profile
viewpoint
Christian Hammond chipx86 Beanbag, Inc. Palo Alto, CA https://www.chipx86.com/ President/CEO of Beanbag, Inc. (@beanbaginc), and co-creator of Review Board (@reviewboard).

chipx86/gtkparasite 247

A utility similar to Firebug but for GTK+ applications.

chipx86/cloudplaya 46

Utilities and APIs for working with Amazon Cloud Player

chipx86/reviewboard 10

A powerful Web 2.0 code review tool

chipx86/gitty 7

A fancy-pantsy GTK-based replacement for gitk

chipx86/sbcloudplayer 7

Amazon Cloud Player plugin for Squeezebox

chipx86/djblets 5

A collection of useful extensions for Django.

chipx86/rb-extension-pack 5

An experimental extension pack for ChipX86's Review Board's extensions branch

chipx86/butte-covid19-dashboard 4

Unofficial dashboard for the COVID-19 situation in Butte County, California.

chipx86/git-vmware 4

Git scripts used at VMware

chipx86/baseconfigs 3

Base configurations for my home directories

push eventreviewboard/reviewboard

David Trowbridge

commit sha fbadaea7588cc95cdfb1d434012dfa060b8c0ae6

Backport RemovedInReviewBoard50Warning to 3.0.x We'll be making a small API change to allow SCMTool filename normalization to make use of the extra_data field. We'll want this to be 3.0.19 and newer, but won't actually remove the deprecation of the older APIs until 5.0. The release-4.0.x branch has the `RemovedInReviewBoard50Warning` and the `warn` helper method, but 3.0.x didn't. This change backports it. Testing Done: Used in conjunction with other changes. Reviewed at https://reviews.reviewboard.org/r/11272/

view details

David Trowbridge

commit sha d53fc2d339bccb9b186a87b24b1cd4ae5d9fdac3

Pass through FileDiff extra_data into filename normalization. While most version control systems are relatively sane and treat a filename as a filename, there are a few that don't. We first encountered this with CVS, which loves to add a `,v` to the end of filenames depending on the method of access. To address this, we added a normalization method that SCMTools could override. Impending work on improving support for ClearCase requires even fancier normalization, and one that requires being able to use some data about the repository. This change adds the FileDiff extra_data as an argument to the normalization method. This is a soft API break. Existing SCMTools which provide a normalize method that does not take any additional arguments will continue to function, but emit a deprecation warning. Testing Done: Ran unit tests. Reviewed at https://reviews.reviewboard.org/r/11282/

view details

David Trowbridge

commit sha d6b9874f16b9a16f15b9f7d2186188c1a53cad24

Plumb through extra_data provided by diff parsers. This change makes it so custom diff parsers (associated with specific SCMTools) can provide additional extra_data to be set on the FileDiff. This change does the plumbing to take the extra_data from the parser and store it in the database. Testing Done: - Ran unit tests. - Used in conjunction with other changes. Reviewed at https://reviews.reviewboard.org/r/11295/

view details

David Trowbridge

commit sha 0e82c5ec75d985509bef1bb8325dd920ee8e74a5

Fix up code style and documentation in ClearCase implementation. This change fixes up a bunch of style issues and code documentation in the old ClearCase implementation. I've also removed an old fallback for a no-longer-supported version of Python. Testing Done: Ran unit tests. Reviewed at https://reviews.reviewboard.org/r/11288/

view details

David Trowbridge

commit sha dda2c32c74d67ba002e6f6231a6db34861df914c

Make cleartool location configurable. In some cases, the cleartool executable may be in an unusual location, have an alternate name, or users may want to use a wrapper script of some sort. This change adds an optional settings key to allow people to set the location or name of cleartool. Testing Done: Ran unit tests. Reviewed at https://reviews.reviewboard.org/r/11290/

view details

David Trowbridge

commit sha 6ee48355fb7f1eaf3fbd3af1bc4871d5f0c28731

Centralize cleartool invocation and fix up windows shell issues. The ClearCase SCM backend invokes `cleartool` a lot, and every time was assembling a big Popen call and duplicating error handling. This change centralizes all of that into a single method, while also improving unicode/bytes compatibility, logging, and console shell handling for windows. The windows issue is the most interesting one here. Some windows applications are marked as requiring a console (as opposed to using windows), and invoking them would always pop up a console window, regardless of whether input and output was actually possible. `cleartool` is one such executable. An existing workaround had been added to use `shell=True`, but this was only enabled when using Python 2.7 and Windows 7 (the initial patch was not written with future-proofing in mind), and relying on the shell opens up a variety of other potential issues, including security concerns. The `startupinfo` flag to `Popen` allows actually controlling this at a lower level. Testing Done: - Ran unit tests. - Tested that startupinfo changes to Popen invocation did the correct thing for Windows hosts. Reviewed at https://reviews.reviewboard.org/r/11298/

