Skip to content

Commit b5658de

Browse files
committed
Merge remote-tracking branch 'origin/main' into claude/project-thread-detgab
2 parents b0ef42a + a19759a commit b5658de

319 files changed

Lines changed: 15622 additions & 1401 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: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,10 @@
1+
---
2+
'@objectstack/spec': patch
3+
---
4+
5+
docs(spec): the tenancy-posture docblocks name `@objectstack/organizations` as open code and the licence as the commercial entitlement
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
The file header and the docblocks in `packages/spec/src/security/tenancy-posture.ts` still called the `@objectstack/organizations` runtime "enterprise", although it ships from this repository under Apache-2.0 since ADR-0132 D1. They now name the package without that adjective, and say "commercial entitlement" where they mean the licence gate that the commercial package of the same name adds by subclassing the open one (ADR-0132 D2, D3). What stays true is unchanged: a walled posture (`group` or `isolated`) still needs that runtime mounted to activate (ADR-0093 D5). Comment-only: no schema, type or runtime behaviour changes.
Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
---
2+
'@objectstack/metadata-protocol': patch
3+
'@objectstack/cli': patch
4+
---
5+
6+
fix(metadata-protocol): `os migrate packageless-overlays` moves a package-less email template row of a locale the base package does not ship into the base's `.custom` pair as the pair's own member
7+
8+
Clause-②: no
9+
10+
Metadata-protocol: none
11+
12+
A Setup translation stored bound to no package before ADR-0146 D11, of a template the installed package ships in other locales only (a `zh-CN` row of a template shipped `en-US` only), was saved into the pair declaring `overrides`, which the override write gate refuses for a locale the base does not ship. The move failed and the row was kept. The ceremony now saves such a row into the pair with no `overrides` declaration, as the pair's own member at that locale (the base's address serves it as the pair's addition), reads it back and only then deletes the unbound row. Its subject, bodies and envelope columns move unchanged, and it keeps sending in that locale. A row of a locale the base ships moves as the declared override, as before; a declared override of a locale the base does not ship is still refused.
13+
14+
The preview names such a row's locale (`addedMember` in `--json`, "its own member" in the listing), and `--apply` reports it `moved` with that `addedMember` in place of a `basedOn`, since the pair's own member records none.
Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,8 @@
1+
---
2+
'@objectstack/mcp': patch
3+
---
4+
5+
Clause-②: no
6+
Metadata-protocol: none
7+
8+
The stdio MCP server lists `agent_prompt` only on a boot that registers at least one agent. A boot with no agent, which is every open-framework boot, used to list it anyway. Its `agentName` argument suggested `"ask"` and `"build"`, and every `prompts/get` came back as a successful prompt whose one message read `Error: Agent "ask" not found`, which an MCP client passes to its model as instructions. On such a boot the prompt is now left out of `prompts/list`, a get on it is the protocol's own unknown-prompt error, and the boot log says once why. On a boot with agents nothing changes for a client: the prompt is listed, its argument suggests names the boot actually holds, and a miss still answers `Error: Agent "X" not found`. If the agent metadata cannot be fully read at boot, the prompt stays listed, because an outage is not proof that no agent exists. Its argument then suggests no names.
Lines changed: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,10 @@
1+
---
2+
'@objectstack/spec': patch
3+
---
4+
5+
docs(spec): `ConnectorConnectionState.status` no longer says a connection with no refresh token is marked because its access token expired
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
Comment only. The `reconnect_required` JSDoc said such a connection "keeps nothing to renew an expired access token with". A refresh also runs on a cache miss (a fresh replica, or a `revision` another replica moved) and on a dispatch's 401, so a connection with no refresh token, or no client to refresh with, is marked even while its access token is still valid. The JSDoc now reads "keeps nothing to renew an access token it has to renew with", the wording the runtime's own `markReconnectRequired` doc and the `sys_connector_connection.status` description use, and the sense of the `status_reason` the runtime stores ("an access token the connection has to renew cannot be renewed").
Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
---
2+
'@object-ui/plugin-list': patch
3+
'@object-ui/plugin-tree': patch
4+
---
5+
6+
