fix(ci): hoist effect-language-service prepare hook to root (#2497)#79
Open
aaditagrawal wants to merge 4 commits intomainfrom
Open
fix(ci): hoist effect-language-service prepare hook to root (#2497)#79aaditagrawal wants to merge 4 commits intomainfrom
aaditagrawal wants to merge 4 commits intomainfrom
Conversation
…pingdotgg#2497) Co-authored-by: Peter van der Heijden <contact@peterjarian.dev>
📝 WalkthroughWalkthroughCentralizes the ChangesDependency Management & Workspace Scripts
Server Persistence, Migrations & Projection Query
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
github-actions
Bot
added
vouch:trusted
Pre-existing format drift on fork files that don't get linted by upstream formatters (provider drivers, adapters, text generation, kiloServerManager tests, migration test, settings.ts, Icons.tsx). Caught by the now-unblocked CI run since the bun install hang was fixed. No semantic changes — purely whitespace / line wrapping per oxfmt.
github-actions
Bot
added
size:L
PR #79's install fix made these visible. Five distinct issues, all from PR #77's manual reconciliation of upstream's multi-provider SPI: 1. ProjectionSnapshotQuery.listThreadRows had a fork-added WHERE clause filtering by `json_extract(model_selection_json, '$.provider')` — but upstream's new SPI stores `$.instanceId` instead. The filter dropped every thread silently, so projection_threads was always empty in the snapshot and every integration-engine test hung waiting for projections that would never arrive. Removing the filter (matches upstream). 2. Migration test bounds were off-by-two. Fork's filenames 026/027/028/029 stayed put but were registered as IDs 28/29/30/31 (because fork's old 023/024 displaced them). The tests still passed numeric IDs matching the filenames, so they ran the canonicalize migration before its deps existed and blew up with `projection_thread_sessions has no column named provider_instance_id`. 3. CodexAdapter native-flush regex /NTIVE: .../ was missing the leading `A` (re-introduced by the merge from a stale upstream revision; we already fixed it once in PR #75 review). Restoring /NATIVE: .../. 4. ProviderRegistry.test asserted only the 4 upstream providers, but BUILT_IN_DRIVERS now contains all 8 (incl. fork's amp/copilot/ geminiCli/kilo). Updated the expected list. Local: `bunx vitest run integration/orchestrationEngine.integration.test.ts` goes from 11 failed/1 skipped → 11 passed/1 skipped.
- 021_Repair: cap second `runMigrations()` at id 24 (the repair migration itself). Running the rest of the chain hits migration #28 (CanonicalizeModelSelectionOptions, file 026) which references `model_selection_json` — but that column is added by migration #16 which the test pre-marks as 'ran' without actually executing. - GeminiCliAdapter delegate test: the local `manager` was being asserted on, but a different manager instance was being provided to the layer. Adapter was talking to the layer's manager; assertion reads the body's manager. Pass the same manager in both places. - GeminiCliAdapter forwards-events test: same bug, same fix.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
🧹 Nitpick comments (1)
apps/server/src/persistence/Migrations/029_BackfillForkProviderInstanceIds.test.ts (1)
282-380: ⚡ Quick winIdempotency test does not cover
provider_session_runtime.The test seeds
projection_thread_sessionswithprovider_instance_id = 'amp-custom'and verifies it is preserved after migration 31, but inserts no correspondingprovider_session_runtimerow. This means the "don't overwrite non-NULL" branch of the backfill forprovider_session_runtimeis untested — if migration 31'sprovider_session_runtimeUPDATE is ever changed to omit theWHERE provider_instance_id IS NULLguard, this test would not catch it.♻️ Suggested addition to the idempotency test
yield* sql` INSERT INTO projection_thread_sessions (...) VALUES ('thread-amp-custom', 'running', 'amp', NULL, NULL, NULL, NULL, '2026-01-01T00:00:00.000Z', 'full-access', 'amp-custom') `; + yield* sql` + INSERT INTO provider_session_runtime ( + thread_id, provider_name, adapter_key, runtime_mode, status, + last_seen_at, resume_cursor_json, runtime_payload_json, provider_instance_id + ) + VALUES ( + 'thread-amp-custom', 'amp', 'amp', 'full-access', 'running', + '2026-01-01T00:00:00.000Z', NULL, NULL, 'amp-custom' + ) + `; yield* runMigrations({ toMigrationInclusive: 31 }); const rows = yield* sql<{ readonly providerInstanceId: string }>` SELECT provider_instance_id AS "providerInstanceId" FROM projection_thread_sessions WHERE thread_id = 'thread-amp-custom' `; assert.deepStrictEqual(rows, [{ providerInstanceId: "amp-custom" }]); + const runtimeRows = yield* sql<{ readonly providerInstanceId: string }>` + SELECT provider_instance_id AS "providerInstanceId" + FROM provider_session_runtime + WHERE thread_id = 'thread-amp-custom' + `; + assert.deepStrictEqual(runtimeRows, [{ providerInstanceId: "amp-custom" }]);🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@apps/server/src/persistence/Migrations/029_BackfillForkProviderInstanceIds.test.ts` around