view details

David Trowbridge

commit sha 99ade8910940aff53138675be527863474889350

Merge branch 'release-3.0.x' into release-4.0.x

view details

push time in 3 days

push eventreviewboard/reviewboard

David Trowbridge

commit sha 0e82c5ec75d985509bef1bb8325dd920ee8e74a5

Fix up code style and documentation in ClearCase implementation. This change fixes up a bunch of style issues and code documentation in the old ClearCase implementation. I've also removed an old fallback for a no-longer-supported version of Python. Testing Done: Ran unit tests. Reviewed at https://reviews.reviewboard.org/r/11288/

view details

David Trowbridge

commit sha dda2c32c74d67ba002e6f6231a6db34861df914c

Make cleartool location configurable. In some cases, the cleartool executable may be in an unusual location, have an alternate name, or users may want to use a wrapper script of some sort. This change adds an optional settings key to allow people to set the location or name of cleartool. Testing Done: Ran unit tests. Reviewed at https://reviews.reviewboard.org/r/11290/

view details

David Trowbridge

commit sha 6ee48355fb7f1eaf3fbd3af1bc4871d5f0c28731

Centralize cleartool invocation and fix up windows shell issues. The ClearCase SCM backend invokes `cleartool` a lot, and every time was assembling a big Popen call and duplicating error handling. This change centralizes all of that into a single method, while also improving unicode/bytes compatibility, logging, and console shell handling for windows. The windows issue is the most interesting one here. Some windows applications are marked as requiring a console (as opposed to using windows), and invoking them would always pop up a console window, regardless of whether input and output was actually possible. `cleartool` is one such executable. An existing workaround had been added to use `shell=True`, but this was only enabled when using Python 2.7 and Windows 7 (the initial patch was not written with future-proofing in mind), and relying on the shell opens up a variety of other potential issues, including security concerns. The `startupinfo` flag to `Popen` allows actually controlling this at a lower level. Testing Done: - Ran unit tests. - Tested that startupinfo changes to Popen invocation did the correct thing for Windows hosts. Reviewed at https://reviews.reviewboard.org/r/11298/

view details

push time in 8 days

push eventreviewboard/reviewboard

David Trowbridge

commit sha d6b9874f16b9a16f15b9f7d2186188c1a53cad24

Plumb through extra_data provided by diff parsers. This change makes it so custom diff parsers (associated with specific SCMTools) can provide additional extra_data to be set on the FileDiff. This change does the plumbing to take the extra_data from the parser and store it in the database. Testing Done: - Ran unit tests. - Used in conjunction with other changes. Reviewed at https://reviews.reviewboard.org/r/11295/

view details

push time in 8 days

push eventreviewboard/reviewboard

David Trowbridge

commit sha d53fc2d339bccb9b186a87b24b1cd4ae5d9fdac3

Pass through FileDiff extra_data into filename normalization. While most version control systems are relatively sane and treat a filename as a filename, there are a few that don't. We first encountered this with CVS, which loves to add a `,v` to the end of filenames depending on the method of access. To address this, we added a normalization method that SCMTools could override. Impending work on improving support for ClearCase requires even fancier normalization, and one that requires being able to use some data about the repository. This change adds the FileDiff extra_data as an argument to the normalization method. This is a soft API break. Existing SCMTools which provide a normalize method that does not take any additional arguments will continue to function, but emit a deprecation warning. Testing Done: Ran unit tests. Reviewed at https://reviews.reviewboard.org/r/11282/

