` It works on my local machine
2025-04-01T06:41:16.298115
2015-09-17T02:29:35
106899770
{ "authors": [ "barrettj", "chenzeyu", "haducloc", "kevin0571", "m-t-space" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12840", "repo": "kevin0571/STPopup", "url": "https://github.com/kevin0571/STPopup/issues/4" }
gharchive/issue
Swift version please! Thanks! @haducloc Alright. But have you tried bridge the STPopup.h into your swift project? @haducloc if you are using Cocoapods to manage your dependencies, it is very easy to integrate an obj-c project to your swift project. I'm using this via cocoapods in my swift project and it works great I would love an example using swift. My Obj-C is sub par...
2025-04-01T06:41:16.299636
2018-03-29T00:26:22
309578113
{ "authors": [ "mmaxim" ], "license": "BSD-3-Clause", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12841", "repo": "keybase/client", "url": "https://github.com/keybase/client/pull/11132" }
gharchive/pull-request
2025-04-01T06:41:16.352648
2021-02-21T08:13:26
812800876
{ "authors": [ "nake89" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12842", "repo": "kuka-js/kuka", "url": "https://github.com/kuka-js/kuka/pull/146" }
gharchive/pull-request
Fix/integration test :tada: This PR is included in version @kuka-js/serverless-v1.1.2 :tada: The release is available on: npm package (@latest dist-tag) GitHub release Your semantic-release bot :package::rocket:
2025-04-01T06:41:16.357593
2018-09-20T19:48:04
362339209
{ "authors": [ "coveralls", "huafu" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12843", "repo": "kulshekhar/ts-jest", "url": "https://github.com/kulshekhar/ts-jest/pull/743" }
gharchive/pull-request
23.10.1 Pull Request Test Coverage Report for Build 1874 0 of 0 changed or added relevant lines in 0 files are covered. No unchanged relevant lines lost coverage. Overall coverage remained the same at 90.771% Totals Change from base Build 1873: 0.0% Covered Lines: 973 Relevant Lines: 1022 💛 - Coveralls
2025-04-01T06:41:16.377305
2018-10-01T18:19:37
365582704
{ "authors": [ "intrepidws" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12844", "repo": "laravel-frontend-presets/tailwindcss", "url": "https://github.com/laravel-frontend-presets/tailwindcss/pull/39" }
gharchive/pull-request
Fix screenshot URL references I found a better way of doing this across all of the presets, on the main repo. Will open that shortly.
2025-04-01T06:41:16.422259
2020-04-19T23:31:58
602851343
{ "authors": [ "coveralls", "fisuda" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12845", "repo": "lets-fiware/ngsi-entity-sort-operator", "url": "https://github.com/lets-fiware/ngsi-entity-sort-operator/pull/1" }
gharchive/pull-request
Fix typos Coverage remained the same at 90.0% when pulling 6b151a536af3389a772a7f3577a9d068ccff9926 on fisuda:patch into 34821ee61413ba191ec05005411af68bff2686a0 on lets-fiware:master.
2025-04-01T06:41:16.559947
2017-03-17T12:11:49
214986098
{ "authors": [ "lucono", "seckwei" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12846", "repo": "lucono/xtypejs", "url": "https://github.com/lucono/xtypejs/pull/14" }
gharchive/pull-request
Minor typo correction Thanks @seckwei @seckwei, your fix was actually on the auto-generated file. The source files for the site are under the /app folder, from which all the files under the /site directory are autogenerated. I've applied your fix to the source file instead, with 8b0e4c8d1b723c612f9be4324f16fa0834a14f2e - thanks. Sorry, I should have noticed that! Thanks for letting me know 😄
2025-04-01T06:41:16.622562
2020-04-26T06:52:44
606938071
{ "authors": [ "codecov-io", "famoser" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12847", "repo": "mangelio/web", "url": "https://github.com/mangelio/web/pull/313" }
gharchive/pull-request
Dev Codecov Report Merging #313 into master will increase coverage by 0.00%. The diff coverage is 81.81%. @@ Coverage Diff @@ ## master #313 +/- ## ========================================= Coverage 58.44% 58.44% - Complexity 2783 2788 +5 ========================================= Files 336 336 Lines 7068 7078 +10 ========================================= + Hits 4131 4137 +6 - Misses 2937 2941 +4 Impacted Files Coverage Δ Complexity Δ src/Entity/Issue.php 81.48% <ø> (ø) 64.00 <0.00> (ø) src/Entity/Craftsman.php 92.15% <60.00%> (-3.50%) 30.00 <3.00> (+3.00) :arrow_down: src/Api/Entity/Edit/Craftsman.php 100.00% <100.00%> (ø) 10.00 <2.00> (+2.00) src/Api/Transformer/Edit/CraftsmanTransformer.php 100.00% <100.00%> (ø) 2.00 <0.00> (ø) src/Controller/Api/EditController.php 76.34% <100.00%> (-0.72%) 85.00 <0.00> (ø) 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 df9d438...4341b09. Read the comment docs.
2025-04-01T06:41:16.625468
2018-04-25T09:55:48
317557581
{ "authors": [ "manti-by" ], "license": "bsd-3-clause", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12848", "repo": "manti-by/M2-Blog-Engine", "url": "https://github.com/manti-by/M2-Blog-Engine/issues/6" }
gharchive/issue
Do not run UWSGi as root https://stackoverflow.com/questions/15140805/can-only-run-uwsgi-with-root Update filesystem access permissions for newly uploaded and converted files
2025-04-01T06:41:16.630686
2018-03-07T11:42:16
303066094
{ "authors": [ "chouhan-deepika", "ssmlee04" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12849", "repo": "marcbachmann/node-html-pdf", "url": "https://github.com/marcbachmann/node-html-pdf/issues/409" }
gharchive/issue
Can we create hyperlinks that will help us to navigate within generated pdf? Wondering if that's supported yet? Thanks.
2025-04-01T06:41:16.633489
2015-05-18T17:14:37
77732688
{ "authors": [ "Nedudi", "marcinkrysiak1979" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12850", "repo": "marcinkrysiak1979/marionette.showAnimated", "url": "https://github.com/marcinkrysiak1979/marionette.showAnimated/issues/4" }
gharchive/issue
it would be great to have bower package Good point I will create bower package soon. Hi Nedudi The bower package is now created. Install by: bower install marionette-show-animated Thanks Marcin
2025-04-01T06:41:16.635302
2015-09-15T01:11:41
106456864
{ "authors": [ "TimothyGu", "marijnh" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12851", "repo": "marijnh/acorn", "url": "https://github.com/marijnh/acorn/pull/311" }
gharchive/pull-request
Update documentation and usage of the tokenizer API What a mess. Sorry about that. Merged as 0c4d5ad and 0236aee
2025-04-01T06:41:16.729536
2019-10-23T21:38:56
511586158
{ "authors": [ "Willyfrog", "amyblais", "esethna" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12852", "repo": "mattermost/desktop", "url": "https://github.com/mattermost/desktop/pull/1089" }
gharchive/pull-request
Update CHANGELOG.md @esethna @deanwhillier @Willyfrog Can you help merge this as a priority? The desktop changelog is out of date because this PR is still open: https://github.com/mattermost/desktop/pull/1081. The desktop changelog in the documentation site is up-to-date and is easier to maintain. @deanwhillier or @Willyfrog please help merge does this need cherry picking on 4.3? @Willyfrog Honestly I don't know if it needs to be cherry-picked or not. I guess not, but I'd love confirmation. Anyways it should be easy if the need arises.
2025-04-01T06:41:16.738372
2016-03-26T20:51:38
143748466
{ "authors": [ "eddieantonio", "mdfeist" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12853", "repo": "mdfeist/TypeV", "url": "https://github.com/mdfeist/TypeV/issues/3" }
gharchive/issue
Get Class Edited in Each Commit @mdfeist Could you describe this issue a bit more? Right now I don't exactly understand what you mean. In fact, if it's what I think, this is already done (in JavaScript, you can iterate by commit, and get the set of types changed in the commit). Example: class MyClass { private MyType t; } Currently: #DECLARE | INSERT | MyType | 1 What I want: #DECLARE | INSERT | MyType | 1 #EDIT | INSERT | MyClass | 1 Therefore we can see what types are being changed. (Their functionality is changing.) So currently we are only looking at external changes (declarations and invocations). But I want to add internal changes as well. Ah, yes! This is very important!
2025-04-01T06:41:16.745187
2018-04-25T20:25:42
317780912
{ "authors": [ "coveralls", "omenocal" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12854", "repo": "mediarain/voxa-dashbot", "url": "https://github.com/mediarain/voxa-dashbot/pull/7" }
gharchive/pull-request
Changed to v0.1.7 Pull Request Test Coverage Report for Build 12 0 of 0 (NaN%) changed or added relevant lines in 0 files are covered. 6 unchanged lines in 1 file lost coverage. Overall coverage decreased (-21.4%) to 78.571% Files with Coverage Reduction New Missed Lines % lib/Voxa-Dashbot.js 6 78.57% Totals Change from base Build 10: -21.4% Covered Lines: 18 Relevant Lines: 24 💛 - Coveralls
2025-04-01T06:41:16.751359
2020-04-14T11:14:35
599490771
{ "authors": [ "MengHao666", "canuck6ix", "melonwan" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12855", "repo": "melonwan/sphereHand", "url": "https://github.com/melonwan/sphereHand/issues/4" }
gharchive/issue
Code is never coming??? Could you publish your code please? apologies for the delay. The paper was published right before my graduation, and I failed to publish the code before leaving the University. After that, I didn't manage to buy a PC and clean the code to publish. Sorry for the procrastination. I'll buy the PC and publish the code this summer.
