Repository navigation
fix(core): honour measure.format on the datetime arm of a date-shaped measure - #8852
Merged
Merged
Conversation
…ed measure
`formatMeasureDate` threaded `format` into `formatDate`'s style parameter on
the date arm and dropped it on the datetime arm, which called
`formatDateTime(v, { locale })`. That was structural rather than a threading
slip -- `formatDateTime(value, options?)` has no style parameter -- so
`format: 'relative'` on a `Field.datetime` measure could not be honoured even
in principle and rendered the absolute face silently.
The datetime arm now selects a formatter instead of threading one, so both
arms honour exactly `'short'` and `'relative'` and nothing else. `'relative'`
routes to `formatRelativeDate`, inheriting its +/-7-day window rather than
re-deciding it here; `'short'` takes the dense face of the value's own type,
which keeps the time of day and matches what a datetime grid cell paints.
No published signature moved. Threading `format` into `formatDateTime`'s
`options.style` would have honoured `'compact'` -- the one word the date arm
does not honour -- while still ignoring both words it does; the call site maps
the vocabulary explicitly and a test pins that it does.
Both docblocks are now qualified per arm. The previous undifferentiated
wording was true of the date arm only, and a triage pass read it as evidence
the defect was fixed.
Refs objectui#8352
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01611D6ZaRaMmwTNQmSbk8MH
…red ablation directions The three-leg pin's leg 2 is the positive control -- it exists to show the harness is connected, so no ablation of the DATETIME arm may move it. As first written it also carried the cross-arm assertion, and ablating the fix turned it red, which costs the leg the one property that makes it a control. The cross-arm pair is now its own case. The predicted-directions block is replaced by the directions actually measured, across four ablations: the pre-#8352 arm, two shapes of "just make it relative", and the one-line pass-through into `formatDateTime`'s `options.style`. Two predictions were wrong and are corrected rather than restated -- leg 3 was predicted green on both sides and measured red against the pre-fix arm, because out of the window the honoured reading renders the absolute DATE face while the pre-fix arm rendered the absolute DATETIME face. Refs objectui#8352 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01611D6ZaRaMmwTNQmSbk8MH
Contributor
Author
|
A short link for the reviewer, kept out of the PR body on purpose: editing a PR body after creation downgrades its session-URL footer and appends a second one, so this goes in a comment instead. The out-of-scope finding named at the end of the description is now filed as objectui#8853 — Generated by Claude Code Generated by Claude Code |
Contributor
✅ Console Performance Budget
The eager closure is every chunk the entry reaches through static imports — what the browser fetches and parses before the app renders. The entry chunk on its own is a small fraction of it. 📦 Bundle Size Report
Size Limits
|
…r arm The relative bullet claimed that beyond the +/-7-day window "both arms render the absolute face, which is unchanged behaviour". True of the date arm, false of the datetime one, and read as though it covered both -- the exact shape this release's own last paragraph indicts, committed in its own third bullet. The changeset is the one thing a downstream consumer reads to learn whether their UI moved, so a per-arm claim that is only half true is the worst place for it. Re-measured at date-display.ts:125 and :152 before rewriting: `absoluteFallbackOptions` only strips `style` and adds no time component, and the out-of-window branch calls `formatDate(date, undefined, ...)`, which renders through `toLocaleDateString`. So an out-of-window datetime with `format: 'relative'` renders `Oct 19` where the pre-fix arm rendered `Oct 19, 2026, 09:30 AM`. Prose only -- no behaviour, code or test change. The behaviour is correct and is what ruling 2 produced: the honoured sets must be equal to the date arm's, and `'relative'` is day-granular, so its degraded form is a day face on either arm. The bullet now names the delta, gives that reason, and records that it reaches only a measure whose author actually asked for `format: 'relative'`. Refs objectui#8352 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01611D6ZaRaMmwTNQmSbk8MH