view details

push time in 15 days

startedmarler8997/reloader

started time in 16 days

startedzig-community/Zig-Showdown

started time in 16 days

delete branch openautocomplete/openautocomplete

delete branch : dependabot/npm_and_yarn/markdownlint-cli-0.24.0

delete time in 18 days

PR closed openautocomplete/openautocomplete

Bump markdownlint-cli from 0.23.1 to 0.24.0 dependencies

Bumps markdownlint-cli from 0.23.1 to 0.24.0. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/igorshubovych/markdownlint-cli/releases">markdownlint-cli's releases</a>.</em></p> <blockquote> <h2>0.24.0</h2> <ul> <li>Update <code>markdownlint</code> dependency to 0.21.0 <ul> <li>Improve <code>MD042</code>/<code>MD044</code></li> </ul> </li> </ul> <h2>0.23.2</h2> <ul> <li>Update markdownlint dependency to 0.20.4</li> </ul> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/b0fe91bb57e3872754d0c46323d859f413909d1e"><code>b0fe91b</code></a> Bump version 0.24.0</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/cd3624ebb0881300ffbdbd2ce81836fceae377f1"><code>cd3624e</code></a> Update dependencies: markdownlint to 0.21.0, markdownlint-rule-helpers to 0.1...</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/5d2a7420e4afe22ec6f93c87dc1f23adb42f0241"><code>5d2a742</code></a> Bump elliptic from 6.5.2 to 6.5.3 (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/110">#110</a>)</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/21e1fb73834688829cd520f6df254a776329e66c"><code>21e1fb7</code></a> Bump lodash from 4.17.15 to 4.17.19 (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/107">#107</a>)</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/8f46fb3ab7e84bc7e5170eb371a4d779474a632d"><code>8f46fb3</code></a> Bump version 0.23.2</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/3fbf4eca7e3cb4bfeab833a262ac08c70529a4be"><code>3fbf4ec</code></a> Update dependencies: markdownlint to 0.20.4, markdownlint-rule-helpers to 0.1...</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/e857b964a04c8d81f660f65951573d52294fcc82"><code>e857b96</code></a> Update README links to point to the new main branch of markdownlint (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/102">#102</a>)</li> <li>See full diff in <a href="https://github.com/igorshubovych/markdownlint-cli/compare/v0.23.1...v0.24.0">compare view</a></li> </ul> </details> <br />

Dependabot compatibility score

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.


<details> <summary>Dependabot commands and options</summary> <br />

You can trigger Dependabot actions by commenting on this PR:

  • @dependabot rebase will rebase this PR
  • @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
  • @dependabot merge will merge this PR after your CI passes on it
  • @dependabot squash and merge will squash and merge this PR after your CI passes on it
  • @dependabot cancel merge will cancel a previously requested merge and block automerging
  • @dependabot reopen will reopen this PR if it is closed
  • @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
  • @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

</details>

+26 -26

1 comment

2 changed files

dependabot[bot]

pr closed time in 18 days

pull request commentopenautocomplete/openautocomplete

Bump markdownlint-cli from 0.23.1 to 0.24.0

Superseded by #13.

dependabot[bot]

