safe-global / safe-client-gateway / 28576326996

87%

push

github

chore(tests): reduce CI flakiness (clearMocks + seeded faker) (#3139)

* chore(tests): clear mocks and seed faker between runs to reduce flakiness

Enable clearMocks in all jest configs so recorded mock calls do not leak
between tests, and add a per-file faker seed (logged, overridable via
FAKER_SEED) so the data behind a failing run can be reproduced.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* test(job-queue): wait for terminal events before clearing state

The multi-job test only waited for handledJobs, but BullMQ emits the
completed/failed worker events asynchronously afterward. afterEach cleared
the consumer's arrays before those events arrived, so a prior job's
completed event leaked into 'should handle job failure' and broke its
completedJobs.toHaveLength(0) assertion. Drain unstarted jobs, then wait
until every handled job has emitted its terminal event before clearing.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* fix(transactions): guard getAddress in address-info index

getAddress throws synchronously on a malformed string, before the promise
(and its .catch) exists. A decoded parameter typed 'address' but carrying a
non-address value therefore surfaced as a 500 in transaction preview. Guard
with isAddress (non-strict, matching getAddress's case-insensitive acceptance).

This removes the intermittent 500 in the kiln preview integration test and
the same latent flake in any spec that decodes randomly-built data.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* chore(tests): add SPDX license header to faker-setup

The license-headers CI job (pre-commit insert-license) failed on the PR
because the new test/faker-setup.ts lacked the required header.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* fix(cache): floor Redis TTL to integer seconds

deviateRandomlyByPercentage returns value+delta; delta is integer but value
may be fractional, so t... (continued)

3514 of 4302 branches covered (81.68%)

Branch coverage included in aggregate %.

2 of 2 new or added lines in 2 files covered. (100.0%)

9295 of 10497 relevant lines covered (88.55%)

490.84 hits per line