Commit Graph

43 Commits

Author SHA1 Message Date
gsxdsm
1e81f99f72 FN-6877: stabilize dashboard changed-mode affected runs
Stabilize changed-mode affected runs by giving dashboard its own memory envelope.

- Generalize scoped affected memory-envelope handling beyond the engine lane.
- Run @fusion/dashboard changed-mode Vitest scopes in a 6144 MB, single-worker envelope.
- Cover dashboard and engine scoped partitioning/env contracts with script tests.
- Document the dashboard changed-mode OOM/SIGKILL guard and remeasurement expectations.

Files changed:
 docs/testing.md                         |   6 +-
 scripts/__tests__/test-changed.test.mjs | 112 +++++++++++++++++++++++++++-----
 scripts/test-changed.mjs                |  67 +++++++++++++++----
 3 files changed, 155 insertions(+), 30 deletions(-)

Fusion-Task-Id: FN-6877

Fusion-Task-Lineage: 084df464-ebda-4578-bfac-305800da5b6c
2026-06-21 16:48:20 -07:00
gsxdsm
d57538581d FN-6854: isolate engine affected tests
Stabilize changed-test runs by giving the engine affected Vitest lane its own memory envelope.

- Split @fusion/engine out from other scoped affected-package Vitest runs.
- Apply a capped heap and single-worker settings to the isolated engine lane while preserving watchdog timing.
- Cover the partitioning and environment behavior with script tests and document the envelope.

Files changed:
 docs/testing.md                         |  3 +++
 scripts/__tests__/test-changed.test.mjs | 45 +++++++++++++++++++++++++++++++++
 scripts/test-changed.mjs                | 40 ++++++++++++++++++++++++++---
 3 files changed, 84 insertions(+), 4 deletions(-)

Fusion-Task-Id: FN-6854

Fusion-Task-Lineage: 144acf15-4ccb-4974-b06c-99cdcc2814dd
2026-06-21 13:25:41 -07:00
gsxdsm
ce90cc9b62 fix(FN-5048): keep Fusion test verification bounded 2026-06-21 10:05:12 -07:00
gsxdsm
6e2e5c6b62 WIP: scope affected test runs to changed-related tests (vitest --changed)
Affected packages with a vitest config now run `vitest run --changed <base>`
(only tests in the changed files' module graph) under the tight "changed"
watchdog, instead of each package's ENTIRE suite under a 60-min backstop.
Packages without a vitest config fall back to their full `test` script. The
curated gate suite still runs as the cross-cutting safety net.

Measured: a dashboard hub-component change drops from 822 files (5-8min) to
40 files (~98s wall incl. gate). Common/leaf changes scope to far fewer.
NOT yet <60s for hub-component changes — the fixed jsdom/setup floor still
dominates (needs the overhead-reduction follow-up), and one scoped-run test
failure needs isolation triage before this gates verification.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-19 22:05:09 -07:00
gsxdsm
24ff12471f Stop quarantine-list edits from forcing pnpm test into gate mode
scripts/lib/test-quarantine.json is runtime data (which tests are
quarantined), not executable test infra, but it tripped the shared-infra
catch-all in isSharedInfraChange. That forced mode=gate, which runs only
the fixed engine-core + cli-shape gate suite and returns before the
affected packages -- so a dev's real changes (e.g. @fusion/core,
@fusion/dashboard) got zero coverage whenever they also touched the
quarantine list. Classify the file as test-irrelevant so the diff stays
in changed mode and the affected packages run.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-19 19:05:49 -07:00
gsxdsm
822d5d1872 fix(FN-6587): stabilize test lanes under production env
Normalize spawned Vitest processes to NODE_ENV=test so React/jsdom lanes do not inherit release-shell production mode. Split the compound-engineering plugin Vitest projects so Node-only setup and teardown only run in Node lanes, and quarantine the remaining broad-lane CE timeout flakes under the deletion ratchet.
2026-06-17 17:13:13 -07:00
gsxdsm
2bffad1784 Merge remote-tracking branch 'origin/main' into fix/test-timeout-failures
# Conflicts:
#	scripts/test-changed.mjs
2026-06-15 12:13:40 -07:00
gsxdsm
80fbcdd5a3 FN-6396: harden test worker temp cleanup
Prevent stale Fusion test worker roots from leaking across merge-gate runs.

- Add per-run tokens to worker-root owner markers and pruning checks.
- Remove self-minted fallback worker roots during Vitest exit cleanup.
- Cover stale pid reuse, markerless redir roots, and SIGKILL-style orphan pruning with regression tests.

