The five internal/app partitions ran end to end inside one job, so the app
shard's wall clock was the sum of all five: 780s in CI, of which the schema
partition owned 357s. Each partition is now its own matrix shard, so they run
concurrently and the shard's wall clock is set by its slowest partition rather
than by their total. Every partition shard still selects the same single
internal/app package, so the impacted-package query maps the shard name back to
app and the partition only chooses which tests run.
The helper gains a partition argument and a list-partitions mode. APP_PARTITIONS
is the single source of truth for the set, and the discovery pass still runs in
every job, so each one independently verifies that the partition patterns cover
every top-level test exactly once before running the one it was asked for.
Two fail-closed checks guard the split, because the helper's own coverage check
can no longer prove the whole package ran once the partitions are separate jobs:
- The helper cross-checks APP_PARTITIONS against the coverage counters in both
directions, so a counted partition that nothing dispatches and a dispatchable
partition with no counter both fail instead of silently skipping tests.
- TestCIAppRacePartitionMatrixMatchesHelper pins the workflow's app-<partition>
shards to list-partitions output in both directions, so a partition cannot
lose its job while every job stays green.
The discovery loop variable is renamed from partition to spec: it would
otherwise shadow the partition requested on the command line, which run mode
reads after the discovery pass completes.
The workflow contract pinned the focused path by literal: the job name
`Test (changed packages)`, the unsharded
`list "$TEST_BASE_REF" "$TEST_HEAD_REF"` call, and a single
`go test -timeout=15m` line standing in for internal/app's package-level
headroom. Sharding the job changed all three literals, so `Test (workflow
and release contracts)` failed on this branch even though every shard
selection test passed.
Each invariant the contract guarded still holds, so the assertions are
updated to the new shape rather than relaxed:
- the focused job must still exist, now as the matrix job, named the way
the contract already names `Test (race: ${{ matrix.shard }})`;
- package selection must still derive from the authoritative synthetic
merge base/head, now with an explicit shard argument, so pointing it at
any other ref still fails the contract;
- internal/app's headroom is asserted through the process-isolating
helper and the per-shard budgets, mirroring the assertions already
applied to test-race. That is stronger than the old single -timeout: it
pins the mechanism that keeps the suite inside its budget rather than
the number alone. release-scripts membership is asserted too, because
its dedicated job only runs at full-suite or release-sensitive scope,
so losing it here would silently stop testing test/scripts changes.
The shard comparisons in the focused job are quoted so that job reads
verbatim like test-race's.
Ablating the implementation one change at a time turns the contract red
in all five cases: removing the app helper call, dropping release-scripts
from the matrix, selecting from HEAD~1, collapsing the matrix back to a
single unsharded job, and dropping the cli/smoke timeout budget.
release_version was interpolated into an awk regex, where '.' matches any
character. Version 1.0.1-beta.1 therefore also admitted
.changes/released/1x0x1-betaX1/, letting the archive drift from the
CHANGELOG version while every other seal assertion still passed and
breaking the documented audit trail.
Compare the archive prefix with index() and split the basename off with
substr(), matching the literal-comparison idiom already used throughout
check-changelog-pr.sh. Only the basename, whose character class is fixed,
stays a pattern.
Git records no diff entry for a directory itself, so adding
.changes/foo/bar.md only surfaced the nested path, which the single-level
trigger regex skipped. The entry validation and the renderer were both
bypassed, letting a nested directory reach main and break every later
fragment render with 'unexpected directory'.
Trigger the top-level tree validation on any .changes change outside
.changes/released/ (which keeps its own immutability and release-seal
checks), and assert .changes itself is still a tree so replacing it with a
blob or symlink cannot empty the child listing unnoticed.
Re-rendering stays keyed on fragment changes so a README-only edit does
not fail on an empty fragment set.