fix(ui): a tree list view draws the view's authored empty state, and the tree's load failure is translated
7+
8+
Clause-②: no
9+
Metadata-protocol: none
10+
11+
- **Empty tree views.** A list view shown as a tree now draws the same empty state as the grid and the gallery: the view's authored `emptyState` (title, message, icon, link) first, otherwise the localized "Nothing here yet" copy, with the same Create, Clear search and Clear filters actions. It used to draw the tree's generic "No data" line, so guidance authored on the view was lost on that layout. The list toolbar stays above it.
12+
- **Tree load failures.** A tree whose read fails shows the translated "Couldn't load records" title with the error's own message under it, in the alert panel the list uses, instead of the English line "Failed to load tree: …" on every page.
13+
14+
No metadata key or option is added.
Lines changed: 21 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,21 @@
1+
---
2+
'@objectstack/service-analytics': patch
3+
---
4+
5+
fix(service-analytics)!: the /analytics/sql dry run refuses a date bucket over a column that is not a date or datetime, as the query does, instead of rendering it
6+
7+
Clause-②: no (narrowing) — `/analytics/sql` (`AnalyticsService.generateSql`) now refuses `INVALID_FIELD` / 400 at `groupBy.N.dateGranularity` a time-dimension granularity over a field declared neither a date nor a datetime (text, select, number, boolean, time, …), with the refusal `/analytics/query` gives the same selection, field for field; it rendered a statement for it (or, at a non-UTC zone, answered `NOT_IMPLEMENTED` / 501 saying the query is served). Nothing is widened: no export, public member, answer member, error code or status is added.
8+
Metadata-protocol: none
9+
10+
<!-- adr-0087: not-required (no-migration-prescription) No key of any schema is added, removed, renamed or re-shaped, and no export or type of the package root changes: a dry run that rendered a statement the query refuses is refused with the query's refusal, already on the wire. There is nothing for a migration to rewrite and no tombstone. -->
11+
12+
**BREAKING** accept-set narrowing on the `/analytics/sql` dry run (`@objectstack/service-analytics`). `/analytics/query` is unchanged.
13+
14+
- **Before.** Measured on real SQLite on `main` `ab085a04`. `/analytics/query` refused a month bucket over a text, select or boolean column, a year bucket over a number and a day bucket over a `time` column with the engine's `INVALID_FIELD` / 400 at `groupBy.N.dateGranularity`. The dry run rendered `strftime('%Y-%m', region)` and its kin, on the direct and the cross-object path, and for a cube dimension's one declared granularity asked raw. At `America/New_York` it answered `NOT_IMPLEMENTED` / 501.
15+
- **Now.** The dry run judges the keys `execute()` hands the engine (its cross-object path's when the selection routes there) with the reading the draft preview already used (#28852), now `dataset-date-bucket-door.ts`. It runs after the dry run's `dateRange` read and its column-asked-twice refusal, as `query()` refuses, and gives the engine's refusal: the same message, `object`, `param`, `path`, `field` and `fields`. A field with no declared type, or a `formula`, is not judged, as in the engine.
16+
- **Unchanged.** A date or datetime bucket renders byte for byte, the statement the query echoes.
17+
- **Who sees it.** A caller of the dry run for a selection the query refuses: the query's refusal, instead of SQL for a statement that never runs.
18+
19+
**The remedy.** Group by the column without a granularity, or bucket a date or datetime field.
20+
21+
Also in this change, the draft preview of `queryDataset` lowers its row filters (the `where`, then the time-dimension windows) once, before its column-asked-twice and date-bucket refusals, where the ObjectQL strategy lowers them. So a selection carrying an unrecognised `dateRange` beside one column asked twice or beside a bucket over a non-date column is refused about the `dateRange` (`ANALYTICS_DATE_RANGE_UNRECOGNIZED` / 400) on every face, and a refused filter beside a non-date bucket is refused about the filter. The preview answered the other refusal.
Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,18 @@
1+
---
2+
'@objectstack/metadata-protocol': minor
3+
---
4+
5+
fix(metadata-protocol)!: a runtime permission-set save refuses a grant on an object that does not exist
6+
7+
Clause-②: no (narrowing)
8+
Metadata-protocol: narrowed
9+
10+
**BREAKING** for one shape only: a `permission` set whose `objects` map grants on an object this runtime does not know. It was stored and published (200), and the grant applied to nothing, because the permission evaluator keys a grant by the object's name: a short `lead` where the object is `crm_lead` silently granted nothing. `defineStack` already refuses the same grant at build; the metadata save doors now refuse it too.
11+