2025-04-01T06:41:16.753455
2017-06-29T16:58:22
239547088
{ "authors": [ "meness", "the94air" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12856", "repo": "meness/verifi", "url": "https://github.com/meness/verifi/pull/11" }
gharchive/pull-request
Publishing the config file fix @the94air Thank you so much, I didn't know that 👍 No problem :+1:
2025-04-01T06:41:16.755735
2017-03-06T15:38:17
212159375
{ "authors": [ "gabrielhartmann", "keithchambers", "mbdcos" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12857", "repo": "mesosphere/dcos-commons", "url": "https://github.com/mesosphere/dcos-commons/pull/654" }
gharchive/pull-request
[KAFKA][KAF-297] Disable JMX and STATSD options Do not merge this yet. LGTM @mbdcos: Please resolve conflicts and merge. retest this please retest this please
2025-04-01T06:41:16.756500
2020-09-14T17:29:03
701290013
{ "authors": [ "richardgoater" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12858", "repo": "metaspace2020/metaspace", "url": "https://github.com/metaspace2020/metaspace/pull/679" }
gharchive/pull-request
Multiple ion images @LachlanStuart this is "ready" except for one test, don't think the filter is changing when I want it to. Will look into it more tomorrow.
2025-04-01T06:41:16.766627
2020-02-28T06:25:39
572547513
{ "authors": [ "bishal-pdMSFT", "issacnitin" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12859", "repo": "microsoft/azure-pipelines-tasks", "url": "https://github.com/microsoft/azure-pipelines-tasks/pull/12447" }
gharchive/pull-request
Updating ARG tasks with typed-rest-client 1.7.2 All the APIs we use from older typed-rest-client is intact with the newer version. I think you will need to bump up version of all tasks which depend on azure-arm-rest-v2
2025-04-01T06:41:16.768056
2020-07-02T19:34:33
650160508
{ "authors": [ "lilgreenbird", "peterbae" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12860", "repo": "microsoft/mssql-jdbc", "url": "https://github.com/microsoft/mssql-jdbc/pull/1379" }
gharchive/pull-request
Fixed javadoc issue Actually I just realized that this issue will be fixed in your PR 1373, so either remove this PR or remove it from the other PR, otherwise there will be a conflict.
2025-04-01T06:41:16.772625
2021-04-22T07:09:04
864618860
{ "authors": [ "zhengchang907" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12861", "repo": "microsoft/weblogic-on-azure", "url": "https://github.com/microsoft/weblogic-on-azure/pull/6" }
gharchive/pull-request
Update Application Gateway doc to use auto-generated self-assigned certificate @edburns Hi Ed, can you please review this PR? I removed the certificate and update to documentation to let users use auto-generated self-assigned cert.
2025-04-01T06:41:17.033312
2015-09-27T01:03:07
108505545
{ "authors": [ "citricsquid" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12862", "repo": "minecraftforum/widget", "url": "https://github.com/minecraftforum/widget/issues/35" }
gharchive/issue
Rewrite stylesheets as SASS and add asset compilation with GULP Resolved https://github.com/minecraftforum/widget/commit/a90ef6b808e29ecdc8c913b169762c61495458d2
2025-04-01T06:41:17.050103
2017-08-31T08:32:31
254256514
{ "authors": [ "cht8687", "xcatliu" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12863", "repo": "mobi-css/mobi.css", "url": "https://github.com/mobi-css/mobi.css/pull/89" }
gharchive/pull-request
Use lerna and cssnext Travis failed: $ rake rake aborted! No Rakefile found (looking for: rakefile, Rakefile, rakefile.rb, Rakefile.rb) @xcatliu @cht8687 Please ignore this. The new version doesn't require ruby or sass anymore. @cht8687 Fixed travis issue.
2025-04-01T06:41:17.053787
2017-08-16T13:25:48
250626075
{ "authors": [ "chris-erickson", "ovangle" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12864", "repo": "moment/moment", "url": "https://github.com/moment/moment/pull/4123" }
gharchive/pull-request
Re-enable disabled tests Thanks for the PR! But the tests fail when --strictNullChecks is passed as a compiler option. It's not enabled by default in the project's tsconfig.json, because officially we only support an older version of typescript, but the tests were commented out for this reason. We're also rewriting the typescript declarations for 3.0 from scratch in #4121, so...
2025-04-01T06:41:17.078929
2015-03-23T18:05:03
63791836
{ "authors": [ "akatsoulas", "coveralls", "johngian" ], "license": "bsd-3-clause", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12865", "repo": "mozilla/mozillians", "url": "https://github.com/mozilla/mozillians/pull/1184" }
gharchive/pull-request
Upgrade Django to version 1.6.11 @akatsoulas or @comzeradd r? Coverage remained the same at 93.21% when pulling 4ab24c62647fa26eb2e44c1eacf9e0e6068dcd71 on johngian:playdoh-upgrade into c69ef80e2e50d296e0c649fb9c1f86fdf17dedd6 on mozilla:master. Coverage remained the same at 93.21% when pulling 4ab24c62647fa26eb2e44c1eacf9e0e6068dcd71 on johngian:playdoh-upgrade into c69ef80e2e50d296e0c649fb9c1f86fdf17dedd6 on mozilla:master. Tested locally and works fine r+ :coffee:
2025-04-01T06:41:17.086342
2021-01-14T09:53:10
785845501
{ "authors": [ "coveralls", "ladisalves" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12866", "repo": "mpneuried/rsmq-worker", "url": "https://github.com/mpneuried/rsmq-worker/pull/53" }
gharchive/pull-request
feat: package update Coverage increased (+0.08%) to 82.34% when pulling 8b0634307c3e56444f21cde453df5bc28a750aaf on ladisalves:feat/update into f5320847ba0324759ac43876224efb83c720be52 on mpneuried:master. Coverage increased (+0.08%) to 82.34% when pulling 8b0634307c3e56444f21cde453df5bc28a750aaf on ladisalves:feat/update into f5320847ba0324759ac43876224efb83c720be52 on mpneuried:master. I make a forked package on https://www.npmjs.com/package/rsmq-worker-ls with updated dependencies. I don't want to fight with outdated CI in this repo... I make a forked package on https://www.npmjs.com/package/rsmq-worker-ls with updated dependencies. I don't want to fight with outdated CI in this repo...
2025-04-01T06:41:17.088416
2018-07-23T05:42:40
343494509
{ "authors": [ "jslay88" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12867", "repo": "mraza007/videodownloader", "url": "https://github.com/mraza007/videodownloader/pull/3" }
gharchive/pull-request
Added support for argparse, so one could cronjob the script Probably want to squash and merge, as there were a couple issues with the filename_prefix not working on pytube.
2025-04-01T06:41:17.099093
2019-02-08T06:28:59
408023620
{ "authors": [ "MikeWasson", "VSC-Service-Account" ], "license": "CC-BY-4.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12868", "repo": "mspnp/architecture-center", "url": "https://github.com/mspnp/architecture-center/pull/1273" }
gharchive/pull-request
Update metadata OPS Build status updates of commit d6b09af: :clock10: Incremental building: average incremental building time is 23 sec(s) OPS Build status updates of commit d6b09af: :warning: Validation status: warnings File Status Preview URL Details building-blocks/extending-templates/toc.json :warning:Warning Details toc.json :warning:Warning Details bread/toc.json :warning:Warning Details docs/reference-architectures/ai/batch-scoring-databricks.md :white_check_mark:Succeeded View building-blocks/extending-templates/toc.json [Warning] Error happen when converting building-blocks/extending-templates/toc.json to Pdf. Details: Could not find a part of the path 'T:\bgvm\building-blocks\extending-templates\toc.json'. toc.json [Warning] Error happen when converting toc.json to Pdf. Details: Could not find file 'T:\bgvm\toc.json'. bread/toc.json [Warning] Error happen when converting bread/toc.json to Pdf. Details: Could not find a part of the path 'T:\bgvm\bread\toc.json'. For more details, please refer to the build report. Note: If you changed an existing file name or deleted a file, broken links in other files to the deleted or renamed file are listed only in the full build report.
2025-04-01T06:41:17.127037
2018-10-27T11:54:07
374637374
{ "authors": [ "xyz-neory" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12869", "repo": "my-first-pr/hacktoberfest-2018", "url": "https://github.com/my-first-pr/hacktoberfest-2018/pull/980" }
gharchive/pull-request
Added my name to README.md I can't resolve the conflict without creating a new pull request, i hope this doesn't matter i am just leaving it as it is.