lines 282 - 380, Add a seeded projection_thread_session_runtime row with provider_instance_id = 'amp-custom' (tied to the same session used in this test) before calling runMigrations({ toMigrationInclusive: 31 }) so the migration's UPDATE that touches projection_thread_session_runtime is exercised for the "existing non-NULL provider_instance_id" branch; modify the "is idempotent — re-running does not overwrite already-set ids" test to insert into projection_thread_session_runtime a row referencing the same session (use the same provider_session_id or thread/session identifier as the projection_thread_sessions row) with provider_instance_id set to 'amp-custom' prior to running migration 31.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In
`@apps/server/src/persistence/Migrations/029_BackfillForkProviderInstanceIds.test.ts`:
- Around line 282-380: Add a seeded projection_thread_session_runtime row with
provider_instance_id = 'amp-custom' (tied to the same session used in this test)
before calling runMigrations({ toMigrationInclusive: 31 }) so the migration's
UPDATE that touches projection_thread_session_runtime is exercised for the
"existing non-NULL provider_instance_id" branch; modify the "is idempotent —
re-running does not overwrite already-set ids" test to insert into
projection_thread_session_runtime a row referencing the same session (use the
same provider_session_id or thread/session identifier as the
projection_thread_sessions row) with provider_instance_id set to 'amp-custom'
prior to running migration 31.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: eda98d02-87ee-455a-b11c-507c01869717
📒 Files selected for processing (21)
apps/server/src/geminiCliServerManager.test.tsapps/server/src/kiloServerManager.test.tsapps/server/src/orchestration/Layers/ProjectionSnapshotQuery.tsapps/server/src/persistence/Migrations/021_RepairProjectionThreadProposedPlanImplementationColumns.test.tsapps/server/src/persistence/Migrations/026_CanonicalizeModelSelectionOptions.test.tsapps/server/src/persistence/Migrations/027_028_ProviderInstanceIdColumns.test.tsapps/server/src/persistence/Migrations/029_BackfillForkProviderInstanceIds.test.tsapps/server/src/provider/Drivers/GeminiCliDriver.tsapps/server/src/provider/Layers/AmpProvider.tsapps/server/src/provider/Layers/CodexAdapter.test.tsapps/server/src/provider/Layers/CopilotAdapter.tsapps/server/src/provider/Layers/CopilotProvider.tsapps/server/src/provider/Layers/GeminiCliAdapter.test.tsapps/server/src/provider/Layers/GeminiCliAdapter.tsapps/server/src/provider/Layers/KiloAdapter.tsapps/server/src/provider/Layers/KiloProvider.tsapps/server/src/provider/Layers/ProviderRegistry.test.tsapps/server/src/textGeneration/CopilotTextGeneration.tsapps/server/src/textGeneration/KiloTextGeneration.tsapps/web/src/components/Icons.tsxpackages/contracts/src/settings.ts
💤 Files with no reviewable changes (1)
- apps/server/src/orchestration/Layers/ProjectionSnapshotQuery.ts
✅ Files skipped from review due to trivial changes (14)
- apps/server/src/textGeneration/CopilotTextGeneration.ts
- apps/server/src/provider/Layers/AmpProvider.ts
- apps/server/src/provider/Drivers/GeminiCliDriver.ts
- apps/server/src/provider/Layers/CodexAdapter.test.ts
- apps/server/src/provider/Layers/CopilotProvider.ts
- apps/server/src/provider/Layers/KiloProvider.ts
- apps/server/src/provider/Layers/GeminiCliAdapter.ts
- apps/server/src/textGeneration/KiloTextGeneration.ts
- apps/server/src/geminiCliServerManager.test.ts
- apps/server/src/kiloServerManager.test.ts
- apps/server/src/provider/Layers/CopilotAdapter.ts
- packages/contracts/src/settings.ts
- apps/server/src/provider/Layers/KiloAdapter.ts
- apps/web/src/components/Icons.tsx
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
CI has been failing on
Install dependencieswith exit code 143 (SIGTERM) across all jobs (Format/Lint/Typecheck/Test/Build, Release Smoke, Release Desktop). The pattern is identical:10 sub-packages each declared
"prepare": "effect-language-service patch". With Bun workspaces, each install fires every workspace'spreparehook in parallel, so the patcher races to monkey-patch the samenode_modules/typescript/lib/typescript.jssimultaneously. CI either hangs or gets killed by GitHub's runner timeout.This is a cherry-pick of upstream commit
02dd47ea(upstream PR #2497).What this changes
prepare: effect-language-service patchfrom 10 sub-packages.package.json(runs exactly once per install).@effect/language-serviceandtypescriptas root devDependencies (already in catalog, so no version bumps).release-smoke.ts: removes a stalebun.lockfrom the temp install root beforebun install --ignore-scripts.Verification
bun installruns the patcher exactly once:INFO: …typescript.js is already patched … skipped.bun typecheck— 12/12 packages passgrep '"prepare"' package.json filesreturns only the root entryTest plan
bun installcleanbun typecheckpassesFormat/Lint/Typecheck/Test/BuildRelease SmokeSummary by CodeRabbit