Files changed:
 packages/core/src/__test-utils__/vitest-setup.ts   | 83 +++++++++++++++++++---
 .../core/src/__test-utils__/vitest-teardown.ts     |  8 ++-
 .../vitest-teardown-worker-root-cleanup.test.ts    | 19 ++++-
 scripts/__tests__/test-changed.test.mjs            | 74 ++++++++++++++++++-
 scripts/test-changed.mjs                           | 76 ++++++++++++++++----
 5 files changed, 233 insertions(+), 27 deletions(-)

Fusion-Task-Id: FN-6396

Fusion-Task-Lineage: 711d966d-c70e-4cd7-81cd-accd18f17202
2026-06-13 16:30:28 -07:00
gsxdsm
e51d05bf1d Merge remote-tracking branch 'origin/main' into fix/test-timeout-failures
# Conflicts:
#	scripts/test-changed.mjs
2026-06-13 15:44:28 -07:00
gsxdsm
a49450ef5e Address PR review feedback (#1669)
- watchdog: escalate forwarded SIGINT/SIGTERM/SIGHUP to SIGKILL after grace so
  external cancellation can't hang for the full budget (coderabbit major)
- watchdog: route onProcExit through signalGroup for injection consistency (greptile)
- watchdog: add cwd option; test-changed passes rootDir so pnpm runs from repo
  root regardless of invocation cwd (coderabbit major — preserved original run() cwd)
- dashboard runner: validate/clamp FUSION_RUN_VITEST_* env so a malformed value
  can't NaN-disable the watchdog (coderabbit)
- tests: verify exit-listener cleanup, forwarded-signal escalation, cwd passthrough
- plan doc: per-class-ceiling fallback wording (not median); label Output Structure fence

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-13 15:21:00 -07:00
gsxdsm
97a49ac196 FN-6382: unquarantine stabilized flaky tests
Restore quarantined tests by fixing their flaky harness seams instead of extending the deletion ratchet.

- Mark active Vitest worker roots and skip live worker roots during prune cleanup.
- Make bubblewrap backend coverage deterministic with an injectable runner and restore it to the engine gate.
- Remove rescued core and bubblewrap tests from the quarantine ledger and Vitest excludes.

Files changed:
 .../core/src/__test-utils__/vitest-teardown.ts     | 10 +++++-
 packages/core/vitest.config.ts                     |  8 +----
 .../__tests__/sandbox/bubblewrap-backend.test.ts   | 29 +++++++++------
 packages/engine/src/sandbox/bubblewrap-backend.ts  |  9 +++--
 packages/engine/vitest.config.ts                   |  1 -
 scripts/__tests__/test-changed.test.mjs            | 23 ++++++++++++
 scripts/lib/test-quarantine.json                   | 34 ++----------------
 scripts/test-changed.mjs                           | 41 ++++++++++++++++++++++
 8 files changed, 102 insertions(+), 53 deletions(-)

Fusion-Task-Id: FN-6382

Fusion-Task-Lineage: 018dc7ac-1ef1-495e-a5fd-96ea44fcd43b
2026-06-13 11:39:37 -07:00
gsxdsm
73be9f8c5d FN-6371: retry stale test worker pruning
Adds bounded retry handling so stale Fusion test temp roots are reclaimed instead of leaking after transient removal failures.

- Share prefix-scoped pruning between isolated HOME and worker temp roots.
- Retry rm-rf on transient busy/non-empty failures and treat ENOENT as successful cleanup.
- Warn once with bounded child diagnostics after persistent prune failures.
- Cover worker and home pruning retry, failure, and ENOENT behavior in script tests.

Files changed:
 scripts/__tests__/test-changed.test.mjs | 118 ++++++++++++++++++++++++++++++++
 scripts/test-changed.mjs                |  84 ++++++++++++++---------
 2 files changed, 168 insertions(+), 34 deletions(-)

Fusion-Task-Id: FN-6371

Fusion-Task-Lineage: a81e6849-5eb0-4bc0-8030-9f8eaed4e35a
2026-06-13 09:49:23 -07:00
gsxdsm
10972bbdce FN-6360: clean up leaked test worker temp roots
Ensure test isolation removes stale worker temp roots after interrupted or busy Vitest runs.

- Add bounded retry cleanup for Vitest worker roots during teardown.
- Prune orphaned fusion-test-workers-* directories before changed-test isolation checks.
- Cover worker-root retry and pruning behavior with targeted tests.