2025-04-01T06:41:17.128603
2019-04-07T17:43:33
430172511
{ "authors": [ "coveralls", "jeffgbutler" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12870", "repo": "mybatis/mybatis-dynamic-sql", "url": "https://github.com/mybatis/mybatis-dynamic-sql/pull/80" }
gharchive/pull-request
Update dependencies, code polishing Coverage remained the same at 100.0% when pulling e29ce8f36975f26777a97e5c5fb0ffab218606d4 on jeffgbutler:master into 6b16765622cf2f21fd4a7dcdcee1da27ebf02d84 on mybatis:master.
2025-04-01T06:41:17.131206
2014-10-29T11:02:03
47134642
{ "authors": [ "corford", "dtarda", "mickaelandrieu", "paazmaya" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12871", "repo": "n1k0/casperjs", "url": "https://github.com/n1k0/casperjs/pull/1071" }
gharchive/pull-request
Ensure headers["Content-Type"] exists before comparing string. Might be, with a broken server or with a mockup data... We were getting "TypeError: undefined is not an object (evaluating 'settings.headers["Content-Type"].match')" when POSTing to a certain site (using the master branch of Casper and phantomjs-2.0.1). I know phantomjs2 isn't officially supported yet but FYI this patch fixed the issue. Thanks ;)
2025-04-01T06:41:17.133220
2021-02-05T11:40:23
802096400
{ "authors": [ "mortenlj" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12872", "repo": "nais/kafkarator", "url": "https://github.com/nais/kafkarator/pull/12" }
gharchive/pull-request
Add rest-api address to secret Turns out the REST API does not support detailed ACLs, so we will disable it again. Further research needed.
2025-04-01T06:41:17.154485
2019-08-23T05:44:47
484341536
{ "authors": [ "codecov-io", "jaehong-kim" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12873", "repo": "naver/pinpoint", "url": "https://github.com/naver/pinpoint/pull/5913" }
gharchive/pull-request
[#noissue] Fix MethodDescriptor of GrpcStreamListener Codecov Report Merging #5913 into master will increase coverage by <.01%. The diff coverage is 0%. @@ Coverage Diff @@ ## master #5913 +/- ## ========================================== + Coverage 40.91% 40.91% +<.01% ========================================== Files 2951 2951 Lines 88802 88802 Branches 11693 11693 ========================================== + Hits 36329 36330 +1 Misses 49313 49313 + Partials 3160 3159 -1 Impacted Files Coverage Δ ...vercorp/pinpoint/plugin/grpc/GrpcClientPlugin.java 0% <ø> (ø) :arrow_up: ...ptor/GrpcStreamListenerResultMethodDescriptor.java 0% <0%> (ø) :arrow_up: ...nitor/DefaultDataSourceMonitorRegistryService.java 77.77% <0%> (-11.12%) :arrow_down: ...rc/main/java/com/navercorp/pinpoint/test/Item.java 66.66% <0%> (-3.34%) :arrow_down: ...navercorp/pinpoint/rpc/common/SocketStateCode.java 82.92% <0%> (-2.44%) :arrow_down: ...cluster/zookeeper/ZookeeperClusterDataManager.java 52.9% <0%> (-1.94%) :arrow_down: ...java/com/navercorp/pinpoint/rpc/DefaultFuture.java 74.16% <0%> (+1.66%) :arrow_up: ...orp/pinpoint/rpc/server/DefaultPinpointServer.java 80.98% <0%> (+1.76%) :arrow_up: ...ofiler/sender/StandbySpanStreamDataSendWorker.java 83.58% <0%> (+2.98%) :arrow_up: ...vercorp/pinpoint/rpc/packet/ClientClosePacket.java 66.66% <0%> (+11.11%) :arrow_up: 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 f9e131b...2d63d31. Read the comment docs.
2025-04-01T06:41:17.158162
2019-10-28T02:22:31
513063751
{ "authors": [ "F-loat", "daniel-dx" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12874", "repo": "ncform/ncform", "url": "https://github.com/ncform/ncform/pull/111" }
gharchive/pull-request
feat(ncform-common): add the keepEmpty parameter to support empty array You can refer to the showOneIfEmpty configuration in all the array type widgets OK. Sorry for not reading the document carefully.
2025-04-01T06:41:17.171353
2017-06-28T14:17:20
239179078
{ "authors": [ "nenko-tabakov" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12875", "repo": "nenko-tabakov/dev-tools", "url": "https://github.com/nenko-tabakov/dev-tools/issues/1" }
gharchive/issue
Converting class with field that have default values results in broken code The following class public class Sample { private int initialized = 1; private Boolean enabled = Boolean.TRUE; private String name; } is now converted to: public class Sample { private final int initialized; private final Boolean enabled; private final String name; public Sample(int initialized, Boolean enabled, String name) { this.initialized = initialized; if (enabled != null) this.enabled = enabled; else this.enabled = Boolean.TRUE; this.name = name; } public int getInitialized() { return initialized; } public Boolean getEnabled() { return enabled; } public String getName() { return name; } }
2025-04-01T06:41:17.196804
2021-01-07T18:06:45
781505597
{ "authors": [ "nickeyzindgi" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12876", "repo": "nickeyzindgi/github-actions-for-ci", "url": "https://github.com/nickeyzindgi/github-actions-for-ci/pull/2" }
gharchive/pull-request
CI for Node Initializes with two players Starts the game with a random player Sets the current player to be whoever it is not Initializes with two players Starts the game with a random player Sets the current player to be whoever it is not
2025-04-01T06:41:17.220668
2016-10-04T06:28:51
180814299
{ "authors": [ "nolanlawson", "sergejmueller", "vigneshshanmugam" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12877", "repo": "nolanlawson/optimize-js", "url": "https://github.com/nolanlawson/optimize-js/pull/27" }
gharchive/pull-request
Add area for 3rd party plugins Let's add the Gulp one too: https://github.com/prateekbh/gulp-optimize-js Thanks for the PR; merging! :( @nolanlawson https://github.com/vigneshshanmugam/optimize-js-plugin @vigneshshanmugam PRs welcome! :)
2025-04-01T06:41:17.260999
2015-05-24T01:50:11
80015132
{ "authors": [ "Alaya-in-Matrix", "octol" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12878", "repo": "octol/vim-cpp-enhanced-highlight", "url": "https://github.com/octol/vim-cpp-enhanced-highlight/issues/11" }
gharchive/issue
Why is the color of "std::cout" and "std::endl" different? Although usually used together, they are fundamentally different beasts. One is a global object and the other one is a IO manipulator (a function).
2025-04-01T06:41:17.271432
2020-04-23T11:36:55
605461428
{ "authors": [ "mandy-chessell", "planetf1" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12879", "repo": "odpi/egeria", "url": "https://github.com/odpi/egeria/pull/2966" }
gharchive/pull-request
Use correct types for CSVFile properties PR build is failing in Data Platform OMAS API though I appeared to build ok.... CTS(inmem) failed with 16 errors. See https://ibm.box.com/s/7mqfgcbhe18lbbwfjrm9cb4b15d52v5h The errors are EntityNotKnown exceptions in reference copy tests Will repeat without this PR to see if it's a PR_specific regression, or a current master issue - in which case we may as well merge anyway Prior to this PR the CTS (inmem) passes with 0 failures. Retesting with PR again. Something has gone odd in the github - merge . When the PR was initially opened it built ok. I was also able to do this locally, and experienced the cts failures. Later in the day I clicked on 'This branch is out-of-date with the base branch' as we often do -- the result was a failed build I pulled down the PR manually and merged with master. No conflicts detected but the build result is the same locally as seen on the build system - problems in the data platform OMAS. It looks as if something could be a bit odd with the git history or prior merges. @mandy-chessell suggest you pull or rebase from master and test locally, then push/force-push. One of us should then retest the CTS - in case the previous failure was a 1-off or an artifact of merge oddities. Ok. I think what's happened is that commit 8e2627630ca196141beb403faef301e5bc1c75eb made some 'in-flight' changes ie to DataPlatformProperties.java which were pushed to origin, and thus into this PR -- before the PR had been merged, and this commit contains an incomplete (work in progress change) and isn't buildable. Need a stable branch for the PR - an option is to create a new branch and then cherry-pick the selected changes in needed for this fix. once any upstream branch contains the fix I'll cherry-pick for the release I think we need to separate the PR from the specific commit needed to go into the 1.7 branch. On this PR - I thought it was merged yesterday and I have continued with the work on the Data Platform OMAS fix, not realising that my intermittant changes were going into the open PR. I will get it into a buildable state again sometime today so it can be merged. It should be possible to just take the commit for 1.7 and test it on the branch. The specific commit for 1.7 on its own should not break the CTS. The change to OMRS fixed the formatting of an error message. The change to the type of the properties was in the OMAS layer which is not tested by the CTS. 1.7 PR created just from the csv fix commit. One option - If you need to 'freeze' for a PR, whilst concurrently continuing development at pace :-) , you can create a branch from your development branch, and use that for the PR. cherry-picks can be added for any fixes found during review and added back to the PR branch from dev branch. This allows for review time without blocking dev work. Just one possibility of so many with git! I ran a test of CTS (inmem) after merge - all runs cleanly. Unsure why the early test failed - perhaps due to work-in-progress or a glitch.