comment created time in 18 days

PR opened openautocomplete/openautocomplete

Bump markdownlint-cli from 0.23.1 to 0.25.0

Bumps markdownlint-cli from 0.23.1 to 0.25.0. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/igorshubovych/markdownlint-cli/releases">markdownlint-cli's releases</a>.</em></p> <blockquote> <h2>0.25.0</h2> <ul> <li>Update <code>markdownlint</code> dependency to 0.21.1 <ul> <li>Improve <code>MD011</code>/<code>MD031</code></li> </ul> </li> <li>Update all dependencies via <code>Dependabot</code></li> </ul> <h2>0.24.0</h2> <ul> <li>Update <code>markdownlint</code> dependency to 0.21.0 <ul> <li>Improve <code>MD042</code>/<code>MD044</code></li> </ul> </li> </ul> <h2>0.23.2</h2> <ul> <li>Update <code>markdownlint</code> dependency to 0.20.4</li> </ul> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/09c5e38fce5a0beb629b53ec4e7e4b6e9b6d9ea4"><code>09c5e38</code></a> Bump version 0.25.0</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/1ac4c777608e52651d46b15c919ce958a880ba76"><code>1ac4c77</code></a> Update dependencies: glob to 7.1.6, ignore to 5.1.8, markdownlint to 0.21.1, ...</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/43ef786b60185c68c854b4b2d131b6600ea7d4db"><code>43ef786</code></a> Update Dependabot to use versioning-strategy increase for npm.</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/ae669571cc9dfea059a464fe1a34e7a90160b321"><code>ae66957</code></a> Bump execa from 2.1.0 to 4.1.0 (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/128">#128</a>)</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/3907304c0a9b7b95fd4d900323f9bb8e058474c1"><code>3907304</code></a> Bump js-yaml from 3.13.1 to 3.14.0 (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/123">#123</a>)</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/097bd7a6f15a33b4fda6e991a783b74e7ef79037"><code>097bd7a</code></a> Bump xo from 0.33.1 to 0.34.2 (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/126">#126</a>)</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/b3ed5e648f5dd6082b38e72b532da355aae84b75"><code>b3ed5e6</code></a> Bump jsonc-parser from 2.2.1 to 2.3.1 (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/124">#124</a>)</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/e69124fa2ba376cdb1f90e886a8b636e47ede111"><code>e69124f</code></a> Bump husky from 3.1.0 to 4.3.0 (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/125">#125</a>)</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/e5c0b4e5a618ae47fd5a6d6b89a4a70dbb158efd"><code>e5c0b4e</code></a> Bump ignore from 5.1.4 to 5.1.8 (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/127">#127</a>)</li> <li><a href="https://github.com/igorshubovych/markdownlint-cli/commit/5fd7a30c5b4dd678114e4dc4878e32d41c626a9e"><code>5fd7a30</code></a> Bump deep-extend from 0.5.1 to 0.6.0 (<a href="https://github-redirect.dependabot.com/igorshubovych/markdownlint-cli/issues/120">#120</a>)</li> <li>Additional commits viewable in <a href="https://github.com/igorshubovych/markdownlint-cli/compare/v0.23.1...v0.25.0">compare view</a></li> </ul> </details> <br />

Dependabot compatibility score

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.


<details> <summary>Dependabot commands and options</summary> <br />

You can trigger Dependabot actions by commenting on this PR:

  • @dependabot rebase will rebase this PR
  • @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
  • @dependabot merge will merge this PR after your CI passes on it
  • @dependabot squash and merge will squash and merge this PR after your CI passes on it
  • @dependabot cancel merge will cancel a previously requested merge and block automerging
  • @dependabot reopen will reopen this PR if it is closed
  • @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
  • @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

</details>

+49 -64

0 comment

2 changed files

pr created time in 18 days

push eventdjblets/djblets

Hailan Xu

commit sha 34b2d14336b98b0252ed656ea123001ef8c6979b

Fix integration popup not disappearing on button click In RB 4.0.x, the popup that comes up after selecting an integration's `Delete` button fails to disappear after deleting the integration and after selecting the cancel button. The issue was that in RB 3.0.x, we are using `<input type="button">` to construct the buttons and from RB 4.0.x and onwards we are using the `<button>` tags instead. I've added a condition so that the popup can be destroyed even if the `<button>` tags are used. Testing Done: Manual testing done with the following repo pairs: - Tested Djblets release-2.0.x with RB release 4.0.x - Tested Djblets master with RB master Reviewed at https://reviews.reviewboard.org/r/11276/

view details

David Trowbridge

commit sha c0360e24e3ed87d2941c9c6d31762017bbecf6c6

Merge branch 'release-2.0.x'

view details

push time in 19 days

push eventdjblets/djblets

Hailan Xu

commit sha 34b2d14336b98b0252ed656ea123001ef8c6979b

Fix integration popup not disappearing on button click In RB 4.0.x, the popup that comes up after selecting an integration's `Delete` button fails to disappear after deleting the integration and after selecting the cancel button. The issue was that in RB 3.0.x, we are using `<input type="button">` to construct the buttons and from RB 4.0.x and onwards we are using the `<button>` tags instead. I've added a condition so that the popup can be destroyed even if the `<button>` tags are used. Testing Done: Manual testing done with the following repo pairs: - Tested Djblets release-2.0.x with RB release 4.0.x - Tested Djblets master with RB master Reviewed at https://reviews.reviewboard.org/r/11276/

view details

push time in 19 days

push eventreviewboard/reviewboard

David Trowbridge

commit sha fbadaea7588cc95cdfb1d434012dfa060b8c0ae6

Backport RemovedInReviewBoard50Warning to 3.0.x We'll be making a small API change to allow SCMTool filename normalization to make use of the extra_data field. We'll want this to be 3.0.19 and newer, but won't actually remove the deprecation of the older APIs until 5.0. The release-4.0.x branch has the `RemovedInReviewBoard50Warning` and the `warn` helper method, but 3.0.x didn't. This change backports it. Testing Done: Used in conjunction with other changes. Reviewed at https://reviews.reviewboard.org/r/11272/

view details

push time in 21 days

push eventreviewboard/reviewboard

Christian Hammond

commit sha 686abca3496092d0041cde96391c361643e8064e

Add the missing features.py for interdiff filtering v2. The `features.py` file was left out of the previous commit to backport interdiff filtering version 2. This is a trivial change that adds it back in.

