Skip to content

Commit 7cd9098

Browse files
committed
Merge origin/main into claude/project-thread-oviakg
Claude-Session: https://claude.ai/code/session_01FEkQ5nhhuhjVQJ8R3qK9UB Co-authored-by: Claude <noreply@anthropic.com>
2 parents 98e0fde + f1cd529 commit 7cd9098

267 files changed

Lines changed: 14465 additions & 5224 deletions

File tree

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.
Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
---
2+
'@objectstack/spec': patch
3+
---
4+
5+
fix(spec): `os lint` tells an author that a field's `externalId: true` does nothing
6+
7+
Clause-②: no (a liveness-ledger re-grade: no schema, accept set, served body or error code moves; only the advisory `liveness-dead-property` warning on an authored `externalId: true` starts, and the flag's description is reworded)
8+
Metadata-protocol: none
9+
10+
The shipped liveness ledger (`liveness/field.json`) graded a field's `externalId` flag `live` on the strength of one renderer module that was deleted for having no importers. Re-measured across `packages/**` and `ui/**`, nothing reads the flag: the Studio field inspector edits it and a metadata summary copies it, and no upsert matches on it. The row is now `dead`, so `os lint` warns on a field that sets `externalId: true` (the default `false` is not warned). The flag's schema description, and the reference pages generated from it, no longer say it drives upserts.
11+
12+
- **What to do.** Remove the flag. To match seed records on a field, name the field in the seed dataset's `externalId` (`name` when it is left out). For an import, name it in the mapping's `upsertKey`.
13+
- **Unchanged.** Parsing (the key is still accepted, default `false`), seed loading, imports and every upsert.
Lines changed: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,10 @@
1+
---
2+
'@object-ui/i18n': patch
3+
---
4+
5+
feat(i18n): the Studio inspectors' page block, action, dataset, flow node, app navigation and dashboard widget strings move into the `studio` namespace
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
- **Second slice of the move ruled on objectstack#25922** (6431032267, Q2 = A; kept by 6431234244). The 520 keys under `engine.inspector.pageBlock.*`, `engine.inspector.action.*`, `engine.inspector.dataset.*`, `engine.inspector.flowNode.*`, `engine.inspector.appNav.*` and `engine.inspector.widget.*` now live in `@object-ui/i18n/namespaces/studio` instead of `@object-ui/app-shell`'s private table, text unchanged in both `en` and `zh`. Studio reads the same strings; a Chinese viewer sees these labels in English for the moment the `zh` catalogue takes to load, as with the first slice. No export changes.
Lines changed: 25 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,25 @@
1+
---
2+
'@object-ui/types': minor
3+
---
4+
5+
refactor(types)!: `object-form` and `object-calendar` declare the four spec keys their renderers read, and `@object-ui/types/zod` renames its legacy `ActionSchema` to `CRUDActionSchema` (#25923)
6+
7+
Clause-②: yes (narrowing)
8+
Metadata-protocol: narrowed
9+
10+
<!-- adr-0087: not-required (no-migration-prescription) No `@objectstack/spec` key, object definition, stored row or stored shape moves. The four keys are members the spec's `ComponentPropsMap['object-form']` and `['object-calendar']` rows already declare, now declared by reference on the `@object-ui/types` mirror of those rows, so `objectui validate` refuses exactly the values the spec refuses and `objectstack migrate meta` has nothing to rewrite. The renamed zod const validates a legacy `type: 'action'` node that is not metadata and has no stored representation; the affected party is a JavaScript or TypeScript consumer importing it by name, reached by its own compiler or bundler. `runtime-export-retired` would be the natural category, but it reads only a package's main entry and refuses an operand declared in a `*.zod.ts`, and this export lives on the `./zod` subpath in `crud.zod.ts`. The package publishes (not unpublished); no ADR-0087 id covers this change (not registered or already-registered); and the symbols are zod-backed (not runtime-interface-only or type-surface-only). -->
11+
12+
`@object-ui/types` mirrors `@objectstack/spec` schemas so that `objectui validate` and `os validate` give one verdict. Over every schema a mirror pairs with (same-name exports, and the `ComponentPropsMap` row of each node type), eleven spec keys were neither declared nor refused on objectui's side. Four of them are read by a renderer, and those are now declared:
13+
14+
- **`ObjectFormSchema`** (TypeScript and zod) declares `contentLayout` and `confirmOnDiscard`, the `ComponentPropsMap['object-form']` members by reference. `ModalForm` reads `contentLayout`; `ModalForm` and `DrawerForm` read `confirmOnDiscard`.
15+
- **`ObjectCalendarSchema`** (TypeScript and zod) declares `locale` and `loading`, the `ComponentPropsMap['object-calendar']` members by reference. `plugin-calendar` reads both off the node. They are node keys, not the calendar field-name keys that live in the `calendar` block.