Files changed:
 .../core/src/__test-utils__/vitest-teardown.ts     | 50 ++++++++++--
 .../vitest-teardown-worker-root-cleanup.test.ts    | 88 ++++++++++++++++++++++
 scripts/__tests__/test-changed.test.mjs            | 33 ++++++++
 scripts/test-changed.mjs                           | 32 ++++++++
 4 files changed, 196 insertions(+), 7 deletions(-)

Fusion-Task-Id: FN-6360

Fusion-Task-Lineage: d537941d-dd58-403a-a82e-f0aeee9c1eb0
2026-06-13 08:11:46 -07:00
gsxdsm
414e1a3bfb feat(test-infra): run changed-file test invocations under the watchdog (U1)
Convert test-changed.mjs's invocation path from blocking spawnSync to async
spawn through the shared watchdog. runWatchedTest throws on failure/timeout with
.exitCode so the existing catch and the runMaybeIsolated finally (prune +
isolation post-check) still run — a watchdog kill reaps any leaked isolated HOME
instead of leaving it for the guard to flag. Full/affected runs use a generous
60min backstop (dashboard lanes are already inner-watchdog'd); the quick gate
uses the changed-class ceiling.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-13 00:43:38 -07:00
gsxdsm
5aa6cad168 Address PR review feedback (#1453)
- boot-smoke: shutdown verdict now requires SIGTERM actually delivered and a
  clean exit (code 0 or SIGTERM); EADDRINUSE port race retries with a fresh
  port (3 attempts)
- test-changed: rename shouldForceFullSuite -> isSharedInfraChange (it routes
  to gate mode, not full); run the changed-mode gate under the isolation guard
- workflows: least-privilege permissions (contents: read) on pr-checks and
  full-suite
2026-06-05 13:58:34 -07:00
gsxdsm
3387ac8096 fix(review): apply autofix feedback 2026-06-05 09:23:15 -07:00
gsxdsm
5a4bb9eeeb feat(test): pnpm test runs gate + affected set; remove implicit full-suite escalation (the local OOM path)
- decideExecutionPlan: implicit wide-blast reasons (missing base, diff failed,
  no changes, shared infra, unmapped package) route to new gate mode instead of full
- CI===true force-full branch removed (CI no longer calls test-changed.mjs)
- changed mode runs pnpm test:gate before the affected set
- full suite reachable only via explicit --full / FUSION_TEST_FULL=1
- characterization tests updated + new only-explicit-full invariant test
2026-06-05 09:08:20 -07:00
gsxdsm
610d473e77 refactor(test): simplify-pass cleanups from 4-angle review
- content-hash: createRepoContentSnapshot — 2 repo-wide git spawns shared across all hash computations (was ~2 spawns x N packages x 2 passes, ~0.6-1.6s per cache-miss run); snapshot-equivalence test pins zero-spawn path
- test-changed: one hash memo + snapshot shared between applyCacheToPlan and recordCachePass (record pass now re-hashes nothing)
- ci-test-shard: listPackageTestFiles single source of truth for the test-file glob (was triplicated)
- check-test-inventory: curatedProjects defaults to projects (removes duplicated 11-entry list in spec)
- ensure-test-artifacts: drop detectMissingArtifacts passthrough alias
- drop dead statSync re-export; cross-reference comments on the two shared-input path lists

Skipped deliberately: --cold-start-probe/--check-timings-staleness removal (plan artifacts for U8 re-eval + scheduled refresh), best-fit dedup + threshold² (behavior-preserving refactor of verified shard math — follow-up), worker-budget duplication (pre-existing on main)
2026-06-03 19:12:28 -07:00
gsxdsm
2c41695e1f fix(test): dependency-aware and dirty-aware test cache invalidation
- cache key v2 = own hash + sorted transitive workspace dep hashes + shared inputs (lockfile, tsconfig.base, core __test-utils__ tree)
- working-tree-dirty files hashed by content (fixes false cache HIT on unstaged edits)
- core __test-utils__ folded globally: 16+ packages import it without a workspace dep on core
- dep folding adds ~5ms to the inner loop (memoized own-hashes)
- docs: cache semantics, --no-cache / FUSION_TEST_NO_CACHE, TTL rationale
2026-06-03 18:06:05 -07:00
gsxdsm
211c0fd557 perf(test): cut inner-loop fixed overhead to sub-second on cache-fresh runs
- skill-sync check conditioned on content hash of its inputs (skips ~0.3s spawn)
- ensure-test-artifacts: git-blob content-hash staleness; branch switches no longer trigger spurious ~2.6s tsc rebuilds (mtime fallback when dirty)
- isolation guard: cheap --before-fast reusing prior post-run baseline (~2.1s -> ~0.07s); detection proven preserved via injected-leak failure test
- vitest-setup: CI skips 4040-4045 discovery probe unless FUSION_RESERVED_PORTS set; kill-guard wrapper untouched, asymmetry pinned by port-probe-policy tests
- cache-fresh fast path skips sync/artifacts/HOME-prune entirely (mode line: fast-path=cache-fresh)
2026-06-03 17:57:29 -07:00
gsxdsm
c69d384e67 perf(test): add timing telemetry, cold-start probe, and baseline snapshot
- ci-test-shard.mjs: --write-timings aggregation into scripts/test-timings.json (bucketed, newer-snapshot-protected, corrupt-shard tolerant) and --cold-start-probe; CI shard invocations emit vitest json timings
- test-changed.mjs: structured mode/reason telemetry line (+ --print-mode)
- pr-checks.yml: upload per-shard timing artifacts
- baseline: docs/test-speed-baseline-2026-06-03.md (core 41s, engine 179s, cli 49s; cold-start ~1.3-1.8s/process => U8 gate: worthwhile-not-urgent)
2026-06-03 16:48:50 -07:00
Fusion (runfusion.ai)
82c8bced60 feat(FN-5144): add create-pr metadata and preflight routes
Added GitHub create-PR metadata and preflight routes to the dashboard, wiring up `submit-github.pr-options-preflight-metadata` contract and integration tests for the new preflight/confirm flow alongside supporting test configuration updates.