view details

Christian Hammond

commit sha 0d8950d342d719c9d738818c5af3e1d223efe72a

Add support for Python 3.9. This updates a few pieces of code to fix up compatibility with Python 3.9. This consisted of a deprecated method use (`HTMLParser.unescape()`) and some code that passed a byte string into `urlopen()`. Since Python 3 support is now official, the package-level checks to enable it only if the `$RB_PY3_HAS_NO_OFFICIAL_SUPPORT` variable is set have been removed, and package metadata has been updated appropriately. Testing Done: Unit tests pass for all versions of Python. Verified that I could build packages on Python 2 and 3. Reviewed at https://reviews.reviewboard.org/r/11271/

view details

David Trowbridge

commit sha 0379aef66cac9727f2756c7d0f9d0cd4a68a865f

Make GitHub post-commit more reliable with "main" branch. Now that Git supports it, more and more people are choosing to name their mainline branch something other than "master". While before we were relatively confident that one of the branches fetched would have this name, that's no longer the case. This is particularly bad for the case where there's only one branch, because if none of the branches were marked as default, the frontend would never fetch the commits. This change updates the GitHub `get_branches` method to make sure that there's always a default branch set. If a branch is found named "master", we'll use that one, then it will try "main", and finally if neither is chosen, we'll just mark the first branch as the default. At the moment GitHub doesn't have a way for us to know what's the default (there's a user setting for when new repositories are created, but that doesn't tell us anything about existing repositories, especially those that are part of organizations rather than personal repos). Hopefully they'll add this to their API at some point and we'll be able to simplify this then. Testing Done: Created a new repository with only one branch, named "main". Saw that I could open the "New Review Request" page and that the main branch was selected and commits were properly loaded. Reviewed at https://reviews.reviewboard.org/r/11251/

view details

David Trowbridge

commit sha aa65dccafbad7641308e5aed3406f0a878bf3309

Merge branch 'release-3.0.x' into release-4.0.x

view details

David Trowbridge

commit sha f21586a28497de9a7c18fed5789ecf168f86aef6

Fix JS errors when closing the "Add File" dialog. DialogView subclasses (such as "Add File") could sometimes cause JS errors when they were closing. The issue here was a race between multiple event handlers. Inside the `modalBox` implementation (which is all our own jQuery-UI based code), a generic handler for clicks in the button area would cause the modal box to self-destruct. This could then conflict with the higher-level DialogView code which would additionally try to call destroy. In previous versions of jQuery-UI this didn't matter, but the version we're using now will error out when calling methods on elements which are not widgets. This change adds a check to make sure we don't try to destroy the modalBox widget after it's already self-destructed. Testing Done: - Manually tested the "Add File" dialog. - Ran js-tests without any modalBox-related errors. Reviewed at https://reviews.reviewboard.org/r/11254/

view details

David Trowbridge

commit sha f7c45c8fb6588a462c2cbe0c33d83c7e1035d999

Merge branch 'release-4.0.x'

view details

push time in 21 days

push eventreviewboard/reviewboard

David Trowbridge

commit sha f21586a28497de9a7c18fed5789ecf168f86aef6

Fix JS errors when closing the "Add File" dialog. DialogView subclasses (such as "Add File") could sometimes cause JS errors when they were closing. The issue here was a race between multiple event handlers. Inside the `modalBox` implementation (which is all our own jQuery-UI based code), a generic handler for clicks in the button area would cause the modal box to self-destruct. This could then conflict with the higher-level DialogView code which would additionally try to call destroy. In previous versions of jQuery-UI this didn't matter, but the version we're using now will error out when calling methods on elements which are not widgets. This change adds a check to make sure we don't try to destroy the modalBox widget after it's already self-destructed. Testing Done: - Manually tested the "Add File" dialog. - Ran js-tests without any modalBox-related errors. Reviewed at https://reviews.reviewboard.org/r/11254/

view details

push time in 21 days

push eventreviewboard/reviewboard

Christian Hammond

commit sha 686abca3496092d0041cde96391c361643e8064e

Add the missing features.py for interdiff filtering v2. The `features.py` file was left out of the previous commit to backport interdiff filtering version 2. This is a trivial change that adds it back in.

