Skip to content

fix(ci): full Playwright suite must not echo grep="(full)" - #1301

Merged
joepio merged 1 commit into
developfrom
cursor/fix-e2e-full-grep-quote-b7ed
Aug 21, 2026
Merged

fix(ci): full Playwright suite must not echo grep="(full)"#1301
joepio merged 1 commit into
developfrom
cursor/fix-e2e-full-grep-quote-b7ed

Conversation

@joepio

@joepio joepio commented Aug 21, 2026

Copy link
Copy Markdown
Member

Related Issues

Unbreaks develop CI after #1286 (and every develop push since the Playwright light/heavy split). The product change in #1286 is not the failure.

https://github.com/ontola/atomic-server/actions/runs/32510577920

What failed

develop runs Playwright full. Dagger logged the mode with:

echo "e2e mode grep="(full)" shard=1/4 ..."

JSON.stringify('(full)') put quotes inside a double-quoted bash string, so (full) became a subshell:

syntax error near unexpected token `('

The suite never started. Feature-branch CI stayed green because light mode greps @smoke, which is not a bash metacharacter.

The same quoting bug has been failing develop since that split landed (including #1294 and #1295).

Fix

Log grep=full without extra quotes or parentheses. --grep is still omitted for the unfiltered suite.

Checklist

  • Add or update tests if needed — bash -n on the full and light shard scripts; the broken echo reproduces the CI error
  • Update docs if needed
Open in Web Open in Cursor 

JSON.stringify('(full)') inside a double-quoted bash echo closed the
quote, so `(full)` became a subshell and every shard died before tests
started. Light CI hid it because `@smoke` is not a metacharacter. This
is why develop has been red since the light/heavy split, including the
#1286 merge.

Co-authored-by: joepmeindertsma <joepmeindertsma@gmail.com>
@joepio
joepio marked this pull request as ready for review August 21, 2026 19:24
@joepio
joepio merged commit 4561062 into develop Aug 21, 2026
6 checks passed
@joepio
joepio deleted the cursor/fix-e2e-full-grep-quote-b7ed branch August 21, 2026 19:25
joepio added a commit that referenced this pull request Aug 21, 2026
…2e] (#1302)

saveOffline treated the local genesis cursor as a last-synced baseline.
Reconnect POSTed genesis, then the empty follow-up export looked like
"OPFS not ready" and the outbox retried forever — pendingDirtyCount
stuck at 1, hasSignedGenesis false. That is the develop full-e2e
failure of offline-create-then-online after #1301 unblocked the suite.

Co-authored-by: Cursor Agent <cursoragent@cursor.com>
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