2025-04-01T06:41:17.318472
2021-04-15T08:40:40
858644384
{ "authors": [ "Kyuuki93", "LXXXXR" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12880", "repo": "open-mmlab/mmclassification", "url": "https://github.com/open-mmlab/mmclassification/issues/215" }
gharchive/issue
How to train a custom multi-label dataset? Hi @Kyuuki93 , First, you need to convert your dataset to existing dataset format, e.g. VOC format or implement a new Dataset class inherited from BaseDataset. Pls refer to Tutorial 2: Adding New Dataset for more info. Then, you can write a config as config for VGG. Hi @Kyuuki93 , First, you need to convert your dataset to existing dataset format, e.g. VOC format or implement a new Dataset class inherited from BaseDataset. Pls refer to Tutorial 2: Adding New Dataset for more info. Then, you can write a config as config for VGG. Thanks for ur quick reply, I have prepared a toy FileList dataset , the anno_file looks like 001.jpg 0 002.jpg 1 003.jpg 0 1 And then created a TestDataset which modified from mmcls\datasets\multi_label.py, use mobilenet as backbone, then I got TypeError: ImageClassifier: MultiLabelClsHead: __init__() got an unexpeted keywork argument 'num_classes' TypeError: ImageClassifier: MultiLabelClsHead: __init__() got an unexpeted keywork argument 'in_channels' Does multi-label task only support vgg for now? For your dataset, it seems like imagenet format, which is not supported for multi-label classification task. You might need to convert it to VOC format. For VGG backbone, we put the final fc layer inside the backbone rather than put it in the head. So MultiLabelClsHead does not expect argument num_classes and in_channels. If you want use other backbones, you might need to use MultiLabelLinearClsHead. Thank you.
2025-04-01T06:41:17.330330
2017-09-29T19:43:41
261750629
{ "authors": [ "danamlewis", "parenthetic" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12881", "repo": "openaps/docs", "url": "https://github.com/openaps/docs/pull/1060" }
gharchive/pull-request
Putting image instead of chart since won't render in RTD @parenthetic just FYI your beautiful chart works in markdown, but not ReadTheDocs. So, I put the raw file of the chart as a file in this folder so we can update if needed, and just stuck an image in that section so people could see it. Bummer. Good catch and fix. On Fri, Sep 29, 2017 at 3:51 PM, Dana Lewis<EMAIL_ADDRESS>wrote: @parenthetic https://github.com/parenthetic just FYI your beautiful chart works in markdown, but not ReadTheDocs. So, I put the raw file of the chart as a file in this folder so we can update if needed https://github.com/openaps/docs/pull/1060/commits/72619d15914ab1edb5ffa96c0d221535ee92c4ef, and just stuck an image in that section so people could see it. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/openaps/docs/pull/1060#issuecomment-333222655, or mute the thread https://github.com/notifications/unsubscribe-auth/AUVFUZk2WiT6rodlRMizltbpjjxdzfrBks5snUpIgaJpZM4PpJVA .
2025-04-01T06:41:17.332148
2019-07-25T07:47:26
472715623
{ "authors": [ "adathatri", "gkeishin" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12882", "repo": "openbmc/openbmc-test-automation", "url": "https://github.com/openbmc/openbmc-test-automation/issues/1900" }
gharchive/issue
Fix IPMI Test failure https://github.com/openbmc/openbmc-test-automation/issues/1881, https://github.com/openbmc/openbmc-test-automation/issues/1882, https://github.com/openbmc/openbmc-test-automation/issues/1883 There is #1882 tracking.. 2 are fixed. closing this one.
2025-04-01T06:41:17.353880
2021-04-29T23:02:35
871640461
{ "authors": [ "brianhlin", "matyasselmeci" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12883", "repo": "opensciencegrid/docker-compute-entrypoint", "url": "https://github.com/opensciencegrid/docker-compute-entrypoint/pull/15" }
gharchive/pull-request
Add "name" labels so we know what the images are supposed to be tagged as I'm not sure I understand this. Do you build the images locally without a tag then want to verify which registry repo to push it to after the fact? When you do a local build of a multi-image Dockerfile, you can't specify which image to tag as what -- -t only applies to the last image. Then you need to docker tag the other images by hand which means you need to docker inspect the images to find out which hash is which. The label makes it easier to read the inspection -- otherwise you'd have to figure it out from the steps or something.
2025-04-01T06:41:17.409958
2017-11-13T09:09:52
273356151
{ "authors": [ "adriancole" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12884", "repo": "openzipkin/docker-zipkin", "url": "https://github.com/openzipkin/docker-zipkin/pull/159" }
gharchive/pull-request
Enables UDF in cassandra and adds a few starter functions merging this, but happy to change it later if anyone wants to. It is a demo image we are affecting.
2025-04-01T06:41:17.416237
2019-09-19T20:51:23
496020322
{ "authors": [ "doxiao", "markxnelson" ], "license": "UPL-1.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12885", "repo": "oracle/weblogic-kubernetes-operator", "url": "https://github.com/oracle/weblogic-kubernetes-operator/pull/1264" }
gharchive/pull-request
WIP OWLS-73891 part3 - add FAQ entry for managing namespaces Ryan, please check if the contents match what is in your mind. Mark, how do you create the timestamp for a doc? I just hand-crafted it, which I am not sure is the right way of doing it. Thanks, Tom, for reviewing the doc. Those are all great comments. I'll add links from the introduction area to the individual sections, and add the section for getting the current value of domainNamespaces. I also feel that the content here fit better in the user guide, instead of the FAQ section. also merged to master in 6624688d4
2025-04-01T06:41:17.419806
2019-03-22T04:37:13
424043630
{ "authors": [ "jiegec" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12886", "repo": "oscourse-tsinghua/rcore_plus", "url": "https://github.com/oscourse-tsinghua/rcore_plus/issues/31" }
gharchive/issue
Rust user programs cannot run in release mode After reading assembly and some searching, it is fixed in https://github.com/rcore-os/rcore-user/commit/a8497fb692ca9e9ac1717ebeb05f73508d9d6ca3. Reason: syscall clobbers rcx and r11 registers. Tell compiler about this. Ref: https://stackoverflow.com/questions/50571275/why-does-a-syscall-clobber-rcx
2025-04-01T06:41:17.441218
2016-03-24T17:12:52
143305857
{ "authors": [ "bmccarrie", "owntheweb" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12887", "repo": "owntheweb/VR-Solar-System-Tour", "url": "https://github.com/owntheweb/VR-Solar-System-Tour/pull/9" }
gharchive/pull-request
uploaded planet images for navigation Thanks Brian! I have the start of the ship nav menu happening now. I'll pull these graphics in immediately.
2025-04-01T06:41:17.454205
2020-09-03T09:25:05
691802340
{ "authors": [ "TomasTomecek", "dhodovsk" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12888", "repo": "packit/dist-git-to-source-git", "url": "https://github.com/packit/dist-git-to-source-git/pull/44" }
gharchive/pull-request
copy only sources defined in specfile I just discovered one issue: rebase-helper.get_source returns also the tarball: " using a temp directory.".format(parent)) > /usr/local/lib/python3.6/site-packages/dist2src/core.py(364)copy_all_sources() 363 import ipdb; ipdb.set_trace() --> 364 for file_ in self.dist_git_spec.get_sources(): 365 file_dest = sg_path / Path(file_).name ipdb> self.dist_git_spec.get_sources() ['/d/dnf/SOURCES/dnf-4.2.23.tar.gz'] I am debating whether this is a problem or not. But since we are not interacting with lookaside cache in source-git, I'd say this is a good thing that we store the tarball alongside.
2025-04-01T06:41:17.464099
2016-06-15T21:23:00
160527180
{ "authors": [ "ArchRoller", "ThiefMaster" ], "license": "BSD-3-Clause", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12889", "repo": "pallets/flask", "url": "https://github.com/pallets/flask/pull/1914" }
gharchive/pull-request
Fix problems with examples and CLI #1902 Please don't use master in your repos for PRs, and if you do rebase instead of doing merge commits. Otherwise you quickly get a mess that'd propagate into the upstream repo if not rebased/squashed before merging.
2025-04-01T06:41:17.473662
2016-10-15T05:05:44
183189250
{ "authors": [ "coveralls", "sinhrks" ], "license": "bsd-3-clause", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12890", "repo": "pandas-ml/pandas-ml", "url": "https://github.com/pandas-ml/pandas-ml/pull/86" }
gharchive/pull-request
TST: Add sklearn.multioutput Coverage increased (+0.04%) to 94.428% when pulling eebb48ede1ee35c0c995bb675aa0180b8fd9a4a7 on sinhrks:multioutput into 16c7c4d0be4657dbf366476967e6600db1c0e649 on pandas-ml:master. Coverage increased (+0.04%) to 94.428% when pulling 86eb52754cb4ef044327fc7bc1d1d9ba5e87798d on sinhrks:multioutput into 16c7c4d0be4657dbf366476967e6600db1c0e649 on pandas-ml:master. Coverage increased (+0.04%) to 94.428% when pulling 67d0349a67619cfd1becadd9dd37a2f23e7845a2 on sinhrks:multioutput into 16c7c4d0be4657dbf366476967e6600db1c0e649 on pandas-ml:master. Coverage increased (+0.02%) to 94.414% when pulling 24a426952669eb005f37bbb13e3dc82bcabd2d28 on sinhrks:multioutput into 16c7c4d0be4657dbf366476967e6600db1c0e649 on pandas-ml:master. Coverage decreased (-0.06%) to 94.33% when pulling 42344e7bd9b797e304d1339cc90263be33a89e76 on sinhrks:multioutput into 16c7c4d0be4657dbf366476967e6600db1c0e649 on pandas-ml:master. Coverage decreased (-0.09%) to 94.303% when pulling 689891bf1819f0ae6a1e807209797cf3fb84ec71 on sinhrks:multioutput into 16c7c4d0be4657dbf366476967e6600db1c0e649 on pandas-ml:master.
