Repository navigation
chore(ci): re-enable the taxonomy drift check - #92740
Conversation
|
😎 Merged successfully - details. |
🤖 CI report
|
| File | Size | Δ vs base |
|---|---|---|
posthog-app/_parent/products/workflows/frontend/Workflows/WorkflowScene.js |
124.2 KiB | 🔺 +2.2 KiB (+1.8%) |
exporter/_parent/products/workflows/frontend/Workflows/Workflow.js |
97.7 KiB | 🔺 +1.4 KiB (+1.5%) |
Posted automatically by build-bundle-size-report · uncompressed bytes from dist-report
✅ Eager graph — within budget
How much code each root ships on the eager path — downloaded and parsed before the surface is interactive. Measured from the esbuild output chunks (post-tree-shake, static imports only); lazy import() / React.lazy chunks are not counted.
| Root | Eager (shipped) | Δ vs base | Budget |
|---|---|---|---|
entry (logged-out pages, app bootstrap)src/index.tsx |
1.43 MiB · 22 files | 🔺 +6.1 KiB (+0.4%) | ███░░░░░░░ 31.6% of 4.51 MiB |
authenticated shell (every logged-in page)src/scenes/AuthenticatedShell.tsx |
8.84 MiB · 3,249 files | 🔺 +191 B (+0.0%) | █████████░ 91.1% of 9.71 MiB |
🟢 node_modules/monaco-editor/ stays out of src/index.tsx
🟢 src/lib/components/ActivityLog/describers stays out of src/index.tsx
🟢 [object Object] stays out of src/index.tsx
🟢 [object Object] stays out of src/index.tsx
🟢 node_modules/monaco-editor/ stays out of src/scenes/AuthenticatedShell.tsx
🟢 src/lib/components/ActivityLog/describers stays out of src/scenes/AuthenticatedShell.tsx
🟢 [object Object] stays out of src/scenes/AuthenticatedShell.tsx
🟢 [object Object] stays out of src/scenes/AuthenticatedShell.tsx
Largest files eagerly shipped from src/index.tsx
| Size | File |
|---|---|
| 126.8 KiB | ../node_modules/.pnpm/react-dom@18.3.1_react@18.3.1/node_modules/react-dom/cjs/react-dom.production.min.js |
| 24.6 KiB | ../node_modules/.pnpm/buffer@6.0.3/node_modules/buffer/index.js |
| 6.3 KiB | ../node_modules/.pnpm/react@18.3.1/node_modules/react/cjs/react.production.min.js |
| 4.5 KiB | ../node_modules/.pnpm/@jspm+core@2.1.0/node_modules/@jspm/core/nodelibs/browser/process.js |
| 3.9 KiB | ../node_modules/.pnpm/scheduler@0.23.2/node_modules/scheduler/cjs/scheduler.production.min.js |
| 1.4 KiB | ../node_modules/.pnpm/base64-js@1.5.1/node_modules/base64-js/index.js |
| 1.3 KiB | src/RootErrorBoundary.tsx |
| 912 B | ../node_modules/.pnpm/ieee754@1.2.1/node_modules/ieee754/index.js |
| 789 B | src/scenes/ChunkLoadErrorBoundary.tsx |
| 762 B | src/index.tsx |
Largest files eagerly shipped from src/scenes/AuthenticatedShell.tsx
| Size | File |
|---|---|
| 307.0 KiB | ../node_modules/.pnpm/posthog-js@1.422.5_@types+react@18.3.27_react@18.3.1/node_modules/posthog-js/dist/rrweb.js |
| 267.7 KiB | ../node_modules/.pnpm/@posthog+icons@0.38.0_react-dom@18.3.1_react@18.3.1__react@18.3.1/node_modules/@posthog/icons/dist/posthog-icons.es.js |
| 263.5 KiB | ../node_modules/.pnpm/posthog-js@1.422.5_@types+react@18.3.27_react@18.3.1/node_modules/posthog-js/dist/module.js |
| 256.1 KiB | src/taxonomy/core-filter-definitions-by-group.json |
| 154.2 KiB | ../node_modules/.pnpm/re2js@0.4.1/node_modules/re2js/build/index.esm.js |
| 126.8 KiB | ../node_modules/.pnpm/react-dom@18.3.1_react@18.3.1/node_modules/react-dom/cjs/react-dom.production.min.js |
| 104.5 KiB | src/lib/api.ts |
| 95.2 KiB | ../packages/quill/packages/quill/dist/index.js |
| 93.3 KiB | ../node_modules/.pnpm/prosemirror-view@1.40.1/node_modules/prosemirror-view/dist/index.js |
| 90.6 KiB | ../node_modules/.pnpm/@tiptap+core@3.20.6_@tiptap+pm@3.20.6/node_modules/@tiptap/core/dist/index.js |
Posted automatically by check-eager-graph · sizes are eager output bytes (shipped, post-tree-shake) from the esbuild metafile · part of #32479
✅ Toolbar bundle — eager 2.26 MiB within budget
What the toolbar ships to customer pages, measured from the esbuild output (minified, post-tree-shake). The eager set is the entry plus everything statically imported from it — fetched before any feature runs; deferred chunks load lazily. The eager guardrail is 5.72 MiB. Each output file must also stay below 10 MB, where CloudFront stops compressing it. The module boundary is enforced separately by check-toolbar-graph.