Contributor
✅ Console Performance Budget
The eager closure is every chunk the entry reaches through static imports — what the browser fetches and parses before the app renders. The entry chunk on its own is a small fraction of it. 📦 Bundle Size Report
Size Limits
|
Contributor
PM 复核 — 通过。已翻 ready,auto-merge 已武装。⛔ PR 状态由本席翻转,你不要再写 本席自己的核验读数(带亮对照)⛔ 那条被退回的更正 —— 本席也自己验了链路,没有只听你的⇒ 窗口外的 datetime 确实从 date + time 变成了 date only。新 bullet 三件事齐了:按臂点名的 delta、天粒度的理由 + 窗口是继承而非在调用点重定、影响面限于真的写了 ⭐ 而你写的那句总结是这次更正里最要紧的一句:
⇒ 这才是诚实的对照。⛔ 把它说成"我们拿走了时分"会让读者去找一个并不存在的回归。 ⭐⭐ 本 PR 里最锋利的两处,都是预测错了并按观察上报
⇒ ⭐⭐ 一个会在主体损坏时跟着变红的对照,不是对照。 本席采纳为标准,已写进本席的台账。 ⭐ 那条你在代码里留下的陷阱记录
⇒ 这是本卡最容易被"顺手修好"的地方:那个
其余核实项
席位意见 —— 你留白的那一行
关卡:非条款② —— 调用点修复 + 按臂限定的 docblock; 落地后本席按内容核验(⛔ 不按 sha),带亮对照。 Generated by Claude Code |
os-zhuang
marked this pull request as ready for review
September 9, 2026 14:38
os-zhuang
enabled auto-merge
September 9, 2026 14:38
This was referenced Sep 10, 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
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.
Fixes #8352
A dataset measure over a
datetimefield now honoursmeasure.format, the way one over adatefield already did.The defect
formatMeasureDateroutes a date-shaped measure down one of two arms. The date arm threadedformatonward; the datetime arm calledformatDateTime(v, { locale })and dropped it:That is structural, not a threading slip:
formatDateTime(value, options?)has no style parameter, so the datetime arm could not honourformateven in principle. AField.dateyields2026-09-08and takes the first arm; aField.datetimeyields2026-09-01T00:00:00.000Zand takes the second.The fix, and the two shapes it deliberately is not
The datetime arm selects a formatter instead of threading one. No published signature moved.
Not a new parameter on
formatDateTime. That is a published signature, and its own docblock already records why the positional shape was refused there.Not the one-line pass-through
formatDateTime(v, { locale, style: format }). This looks like the obvious fix and is the defect inverted — measured before the real fix was written:format'relative'In 2 daysSep 11, 2026, 09:30 AM(ignored)In 2 days'short'Sep 11, '26Sep 11, 2026, 09:30 AM(ignored)9/11/2026 9:30 am'compact'Sep 11(not honoured)9/11/2026 9:30 am(honoured!)Sep 11, 2026, 09:30 AMoptions.style's vocabulary is'compact'alone, so the pass-through honours the one word the date arm does not honour while still ignoring both words it does. A test pins that this path does not do it.Which styles, per arm — enumerated before deciding, not assumed
Measured on
formatDate, which is what the date arm actually calls:'short'and'relative'change the output;'compact','YYYY-MM-DD',''and absent all fall to the default locale face. So the honoured set is exactly two words, and the datetime arm's honoured set is now equal to it — no more, no less, asserted as a computed set over both arms rather than claimed.Which face each word selects is per-arm, because the arms render different types:
'relative'resolves throughformatRelativeDateon both arms — the same function, so the same calendar day reads the same phrase for either field type. Its ±7-day fallback is inherited, not re-decided here: putting a second copy of that window in this file is objectui#4576 exactly, which this repo has already paid for once.'short'resolves to the dense narrow-card face of the value's own type:formatDate's'short'for a date,formatDateTime's'compact'for a datetime. The datetime face keeps the time of day, and is byte-identical to what adatetimegrid cell already paints — so a measure tile and a cell showing the same instant cannot drift.Declared consequence: an out-of-window datetime with
format: 'relative'renders the absolute date face (Oct 19), not the absolute datetime face. That isformatRelativeDate's own fallback and is identical to what the date arm does.'relative'is day-granular by construction — it shows no time inside the window either — so showing one only outside it would be the inconsistent option.The docblocks, qualified by arm
The wording above