view details

David Trowbridge

commit sha 0379aef66cac9727f2756c7d0f9d0cd4a68a865f

Make GitHub post-commit more reliable with "main" branch. Now that Git supports it, more and more people are choosing to name their mainline branch something other than "master". While before we were relatively confident that one of the branches fetched would have this name, that's no longer the case. This is particularly bad for the case where there's only one branch, because if none of the branches were marked as default, the frontend would never fetch the commits. This change updates the GitHub `get_branches` method to make sure that there's always a default branch set. If a branch is found named "master", we'll use that one, then it will try "main", and finally if neither is chosen, we'll just mark the first branch as the default. At the moment GitHub doesn't have a way for us to know what's the default (there's a user setting for when new repositories are created, but that doesn't tell us anything about existing repositories, especially those that are part of organizations rather than personal repos). Hopefully they'll add this to their API at some point and we'll be able to simplify this then. Testing Done: Created a new repository with only one branch, named "main". Saw that I could open the "New Review Request" page and that the main branch was selected and commits were properly loaded. Reviewed at https://reviews.reviewboard.org/r/11251/

view details

David Trowbridge

commit sha aa65dccafbad7641308e5aed3406f0a878bf3309

Merge branch 'release-3.0.x' into release-4.0.x

view details

push time in 21 days

push eventreviewboard/reviewboard

David Trowbridge

commit sha 0379aef66cac9727f2756c7d0f9d0cd4a68a865f

Make GitHub post-commit more reliable with "main" branch. Now that Git supports it, more and more people are choosing to name their mainline branch something other than "master". While before we were relatively confident that one of the branches fetched would have this name, that's no longer the case. This is particularly bad for the case where there's only one branch, because if none of the branches were marked as default, the frontend would never fetch the commits. This change updates the GitHub `get_branches` method to make sure that there's always a default branch set. If a branch is found named "master", we'll use that one, then it will try "main", and finally if neither is chosen, we'll just mark the first branch as the default. At the moment GitHub doesn't have a way for us to know what's the default (there's a user setting for when new repositories are created, but that doesn't tell us anything about existing repositories, especially those that are part of organizations rather than personal repos). Hopefully they'll add this to their API at some point and we'll be able to simplify this then. Testing Done: Created a new repository with only one branch, named "main". Saw that I could open the "New Review Request" page and that the main branch was selected and commits were properly loaded. Reviewed at https://reviews.reviewboard.org/r/11251/

view details

push time in 21 days

push eventreviewboard/reviewboard

André Klitzing

commit sha c7d8d0d4664b653ed6a95bc8b0394ea1d39d49ad

Avoid internal retry of jira class If the jira object cannot connect to jira it will try a re-connect after a period of time. This could lead to a timeout in a long running RB process of a WSGI server (like uwsgi). Let's disable internal retry of the jira class and let Review Board handle that. Error of jira class: Got recoverable error from GET https://jira.company.de/rest/api/2/issue/PROJECT-123, will retry [1/3] in 19.0767479194s. Err: 401 Testing Done: Added the fix and saw that it fails immediately and don't block any process. Reviewed at https://reviews.reviewboard.org/r/11269/

view details

push time in a month

fork andrewrk/fetch-them-macos-headers

A simplistic utility for fetching minimal macOS libc headers

fork in a month

push eventreviewboard/rbintegrations

Hailan Xu

commit sha 2a416344221d901ba2997239e59a6d2d3ddf105f

Change 'Ship It!' color 'success' to 'good' for chat integrations Currently, when reviewers select the "Ship It!" option, the notification sent to Slack has a bar that is colored grey. I found that a value of 'success' is passed to color. Upon investigation it seems that Slack and Mattermost doesn't understand what color corresponds to 'success'. However, it knows 'good' corresponds to green. Evidence that Mattermost uses ‘good’ and not ‘success’: https://mattermost.atlassian.net/browse/MM-3181 https://mattermost.com/blog/getting-started-with-mattermost-integrations/ Evidence that Slack uses ‘good’: https://api.slack.com/reference/messaging/attachments#fields Testing Done: * I ran the unit tests on Slack and Mattermost. Two unrelated tests are failing because they cannot find `logo.png` at `/var/folders/...`. See #integrations * Confirmed that "Ship It!" message has a green bar instead of a grey one in Slack and Mattermost for Review Board version 3.0 and 4.0 after the change Reviewed at https://reviews.reviewboard.org/r/11211/