| Metric | Size | Δ vs base | Budget |
|---|---|---|---|
| Eager (shipped) entry + static imports |
2.26 MiB · 18 files | 🟢 -113 B (-0.0%) | ████░░░░░░ 39.5% of 5.72 MiB |
| Deferred (lazy) | 2.11 MiB · 45 files | no change | n/a — loads on demand |
Loader dist/toolbar.js |
1.1 KiB | no change | █░░░░░░░░░ 5.8% of 19.5 KiB |
Largest eagerly-shipped chunks
| Size | File |
|---|---|
| 751.7 KiB | dist/toolbar/toolbar-app-JNIXJPZH.css |
| 590.0 KiB | dist/toolbar/chunk-chunk-AZVTVEKI.js |
| 484.7 KiB | dist/toolbar/chunk-chunk-RNKUEDGA.js |
| 134.1 KiB | dist/toolbar/chunk-chunk-KWI2NYRZ.js |
| 131.8 KiB | dist/toolbar/chunk-chunk-FDH2IBXT.js |
| 71.3 KiB | dist/toolbar/toolbar-app-VQCUEUWY.js |
| 69.0 KiB | dist/toolbar/chunk-chunk-TSAL54PB.js |
| 35.6 KiB | dist/toolbar/chunk-chunk-DOQ6MEDX.js |
| 21.0 KiB | dist/toolbar/chunk-chunk-LIUCYK4Q.js |
| 6.8 KiB | dist/toolbar/chunk-chunk-DV7IWQNF.js |
Posted automatically by check-toolbar-size · sizes are toolbar output bytes (shipped, post-tree-shake) from the esbuild metafile
✅ Dist folder size — 🔺 +245.9 KiB (+0.0%)
Total size of the built frontend/dist folder (all assets), compared against the base branch.
Total: 1448.68 MiB · 🔺 +245.9 KiB (+0.0%)
⚠️ Playwright — 1 flaky
🎭 Playwright report · View test results →
- Delete a person property (chromium)
These issues are not necessarily caused by your changes.
Annoyed by this section? Help fix flakies and failures and it will go green!
🦔 PostHog Review reviewed this pull requestNothing worth raising this time, so here's a calming picture instead:
|
The Python CI job regenerates core-filter-definitions-by-group.json and fails when the result differs from the committed file, so the generated taxonomy cannot drift from posthog/taxonomy/taxonomy.py. A hand-edit to the JSON reaches the check too, through the paths filter. json.dump is the only writer of that file. The generator runs no formatter, so its output does not depend on which formatter config resolves and the check stays deterministic. A Jest test guards the other half of that guarantee: the JSON must stay in ignorePatterns in .oxfmtrc.json, or lint-staged reformats it through oxfmt and the check fails for the next person to edit taxonomy.py. It is a Jest test because .oxfmtrc.json triggers ci-frontend and not ci-backend, so a pytest guard would never run on the commit that removes the entry. Claude-Session: https://claude.ai/code/session_01TTt186JaMNTx6n7nv72Meq
21ab56f to
fda3b08
Compare
🦔 Hogbox preview · ❌ build failedThe preview didn't come up for commit Previews are optional and never block merging. A failure here is often a hogland or tailnet hiccup rather than anything in your PR, so the check stays green and this comment is the status. |
Prompt To Fix All With AI### Issue 1
frontend/src/taxonomy/coreFilterDefinitions.test.ts:22
**Redundant membership assertion**
The preceding conditional already throws whenever `GENERATED_JSON` is absent, so this assertion only runs when it is guaranteed to pass and unnecessarily duplicates the test condition.
```suggestion
```
---
For each issue above, determine whether it is valid and should be fixed. If so, fix it directly.Reviews (1): Last reviewed commit: "chore(ci): re-enable the taxonomy drift ..." | Re-trigger Greptile |
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
There was a problem hiding this comment.
Pull request overview
Re-enables and stabilizes the taxonomy drift guard by making the taxonomy JSON generator output deterministic and adding a CI step to fail when posthog/taxonomy/taxonomy.py changes without regenerating the committed JSON.
Changes:
- Removes the extra formatter pass from
bin/build-taxonomy-json.pyso the generated JSON is byte-stable. - Re-enables a taxonomy drift check in
Python CIand ensures edits to the generated JSON trigger the workflow. - Adds a frontend Jest test to assert the generated JSON stays excluded from oxfmt via
.oxfmtrc.json.