Fusion-Task-Id: FN-5144
2026-05-19 10:33:15 -07:00
Fusion (runfusion.ai)
15bb84f862 feat(FN-5124): add soft-delete QA boundary regression coverage
Adds a regression test for FN-5124 covering soft-delete QA boundary cases, targeting the engine reliability interaction suite. The single new test file exercises edge conditions in the soft-delete path to guard against future regressions.

Fusion-Task-Id: FN-5124

Fusion-Task-Lineage: 7166e252-9c98-4f51-956a-1402dae6f402
2026-05-19 05:52:09 -07:00
gsxdsm
ae3fb0d991 fix(FN-3947): strip forwarded vitest silent flags 2026-05-10 15:33:25 -07:00
Fusion
2ba9b00704 feat(FN-3723): harden isolated HOME cleanup and shell connection visibility
- Harden test isolation cleanup retries and allow-list behavior in scripts/test-changed.mjs
- Expand test coverage for cleanup retry logic and allow-list handling in scripts/__tests__/test-changed.test.mjs
- Add shell-native connection status plumbing and UI coverage across dashboard, desktop preload/types, and mobile typings
- Add ShellConnectionStatus and mobile nav wiring plus related app/component test coverage
- Document shell connection contracts and add FN-3568 diagnosis/recovery notes

Fusion-Task-Id: FN-3723
2026-05-08 01:16:54 -07:00
Fusion
34a599f0b6 feat(FN-3725): add --accent-text token per theme for chat message readabili
Adds a `--accent-text` CSS token defined per theme for consistent chat text readability, and highlights the fast mode bolt icon in the task detail modal, with corresponding theme-color and rendering tests.

Fusion-Task-Id: FN-3725
2026-05-08 00:37:39 -07:00
Fusion
9e04467b57 feat(FN-3406): add shell context hook plumbing for dashboard
Adds a new `useShellContext` hook with tests to the dashboard, integrated into `App.tsx` and `Header.tsx` to provide shell context plumbing throughout the UI.

Fusion-Task-Id: FN-3406
2026-05-07 23:21:05 -07:00
gsxdsm
ae97fca340 fix(test-isolation): allow-list known isolated HOMEs to stop FN-3711-style flakes
The leak guard in scripts/check-test-isolation.mjs reported
fusion-test-home-root-* dirs created by scripts/test-changed.mjs as
leaks when transient EBUSY on /var/folders prevented cleanup or the
baseline file got rotated. Track every HOME basename the script mints
and pass it via FUSION_TEST_ISOLATION_IGNORE_NAMES so the check
allow-lists them unconditionally. Surface cleanup rm failures via
console.warn instead of swallowing.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-07 12:25:45 -07:00
Aron Prins
aaa4c59c54 test(FN-000): include reverse dependents in changed tests 2026-05-07 18:35:32 +02:00
Aron Prins
10ebfea17f test(FN-000): fix changed-test package cache dirs 2026-05-07 18:26:48 +02:00
Aron Prins
1ed390052c test(FN-000): address PR review feedback 2026-05-07 17:32:21 +02:00
Fusion
f3c99eb4fa feat(FN-3577): add native shell guide and bridge contract documentation
Adds a native shell integration guide (FN-3577) with fusionShell bridge contract documentation, wired into the docs navigation, plus a regression test; also hardens the dependency graph's task filtering (FN-3085) to include in-review tasks and handle orphan edges, with corresponding test coverage ac

Fusion-Task-Id: FN-3577
2026-05-07 03:59:43 -07:00
Fusion
fa57034c05 feat(FN-3632): remove quick-entry branch handling from dashboard and docs
Removes the `QuickEntryBox` component and its styles as part of FN-3632 cleanup, with its tests refactored accordingly. Also updates the test-changed script and fixes an isolated test home cleanup ordering issue.