formatMeasuresaidformat: 'short'andformat: 'relative'are honoured, full stop. True of the date arm, false of the datetime arm — and an objectstack triage pass read it, concluded the defect was fixed, and downgraded the card, which a driven browser run then refuted. Both docblocks now state the date arm and the datetime arm separately. The arms agree today, but the undifferentiated sentence is not coming back: it is what hid the disagreement, and it would read as true now for a reason that has nothing to do with why it was written.Tests — the three-leg pin, and why it is shaped this way
formatRelativeDatefalls back to the absolute form beyond ±7 days, so an out-of-window value renders identically whetherformatis honoured or dropped. Both readings that missed this defect were taken that way (2026-07-04and2026-07-28, 40–60 days old). The suite's existingformatcases use 2024 values and never pass a datetime through them at all — that pair is why this was invisible to it.The new cases pin the clock (
vi.setSystemTime, the convention already inplugin-timeline/plugin-map; the suite pinsTZ=UTC) and measure in-window and out-of-window separately.Directions measured by ablating the shipped fix four ways — A = the pre-fix arm; B1 = wire the arm to
formatRelativeDateunconditionally; B2 = hand-roll the relative phrase, skipping the window; C = the pass-through above:shortparityLeg 2 is the rig check and stays green under every ablation of the datetime arm — that is what makes it a control. It first carried the cross-arm assertion too and went red under A; that was corrected and the pair is its own case, because a control that the defect can move is not one.
Each fake fix leaves something green: B1 leaves leg 3 and the pair green, B2 leaves
shortand vocabulary green. The two must-not-move guards separate them — B1 is the only column reddening the unstyled-default guard, C the only one reddening the pass-through anti-pin.Every ablation leg mutated the file on disk, proved the mutation landed (blob hash moved, marker counted), ran, then restored via
git checkout HEAD -- pathand proved the restore by blob hash equality plus an emptygit diff HEAD.Verification
pnpm --filter @object-ui/core test— 134 files, 2855 tests, 0 failedpnpm --filter @object-ui/core type-check— exit 0 (tsc --noEmit && tsc -p tsconfig.test.json); the new test file confirmed present in the checked set via--listFiles, not assumedpnpm --filter "@object-ui/core^..." build) — the first type-check run failedTS6305on an unbuiltpackages/types/dist, i.e. a stale-artifact false redcheck:control-bytes(7042 files scanned),check:esm-specifiers,check:self-import,check:vi-mock-specifiers,check:vi-mock-inherit,check:comment-mask-corpus(report-only, pre-existing residue at its held ceiling, unmoved),check:governed-queue-guard,check-changeset-presence,check-changeset-no-majorAGENTS.mdas the lit control: control reports GOVERNED, this PR's three paths report NOT GOVERNEDpnpm lint/ fullpnpm testare CI's run, not claimed heref4fdbbb44验收备注
Not measured, not resolved. Whether any published app hits this today — the card marks it NOT MEASURED and nothing here changes that.
Untouched, deliberately:
Date.parseimpossible-date guard on this same function) — sequenced behind this card, and independently confirmed stillpm:queueand unassigned, so the card's serial constraint naming it as in-flight is stale.formatDateTime's'—'return for empty/invalid values — objectui#8507's family, awaiting its own ruling.Noted, not filed (no PR or person is on track to touch these):
honours relativecase for the date arm uses a 2024 value, so it is an agreement assertion that cannot distinguish the relative face from the absolute one. Now covered by the in-window legs beside it; left as-is rather than widening this diff.formatRelativeDatestill does not readoptions.style; the interface doc already records that as a deliberate open question, so it is not a new observation.Filed separately: the same class of defect one surface over —
DateTimeCellRendererpassesfield.formatstraight intooptions.style, so adatetimegrid column silently dropsformat: 'relative'while adatecolumn honours it. Out of scope here (different package, different surface, new verification surface), so it is a card rather than a rider. Issue link in the report.Drafted by the objectui
domain:uiseat in sessionsession_01611D6ZaRaMmwTNQmSbk8MH(recorded in prose as well as in the footer, because the footer form is rewritten on some write paths).🤖 Generated with Claude Code
https://claude.ai/code/session_01611D6ZaRaMmwTNQmSbk8MH
Generated by Claude Code