2025-04-01T06:41:17.482506
2017-12-27T18:44:05
284775505
{ "authors": [ "paragonie-scott" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12891", "repo": "paragonie/easydb", "url": "https://github.com/paragonie/easydb/pull/65" }
gharchive/pull-request
Fix boolean handling for SQLite Currently, this causes problems on SQLite: $db->update( 'table', ['col2' => true], ['col2' => false, 'col1' => 'test'] ); SQLite doesn't have TRUE/FALSE, it only has 1/0.
2025-04-01T06:41:17.503541
2016-04-19T05:17:26
149350381
{ "authors": [ "codecov-io", "pbremer" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12892", "repo": "pbremer/ice-cream-manager", "url": "https://github.com/pbremer/ice-cream-manager/pull/41" }
gharchive/pull-request
added first batch step Current coverage is 39.35% Merging #41 into master will increase coverage by +7.34% as of 62277dd @@ master #41 diff @@ ====================================== Files 93 100 +7 Stmts 1062 1108 +46 Branches 73 82 +9 Methods ====================================== + Hit 340 436 +96 - Partial 9 15 +6 + Missed 713 657 -56 Review entire Coverage Diff as of 62277dd Powered by Codecov. Updated on successful CI builds.
2025-04-01T06:41:17.504675
2019-07-10T21:35:42
466533856
{ "authors": [ "Jason-Cooke", "serene" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12893", "repo": "pdfkit/pdfkit", "url": "https://github.com/pdfkit/pdfkit/pull/444" }
gharchive/pull-request
docs: fix typo @Jason-Cooke Update from master for CI? @serene Sorry for the late reply. I don't quite follow?
2025-04-01T06:41:17.559194
2020-04-26T15:21:52
607037673
{ "authors": [ "at-wat" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12894", "repo": "pion/mediadevices", "url": "https://github.com/pion/mediadevices/pull/153" }
gharchive/pull-request
Fix go.sum PR generated by Renovate bot sometimes lack package sum entries in go.sum. https://github.com/pion/.goassets/issues/14 This fixes go.sum of the current master branch and also add a GitHub Action to automatically fix future problem. oh, I have to fix to ignore examples/go.mod.
2025-04-01T06:41:17.565908
2017-03-16T12:26:39
214687969
{ "authors": [ "codecov-io", "pjan" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12895", "repo": "pjan/akka-d3", "url": "https://github.com/pjan/akka-d3/pull/11" }
gharchive/pull-request
updates dependencies Codecov Report Merging #11 into master will not change coverage. The diff coverage is n/a. @@ Coverage Diff @@ ## master #11 +/- ## ==================================== Coverage 0% 0% ==================================== Files 12 12 Lines 241 241 Branches 6 6 ==================================== Misses 241 241 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 39e98fd...6a22043. Read the comment docs.
2025-04-01T06:41:17.572280
2018-02-13T20:04:20
296867602
{ "authors": [ "coveralls", "robgietema" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12896", "repo": "plone/plone-react", "url": "https://github.com/plone/plone-react/pull/75" }
gharchive/pull-request
Upgrade to node 8 Coverage increased (+0.06%) to 58.272% when pulling 6d04600218b338febc9d5d70f7a5c036126aa3ee on upgrade-node8 into 17e7ff5e1b15a86f2ea9813b1441e785c75cffa9 on master.
2025-04-01T06:41:17.581386
2016-12-04T03:14:13
193328741
{ "authors": [ "chrisfilo" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12897", "repo": "poldracklab/fmriprep", "url": "https://github.com/poldracklab/fmriprep/pull/210" }
gharchive/pull-request
[RTM] remove unnecessary fslroi I have verified that the two files have the same affine matrices. I'll merge it because I'm about to propose some changes that build on top of this one.
2025-04-01T06:41:17.646430
2021-03-31T16:18:17
846924402
{ "authors": [ "alastairreid" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12898", "repo": "project-oak/rust-verification-tools", "url": "https://github.com/project-oak/rust-verification-tools/pull/125" }
gharchive/pull-request
Fix sse2 problem I forgot to run the regression tests before pushing this. Unfortunately, it fails almost all regression tests because it causes linking errors. I will need to revisit this and try again.
2025-04-01T06:41:17.661720
2017-11-17T16:19:27
274918122
{ "authors": [ "evgenyrodionov", "peterjwest", "tuckbick" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12899", "repo": "ptshih/images-require-hook", "url": "https://github.com/ptshih/images-require-hook/pull/2" }
gharchive/pull-request
Update README.md oh please merge it ASAP @ptshih I've just realised my PR is a duplicate, I'll leave it open for now, so hopefully one of them makes it in 🤔
2025-04-01T06:41:17.662909
2020-11-02T10:35:32
734390176
{ "authors": [ "battaodineucothe", "shanev" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12900", "repo": "public-awesome/stakebird-testnets", "url": "https://github.com/public-awesome/stakebird-testnets/pull/17" }
gharchive/pull-request
Create gentx-e2d961c91379f6a09869a731c8022e0795ba1866.json Thanks for the submission, but Github accounts have to be at least a year old to participate: https://github.com/public-awesome/stakebird-testnets#submit-genesis-transaction. This is to prevent Sybil attacks and ensure a fair and decentralized launch.
2025-04-01T06:41:17.668636
2017-03-02T20:33:55
211509092
{ "authors": [ "mullr", "puppetcla", "puppetlabs-jenkins", "rbrw" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12901", "repo": "puppetlabs/puppetdb", "url": "https://github.com/puppetlabs/puppetdb/pull/2183" }
gharchive/pull-request
(PDB-3298) Don't carry fact values as valuemaps Test PASSed CLA signed by all contributors. Test PASSed Test PASSed Test PASSed Test PASSed Test PASSed travis is effectively green, merging.
2025-04-01T06:41:17.670474
2015-05-12T22:41:09
75765655
{ "authors": [ "MikaelSmith", "branan" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12902", "repo": "puppetlabs/puppetlabs-agent_upgrade", "url": "https://github.com/puppetlabs/puppetlabs-agent_upgrade/pull/3" }
gharchive/pull-request
(PUP-4557) Extend unit testing for 100% coverage This conflicts with the other spec pull. I probably should've merged this one first, since it's bigger. Oh well. This is ready as soon as it gets a rebase, I think Rebased.
2025-04-01T06:41:17.675797
2019-06-28T14:57:37
462075504
{ "authors": [ "callum-oakley" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12903", "repo": "pusher/chatkit-client-js", "url": "https://github.com/pusher/chatkit-client-js/pull/185" }
gharchive/pull-request
upgrade to v6 for create room with supplied ID Answered offline, but for posterity: just committed to mimir, and yeah let's merge all at the same time.
2025-04-01T06:41:17.697736
2015-05-31T10:54:18
83062465
{ "authors": [ "coveralls", "emmanuel-keller" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12904", "repo": "qwazr/QWAZR", "url": "https://github.com/qwazr/QWAZR/pull/36" }
gharchive/pull-request
Work on packaging Changes Unknown when pulling adf086e51c49f2897eeb30d3dedc3b13ec7439b6 on emmanuel-keller:master into ** on qwazr:master**.
2025-04-01T06:41:17.717134
2015-04-14T13:37:06
68377588
{ "authors": [ "rfecher" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12905", "repo": "radiantbluetechnologies/ossim-geotools", "url": "https://github.com/radiantbluetechnologies/ossim-geotools/issues/86" }
gharchive/issue
configurable usage of geowave for tile cache submitting a pull request for an initial level of integration... I made sure that everything is wired as it was, to wire in geowave support, modify line 175 of TileCacheHibernate.groovy to swap bean definitions This initial cut is meant to be 0 impact...the reality is you will get nothing from this the way it is currently being used (what you may get is the ability to use GeoServer for a different usage) - it just introduces slight inefficiency for the current usage... because PostGIS is relied on for all the indexing, and the current spatial query is tied to PostGIS to bring back a series of keys, the "0 impact" way I tied in GeoWave was to take the bounds from PostGIS results and use it to grab the tiles - kinda counter-intuitive, but to do more would probably impact the current query logic Because GeoWave is the spatial index on the raster data, the way it is supposed to be used is to write data with a spatial extent and query the data with a spatial extent. this initial cut will query postGIS, bring back results with the bounds and use each tile's bounds to then get the data. In particular when a user wants to query spatially, postGIS would typically be unnecessary with GeoWave - just go straight to geowave with the query geometry ...so yeah, you could put the data in geowave, add geoserver, and try out geotools' build-in support for geopackage (http://docs.geoserver.org/latest/en/user/community/geopkg/output.html) ...or not...
