fix(api): correct baseline-snapshot sample + tech-events omitted-paging default (#4599 adversarial follow-ups A+B)#4666
Merged
Conversation
…ng default (adversarial follow-ups #4599) Two defects surfaced by an adversarial re-audit of the merged OpenAPI docs (#4657): A — RecordBaselineSnapshot request sample used `type: "all"`, which the handler rejects (VALID_BASELINE_TYPES has no 'all' → record-baseline-snapshot.ts:43 silently `continue`s), so a copy-paste yields `updated: 0` while the 200 example claims `updated: 1`. The examples injector's `endsWith('type')` heuristic emitted the literal 'all' with no curated override. Add a drift-anchored BASELINE_TYPE_EXAMPLE_ID (= infrastructureTemporalBaselineTypes[0], 'military_flights', matching the sibling GetTemporalBaseline enum example) + an overrideStringExample case scoped to baseline ops. Re-ran only the examples injector (version-independent). B — list-tech-events documented `limit`/`days` "defaults to 50/90 when omitted", but the REST decoder maps an omitted query param to 0, and clampInt treats a finite 0 as a value (→ min 1), not the fallback — so a no-params call returned 1 event within 1 day, not 50/90. Extract a pure, tested resolveTechEventsPaging() that guards with `|| default` (mirrors seismology `|| 500` and BLS `>0 ? : 60`); wire both handler call sites (fetchTechEvents + filterEvents) to it. The primary documented claim is now true. (The residual "explicit 0 clamps to 1" wording is a proto-comment change and rides the follow-up regen PR with #4599 items C/D.) Guard tests: baseline sample type ∈ VALID_BASELINE_TYPES; resolveTechEventsPaging omitted/0 → 50/90 (reproduced failing first, then fixed). 384 contract tests pass; tsc clean; examples injector --check idempotent. Claude-Session: https://claude.ai/code/session_01EsjtheWqt8GxEZfd3DyTMu
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Contributor
|
Preview deployment for your docs. Learn more about Mintlify Previews.
💡 Tip: Enable Workflows to automatically generate PRs for you. |
Contributor
Greptile SummaryThis PR fixes two doc-to-handler drift defects caught by an adversarial re-audit of the OpenAPI specs. Both fixes are narrowly scoped, version-independent, and accompanied by drift-anchored tests.
Confidence Score: 5/5Safe to merge — both changes are narrowly scoped bug fixes with targeted tests and no observable side-effects on other endpoints. The handler fix correctly mirrors the established sibling pattern (|| default) and is exercised by five unit tests covering the critical 0/undefined case. The OpenAPI sample correction is regenerated output guarded by a new drift-anchored contract test. The only rough edge is a test description that says 'below the minimum' while its input is at the minimum, which is a labelling issue with no impact on correctness. tests/tech-events-paging.test.mts — the 'below minimum' test case exercises the boundary value (1), not a sub-minimum value; worth a one-line fix before the follow-up PR adds more paging tests. Important Files Changed
Flowchart%%{init: {'theme': 'neutral'}}%%
flowchart TD
A[REST decoder\nparams.get limit ?? 0] -->|decoded value| B{resolveTechEventsPaging}
B -->|req.limit = 0 or undefined| C["req.limit || 50 → 50\nclampInt(50, 50, 1, 200) → 50"]
B -->|req.limit = 25| D["req.limit || 50 → 25\nclampInt(25, 50, 1, 200) → 25"]
B -->|req.limit = -5| E["req.limit || 50 → -5\nclampInt(-5, 50, 1, 200) → 1"]
C --> F[fetchTechEvents / filterEvents]
D --> F
E --> F
G[examples injector\noverrideStringExample] -->|key='type', path includes 'baseline'| H[BASELINE_TYPE_EXAMPLE_ID\n= military_flights]
H --> I[record-baseline-snapshot\nrequest sample]
I -->|handler reads type| J{VALID_BASELINE_TYPES\nincludes?}
J -->|yes| K[updated += 1]
J -->|no — old: 'all'| L[silently continue\nupdated = 0]
%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
flowchart TD
A[REST decoder\nparams.get limit ?? 0] -->|decoded value| B{resolveTechEventsPaging}
B -->|req.limit = 0 or undefined| C["req.limit || 50 → 50\nclampInt(50, 50, 1, 200) → 50"]
B -->|req.limit = 25| D["req.limit || 50 → 25\nclampInt(25, 50, 1, 200) → 25"]
B -->|req.limit = -5| E["req.limit || 50 → -5\nclampInt(-5, 50, 1, 200) → 1"]
C --> F[fetchTechEvents / filterEvents]
D --> F
E --> F
G[examples injector\noverrideStringExample] -->|key='type', path includes 'baseline'| H[BASELINE_TYPE_EXAMPLE_ID\n= military_flights]
H --> I[record-baseline-snapshot\nrequest sample]
I -->|handler reads type| J{VALID_BASELINE_TYPES\nincludes?}
J -->|yes| K[updated += 1]
J -->|no — old: 'all'| L[silently continue\nupdated = 0]
Reviews (1): Last reviewed commit: "fix(api): correct baseline-snapshot samp..." | Re-trigger Greptile |
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
Comment on lines
+27
to
+29
| it('clamps explicit values below the minimum to 1', () => { | ||
| assert.deepEqual(resolveTechEventsPaging({ limit: 1, days: 1 }), { limit: 1, days: 1 }); | ||
| }); |
Contributor
There was a problem hiding this comment.
