Repository navigation
Regenerate the stdlib map on a weekly schedule - #61
Merged
Merged
Conversation
This workflow only ran when somebody dispatched it, so the map is correct about a new Julia release only if a person happens to act. Julia 1.13.1 was released on 2026-09-26, having bumped LLD_jll and libLLVM_jll to 20.1.8+2 and LibSSH2_jll to 1.11.104+0. The map had last been regenerated on 2026-09-08, so it was wrong about 1.13.1 from the moment that version existed, and would have stayed wrong for as long as nobody looked. It was fixed two days later because a downstream consumer pinning this package resolved 1.13.1 against 1.13.0's versions, noticed, and dispatched this workflow by hand. A Monday run bounds that: wrong for at most a week, and nobody has to think of it. The generator is unchanged, and create-pull-request updates the existing bot branch rather than opening a second one, so a week with no new Julia costs a no-op run. Two things this does not fix. The bot's pull requests still cannot start CI, since GITHUB_TOKEN-created ones never do, so each still wants a close/reopen to run check_hsg. And releasing is still manual -- the version bump and registration are not automated here -- so this shortens the gap between a Julia release and a pull request, not between a Julia release and a registered version. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_011hjZ6pHxaSNseY8pa5SKBx
Member
Author
|
@IanButterworth, what do you think — is this a good idea? |
Member
|
SGTM. Might be worth adding to the release todo list too |
Member
Author
|
Would you merge this if it looks good and then make that change? Or do you want me to ask my 🤖 to do it? |
Member
|
I don't know where the release TODO is. @ararslan @KristofferC what do you think about adding regenerating this once a release is out?
A lot of that is obvious, just listing so the reality of the task is clear. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This workflow only runs when somebody dispatches it, so the map is correct about a new Julia release only if a person happens to act.
Julia 1.13.1 was released 2026-09-26, bumping
LLD_jllandlibLLVM_jllto 20.1.8+2 andLibSSH2_jllto 1.11.104+0. The map had last been regenerated 2026-09-08, so it was wrong about 1.13.1 from the moment that version existed — and would have stayed wrong for as long as nobody looked.It got fixed two days later, but only by accident: a downstream consumer (Resolver.jl, which pins this package to answer "what does Julia X bundle") resolved 1.13.1 against 1.13.0's versions, its CI caught the mismatch, and I dispatched this workflow by hand. That became #60, then 2.0.8.
A Monday cron bounds the staleness at a week and means nobody has to think of it:
The generator is untouched.
create-pull-requestupdates the existingbot/update-historical-stdlibsbranch rather than opening a second one, so a week with no new Julia release costs a no-op run and no PR churn.What this doesn't fix
Worth being explicit, since a schedule can look like more of a solution than it is:
GITHUB_TOKEN-created PRs never trigger workflows — the workflow's own PR body says so — so each still wants a close/reopen to runcheck_hsg. Solving that properly needs a PAT orpull_request_target, which felt like a separate decision rather than something to bundle in here.Happy to drop the cron to monthly, or to pair it with an automated version bump, if either fits better with how you'd rather run releases here.
🤖 Generated with Claude Code
https://claude.ai/code/session_011hjZ6pHxaSNseY8pa5SKBx