ci: Add a conv. commit lint action+config #5598

dougch · 2025-10-31T21:45:29Z

Release Summary:

Resolved issues:

resolves none

Description of changes:

Adds a GitHub Action to check the last commit for compliance with conventional commits.

Call-outs:

Shooting for parity with the configuration in https://github.com/aws/s2n-quic.

Testing:

In CI this run should pass...as this PR commit is using conventional commits.

Locally, we can go back in time and find a commit that didn't follow the format:

cd .github
npm install --save-dev @commitlint/config-conventional @commitlint/cli
#Check the last commit
npx commitlint -g ./config/commitlint.config.js --last
npx commitlint -g ./config/commitlint.config.js --from a214bcb9d5e73bec9b760c714c4a57705c9a754f --to e8f51528dece033d7627cf557a9b31fa290d5f77
⧗   input: refactor 2/2: Fix security policy version in tests to numbered string (#5553)
✖   subject may not be empty [subject-empty]
✖   type may not be empty [type-empty]

✖   found 2 problems, 0 warnings
ⓘ   Get help: https://github.com/conventional-changelog/commitlint/#what-is-commitlint

⧗   input: refactor 1/2: Fix security policy version in tests to numbered string (#5549)
✖   subject may not be empty [subject-empty]
✖   type may not be empty [type-empty]

✖   found 2 problems, 0 warnings
ⓘ   Get help: https://github.com/conventional-changelog/commitlint/#what-is-commitlint

⧗   input: bindings(rust): bump extended crates MSRV to 1.72.0 (#5534)
✖   type must be one of [build, chore, ci, docs, feat, fix, perf, refactor, revert, style, test, opt, internal, tests] [type-enum]

✖   found 1 problems, 0 warnings
ⓘ   Get help: https://github.com/conventional-changelog/commitlint/#what-is-commitlint

If there are no issues found, the command exits with no output.
The issues were:

refactor 1/2: isn't quite right, if the 1/2 had been in () this would have passed...
bindings(): is not a valid type.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

.github/config/commitlint.config.js

jmayclin · 2025-11-04T00:05:03Z

.github/config/commitlint.config.js

@@ -0,0 +1,20 @@
+module.exports = {
+  extends: ['@commitlint/config-conventional'],
+  rules: {


Why these particular rules? What does the [0] do? Can we just use someone else's config? Like what if we didn't "extend" the config-conventional

We already add this to s2n-quic repo: aws/s2n-quic#2881

The doc for this rule can be found in https://commitlint.js.org/reference/rules-configuration.html#rules-configuration.

It seemed like a reasonable starting point... here is the complete list

@commitlint/config-angular
@commitlint/config-conventional
@commitlint/config-lerna-scopes
@commitlint/config-nx-scopes
@commitlint/config-patternplate
@commitlint/config-workspace-scopes
conventional-changelog-lint-config-atom
conventional-changelog-lint-config-canonical

For subject-case 0 turns this off; I could not sus out a common pattern to past commit subject case; that isn't to say we shouldn't pick one, but I'm trying to ease us into this check. I can also add more doc links if that would help?

Hmm, it seems like the most reasonable starting point is just their default? I understand that it might cause a bit of friction, but one of the benefits of this tooling is the consistency. If we hate it we can always change it 🤷

Some of the ones that we've specified seem less than ideal e.g. allowing test and tests.

Can you add this link https://commitlint.js.org/reference/rules-configuration.html#rules-configuration to this file?

boquan-fang · 2025-11-04T00:22:53Z

.github/config/commitlint.config.js

@@ -0,0 +1,20 @@
+module.exports = {
+  extends: ['@commitlint/config-conventional'],
+  rules: {


Can you add this link https://commitlint.js.org/reference/rules-configuration.html#rules-configuration to this file?

jmayclin · 2025-11-04T00:22:39Z

.github/config/commitlint.config.js

@@ -0,0 +1,20 @@
+module.exports = {
+  extends: ['@commitlint/config-conventional'],
+  rules: {


Hmm, it seems like the most reasonable starting point is just their default? I understand that it might cause a bit of friction, but one of the benefits of this tooling is the consistency. If we hate it we can always change it 🤷

Some of the ones that we've specified seem less than ideal e.g. allowing test and tests.

jmayclin · 2025-11-04T00:24:49Z

.github/config/commitlint.config.js

+      issuePrefixes: ['#']
+    }
+  },
+  ignores: [


Why do we need to ignore this? And what is actually getting ignored? Like conventional commits aren't enforced if the commit includes "Co-authored-by"?

It's a line by line check, from my local testing. If someone adds a "recommended change" to your PR and you accept it, Co-authored-by gets added to the commit message. With-out this present, you would need to rebase and strip these out for this check to pass; example.

Hmm, why does it fail that commit?

Co-authored-by: maddeleine <[email protected]>

jmayclin · 2025-11-05T18:56:48Z

.github/config/commitlint.config.js

+  rules: {
+    'subject-case': [0],
+    'body-max-line-length': [0],
+    'footer-max-line-length': [0],
+    'type-enum': [2, 'always', [
+      'build', 'chore', 'ci', 'docs', 'feat', 'fix', 'perf', 'test'
+    ]]


Suggested change

rules: {

'subject-case': [0],

'body-max-line-length': [0],

'footer-max-line-length': [0],

'type-enum': [2, 'always', [

'build', 'chore', 'ci', 'docs', 'feat', 'fix', 'perf', 'test'

]]

Maybe I'm missing something, but can we just delete these and totally take everything from the predefined stuff?

Subject case is pretty pedantic; I'm in favor of strictness, but I'm trying to ease into this change and avoid the inevitable "why doesn't this pass?" question by relaxing it a bit.

dougch · 2025-11-05T19:33:11Z

Reworking in favor of a PR title checker.

ci: Add a conv. commit lint action+config

3fe4843

github-actions bot added the s2n-core team label Oct 31, 2025

dougch added the type/ci CodeBuild or CI related label Oct 31, 2025

dougch requested a review from maddeleine October 31, 2025 21:46

dougch marked this pull request as ready for review October 31, 2025 21:47

Merge branch 'main' into convcommit

09b61d0

dougch requested review from boquan-fang and jmayclin and removed request for maddeleine November 3, 2025 23:58

boquan-fang reviewed Nov 4, 2025

View reviewed changes

.github/config/commitlint.config.js Outdated Show resolved Hide resolved

jmayclin reviewed Nov 4, 2025

View reviewed changes

dougch added 2 commits November 4, 2025 00:17

ci: PR feedback

4ca0d91

ci: More cr/lf

73ef096

dougch requested review from boquan-fang and jmayclin November 4, 2025 00:19

boquan-fang approved these changes Nov 4, 2025

View reviewed changes

jmayclin reviewed Nov 4, 2025

View reviewed changes

ci: simplify and add lightness

4e6184b

dougch requested review from boquan-fang and jmayclin November 4, 2025 01:23

jmayclin reviewed Nov 5, 2025

View reviewed changes

dougch closed this Nov 5, 2025

dougch mentioned this pull request Nov 5, 2025

ci: PR conventional commit lint GHA #5603

Draft

ci: Add a conv. commit lint action+config #5598

ci: Add a conv. commit lint action+config #5598

Uh oh!

Conversation

dougch commented Oct 31, 2025

Release Summary:

Resolved issues:

Description of changes:

Call-outs:

Testing:

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dougch Nov 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dougch commented Nov 5, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

dougch Nov 4, 2025 •

edited

Loading