Files
clide/ci/test_core.sh
T
jpmschweitzerandClaude Opus 4.8 ceaa8392d7 serialize real-pql tests to fix parallel db contention
The pql integration tests (test/pql/client_test.dart) and pql.* daemon
handler tests (test/daemon/pql_commands_test.dart) each spawn a real
`pql` process against the shared on-disk .pql/pql.db. In the parallel
test pool, concurrent invocations contend for the SQLite lock and flake
with PqlException(69) (db busy) — surfaced reliably by the pql 1.10
record_id migration. They pass one-at-a-time, so isolation is the fix.

Tag both files @Tags(['serial']) and add a --concurrency=1 serial pass
to ci/test_core.sh (pty + serial together), mirroring ci/test.sh's
existing serial handling. The error-path companion uses a fake binary,
so it stays parallel.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-10 10:01:15 +02:00

68 lines
2.6 KiB
Bash
Executable File

#!/usr/bin/env bash
# ci/test_core.sh — run the Flutter-free core Dart tests.
#
# Covers `test/` at the repo root (IPC, daemon, PTY). Wraps `dart test`
# in a hard timeout + process-group kill so a hanging test (typically
# one holding a native fd open) can't wedge CI or pre-push.
#
# Rationale: D-030 makes tests client-side only; a hang here is always
# local — either a real bug or a bad test. Either way we'd rather fail
# loudly at 120s than block a pre-push indefinitely.
set -euo pipefail
HERE="$(cd "$(dirname "$0")" && pwd)"
ROOT="$(cd "$HERE/.." && pwd)"
cd "$ROOT"
if ! command -v dart >/dev/null; then
echo "test-core: dart not on PATH; is Flutter installed?" >&2
exit 2
fi
# Hard timeout (seconds). The PTY tests should finish in <5s; IPC/daemon
# tests are faster still. 120s is generous for CI warmup, tiny for a
# hang.
TIMEOUT_SECONDS=${TIMEOUT_SECONDS:-120}
# failures-only: print failing tests + a final count, not one line per test.
# Override with TEST_REPORTER=expanded when debugging. (T-242)
REPORTER="${TEST_REPORTER:-failures-only}"
# Run dart test in its own process group so we can kill descendants on
# timeout. `setsid` starts a new session; `timeout --kill-after` SIGKILLs
# after SIGTERM if the test ignores it.
CORE_DIRS="test/ipc test/pty test/daemon test/git test/panes test/files test/editor test/pql"
# Run a `dart test` pass under the hard timeout + process-group kill.
run_pass() {
if ! timeout --kill-after=5s "${TIMEOUT_SECONDS}s" \
setsid --wait dart test -r "$REPORTER" "$@" ; then
rc=$?
if [[ $rc -eq 124 ]]; then
echo "test-core: TIMEOUT — killing descendants" >&2
pkill -9 -f "dart test" 2>/dev/null || true
exit 1
fi
exit $rc
fi
}
# Some core tests must not share the parallel pool:
# - pty: spawn real PTYs (posix_openpt + posix_spawn) + a reader isolate;
# in the parallel pool they contend for fds + CPU and the isolate starves,
# flaking 'write sends keystrokes to child' (T-292).
# - serial: spawn real `pql` processes against the shared on-disk
# `.pql/pql.db`; concurrent invocations contend for the SQLite lock and
# flake with PqlException(69) (db busy). (T-193, surfaced by the pql 1.10
# record_id migration.)
# Run both in one --concurrency=1 pass (matching ci/test.sh's serial handling),
# then everything else in parallel.
echo "test-core: dart test (pty + serial; --concurrency=1) (timeout ${TIMEOUT_SECONDS}s)"
run_pass --concurrency=1 --tags "pty || serial" $CORE_DIRS
echo "test-core: dart test (rest; parallel, excludes pty + serial) (timeout ${TIMEOUT_SECONDS}s)"
run_pass --exclude-tags "pty || serial" $CORE_DIRS
echo "test-core: ok"