16+
17+
The other seven are the spec form-view entry's keys on `FormFieldSchema`, which is objectui's runtime form field, a different layer under the same name. `normalizeSectionField` in `@object-ui/plugin-form` translates between the two layers, and each key's handling is now recorded beside the mirrors.
18+
19+
The legacy `type: 'action'` node's zod schema was exported as `ActionSchema`, the name of `@objectstack/spec/ui`'s action metadata schema, with which it shares no vocabulary. It is now `CRUDActionSchema`, beside `CRUDDialogSchema` and `CRUDComponentSchema`; what it accepts is unchanged. Its TypeScript twin keeps the deprecated `ActionSchema` name in `@object-ui/types`.
20+
21+
**BREAKING** for code that imports the zod const by its old name, and for nodes that carry a value the spec refuses on the four keys.
22+
23+
- **Before.** `import { ActionSchema } from '@object-ui/types/zod'` resolved to the legacy action node's schema. `ObjectFormSchema.safeParse({ type: 'object-form', objectName: 'account', mode: 'create', contentLayout: 'tabs' })` succeeded and kept the value, which stacked the sections.
24+
- **Now.** The old import name is gone. The zod mirrors refuse `contentLayout: 'tabs'`, a non-boolean `confirmOnDiscard` or `loading`, and a non-string `locale`, at the key's own path, as the spec does.
25+
- **Fix.** FROM `import { ActionSchema } from '@object-ui/types/zod'` TO `import { CRUDActionSchema } from '@object-ui/types/zod'`. Write `contentLayout` as `'simple'` or `'tabbed'`.
Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
---
2+
'@object-ui/plugin-grid': patch
3+
---
4+
5+
refactor(plugin-grid): `ObjectGrid`'s row-height sync, column state and mobile affordances move into three sibling hooks
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
Three adjacent blocks at the top of the `ObjectGrid` component move verbatim out of `ObjectGrid.tsx`. The effect that follows a changed `schema.rowHeight` goes to `useObjectGridRowHeightSync` (`objectGridRowHeight.ts`). The column-state persistence (the storage key, the `columnState` seed, the sync of a host-supplied `columnState` and `saveColumnState`) goes to `useObjectGridColumnState` (`objectGridColumnState.ts`). Pull to refresh and the card-view breakpoint go to `useObjectGridMobile` (`objectGridMobile.ts`). `ObjectGrid` calls each hook where its block stood, so its hook order, its stored column state and its rendered output do not change. No name is renamed, and the package entry is unchanged.
11+
12+
What to do: nothing.
Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
---
2+
'@object-ui/plugin-grid': patch
3+
---
4+
5+
refactor(plugin-grid): `ObjectGrid`'s column generator, column post-processing and Actions column move into three sibling modules
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
The last three blocks of the move-only split leave `ObjectGrid.tsx` verbatim. `generateColumns`, which turns the authored columns or the object's fields into column drafts, goes to the hook `useObjectGridGenerateColumns` (`objectGridGenerateColumns.tsx`). The column post-processing (field-type enrichment, the fold onto the declared column type, and the user's persisted widths and order) goes to `resolveObjectGridPersistedColumns` (`objectGridPersistedColumns.ts`). The Actions column and the pinned column sets go to `resolveObjectGridActionsColumn` (`objectGridActionsColumn.tsx`). `ObjectGrid` calls each where its block stood, so its hook order, its columns and its rendered output do not change. No name is renamed, and the package entry is unchanged.
11+
12+
What to do: nothing.
Lines changed: 27 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,27 @@
1+
---
2+
'@objectstack/spec': minor
3+
'@objectstack/platform-objects': minor
4+
'@objectstack/service-messaging': minor
5+
---
6+
7+
feat(spec)!: a single-word language name (`Chinese`, `English`, `Japanese`) is refused as a locale
8+
9+
The 18.x locale shape refused `zh_CN`, a padded value and a multi-word name such as `Chinese (Simplified)`, but its 2–8 letter language subtag still admitted any single word of that length. An author who wrote `locale: 'Chinese'` got a green save, and the strings were stored under a locale no session ever asks for, so they never rendered. The refusal text already said a language name is refused.
10+
11+
The language subtag is now 2–3 letters, the ISO 639 codes every tag in use starts with (BCP-47 reserves 4 letters and registers no 5–8 letter language). Everything else in the shape is unchanged: hyphen-separated 1–8 letter-or-digit subtags (`en`, `zh-CN`, `zh-Hans-CN`, `es-419`, `yue-HK`), and case is not judged. The one shape moves in all three places that hold it:
12+
13+
- `LocaleSchema` (`@objectstack/spec/system`): a `translation` item's `locale`, translation bundle keys, `i18n.defaultLocale` / `fallbackLocale` / `supportedLocales`, a seed dataset's `locale` scope and the seed loader's `locale`.
