ci: gate area jobs on a changes filter + single ci aggregator - #64
Merged
Conversation
A cheap `changes` job computes which areas a PR touches; the frontend/rust/ desktop jobs gate on its outputs (needs + job-level if), so an untouched area's job no longer spins up a runner just to skip its steps (notably the Windows one). Since a skipped job doesn't satisfy a required status check, add a `ci` aggregator (always() + needs all) that fails only on a real job failure — point branch protection at that single context instead of the individual jobs.
loss-and-quick
force-pushed
the
ci/gate-job
branch
from
June 21, 2026 14:36
5a710a8 to
a515d4c
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Stop spinning up runners for areas a PR doesn't touch. Today every job runs on
every PR (to keep the required checks reporting) and only its steps skip — so an
irrelevant
rust/desktop-windowsjob still starts a runner (a Windows one) justto no-op.
New shape:
changesjob computesfrontend/rust/desktopflags from thediff (the detect logic, previously copy-pasted into each job, now lives once);
needs: changes+if), so an untouchedarea's job is genuinely skipped — no runner;
ciaggregator (if: always(),needs:all) is the one job that alwaysruns and fails only on a real failure/cancel (skipped = fine).
Rollout (required — read before merging)
A skipped job doesn't satisfy a required status check, so the single required check
must become
ci. Sequence:.github/, so the area jobs skip — may need abypass merge under the current ruleset).
mainrequired checks:frontend/rust/desktop-linux/desktop-windows→ci.mainso their runs emit thecicontext.Affected layer
.github/Verification
actionlint .github/workflows/ci.ymlNotes for reviewers
Trade-off: heavy jobs now wait on
changes(~10–20s added wall-clock on relevantPRs) in exchange for not starting runners on irrelevant ones, and a single,
maintainable required check.