2025-04-01T06:41:17.720113
2019-05-03T09:03:33
439952862
{ "authors": [ "karlb", "pirapira" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12906", "repo": "raiden-network/raiden-contracts", "url": "https://github.com/raiden-network/raiden-contracts/issues/918" }
gharchive/issue
enable mypy strict mode Try enabling checks one by one: https://github.com/raiden-network/raiden-services/blob/master/setup.cfg#L29 In --strict, all generic types require type parameters. Also mypy --strict requires annotating many classes and functions in Web3.py. The --strict flag is extremely strict (at the moment it's 1150 errors). It's probably not worth making the code adhere to it fully, but we can enable a few more checks in the setup.cfg. We're currently missing # disallow_any_generics = True # disallow_subclassing_any = True # disallow_untyped_calls = True # disallow_incomplete_defs = True # disallow_untyped_decorators = True # no_implicit_optional = True # warn_redundant_casts = True # warn_return_any = True # no_implicit_reexport = True
2025-04-01T06:41:17.728933
2021-01-07T18:56:20
781533790
{ "authors": [ "StrongMonkey", "boxboatmatt", "srbala" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12907", "repo": "rancher/build-tekton", "url": "https://github.com/rancher/build-tekton/pull/2" }
gharchive/pull-request
Add user 1000 in base image Its good to see image run using NON root user Its good to see image run using NON root user https://github.com/rancher/fleet/issues/216 https://github.com/rancher/fleet/issues/216 This will also need to be added to rancher/gitjob image This will also need to be added to rancher/gitjob image https://github.com/rancher/gitjob/pull/26 https://github.com/rancher/gitjob/pull/26
2025-04-01T06:41:17.734955
2016-06-16T21:15:59
160762927
{ "authors": [ "codecov-io", "rasmus" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12908", "repo": "rasmus/EventFlow", "url": "https://github.com/rasmus/EventFlow/pull/228" }
gharchive/pull-request
Easier command scheduling [Current coverage][cc-pull] is 84.75% Merging [#228][cc-pull] into [develop][cc-base-branch] will decrease coverage by 0.34% @@ develop #228 diff @@ ========================================== Files 216 217 +1 Lines 4482 4500 +18 Methods 0 0 Messages 0 0 Branches 366 366 ========================================== Hits 3814 3814 - Misses 484 502 +18 Partials 184 184 [][cc-pull] Powered by Codecov. Last updated by [08c231d...1d43ad6][cc-compare] [cc-base-branch]: https://codecov.io/gh/rasmus/EventFlow/branch/develop?src=pr [cc-compare]: https://codecov.io/gh/rasmus/EventFlow/compare/08c231d2b66f793569224a6718f41e2d88163b22...1d43ad69a520883ba56507403b258c71cb714815 [cc-pull]: https://codecov.io/gh/rasmus/EventFlow/pull/228?src=pr
2025-04-01T06:41:17.741063
2015-07-21T19:24:21
96397948
{ "authors": [ "antonmoiseev", "cybrown", "rbuckton" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12909", "repo": "rbuckton/ReflectDecorators", "url": "https://github.com/rbuckton/ReflectDecorators/pull/11" }
gharchive/pull-request
Added typescript.definition entry in package.json for tsd #10 Added entry in package.json, I hope this will be usefull @rbuckton are you accepting PRs in this repo? This PR looks tiny but very useful for TypeScript projects. @antonmoiseev I need to clarify with Microsoft Legal as to whether contributors need to sign a CLA (http://cla.microsoft.com) for this repo. We can add this for tsd support. Thanks! Thank you for the merge, but I hope it will be obsolete as soon as possible with the "typings" field !
2025-04-01T06:41:17.796019
2020-12-17T12:58:47
769997194
{ "authors": [ "mateusz-hobgarski-red", "mhughesacn", "rdkcmf-jenkins" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12910", "repo": "rdkcentral/RDKShell", "url": "https://github.com/rdkcentral/RDKShell/pull/48" }
gharchive/pull-request
RDK-30526: added client_control_extension test app https://jenkins.code.rdkcentral.com/job/github-triggered/5413/ started Copyright scan failure Commit: 9a66dbb6b0bc02f7a4e29ec7ae7f6f8ed3139420 Report detail: https://gist.github.com/cc4d3f23e6c83544f3b1255b225e5c5f https://jenkins.code.rdkcentral.com/job/build-component-rdk-next/24699/ started. Blackduck scan failure details Summary: 0 violations, 0 files pending approval, 2 files pending identification. Protex Server Path: /home/blackduck/github/RDKShell/48/rdkcentral/RDKShell Commit: 9a66dbb6b0bc02f7a4e29ec7ae7f6f8ed3139420 Report detail: gist https://jenkins.cmf.code.rdkcentral.com/job/github-triggered/630/ started https://jenkins.cmf.code.rdkcentral.com/job/build-component-rdk-next-platform/2627/ started. https://jenkins.code.rdkcentral.com/job/github-triggered/5414/ started https://jenkins.code.rdkcentral.com/job/build-component-rdk-next/24700/ started. https://jenkins.cmf.code.rdkcentral.com/job/github-triggered/631/ started Blackduck scan failure details Summary: 0 violations, 0 files pending approval, 1 file pending identification. Protex Server Path: /home/blackduck/github/RDKShell/48/rdkcentral/RDKShell Commit: a621f033b3be3af2a2d5932ff2dd6a283ae291c2 Report detail: gist https://jenkins.cmf.code.rdkcentral.com/job/build-component-rdk-next-platform/2629/ started. Smoke tests for ['build-qemux86-hyb-rdk-next'/'build-qemux86-mc-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/70284/ Smoke tests for ['build-rpi-hybrid-wpe-rdk-next'/'build-rpi-mc-wpe-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/70285/ Smoke tests for ['build-rpi-hybrid-refapp-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/70283/ Smoke tests for ['build-rpi-hybrid-wpe-rdk-next'/'build-rpi-mc-wpe-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/70288/ Smoke tests for ['build-qemux86-hyb-rdk-next'/'build-qemux86-mc-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/70286/ Smoke tests for ['build-rpi-hybrid-refapp-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/70287/ Hi Mateusz, clientcontrolextensiontest.cpp is getting code matches elsewhere i.e. this is not original code. Where did you get the code for setupGL (line 724+) and renderGL (760+), also line 680-696, vert_shader_text and frag_shader_text initialisations? (Depending on your source, this should not be a big amount of extra work.) Hi Martin, clientcontrolextensiontest.cpp is based on westeros-test.cpp from westeros repo, some parts of of the files are identical. Regards, Mateusz Hobgarski I've placed this information in a comment on top of the clientcontrolextensiontest.cpp. Regards, Mateusz Hobgarski Thanks Mateusz. There's a standard RDK way to do this, plus westeros is using some earlier code from wayland/weston. Please will you make the following changes: After the paragraph about westeros, insert: Some code is derived from weston and is: Copyright © 2010 Intel Corporation Copyright © 2011 Benjamin Franzke Copyright © 2012-2013 Collabora, Ltd. Licensed under the MIT License Please append the same block to the end of the NOTICE file at top level. I will clear off the blackduck check when this is sorted. Thanks! https://jenkins.code.rdkcentral.com/job/github-triggered/5429/ started https://jenkins.code.rdkcentral.com/job/build-component-rdk-next/24729/ started. https://jenkins.cmf.code.rdkcentral.com/job/github-triggered/640/ started https://jenkins.cmf.code.rdkcentral.com/job/build-component-rdk-next-platform/2655/ started. Copyright scan failure Commit: a4a3c4089570c0506bf9f2e626fe118f9dff1082 Report detail: https://gist.github.com/e819cfddf4f6912f681aacb268403f44 Blackduck scan failure details Summary: 0 violations, 0 files pending approval, 1 file pending identification. Protex Server Path: /home/blackduck/github/RDKShell/48/rdkcentral/RDKShell Commit: a4a3c4089570c0506bf9f2e626fe118f9dff1082 Report detail: gist Hi Martin, I've added what you said and now the copyright scan is failing :) Something is still wrong? Regards, Mateusz Hobgarski Smoke tests for ['build-rpi-hybrid-wpe-rdk-next'/'build-rpi-mc-wpe-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/70352/ Smoke tests for ['build-qemux86-hyb-rdk-next'/'build-qemux86-mc-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/70350/ Smoke tests for ['build-rpi-hybrid-refapp-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/70351/ Thanks Mateusz. I have to clear findings in our tools to clear the checks. I've done this and everything is clear now - thank you. https://jenkins.code.rdkcentral.com/job/github-triggered/5710/ started https://jenkins.code.rdkcentral.com/job/github-triggered/5710/ started https://jenkins.code.rdkcentral.com/job/build-component-rdk-next/25341/ started. https://jenkins.code.rdkcentral.com/job/build-component-rdk-next/25341/ started. https://jenkins.cmf.code.rdkcentral.com/job/github-triggered/878/ started https://jenkins.cmf.code.rdkcentral.com/job/github-triggered/878/ started https://jenkins.cmf.code.rdkcentral.com/job/build-component-rdk-next-platform/3111/ started. https://jenkins.cmf.code.rdkcentral.com/job/build-component-rdk-next-platform/3111/ started. Smoke tests for ['build-qemux86-hyb-rdk-next'/'build-qemux86-mc-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/71602/ Smoke tests for ['build-qemux86-hyb-rdk-next'/'build-qemux86-mc-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/71602/ Smoke tests for ['build-rpi-hybrid-wpe-rdk-next'/'build-rpi-mc-wpe-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/71604/ Smoke tests for ['build-rpi-hybrid-wpe-rdk-next'/'build-rpi-mc-wpe-rdk-next'] added. https://jenkins.cmf.code.rdkcentral.com/job/test-components-flow/71604/ '['build-rpi-hybrid-wpe-rdk-next'/'build-rpi-hybrid-wpe-rdk-next']' Smoke test FAILURE, investigation pending. Flash MC [SUCCESS] Flash HYB [SUCCESS] MC - Video Test [FAILURE] Video HLS gst-launch [SUCCESS] Video HLS rdkmediaplayer [SUCCESS] Video Streaming using aamp-cli [FAILURE] Video local using gst-play-1.0 [SUCCESS] Video Streaming using rmfApp [SUCCESS] RDKShell CI Tests [FAILURE] checking thunder security Thunder security is disabled getClients : Failed - [clients:[residentapp],success:true] getZOrder : Failed - [clients:[residentapp],success:true] getClients : Failed - [clients:[testapp2,testapp1,testapp,residentapp],success:true] getZOrder : Failed - [clients:[testapp2,testapp1,testapp,residentapp],success:true] add key intercept test failed add key listener test failed getClientsAfterKill : Failed - [clients:[residentapp],success:true] Number of success test - 33 Number of failed test - 7 Video Streaming using aamp-cli [SUCCESS] HYB - Video Test [FAILURE] Video HLS gst-launch [SUCCESS] Video HLS rdkmediaplayer [SUCCESS] Video Streaming using aamp-cli [SUCCESS] Video local using gst-play-1.0 [SUCCESS] RDKShell CI Tests [FAILURE] checking thunder security Thunder security is disabled getClients : Failed - [clients:[residentapp],success:true] getZOrder : Failed - [clients:[residentapp],success:true] getClients : Failed - [clients:[testapp2,testapp1,testapp,residentapp],success:true] getZOrder : Failed - [clients:[testapp2,testapp1,testapp,residentapp],success:true] add key intercept test failed add key listener test failed getClientsAfterKill : Failed - [clients:[residentapp],success:true] Number of success test - 33 Number of failed test - 7 https://jenkins.cmf.code.rdkcentral.com/job/test-rdkv-raspberrypi-flow/21132/ '['build-rpi-hybrid-wpe-rdk-next'/'build-rpi-hybrid-wpe-rdk-next']' Smoke test FAILURE, investigation pending. Flash MC [SUCCESS] Flash HYB [SUCCESS] MC - Video Test [FAILURE] Video HLS gst-launch [SUCCESS] Video HLS rdkmediaplayer [SUCCESS] Video Streaming using aamp-cli [FAILURE] Video local using gst-play-1.0 [SUCCESS] Video Streaming using rmfApp [SUCCESS] RDKShell CI Tests [FAILURE] checking thunder security Thunder security is disabled getClients : Failed - [clients:[residentapp],success:true] getZOrder : Failed - [clients:[residentapp],success:true] getClients : Failed - [clients:[testapp2,testapp1,testapp,residentapp],success:true] getZOrder : Failed - [clients:[testapp2,testapp1,testapp,residentapp],success:true] add key intercept test failed add key listener test failed getClientsAfterKill : Failed - [clients:[residentapp],success:true] Number of success test - 33 Number of failed test - 7 Video Streaming using aamp-cli [SUCCESS] HYB - Video Test [FAILURE] Video HLS gst-launch [SUCCESS] Video HLS rdkmediaplayer [SUCCESS] Video Streaming using aamp-cli [SUCCESS] Video local using gst-play-1.0 [SUCCESS] RDKShell CI Tests [FAILURE] checking thunder security Thunder security is disabled getClients : Failed - [clients:[residentapp],success:true] getZOrder : Failed - [clients:[residentapp],success:true] getClients : Failed - [clients:[testapp2,testapp1,testapp,residentapp],success:true] getZOrder : Failed - [clients:[testapp2,testapp1,testapp,residentapp],success:true] add key intercept test failed add key listener test failed getClientsAfterKill : Failed - [clients:[residentapp],success:true] Number of success test - 33 Number of failed test - 7 https://jenkins.cmf.code.rdkcentral.com/job/test-rdkv-raspberrypi-flow/21132/
2025-04-01T06:41:17.803564
2019-11-22T10:33:21
527116364
{ "authors": [ "abhiilash", "deLTreeTech", "nicolekapp" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12911", "repo": "react-native-sensors/react-native-sensors", "url": "https://github.com/react-native-sensors/react-native-sensors/issues/290" }
gharchive/issue
Sensor accelerometer is not available in ios I find out the solution iOS simulators currently have no support for sensors. In order to retrieve any sensor output, you must develop on a real device I have wasted a lot of time on this. Accelerometer reading on iOS is very wrong compare to Android even when I perform the same action on both I have wasted a lot of time on this. Accelerometer reading on iOS is very wrong compare to Android even when I perform the same action on both what can we do to use accelerometer? in ios
2025-04-01T06:41:17.823674
2016-07-29T12:59:22
168319253
{ "authors": [ "leifmadsen" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12912", "repo": "redhat-nfvpe/ansible-cira", "url": "https://github.com/redhat-nfvpe/ansible-cira/pull/13" }
gharchive/pull-request
Move inventory around to allow for additional inventory files This will all make more sense in a minute once I open another pull request :) Basically, the purpose of this is to allow the creation of additional inventory files in the hosts/ directory, such as the data for a Jenkins slave. The .gitignore changes makes it so you can locally add your Jenkins slave inventory file, without it polluting the repository.
2025-04-01T06:41:17.859024
2017-07-20T07:21:08
244267797
{ "authors": [ "agherzan", "floion", "willnewton" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12913", "repo": "resin-os/meta-resin", "url": "https://github.com/resin-os/meta-resin/pull/763" }
gharchive/pull-request
Adapt to morty on 1.x @resin-jenkins retest this please Can we have this merged? Merged, @floion . Thanks.
2025-04-01T06:41:17.862614
2020-11-30T20:58:02
753798464
{ "authors": [ "jamezp", "ronsigal", "rsearls" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12914", "repo": "resteasy/resteasy", "url": "https://github.com/resteasy/resteasy/pull/2606" }
gharchive/pull-request
[RESTEASY-2710] added new class and test. Hey @rsearls, Note that Andy has made some changes to ServletContainerInitializer in https://github.com/resteasy/Resteasy/pull/2474: https://github.com/resteasy/Resteasy/blob/1bf0e2396d460932b47637a6dd673bf83b2fa8d6/resteasy-servlet-initializer/src/main/java/org/jboss/resteasy/plugins/servlet/ResteasyServletInitializer.java so you should update on top of that. -Ron Ok, I merged https://github.com/resteasy/Resteasy/pull/2474. Thank you for the contribution. We are cleaning up older PR's that seem to be stale. If you feel this is still an issue please feel free to re-open.
2025-04-01T06:41:17.887053
2015-11-26T09:56:10
119008825
{ "authors": [ "dmportella", "karlatkinson", "tegud" ], "license": "apache-2.0", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12915", "repo": "river-styx/river-styx.core", "url": "https://github.com/river-styx/river-styx.core/issues/9" }
gharchive/issue
Readme - basic usage @tegud use the wiki.... may be? It is normal to have a basic usage in the readme and then go into more detail in the API docs/wiki. It usually goes, Installation, Getting Started, Doc links Maybe an /Examples dir too this package is very simple i think steve has done some examples already
2025-04-01T06:41:17.917209
2021-03-13T11:44:02
830881268
{ "authors": [ "togaen" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12916", "repo": "ros-safety/contracts_lite", "url": "https://github.com/ros-safety/contracts_lite/pull/15" }
gharchive/pull-request
add nonzero real type, update readme @gbiggs Do you know what might be causing the build failure?
2025-04-01T06:41:17.926516
2018-07-26T13:29:57
344844658
{ "authors": [ "Alig1493", "ericholscher", "safwanrahman" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12917", "repo": "rtfd/readthedocs.org", "url": "https://github.com/rtfd/readthedocs.org/pull/4431" }
gharchive/pull-request
Fix for issue #4092: Remove unused field from Project model @stsewd @safwanrahman I'll look into it. @stsewd @safwanrahman I've looked into it to the best of my ability. Would you like to take a look into it? Looks good. r+ Bumped the migration name, and will merge this :+1:
2025-04-01T06:41:17.935047
2018-03-02T13:27:21
301765674
{ "authors": [ "mod", "stannak7", "yivo" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12918", "repo": "rubykube/peatio", "url": "https://github.com/rubykube/peatio/issues/623" }
gharchive/issue
On the fly member registration in API by using JWT payload Bittrex sourcecode: https://github.com/stannak7/bittrex-source-code @yivo this is the highest priority for us to continue testing on workbench the integration https://github.com/rubykube/barong/commit/3196b9ca0ff749be7158c6dbbf6b8c13a6ca9999 We will also try to change the login flow on cryptobase @calj Closed via #657.
