Uh oh!
There was an error while loading. Please reload this page.
test(dogfood): field-type capability-matrix round-trip (#2004 + AI-authoring) - #2022
Merged
Conversation
…thoring) Extends the dogfood gate from "pin one example app" toward the platform's real risk: third parties have an AI author arbitrary metadata, so every authorable primitive needs a runtime proof — not a static shape check. `showcase_field_zoo` carries one field of (almost) every protocol FieldType but was only static-checked (the bundle registers it). This adds the first capability-matrix block: write one record covering many field types over the real REST data API, read it back, assert each round-trips with type fidelity. - Headliner coverage = the #2004 regressions: array-typed fields (multiselect/checkboxes/tags) and Field.time (time-of-day) — all proven to round-trip end-to-end on the in-memory (WASM) driver path `objectstack dev` uses for :memory:. - On first run it surfaced three real type-fidelity gaps — rating/slider/toggle persist but read back as strings ('4'/'25'/'1') while number/currency/percent/ boolean round-trip correctly. They are quarantined with `it.fails` (passes while broken, turns RED when fixed) rather than hidden by coercion, and filed as a separate task. Private package — no changeset. 21 passed + 3 expected-fail. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Contributor
📓 Docs Drift CheckNo hand-written docs reference the 1 changed package(s). ✅ |
Uh oh!
There was an error while loading. Please reload this page.
This was referenced Jun 18, 2026
Merged
os-zhuang added a commit
that referenced
this pull request
Jun 18, 2026
…2024) * docs(adr): ADR-0054 prove-it-runs gate for the authorable surface Extends ADR-0049 (enforce-or-remove) with a third leg. The liveness ledger (#1919) classifies every authorable property live/experimental/dead, but "live" means only a static file:line consumer pointer — proof that something reads the property, not that authoring it produces correct runtime behavior. #2018 (tz bucketing: live at every layer, broken in integration) and the field-type fidelity gaps (#2022: rating/slider/toggle read back wrong-typed) fell through that gap — call it "unproven liveness". For a platform whose authors are AI emitting metadata across a combinatorial space the examples never cover, unproven liveness ships silently into third-party apps. ADR-0054 upgrades a `live` classification to optionally carry a `proof` — a @objectstack/dogfood test that authors the property against the real in-process stack and asserts the runtime outcome. Required as a ratchet (not a retrofit) for a high-risk authorable class on change, and for any property implicated in a shipped regression (the fix carries its proof). Generative testing is explicitly deferred (Phase 3, evidence-gated). Proposed — for architect review. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * docs(adr): accept ADR-0054 (prove-it-runs gate) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
os-zhuang added a commit
that referenced
this pull request
Jul 30, 2026
…lock text on every dialect (#3994) (#4020) * feat(spec,ci): temporal hooks onto the IDataDriver contract; conformance job with live non-UTC servers Closes the two debts the datetime storage work (#3912/#3942/#3954) left open in ADR-0053. D-A2 — contract promotion. temporalFilterValue and temporalFilterColumnSql were duck-typed: analytics probed `typeof driver.x === 'function'` against a locally-invented interface, and nothing at the type level said a driver must implement both or neither. #3912's lesson is precisely that coercing the comparand without normalising the column reintroduces half the bug, so a driver implementing one hook alone would silently regress. Both are now optional members of IDataDriver, documented as a PAIR with "absent = identity" semantics for drivers whose storage form is the wire form. SqlDriver `implements IDataDriver`, so its signatures are compile-checked from here on; analytics Picks the contract instead of inventing a local shape, keeping the runtime typeof guards as the correct way to consume an optional member. coerceFilterValueForSql already sits behind the hook as the last-resort boolean/number recovery — the demotion D-A2 asked for — and stays in exactly that role. ADR-0053 records the resolution. D-A3 — the conformance backstop. The live-server suites from #3912/#3942 are opt-in (OS_TEST_POSTGRES_URL / OS_TEST_MYSQL_URL) and skip without a server, so nothing ran them in CI and the seam could regress silently. New `temporal-conformance` job: postgres:16 and mysql:8.0 service containers, servers pointed at +08:00 post-start (services cannot override the image command), the Node process at America/New_York, and assertions in UTC — the three-way skew that caught every bug in this family, with both suites asserting a non-UTC server so a mis-provisioned service fails loudly rather than passing vacuously. The WHOLE driver-sql suite runs under the skewed zone, so a TZ-sensitive assumption anywhere in the driver's tests fails here before it ships. mysql:8.0 also covers the other half of the #3942 compatibility claim — the hands-on verification ran on MariaDB 10.11, the stricter dialect. Verified locally with the exact CI command against live PG 16 (Asia/Shanghai) and MariaDB (+08:00): 47 files, 480/480, zero skips. Full pnpm build + test green (132 tasks); spec doc gates green. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TPxNwPnjcn599ujXpU3ibJ * fix(driver-sql): Field.time gets a canonical storage form — HH:MM:SS[.fff] wall-clock text on every dialect (#3994) Field.time repeated the pre-#3912 Field.datetime pattern: writes were never normalised, only reads were repaired, so one SQLite column accumulated bare time-of-day TEXT, full-timestamp TEXT and INTEGER epoch ms side by side. find() looked right; everything comparing the STORED form was wrong — measured: a business-hours window filter silently dropped 4 of 7 rows, ORDER BY sorted 14:30 before 08:00, a full-ISO write failed the statement on both Postgres and MySQL, a bound Date stored a process-timezone wall clock on pg, MySQL's bare TIME rounded fractional seconds, and NOW() defaults read three different clocks across the three dialects. The #3912→#3942→#3954 construction, transplanted (ADR-0053 D-C1..D-C3): - One canonicalTimeOfDay — HH:MM:SS, .fff only when non-zero; Date/epoch/ full-timestamp fold to the UTC time-of-day — applied on write (formatInput), to filter comparands (coerceFilterValue → temporalFilterValue) and on read (toTimeOnly). - SQLite: backfillCanonicalTimes converges legacy columns at schema sync (IS NOT-guarded UPDATE, log-and-swallow); until then filters wrap the column in sqliteCanonicalTimeSql — correct, just unindexed. os migrate plan lists the work as normalize_time_storage with a row count. - MySQL: new time columns are TIME(3); legacy TIME(0) widens at schema sync (migrateMysqlTimeColumns, plan kind widen_time_columns) since zero-precision TIME rounds fractional writes. - NOW() defaults read the UTC clock on every dialect; MySQL 8.0 rejects a plain CURRENT_TIMESTAMP default on TIME entirely, so the expression default is also a compatibility fix. - distinct()/aggregate() present time columns exactly as find() does (ReadPresentationKind gains 'time'). HH:MM:SS writes round-trip byte-identically (field-zoo f_time, #2022); a minutes-only HH:MM completes to HH:MM:00; uninterpretable values pass through untouched. Verified on live servers: SQLite, PG 16 at Asia/Shanghai, MariaDB 10.11 at +08:00, Node at America/New_York — 505/505 driver tests with zero skips. Closes#3994 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TPxNwPnjcn599ujXpU3ibJ --------- Co-authored-by: Claude <noreply@anthropic.com>
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 freeto 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.
Why
This is a development platform where third parties have an AI author arbitrary metadata. The real runtime risk isn't "a platform change broke the CRM example" — it's "the AI used a valid primitive the examples don't exercise, and it silently breaks at runtime." So the dogfood gate (#2020) has to grow from pinning one curated app into a capability matrix: every authorable primitive gets a runtime round-trip proof.
showcase_field_zoocarries one field of (almost) every protocolFieldType, but until now it was only static-checked — the metadata bundle registers it; nothing ever wrote a record and read it back.What
First capability-matrix block: write one record covering many field types over the real REST data API, read it back, assert each round-trips with type fidelity.
multiselect/checkboxes/tags) andField.time(time-of-day). All proven to round-trip end-to-end on the in-memory (WASM) driver path thatobjectstack devuses for:memory:.formula= 31.5) and system (autonumber) fields.The gate earned its keep on day one
On its first run it surfaced three real type-fidelity gaps:
rating/slider/togglepersist but read back as strings ('4'/'25'/'1') whilenumber/currency/percent/booleanround-trip correctly with the same harness. The value isn't lost — the type leaks (schema→column mapping / read coercion doesn't cover these types).Per dogfood philosophy these are not hidden by coercion — they're quarantined with
it.fails, which passes while broken and turns RED the day the gap is fixed, forcing the quarantine to be lifted rather than rotting. Filed as a separate task to fix the mapping/coercion and lift the quarantine.Notes
21 passed | 3 expected fail.SqliteWasmDriver(':memory:')(what dev uses for in-memory), so this guards the shared validator + WASM driver array path and complements fix: array field types persist + Field.time accepts time-of-day #2004'sdriver-sqlunit test by covering the integration + a second driver.🤖 Generated with Claude Code