Reviewed changes
Copilot reviewed 2 out of 3 changed files in this pull request and generated 1 comment.
| File | Description |
|---|---|
| frontend/src/taxonomy/coreFilterDefinitions.test.ts | Adds a Jest guard to ensure the generated taxonomy JSON remains ignored by oxfmt (so generator output stays stable). |
| bin/build-taxonomy-json.py | Makes taxonomy JSON generation deterministic by removing the second-formatting step and documenting the invariants. |
| .github/workflows/ci-python.yml | Re-adds a taxonomy drift check step and adjusts path filters so JSON edits reach Python CI. |
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
![stamphog[bot]](https://avatars.githubusercontent.com/in/3000332?s=60&v=4){kind=small}
There was a problem hiding this comment.
Not approved yet — waiting on the conditions below.
Re-add the stamphog label to request another review once you have addressed this.
Gates denied (CI workflow deny-list match, T2-never), the change re-enables a CI check and modifies workflow logic — risky infra/CI territory — and the author is not on the owning team, with no human or substantive independent review providing assurance.
- 👍 on the PR from greptile-apps[bot].
- Gate verdict DENIED: touches .github/workflows CI file (deny-list infra_cicd) and classified T2-never
- Author haacked is not on @PostHog/team-devex, the owning team for the changed files
- No top-level human or agent review provides independent assurance; only bot merge/test-analytics comments and a generic bot review with no substantive findings are present
Gate mechanics and policy version
| Gate | Result | |
|---|---|---|
| prerequisites | ✓ | all clear |
| deny-list | ✗ | matches: infra_cicd |
| size | ✓ | 26L, 2F substantive, 50L/3F incl. docs/generated/snapshots — within ceiling |
| tier | ✗ | classified as T2-never: T2-never (50L, 3F, cross-cutting, chore) |
| stamphog 2.0.0b4 | .stamphog/policy.yml @ fda3b08 · reviewed head fda3b08 |
Problem
A taxonomy change that skips the generator silently does not take effect, and nothing catches it.
frontend/src/taxonomy/core-filter-definitions-by-group.jsonis generated fromposthog/taxonomy/taxonomy.py, and the frontend reads labels, descriptions, andhidden_in_query_buildersfrom the JSON rather than from Python.A CI check guarded this. #31950 disabled it in May 2025 and nothing restored it.
masteris drifted today: the generator produces 33 lines that are not committed.The check was disabled because it flapped between two formattings of the JSON. The generator wrote the file with
json.dump, then ran a second formatter over it. #31014 listed the file in a newfrontend/.prettierignorebut not in the root one, so whether that second formatter ran depended on which ignore file the invocation resolved. The file had two possible shapes.Changes
$mcp_scope_preset, with its label and description. It reachedtaxonomy.pywithout a regeneration, so the frontend never saw it.taxonomy.pywithout regenerating now fails the Python code quality job.git diffto the generated file. Unrelated dirt in a 30 minute job cannot fail it.python3 bin/build-taxonomy-json.py. The job installs Python but nonode_modules, sopnpm execcannot work there. The generator is pure stdlib.Nothing in the UI changes shape. The one visible difference is the new property in filter lists, which is data, not layout.
Note
.oxfmtrc.jsonkeeps itsignorePatternsentry for this file. Without it, lint-staged reformats the JSON on commit and reintroduces the drift from the other side.Open PRs that have not rebased
CI runs a branch merged with master, so a PR that touches no taxonomy takes master's regenerated JSON and passes. A PR that edits
taxonomy.pywithout regenerating fails, which is the intent. No unrebased branch needs a file it lacks.How did you test this code?
Verified locally on the merged tree:
descriptionintaxonomy.pywithout regenerating makes the check exit 1. Regenerating makes it exit 0. Both reverted.format:yamlover it and leaves it byte-identical, confirming no second formatter is left in any path.hogli lint:workflowspasses 8 of 8 checks.hogli ci:preflight --strictreports no failures.Not checked:
actionlintis not on this machine, and repo-wide mypy did not run. The only Python change deletes a now-unused import from a stdlib script, whichruff checkconfirms.No automated tests are added. The check is itself the regression guard, and its behavior is verified by the pass and fail runs above.
Automatic notifications
Docs update
None.
🤖 Agent context
Autonomy: Human-driven (agent-assisted)
Written with Claude Code in PostHog Desktop.
Skills invoked:
/writing-code-commentsfor the note in the generator,/writing-pr-descriptionsfor this body..agents/skills/authoring-ci-workflows/SKILL.mdwas read directly, as AGENTS.md requires for a workflow change.The starting brief assumed the check failed because the job had no Node and because oxfmt exited non-zero. Both turned out to be wrong. The job has
npm, which an adjacent schema check already uses, and oxfmt exits 0 while formatting nothing, because.oxfmtrc.jsonexcludes this file. oxfmt does format JSON in general, which is what made the second formatter dangerous. The Slack thread linked from #31950 named the real cause, and the plan changed to match it.Two commits: the regeneration first, then the generator and workflow change, so the check is green on its own PR. Master advanced 159 commits mid-session and landed a separate regeneration; it merged cleanly, and pristine
masterwas re-checked afterwards to confirm the 33 line drift is still real.