2025-04-01T06:41:17.946911
2018-09-23T03:53:07
362911172
{ "authors": [ "GMaiolo", "jasp402", "toddmantell" ], "license": "mit", "license_source": "bigquery", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12919", "repo": "ryanmcdermott/clean-code-javascript", "url": "https://github.com/ryanmcdermott/clean-code-javascript/pull/246" }
gharchive/pull-request
Fix for #245 - Redundant information in dont-add-unneeded-context @toddmantell https://github.com/ryanmcdermott/clean-code-javascript/issues/245 is open for discussion, I just left the PR open in case it's needed to close it. I think that this is more a philosophical precept than a technicalism. Let's see the following example. suppose we have 3 objects. (table, car and house). Now in real life I have a paint gun that works for all 3 things. I can paint any of the objects with the same tool. but in this case the clean code tells me that I must necessarily have a paint gun for each object even if they are the same. I can or use a single function for everything, passing the parameter (object, color). or I can use one for each and internally each function has the color. I believe that the clean code is not focused on simplifying the code but on clarifying it. so that if we follow the rules of clearCode. we must predispose our senses to write things that we do not think are good practices. because 1 single function is more effective, scalable and maintainable. that "n" functions. then at this point it becomes counter-productive and you have to choose between something readable. and something productive. `const table = { color : 'light', size : '90x180' }; const home = { bedrooms : 3, bathrooms : 1, garage : 'Yes', color : 'turquoise' }; const Car = { make : 'Honda', model : 'Accord', color : 'Blue' }; function paintCar(car) { car.color = 'Red'; } function paintHome(home) { home.color = 'Red'; } function paintTable(table) { table.color = 'Red'; } paintCar(Car) paintHome(home) paintTable(table) console.log(Car) console.log(home) console.log(table)` @jasp402 I feel I didn't entirely address your point. I totally agree that if you have a more general function that paints multiple different kinds of objects (like paint(something)), it would be excess duplication to have paintHouse, paintCar, etc. and I think you would want a general paint function to avoid unnecessary duplication. What I was merely objecting to was the idea that one should mix the function name with the parameter name, like paint(car) in the function signature instead of paintCar.
2025-04-01T06:41:17.955101
2016-08-18T18:28:30
171968549
{ "authors": [ "buckett", "ern", "jonespm", "ottenhoff" ], "license": "ECL-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12920", "repo": "sakaiproject/sakai", "url": "https://github.com/sakaiproject/sakai/pull/3219" }
gharchive/pull-request
SAK-31692 - Upgrade Sakai to Spring 4.2 (4.2.7 currently available) LGTM Build failure looks related. Thanks Matthew, looks like there's this one test that's still using a deprecated class I missed, I'll update it. Okay, two tests, but it's all passing locally now, lets see what travis thinks! Right on. Thanks @jonespm
2025-04-01T06:41:17.976892
2019-12-17T13:53:52
539080156
{ "authors": [ "harrietc52", "stevieing" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12921", "repo": "sanger/traction-ui", "url": "https://github.com/sanger/traction-ui/issues/252" }
gharchive/issue
Check Run comments Need to check with Craig. Runs 1,3 and 4 seem to have the standard comments. Runs 2 and 3 have no comments. Run 6 seems to have a different style of comment? using the trac barcodes.
2025-04-01T06:41:17.980029
2021-06-17T14:32:38
923997536
{ "authors": [ "LucasBrazi06", "OliveGerste" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12922", "repo": "sap-labs-france/ev-server", "url": "https://github.com/sap-labs-france/ev-server/issues/2570" }
gharchive/issue
Assets > provide an API Endpoint for pushing Asset consumptions @LucasBrazi06 Just to be sure. I would introduce the new Routing Framework and Validation for assets with this issue. So the AssetRouter and AssetValidator would be implemented containing the new endpoint. @OliveGerste Yes, exactly! If you need help, @Ninjeneer can help you :-)
2025-04-01T06:41:17.997836
2016-03-18T15:13:38
141894426
{ "authors": [ "nicolasstucki", "scala-jenkins", "sjrd" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12923", "repo": "scala-js/scala-js", "url": "https://github.com/scala-js/scala-js/pull/2303" }
gharchive/pull-request
Add DOMJSEnv based on 'jsdom' in Node.js. Review by @sjrd. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2941/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3807/ Test FAILed. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2943/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3812/ Test FAILed. This PR will have merge conflicts with #2289 due to the changes in the CI matrix. Rebased. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2949/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3825/ Test FAILed. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2952/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3828/ Test FAILed. Oops, forgot to check if Jenkins had jsdom installed... it does not. Jenkins, please retest this. Jenkins, retest this please. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2954/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3831/ Test FAILed. Note that I had to add some binary incompatibilities on the on the Node.js jsEnvs. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2955/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3832/ Test PASSed. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2956/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3833/ Test PASSed. That's all. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2977/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3863/ Test PASSed. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2981/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3870/ Test PASSed. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2982/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3871/ Test PASSed. Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2984/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3874/ Test PASSed. LGTM Refer to this link for build results (access rights to CI server needed): https://scala-webapps.epfl.ch/jenkins/job/scalajs-pr/2986/ https://scala-webapps.epfl.ch/jenkins/job/scalajs-matrix-build/3876/ Test PASSed. :tada:
2025-04-01T06:41:18.021776
2020-03-07T14:05:08
577339215
{ "authors": [ "hbs", "pi-r-p" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12924", "repo": "senx/warp10-platform", "url": "https://github.com/senx/warp10-platform/pull/686" }
gharchive/pull-request
Added support for listing and aborting WarpScript executions First implementation done in VScode extension. As discussed, it could be great to have a "session info" function that returns information about user scripts for one session, without knowing the server stackps.secret. For better security, session name could also have a minimum length. PTAL PTAL
2025-04-01T06:41:18.033009
2020-07-11T12:04:04
655196544
{ "authors": [ "sharkbit", "tdasto" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12925", "repo": "sharkbit/BadgeApp", "url": "https://github.com/sharkbit/BadgeApp/issues/49" }
gharchive/issue
undefined error on Check availability fixed @sharkbit how can I archive this if I can't check it? What was the fix? I liked the reference to the fixed code in a previous Issue
2025-04-01T06:41:18.043735
2020-10-02T08:44:31
713454824
{ "authors": [ "manpreet2000", "shivangdubey" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12926", "repo": "shivangdubey/HacktoberFest2020", "url": "https://github.com/shivangdubey/HacktoberFest2020/pull/57" }
gharchive/pull-request
Added two pyhton files. Nice! Please edit and give some details about your scripts. Snake game is a pure python-based game which uses pygame module, It is working in Linux and windows operating system.
2025-04-01T06:41:18.048326
2020-07-17T18:43:43
659521616
{ "authors": [ "codecov-commenter", "shkumagai" ], "license": "Apache-2.0", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12927", "repo": "shkumagai/drf_renderer_svgheatmap", "url": "https://github.com/shkumagai/drf_renderer_svgheatmap/pull/79" }
gharchive/pull-request
Add codecov Codecov Report :exclamation: No coverage uploaded for pull request base (master@a2732ee). Click here to learn what that means. The diff coverage is n/a. @@ Coverage Diff @@ ## master #79 +/- ## ========================================== Coverage ? 100.00% ========================================== Files ? 1 Lines ? 53 Branches ? 0 ========================================== Hits ? 53 Misses ? 0 Partials ? 0 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 a2732ee...ad265c4. Read the comment docs.
2025-04-01T06:41:18.084086
2019-08-15T16:38:52
481235736
{ "authors": [ "simondlevy" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12928", "repo": "simondlevy/MulticopterSim", "url": "https://github.com/simondlevy/MulticopterSim/issues/73" }
gharchive/issue
Detect crash into wall and ceiling Should check for beam hits on six cardinal directions from sphere. Anything should count as a hit. Then zero-out velocities in all three directions (not just Z).
2025-04-01T06:41:18.094113
2017-09-14T10:27:08
257669949
{ "authors": [ "DerGoogler", "huge818", "jjqq2013" ], "license": "MIT", "license_source": "github-api", "license_type": "permissive", "provenance": "gharchive-dolma-0000.json.gz:12929", "repo": "sjitech/build-nodejs-for-android", "url": "https://github.com/sjitech/build-nodejs-for-android/issues/5" }
gharchive/issue
how is the p[assword of docker image? osexp2000/build-nodejs-for-android how is the password of docker image? Sorry i did not mention that because i think that is a convension about docker image usage. You do not need use password to access my private docker hub page. Please run "docker run osexp2000/build-nodejs-for-android" Then you will will be automatically dowliad the image for the first time. Sorry i did not mention that because i think that is a convension about docker image usage. You do not need use password to access my private docker hub page. Please run "docker run osexp2000/build-nodejs-for-android" Then you will will be automatically dowliad the image for the first time. not work