view details

David Trowbridge

commit sha 97764296bb6b81760f33753e6e8f3c911e041713

Merge branch 'release-1.0.x'

view details

push time in a month

push eventreviewboard/rbintegrations

Hailan Xu

commit sha 2a416344221d901ba2997239e59a6d2d3ddf105f

Change 'Ship It!' color 'success' to 'good' for chat integrations Currently, when reviewers select the "Ship It!" option, the notification sent to Slack has a bar that is colored grey. I found that a value of 'success' is passed to color. Upon investigation it seems that Slack and Mattermost doesn't understand what color corresponds to 'success'. However, it knows 'good' corresponds to green. Evidence that Mattermost uses ‘good’ and not ‘success’: https://mattermost.atlassian.net/browse/MM-3181 https://mattermost.com/blog/getting-started-with-mattermost-integrations/ Evidence that Slack uses ‘good’: https://api.slack.com/reference/messaging/attachments#fields Testing Done: * I ran the unit tests on Slack and Mattermost. Two unrelated tests are failing because they cannot find `logo.png` at `/var/folders/...`. See #integrations * Confirmed that "Ship It!" message has a green bar instead of a grey one in Slack and Mattermost for Review Board version 3.0 and 4.0 after the change Reviewed at https://reviews.reviewboard.org/r/11211/

view details

push time in a month

push eventreviewboard/reviewboard

David Trowbridge

commit sha 2400584f517cdc83c7cffeb3b13af656d209175e

Revert "Fix some errors in js-tests." This reverts commit c001ff2c4b6a40e456ee212fa3df8f74647664ca.

view details

David Trowbridge

commit sha 47c48d697d73fb1d3564bf11f49504cc45001583

Merge branch 'release-4.0.x'

view details

push time in a month

push eventreviewboard/reviewboard

David Trowbridge

commit sha 2400584f517cdc83c7cffeb3b13af656d209175e

Revert "Fix some errors in js-tests." This reverts commit c001ff2c4b6a40e456ee212fa3df8f74647664ca.

view details

push time in a month

push eventreviewboard/reviewboard

Amir Yalon

commit sha b122d1e7842cb244d39899b90bf3727a0e10870d

Fix error dispatching webhook with auth credentials The error was: AttributeError: 'SplitResult' object has no attribute 'params' The 5-tuple used for rebuilding the URL was missing the `fragment` element instead. Bugs closed: 4879 Reviewed at https://reviews.reviewboard.org/r/11191/

view details

Amir Yalon

commit sha f5b5a5f96d0090f91d3e56ec9461d68879971a2e

Log webhook server responses on error. If the exception from urlopen represents an HTTP response, then the included information may be useful to the site admin in troubleshooting. This change adds it to the server log. Reviewed at https://reviews.reviewboard.org/r/11192/

view details

Amir Yalon

commit sha 5f34cfebc0f22505b73c7af474ef7785bca46a79

Always send Authorization header on authenticated webhooks The problem with `HTTPBasicAuthHandler` is that it will first attempt to make an unauthenticated request, and will retry with authentication only if a 401 response was seen. However, some servers respond with 403 if the right credentials are not supplied right away. This change alters the webhook notification to unconditionally send the HTTP Basic header if credentials are present in the webhook URL. Testing Done: Ran unit tests. Reviewed at https://reviews.reviewboard.org/r/11193/

view details

David Trowbridge

commit sha b960ee6bdd189f70f6bc9b17646d18ee204345b9

Merge branch 'release-3.0.x' into release-4.0.x

view details

Marcus Boay

commit sha 55b566440af48d31f6ea0afa7ba8a0523863dc25

Fix My Privacy Rights showing empty label. The "My Privacy Rights" page in the user account settings was showing an orphaned `:*`. This was for the field label for the consent items, but that form field didn't have a label defined. The cause here was that the Review Board form field template was checking if the form had a `fields_no_label` attribute, but that was deprecated and removed from Djblets a long time ago. The correct fix is to check if the field has a label defined. Testing Done: Manual testing by refreshing the My Privacy Rights page and checking that the empty label is no longer visible. Other labels on the account forms that are supposed to be there have been verified to be visible. Doing a search of `fields_no_label` on all the reviewboard repositories show no usage of that particular variable. Bugs closed: 4882 Reviewed at https://reviews.reviewboard.org/r/11180/