14+
- `SYS_USER_LOCALE_TAG_PATTERN` (`@objectstack/platform-objects/identity`): the `locale_bcp47_shape` rule on a user's own `sys_user.locale` column.
15+
- `LOCALE_TAG_SHAPE` (`@objectstack/service-messaging`): the read side that picks a recipient's notification locale, byte-identical to the column rule.
16+
17+
Clause-②: no (narrowing) — the locale shape refuses values it used to accept. Nothing is widened.
18+
19+
Metadata-protocol: narrowed (a locale whose language subtag is 4–8 letters, in practice a language name, is refused on translation items, bundle keys, the i18n configuration and seed datasets)
20+
21+
<!-- adr-0087: not-required (already-registered translation-locale-non-tag-refused) the 18.x D3 entry already prescribes rewriting a language name by hand, and its conversion already lists such a stored row as a TODO through LocaleSchema; this diff narrows the same shape and corrects that entry's subtag length -->
22+
23+
**BREAKING** for an author or user who wrote a language name as a locale.
24+
25+
- **Before.** `defineTranslation({ name: 'qa_x', locale: 'Chinese', … })` parsed and its strings never rendered; `PATCH` of a user's `locale` to `Japanese` was stored, and that user's notifications fell to the delivery ladder's floor.
26+
- **Now.** The value is refused at `locale` (or at the bundle key) with a message naming the tag shape, and the user's write is refused by the column's `locale_bcp47_shape` rule.
27+
- **Fix.** Write the tag, not the name: `Chinese` → `zh-CN`, `English` → `en`, `Japanese` → `ja-JP`. A language name has no single tag it can have meant, so `os migrate meta --stored` lists a stored translation row as a TODO (D3 entry `translation-locale-non-tag-refused`) and you rewrite it by hand. A user whose stored `locale` is already a name such as `Japanese` keeps loading: their notifications fall to the deployment default, and saving that value again is refused until it is replaced with a tag.
Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,18 @@
1+
---
2+
'@object-ui/plugin-detail': patch
3+
---
4+
5+
fix(plugin-detail): `record:path` leaves out a stage the record's status option rules out, instead of ticking it
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
Ruling-ref: #26581 ruling `6442747364` letter A, with its guardrail
10+
11+
A path ticked every stage before the current one by position, so a stage the record could never have stood on read as passed: an order contract that skips legal review by type showed 「审查中 ✓」 after approval. The status field's option already says which records it applies to. Its `visibleWhen` means "offered only when TRUE (else omitted)", and the server refuses a write of a value whose predicate is false. So `record:path` now omits, on both the desktop and the mobile row, any stage whose matching option declares a `visibleWhen` that is FALSE for the record.
12+
13+
- Stages match options by value, so authored `stages[]` are filtered the same way as stages derived from the picklist.
14+
- A predicate that reads the acting user (`current_user`, `user`, `ctx` or `os`) hides nothing. It says who may pick the value, not which route the record takes.
15+
- A predicate that cannot be evaluated keeps its stage, as an option list keeps an option (render side fail-open, ADR-0137 D3).
16+
- The stage the record stands on is always drawn.
17+
18+
No new key and no new read: the predicate is evaluated against the record and object definition the record page already holds.
Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,16 @@
1+
---
2+
'@objectstack/formula': minor
3+
'@objectstack/objectql': minor
4+
---
5+
6+
feat(formula): `label(record.<field>)` reads a select field's option label in the reader's locale, so a record name composed in a formula renders `v2 · 清稿` to a zh-CN reader and `v2 · Clean` to an en-US one (#26608)
7+
8+
Clause-②: yes (widening)
9+
Metadata-protocol: added
10+
11+
A select field stores its option's value, so a formula `nameField` such as `'v' + string(record.version_no) + ' · ' + record.kind` printed `v2 · clean` to every reader. Wrap the field in `label()` and the name prints the label the select control shows that reader.
12+
13+
- **The function (`@objectstack/formula`).** `label(record.<field>)` is a new CEL stdlib function, advertised in `CEL_STDLIB_FUNCTIONS`. It takes exactly `record.<field>`; any other argument is refused when the expression compiles. It reads only data: the new `EvalContext.optionLabels` (field → stored value → label, type `EvalOptionLabels`), which the caller resolves in the reader's locale before evaluating. An empty value reads `''`; a value no option declares reads as itself; a field with no options, a multi-value select, and an evaluation whose context carries no `optionLabels` fail loudly.
14+