{ limit: 1, days: 1 } are AT the minimum, not below it. The output { limit: 1, days: 1 } confirms this — no clamping actually occurs. As-is the test verifies that the lower bound is inclusive (fine), but the name implies a negative/sub-minimum value (-5 etc.) would be tested. This could mislead a future reader who changes clampInt's lower boundary and wonders why this test still passes.
Suggested change
| it('clamps explicit values below the minimum to 1', () => { | |
| assert.deepEqual(resolveTechEventsPaging({ limit: 1, days: 1 }), { limit: 1, days: 1 }); | |
| }); | |
| it('floor-clamps the minimum-boundary values (limit=1, days=1 are the minimums)', () => { | |
| assert.deepEqual(resolveTechEventsPaging({ limit: 1, days: 1 }), { limit: 1, days: 1 }); | |
| }); | |
| it('clamps explicit values below the minimum to 1', () => { | |
| assert.deepEqual(resolveTechEventsPaging({ limit: -5, days: -3 }), { limit: 1, days: 1 }); | |
| }); |
… fixed default (review #B) Follow-on to the resolveTechEventsPaging fix in this PR: with the handler now treating omitted/0 as the 50/90 default, the proto field comment's residual "0 is not unlimited (it clamps up to 1)" clause was stale (0 now → 50, since an omitted query param and an explicit 0 are the same value on the wire). Reword to "defaults to 50/90 when omitted or 0" and regenerate the ResearchService specs + bundle with the pinned sebuf v0.11.1 plugin (scoped diff: 3 artifacts, param descriptions only). make generate idempotent; contract + description-guard tests pass.
This was referenced Jul 4, 2026
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
1 participant
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.
What & why
An adversarial re-audit of the merged OpenAPI docs (#4657, umbrella #4599) — 5 lenses (samples, auth, completeness, contract-validity, doc-truth) each verified against the real gateway/handlers — surfaced two defects the guards missed. This PR fixes both (the version-independent pair; the plugin-gated items C/D follow in a second PR).
A —
RecordBaselineSnapshotsample rejected by the handlerThe request sample used
type: "all", butVALID_BASELINE_TYPES=[military_flights, vessels, protests, news, ais_gaps, satellite_fires]—record-baseline-snapshot.ts:43silentlycontinues on a non-member, so a copy-paste yieldsupdated: 0while the200example claimsupdated: 1. Root cause: the examples injector'sendsWith('type')heuristic emitted the literal'all'with no curated override (the siblingGetTemporalBaselinequery enum correctly showsmilitary_flights).Fix: drift-anchored
BASELINE_TYPE_EXAMPLE_ID(=infrastructureTemporalBaselineTypes[0]) + anoverrideStringExamplecase scoped to baseline ops. Re-ran only the examples injector — version-independent, 3-artifact diff.B — tech-events "defaults to 50/90 when omitted" was false
The REST decoder maps an omitted query param to
0(Number(params.get("limit") ?? "0")), andclampInttreats a finite0as a value (→ min1), not the fallback — so a no-params call returned 1 event within 1 day, not the documented 50/90. Siblings guard correctly (seismology|| 500, BLS> 0 ? : 60); tech-events didn't.Fix: extract a pure, tested
resolveTechEventsPaging()that guards with|| default, and wire both handler call sites (fetchTechEvents+filterEvents) to it — de-duplicating the two identical resolutions. The primary documented claim is now true.Tests (Bug Fix Protocol — reproduced first)
resolveTechEventsPagingomitted/0→ 50/90 (added the failing test, confirmed it returned{1,1}, then fixed).record-baseline-snapshotsampletype∈VALID_BASELINE_TYPES(drift-anchored guard).tscclean; examples injector--checkidempotent; biome clean.Follow-up: #4658 tracks the handler-behavior notes.