view details

David Trowbridge

commit sha f99ea82ee5877871a01cd15b2182f3ca111d9a9b

Merge branch 'release-4.0.x'

view details

push time in a month

push eventreviewboard/reviewboard

Marcus Boay

commit sha 55b566440af48d31f6ea0afa7ba8a0523863dc25

Fix My Privacy Rights showing empty label. The "My Privacy Rights" page in the user account settings was showing an orphaned `:*`. This was for the field label for the consent items, but that form field didn't have a label defined. The cause here was that the Review Board form field template was checking if the form had a `fields_no_label` attribute, but that was deprecated and removed from Djblets a long time ago. The correct fix is to check if the field has a label defined. Testing Done: Manual testing by refreshing the My Privacy Rights page and checking that the empty label is no longer visible. Other labels on the account forms that are supposed to be there have been verified to be visible. Doing a search of `fields_no_label` on all the reviewboard repositories show no usage of that particular variable. Bugs closed: 4882 Reviewed at https://reviews.reviewboard.org/r/11180/

view details

push time in a month

push eventreviewboard/reviewboard

Amir Yalon

commit sha b122d1e7842cb244d39899b90bf3727a0e10870d

Fix error dispatching webhook with auth credentials The error was: AttributeError: 'SplitResult' object has no attribute 'params' The 5-tuple used for rebuilding the URL was missing the `fragment` element instead. Bugs closed: 4879 Reviewed at https://reviews.reviewboard.org/r/11191/

view details

Amir Yalon

commit sha f5b5a5f96d0090f91d3e56ec9461d68879971a2e

Log webhook server responses on error. If the exception from urlopen represents an HTTP response, then the included information may be useful to the site admin in troubleshooting. This change adds it to the server log. Reviewed at https://reviews.reviewboard.org/r/11192/

view details

Amir Yalon

commit sha 5f34cfebc0f22505b73c7af474ef7785bca46a79

Always send Authorization header on authenticated webhooks The problem with `HTTPBasicAuthHandler` is that it will first attempt to make an unauthenticated request, and will retry with authentication only if a 401 response was seen. However, some servers respond with 403 if the right credentials are not supplied right away. This change alters the webhook notification to unconditionally send the HTTP Basic header if credentials are present in the webhook URL. Testing Done: Ran unit tests. Reviewed at https://reviews.reviewboard.org/r/11193/

view details

David Trowbridge

commit sha b960ee6bdd189f70f6bc9b17646d18ee204345b9

Merge branch 'release-3.0.x' into release-4.0.x

view details

push time in a month

push eventreviewboard/reviewboard

Amir Yalon

commit sha 5f34cfebc0f22505b73c7af474ef7785bca46a79

Always send Authorization header on authenticated webhooks The problem with `HTTPBasicAuthHandler` is that it will first attempt to make an unauthenticated request, and will retry with authentication only if a 401 response was seen. However, some servers respond with 403 if the right credentials are not supplied right away. This change alters the webhook notification to unconditionally send the HTTP Basic header if credentials are present in the webhook URL. Testing Done: Ran unit tests. Reviewed at https://reviews.reviewboard.org/r/11193/

view details

push time in a month

push eventreviewboard/reviewboard

Amir Yalon

commit sha f5b5a5f96d0090f91d3e56ec9461d68879971a2e

Log webhook server responses on error. If the exception from urlopen represents an HTTP response, then the included information may be useful to the site admin in troubleshooting. This change adds it to the server log. Reviewed at https://reviews.reviewboard.org/r/11192/

view details

push time in a month

push eventreviewboard/reviewboard

Amir Yalon

commit sha b122d1e7842cb244d39899b90bf3727a0e10870d

Fix error dispatching webhook with auth credentials The error was: AttributeError: 'SplitResult' object has no attribute 'params' The 5-tuple used for rebuilding the URL was missing the `fragment` element instead. Bugs closed: 4879 Reviewed at https://reviews.reviewboard.org/r/11191/

view details

push time in a month

more