12+
- **Where.** `PUT /api/v1/meta/permission/NAME`, live and `?mode=draft`; the publish of a draft (`POST /meta/permission/NAME/publish`, `POST /packages/ID/publish-drafts`); and the restore doors (`POST /meta/permission/NAME/rollback`, a commit revert).
13+
- **The refusal.** `422 INVALID_METADATA`, the save door's existing envelope, with one finding per unknown object at `objects.NAME`: "Permission 'SET' grants on object 'NAME' which is not defined in objects: …". Nothing is stored; a package publish aborts the batch.
14+
- **What counts as known.** The `'*'` wildcard; an object the runtime registers, from any package, platform objects (`sys_*`, the RBAC link tables) included; an object stored in this environment or defined in code; on a draft save, any object draft; on a package publish, an object drafted in the same batch.
15+
- **What is not judged.** A grant the set already holds: a re-save, `os migrate meta --stored` or a restore that keeps it is not refused, even if the object has gone since. Editing a shipped set is therefore never refused for a grant it ships with.
16+
- **What to do.** Correct the object name to the object's full name, create the object first (or draft it in the same package), or remove the grant.
17+
18+
<!-- adr-0087: not-required (no-migration-prescription) the diff registers no retirement and prescribes no rewrite: the refused value granted nothing, so there is no stored shape to convert; a set that already holds such a grant is not refused on re-save -->
Lines changed: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,10 @@
1+
---
2+
'@objectstack/spec': patch
3+
---
4+
5+
docs(spec): the record-page `timeline` total is documented as an upper bound of the entries the 动态 panel draws
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
The `RecordPageTimelineRequestSchema` docblock, and the record-page file header that the API reference page is generated from, now say what the `timeline` total counts: the sources' rows as the schema's table defines them. That makes it an upper bound of the entries the 动态 panel draws, not their number. The panel draws fewer where its `showCompleted` / `types` configuration filters entries out, and where a ledger row has no displayable diff. Prose only: no schema shape, `.describe()` text or behaviour changes.
Lines changed: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,10 @@
1+
---
2+
'@objectstack/runtime': patch
3+
---
4+
5+
fix(runtime): a flow clone the engine refuses no longer installs the source's empty `.custom` customization package
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
`POST /api/v1/automation/:name/clone` without `package` lands the copy in the source's paired customization package `<base>.custom`, ensuring it if needed. The pair used to be ensured before the engine validated the copy, so a clone refused `400` for an illegal machine name (or any other definition the engine refuses) still left a newly installed, empty `<base>.custom` in the package list and in Studio. The pair is now ensured only after the engine has accepted the copy, so a refused clone changes nothing. Refusals that can only come later — the ensured pair judged not writable, or the store refusing the save — still leave the pair in place, which is its ordinary state: ensure is idempotent, an empty pair is valid, and the next clone reuses it.
Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,18 @@
1+
---
2+
'@object-ui/console': patch
3+
---
4+
5+
fix(console): a refused form submit says a localized sentence, never the server's response body
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
The Console form page (`/forms/NAME`, the in-app form and the public `/f/SLUG`) drew a refused write as the verb, the status and the raw response body, so a member without create rights read `Create failed (403): {"error":…,"code":"PERMISSION_DENIED","object":…}` in English on the banner and the toast. The page now reads the refusal the way the record dialog does, and the banner and the toast say the same sentence:
11+
12+
- a sentence the producer marked for the end user (`userMessage`), or a delegated administrator's scope refusal, as written;
13+
- any other denial (`403`, `PERMISSION_DENIED`, `FORBIDDEN`) as the localized "You don't have permission to save this record.";
14+
- the per-rule lines no input carries, one per line, as before;
15+
- the server's own sentence for any other `4xx`, read out of the envelope (`error`, or the wrapped `error.message`), never the body itself;
16+
- a `5xx`, or a refusal with no sentence (a proxy's markup page), as the localized "Could not save. Please try again."
17+
18+
The fields a refusal names are still marked under their inputs. No i18n key is added: both lines are the existing `form.noPermissionToSave` and `form.submitFailed`, present in all ten packs.

0 commit comments

Comments
 (0)