- **Where it answers (`@objectstack/objectql`).** The engine pins the map for every read and write it serves formula fields on: `find`, `findOne`, and the insert and update responses. The labels are the object's declared `options` with the i18n service's bundle laid over them by `translateObject`, the function `GET /meta/object/:name` serves the object with, in `ExecutionContext.locale` (the request's `Accept-Language`). The field-level key `objects.<object>.fields.<field>.options.<value>` wins over a shared picklist's `picklists.<picklist>.options.<value>`, and both over the declared label. No new keys and no new bundle. A read with no locale, an engine with no i18n service, and an i18n service that throws all read the declared labels.
15+
- **Stored values.** A stored computed field (a non-formula field with an `expression`) is computed once, on the write, so it stores the declared label whatever the writer's locale. Search never indexes formula fields, so what search indexes is unchanged.
16+
- **`setI18nService`** accepts the optional `getTranslations`, `getFallbackLocale` and `getDefaultLocale` the engine now reads. An `II18nService` already carries all three.
Lines changed: 22 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,22 @@
1+
---
2+
'@objectstack/cli': patch
3+
---
4+
5+
fix(cli): `os dev`'s in-place organization-ownership upgrade names what stopped it, and the exact command that carries on
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
When the organization-ownership boot guard refuses a development database, `os dev` copies it and runs the ceremony on it in place. When that upgrade did not finish, it could exit with 「The upgrade did not run to a report」 and no cause.
11+
12+
What changes:
13+
14+
- The ceremony's own refusal is printed in its own words, also when the terminal forces colour (`FORCE_COLOR`): the refusal lines are read with colour stripped, and the `✗` glyph is no longer printed twice.
15+
- An apply that stopped with no refusal of its own (an uncaught error, a crash, a signal) says how it ended and prints the end of its log, without the kernel's boot narration.
16+
- A report file still holding the apply's reservation text, or one that does not parse, is read as "no report", with the cause, never as "a blocker (above)" with nothing above. A blocked report that names no blocker prints the log's cause too.
17+
- A stale `organization_id` the plan gives the column-drop fate is upgraded as before, but no longer silently: before the apply runs, `os dev` prints per table how many stamped values the drop discards, and names the copy that keeps them.
18+
- Each failed upgrade ends on the exact command that carries on: `os migrate organization-ownership --dev --database-url URL --apply --plan-file PLAN --backup-taken`, with the database and the plan the upgrade wrote.
19+
20+
The copy taken before the upgrade is kept, as before. A development database with a stale `organization_id` column on `sys_flow_credential` and `sys_secret` (objects that no longer declare it) is upgraded and starts under `os dev`, now pinned by a test. The production ceremony is unchanged.
21+
22+
What to do: nothing.
Lines changed: 15 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,15 @@
1+
---
2+
'@objectstack/spec': minor
3+
'@objectstack/plugin-auth': minor
4+
---
5+
6+
feat(spec,plugin-auth): a `demoStaff` entry names its line manager, and the development boot writes it
7+
8+
Clause-②: yes (widening)
9+
Metadata-protocol: added
10+
11+
A `demoStaff` entry declares one optional key, `manager`: the `email` of another entry of the same `demoStaff` list, compared without regard to case. It is named by address because the address identifies an account everywhere the platform provisions one; a display name is a label a person can edit. Before, the entry was strict and refused the key, so a fresh demo's `manager` approval rung reached nobody until someone ran Setup → Users → Set Manager by hand.
12+
13+
`defineStack` (and so `os validate`) refuses, with `STACK_CROSS_REFERENCE_INVALID` 422, a `manager` that is the address of no entry, an entry that names itself, and managers that form a loop. A manager written as a display name is refused at the entry's path, saying it takes an address. `reportsTo`, `managerEmail`, `managerId` (and so `manager_id`) and `supervisor` are pointed to `manager`.
14+
15+
On an `os dev` boot, `@objectstack/plugin-auth` links each declared manager in a second pass, once every entry has its account and still before the application seed runs. Each link goes through the same checks as `POST /api/v1/auth/admin/set-user-manager`. A link is written only when the account has no manager, so a manager someone set by hand is left as it is on every later boot; a manager cleared by hand reads as no manager and is written again. A link that cannot be written (the manager's own entry was not provisioned, or the set-manager checks refuse it) is reported on the boot log under the subordinate's address. `DemoStaffOutcome` gains `managersLinked` and `managersKept`.

0 commit comments

Comments
 (0)