Fusion-Task-Id: FN-3632
2026-05-07 01:30:34 -07:00
Fusion
fb967fe87e feat(FN-3081): replicate task creation across mesh nodes
- Add core mesh task replication flow and task metadata needed for cross-node create propagation
- Wire dashboard task creation routes and mesh routes to trigger/handle replicated task creation
- Prune isolated test homes before isolation checks to avoid false failures in changed-test runs
- Expand core, dashboard, CLI extension, and remote-node client tests; update architecture and task-management docs

Fusion-Task-Id: FN-3081
2026-05-07 00:28:52 -07:00
Fusion
6de69a2ced feat(FN-3449): implement distributed task-id allocator with mesh API endpoi
Implements a distributed task-ID allocator with mesh networking endpoints for multi-project coordination (FN-3449), alongside substantial ChatView and GitManager mobile polish including improved iOS keyboard handling, rubber-band containment, scroll-drift fixes, and visual viewport stabilization. In

Fusion-Task-Id: FN-3449
2026-05-06 21:16:33 -07:00
Fusion
4d39e577fd feat(FN-3615): fix Codex weekly pace parsing
Fixed Codex weekly pace parsing in the usage tracking module, with corresponding test coverage added to verify the fix.

Fusion-Task-Id: FN-3615
2026-05-06 20:54:33 -07:00
Fusion
e7884370f2 fix(FN-3610): isolate test home for changed-package test runs
- Create a disposable HOME/USERPROFILE env for scripts/test-changed.mjs execution
- Run isolation guard checks with the same isolated env, including cache-hit no-op paths
- Clean up temp HOME after test execution to avoid residue
- Update test coverage and contributing docs for the shared isolation behavior

Fusion-Task-Id: FN-3610
2026-05-06 18:57:37 -07:00
Fusion
f6394cc22b test(FN-3607): harden test workflow verification and isolation checks
- Expand test-changed coverage and shard selection assertions for CI workflows
- Improve vitest worker temp-directory utilities and related core/CLI tests
- Refine test isolation guardrails and runtime ignore handling for live .fusion noise
- Update contributing guidance and root test script usage for the verified workflow

Fusion-Task-Id: FN-3607
2026-05-06 11:51:48 -07:00
Fusion
b3aa9f9890 feat(FN-3595): document live reviewer override behavior in settings and tas
Documents the live reviewer override behavior in the settings reference and task management guides, adding two lines to each file for a total of 4 lines of documentation.

Fusion-Task-Id: FN-3595
2026-05-06 10:05:40 -07:00
Fusion
7d02ac81ef feat(FN-3593): add test isolation CI enforcement, fix stuck-requeue race, a
This merge lands five FN-3593 commits establishing a test isolation contract with a new `scripts/check-test-isolation.mjs` guard that scans for accidental `beforeEach`/`afterEach`/`beforeAll`/`afterAll` in setup helpers, plus per-package `setup-test-isolation.ts` bootstraps that canonicalize the pat

Fusion-Task-Id: FN-3593
2026-05-06 09:43:51 -07:00
Fusion
61bf81d6b8 feat(FN-3470): prefer source aliases for runtime plugins and add test artif
Merges FN-3470 in two steps: first, runtime plugins now prefer source path aliases over build artifacts, with hardening tests for vitest alias resolution and plugin view registry updates; second, a new `ensure-test-artifacts.mjs` bootstrap script guarantees required test files exist before test runs

Fusion-Task-Id: FN-3470
2026-05-05 13:25:02 -07:00
Fusion
c9e776cb34 feat(FN-3293): add stabilization docs to test audit report
Documentation for test stabilization was finalized by updating the test audit report with 2 additional lines.

Fusion-Task-Id: FN-3293
2026-05-04 13:17:57 -07:00
Fusion
4d98b12243 feat(FN-3234): refactor test-changed script and add workflow tests
This merge refactors the test-changed script and adds new CLI tests (FN-3234), keeping automation on the full-suite path. It includes a new test file for the root test command, updates to the CI workflow, and documentation improvements.

Fusion-Task-Id: FN-3234
2026-05-03 05:47:11 -07:00