The fragment gate only ran validation when the changed path matched the
legal fragment name pattern, so `.changes/Foo.md`, `.changes/notes.txt`
and a symlinked fragment slipped through untouched and then broke the
next PR that added a legal fragment. The trigger now fires on any
top-level `.changes/` change other than README.md and rejects every
entry that is not README.md, released/, or a 100644 blob named
^[a-z0-9][a-z0-9._-]*\.md$.
The renderer had the same hole from the other side: `find -type f`
is false for symlinks, so a symlinked fragment was silently dropped
from the rendered notes, and the `[a-z0-9]*.md` glob only constrained
the first character so `chat reply.md` passed. It now walks every
top-level entry and fails on symlinks, unexpected directories,
non-regular files and illegal names. Both scripts pin LC_ALL=C so the
ASCII ranges cannot match uppercase under a different collation.
Adds regression coverage for illegal names, non-markdown entries,
symlinks and executable modes on both the gate and the renderer.
Drop Windows-only DPAPI export/import names from the darwin -run filter;
those tests skip on macOS and already run under the Windows job.
Co-authored-by: Cursor <cursoragent@cursor.com>
Keep the darwin workflow contract from re-accepting the merged
keychain+auth+app invocation that main briefly required.
Co-authored-by: Cursor <cursoragent@cursor.com>
Port OA approval form-schema / forecast-process / create-instance from
main via DeclareLeafMetadata and mapping-ledger exclusions; keep retired
schema pin paths deleted and preserve the CI auth/keychain split.
Co-authored-by: Cursor <cursoragent@cursor.com>
Resolve the macOS race budget conflict in favour of the focused scope.
c1f96241 on main extended the whole-package macOS race step from 10m to
12m and pinned that budget in the workflow contract. This branch removes
the whole-package run instead: ./internal/app is already covered by the
Ubuntu "race: app" shard, and macOS only needs the natively-gated tests.
With the focused scope the job drops from 10m47s to ~3m, so the 12m
budget is no longer needed and the two step timeouts (6m + 5m) fit inside
the 15m job budget with headroom.
The contract assertion c1f96241 added is superseded rather than dropped:
pinning both focused commands locks the per-step timeouts, and the
existing checks still block a whole-package regression and require
./internal/app to appear exactly once.
Narrowing the macOS internal/app -run pattern orphaned
TestValidateNewBinary_RecoversFromUnsignedDarwin: it is the only
runtime.GOOS != "darwin" gated test in the package, the Ubuntu race shard
skips it on Linux, and the platform coverage gate only runs
^(TestAllShortcuts|TestCrossPlatformCoverage). No CI job selected it any
more, so it could never run or fail again.
- Add the self-heal test back to the macOS -run pattern.
- Add the (CrossPlatformCoverage)? group the Windows pattern already has,
which also recovers TestCrossPlatformCoverageAuthMigrateKeychainRemainingBranches.
- Attribute vacuous runs: the two skip paths now name the branch that went
unverified, and DWS_REQUIRE_AMFI_SELF_HEAL=1 escalates such a run to a
hard failure on a host that does enforce amfid. GitHub's hosted macOS
runners do not reproduce the amfid kill, so this test has been silently
skipping there all along.
- Restore step-timeout headroom: 10m + 5m exactly equalled the 15m job
budget, leaving none for setup. The keychain/auth step drops to 6m
(measured 2m43s).
- Add a contract test that couples the macOS -run pattern to the set of
darwin-gated tests in internal/app, so the next narrowing fails loudly
instead of silently orphaning one.
race:remaining was SIGTERM'd (exit 143) mid test/smoke after mock_mcp with
no FAIL/DATA RACE; NewRootCommand public-tree smoke under -race is too heavy
to share that shard. Mirror the cli split and give smoke a 15m budget.
Co-authored-by: Cursor <cursoragent@cursor.com>
Pin admission race shards to timeout_budget (12m/cli 15m), cover
runtimeannotate and schema_source_root success paths for platform/main
gates, and make Windows absolute catalog path checks platform-safe.
Co-authored-by: Cursor <cursoragent@cursor.com>
chichuan