Skip to content

Conversation

@Mynacol
Copy link
Contributor

@Mynacol Mynacol commented Nov 1, 2025

  • Also clear hidden files and folders from the PR checkout. This improves security by avoiding unintentional remaining files.
  • Run upload-test-result directly after test-pr and don't wait for comment-pr. There is no dependency on the comment job.

@User123698745 what do you think?

- Also clear hidden files and folders from the PR checkout. This
  improves security by avoiding unintentional remaining files.
- Run upload-test-result directly after test-pr and don't wait for
  comment-pr. There is no dependency on the comment job.
@Mynacol Mynacol merged commit 9e24ad8 into RSS-Bridge:master Nov 2, 2025
14 checks passed
@Mynacol
Copy link
Contributor Author

Mynacol commented Nov 2, 2025

Successfully tested after merge on a force-push (rebase) on #4790.

@Mynacol Mynacol deleted the ci-prhtml-improvements branch November 2, 2025 00:11
@User123698745
Copy link
Contributor

User123698745 commented Nov 2, 2025

Looks good, but there is an small edge case where needs.checks.outputs.BRIDGES > 0 is true but needs.test-pr.outputs.COMMENT_LENGTH > 130 is false. This can happen when whitelist.txt contains a bridge name that does not exist (because of false positives or e.g. one could mention bridges/DoesNotExistOrMesspelledItsNameBridge.php in a git commit message).
In that case the actions/download-artifact step in upload-test-results would now fail after your change.
Easiest fix would be change

    if: needs.checks.outputs.WITH_UPLOAD == 'true'

to

    if: needs.test-pr.outputs.COMMENT_LENGTH > 130 && needs.checks.outputs.WITH_UPLOAD == 'true'

(before the change, this if was implicit by the required needs).

Mynacol added a commit to Mynacol/rss-bridge that referenced this pull request Nov 16, 2025
dvikan pushed a commit that referenced this pull request Nov 16, 2025
* CI: Catch bad whitelist.txt entries

As worked out by @User123698745 in
#4792 (comment)

Introduced in 9e24ad8.

* CI: Switch all to ubuntu-24.04-arm runners

In my experience, the ARM runners are measurably faster on compute-heavy
jobs. PHP should work as before. The container build uses QEMU anyways,
so the host CPU architecture only matters in performance.

This also upgrades old Ubuntu 22.04 runners to 24.04.
wrobelda pushed a commit to wrobelda/rss-bridge that referenced this pull request Nov 18, 2025
* CI: Catch bad whitelist.txt entries

As worked out by @User123698745 in
RSS-Bridge#4792 (comment)

Introduced in 9e24ad8.

* CI: Switch all to ubuntu-24.04-arm runners

In my experience, the ARM runners are measurably faster on compute-heavy
jobs. PHP should work as before. The container build uses QEMU anyways,
so the host CPU architecture only matters in performance.

This also upgrades old Ubuntu 22.04 runners to 24.04.
wrobelda pushed a commit to wrobelda/rss-bridge that referenced this pull request Nov 18, 2025
* CI: Catch bad whitelist.txt entries

As worked out by @User123698745 in
RSS-Bridge#4792 (comment)

Introduced in 9e24ad8.

* CI: Switch all to ubuntu-24.04-arm runners

In my experience, the ARM runners are measurably faster on compute-heavy
jobs. PHP should work as before. The container build uses QEMU anyways,
so the host CPU architecture only matters in performance.

This also upgrades old Ubuntu 22.04 runners to 24.04.
wrobelda pushed a commit to wrobelda/rss-bridge that referenced this pull request Nov 18, 2025
* CI: Catch bad whitelist.txt entries

As worked out by @User123698745 in
RSS-Bridge#4792 (comment)

Introduced in 9e24ad8.

* CI: Switch all to ubuntu-24.04-arm runners

In my experience, the ARM runners are measurably faster on compute-heavy
jobs. PHP should work as before. The container build uses QEMU anyways,
so the host CPU architecture only matters in performance.

This also upgrades old Ubuntu 22.04 runners to 24.04.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants