Skip to content

Commit 5aff45c

Browse files
committed
Merge remote-tracking branch 'origin/main' into claude/project-thread-a2d68g
Claude-Session: https://claude.ai/code/session_01Utp6Cb8MSvSrJz9nXGSj2F Co-authored-by: Claude <noreply@anthropic.com>
2 parents 98ed720 + 0e4dfbe commit 5aff45c

88 files changed

Lines changed: 6867 additions & 1250 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: 20 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,20 @@
1+
---
2+
'@objectstack/plugin-approvals': patch
3+
'@objectstack/plugin-sharing': patch
4+
---
5+
6+
fix(plugin-approvals)!: an approval node refuses to open when a team or position approver's directory read fails, or a queue approver's expansion rejects, and the ownership-queue reads page to completion
7+
8+
Clause-②: yes (narrowing) — an approval node no longer opens when a `team` or `position` approver (static, or an `expression` approver with `resolveAs` `team` / `position`) cannot be expanded because a directory read or the cross-organization read filter failed, or when a `queue` approver's expansion rejects. Under every `onEmptyApprovers` policy, the `fallbackApprovers` slate included, it now fails the node with a retryable `APPROVER_DIRECTORY_UNREADABLE:` error naming the read, where it used to open on the rest of the slate, on the fallback slate or on the administrators, or auto-approve. The registered ownership-queue service does not reject yet: a failed member read inside it still answers nobody until #29604 part 2 (#30251), so a `queue` approver refuses here only when `expandMembers` itself rejects. Every other input answers as before.
9+
Metadata-protocol: none
10+
11+
<!-- adr-0087: not-required (no-migration-prescription) No key of any schema is added, removed, renamed or re-shaped: the approval node config, the approver types and the ownership-queue contract keep their shapes. What moves is the answer a node gives while a directory table cannot be read, and nothing at rest records that answer, so there is nothing for a migration to rewrite. -->
12+
13+
**BREAKING** accept-set narrowing at the approval node (`@objectstack/plugin-approvals`): a failed directory read behind a `team` or `position` approver, or a `queue` approver whose expansion rejects, refuses the node instead of opening it on a slate that is missing people (ADR-0142 D6 as amended, ruling record 6599731698 on #29604).
14+
15+
- **Before.** A `team` or `position` approver whose `sys_team_member`, `sys_user_position` or `sys_member` read failed kept its `type:value` literal slot, which nobody can act on, and logged the failure. A `queue` approver whose expansion threw answered "expanded to nobody" and added no slot. Either way the node went on as if those people did not exist: beside other approvers it opened on them alone, and on an otherwise empty slate the node's `onEmptyApprovers` policy acted, opening on the `fallbackApprovers` or the organization's administrators, or auto-approving.
16+
- **Now.** The node does not open and writes no request, approver or action row. It fails with an `ApproverDirectoryUnreadableError` whose message starts `APPROVER_DIRECTORY_UNREADABLE:`, names the approver, the organization and the read that failed (a failure of the cross-organization read filter is named as that filter's read, on a static approver and on an `expression` one alike), and is retryable: the flow node's failure is a runtime one, so the run can be resumed once the directory can be read. A `fallbackApprovers` slate whose own `team` or `position` read fails refuses the same way. `department` / `business_unit`, `org_membership_level` and `manager` approvers are not named by the ruling and answer exactly as before.
17+
- **Unchanged.** A decision on a request opened before the approver snapshot existed still re-reads its slate on the old path, which keeps the literal slot on a failed read, because the decision has already been recorded when that read happens.
18+
- **Not yet.** The ownership-queue service shipped in `@objectstack/plugin-sharing` answers a failed `sys_queue` / `sys_queue_member`, team or position read with nobody (logged once) rather than rejecting, so a `queue` approver backed by it still adds no slot on such a failure. Its refusal arrives with #29604 part 2, once the contract can carry it (#30251).
19+
20+
`@objectstack/plugin-sharing`: the ownership-queue service reads its `sys_queue` and `sys_queue_member` rows by id, page after page, to the end, where it read one page of 10,000 rows and answered on it. A queue's 10,001st member, an organization's 10,001st queue and the queues naming a recipient past that bound are now found. A read that fails partway answers as a failed read did before (that read contributes nobody, logged once, with the read's reason and without the walk's own "refuses instead" clause), never with the pages it had read.
Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
---
2+
'@objectstack/objectql': patch
3+
---
4+
5+
refactor(objectql): the private docblocks of 36 hand-series stubs in `engine.ts` move into their hand modules, onto the functions that hold their bodies; no behaviour or public declaration changes
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
- A new third pass of the `engine.ts` codemod, `--hand-docs`, moves a private hand-series stub's docblock onto the function its stub calls, as the first two passes already do for the bodies they move. A public or protected method keeps its docblock on the class.
11+
- This is the first of two slices: 8 hand modules here, the other 9 next. `engine.ts` goes from 13,550 to 12,618 lines.
12+
- Nothing to do on upgrade: the exported names, the public `.d.ts` surface and what every method does are the same.
Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,11 @@
1+
---
2+
'@objectstack/spec': patch
3+
---
4+
5+
fix(spec): the liveness ledger's `query.searchFields` row cites `engine-find.ts#expandSearchOnAst`, where the hand-off lives, so `check:liveness` is green again
6+
7+
Clause-②: no
8+
Metadata-protocol: none
9+
10+
- The row cited `engine.ts#expandSearchOnAst`. The body has lived in `engine-find.ts` since the `engine.ts` codemod split, and the last mention of `searchFields` in `engine.ts`, a private docblock, moved out with the first slice of the hand-series docblock move, so the check found no mention of the key in the cited file.
11+
- Nothing to do on upgrade: only the shipped `liveness/query.json` evidence text changes.
Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
---
2+
'@objectstack/metadata-protocol': minor
3+
'@objectstack/rest': minor
4+
---
5+
6+
fix(metadata-protocol,rest): a `.custom` pair's declared override of a page, dataset, app, book, dashboard, view or report is labelled in the ruled order on every `/meta` read — the pair's own translation, then the override's own label, then the base's translation (#30062)
7+
8+
Clause-②: yes (widening) — `ObjectStackProtocolImplementation` gains six public methods: `getPackagedPageBase`, `getPackagedDatasetBase`, `getPackagedAppBase`, `getPackagedBookBase`, `servedOverrideBaseOf` and `getPairTranslationBundle` (measured on `packages/metadata-protocol/dist/index.d.ts`); `@objectstack/rest`'s exported `MetaBookTreeSources` accepts one more optional key, `resolveProtocol` (measured on `packages/rest/dist/index.d.ts`). No accepted key, value or answer is taken away.
9+
Metadata-protocol: semantic
10+
11+
- **Now.** A changed label on a declared override reads the pair's own translation of that key when the pair ships one, else the override's own text, in every locale, on `GET /meta/:type/:name`, `GET /meta/:type` and `GET /meta/book/:name/tree`. An unchanged label still reads the base's translation. Before, a page, dataset, app or book override read the base's translation over its own changed label, and no type read the pair's translation first.
12+
- **The pair's translations.** They are the pair's own active `translation` rows (`package_id` = the pair). A `.custom` package has no code bundle, so it has no code translation source. A package-less authored row is in the deployment catalog but is not the pair's, so it does not rank first.
13+
- **Object.** An object is extended, never replaced (ADR-0146 D10), so it has no declared override and no pair bundle: seven types take one.
14+
- **The book tree.** `GET /meta/book/:name/tree` now judges a book against its packaged base as the list does, so the docs sidebar and the book list serve the same label (#26690).
Lines changed: 22 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,22 @@
1+
---
2+
'@objectstack/plugin-auth': patch
3+
'@objectstack/service-settings': patch
4+
---
5+
6+
fix(plugin-auth)!: saving the auth settings refuses to turn off the last way to sign in (#30202)
7+
8+
Clause-②: yes (narrowing) — a save an administrator could make before is now refused at the settings save door with the existing `SETTINGS_VALIDATION` / `invalid_value` field error: a save after which email and password is off and nothing else signs anyone in, judged on the whole save, including the Google change that removes the last provider while email and password is off. No new error code or settings key. The widening half: `SettingsWritePredicateInput`, exported from `@objectstack/service-settings`, gains an optional `batch` field, the whole write a predicate is asked about.
9+
Metadata-protocol: none
10+
Skills-docs: none — no file under `skills/**` teaches `auth.email_password_enabled` or the settings write predicate; `content/docs/deployment/environment-variables.mdx` already names the `OS_AUTH_EMAIL_PASSWORD_ENABLED` override used for recovery below.
11+
12+
<!-- adr-0087: not-required (no-migration-prescription) No key of any schema is added, removed, renamed or re-shaped: auth.email_password_enabled and the auth.google_* keys keep their declarations, and the save door now refuses only the save that leaves no way to sign in. The predicate input's new batch field is optional and additive. A value stored before is not rewritten (the existing OS_AUTH_EMAIL_PASSWORD_ENABLED override recovers it), so there is nothing at rest for a migration to rewrite. -->
13+
14+
**BREAKING** accept-set narrowing at the auth settings save door (`PUT /api/settings/auth`, `SettingsService.setMany('auth', …)`): a save that would leave no way to sign in is refused, where it was saved before.
15+
16+
- **Before.** Setup → Settings → Authentication, "Enable email/password login" off, save: "Settings saved", with no social provider and SSO off. After the next sign-out nobody could sign in, the administrator included.
17+
- **Now.** A save after which email and password is off and no other way in is in force answers `400 SETTINGS_VALIDATION`. The field error sits at `email_password_enabled` (`code: invalid_value`, `constraint.lastSignInMethod: email_password`), and its message, in the caller's language (en, zh-CN, ja-JP, es-ES), says to keep or set up another sign-in method before or in that save. Nothing in that batch is written, and email and password stays on.
18+
- **The whole save is judged.** `{ email_password_enabled: false, google_enabled: false }` (or with `google_client_secret` cleared) on a deployment signing in with email and password plus Google is refused, with a field error at each of the two keys. Google filled in (switch, client ID and secret) in the same save that turns email and password off is accepted, and so is the switch alone over a client ID and secret already stored.
19+
- **What counts as another way in.** A social or OIDC provider in force whose client ID and client secret are both set; enterprise SSO switched on with at least one identity provider registered (the switch alone does not count); phone-number sign-in; a host sign-in handoff the host declared. `ssoOnlyMode` alone, magic link, passkeys, and a provider listed without its credentials do not count.
20+
- **The other direction.** While email and password is off, `google_enabled: false`, or clearing `google_client_id` / `google_client_secret` when no environment variable or stack config supplies it, is refused (`constraint.lastSignInMethod: google`) if Google is the last way in.
21+
- **Already locked out.** Set `OS_AUTH_EMAIL_PASSWORD_ENABLED=true` and restart: the environment override wins over the stored `false`, and email and password sign-in comes back. Then set up another method before turning it off again.
22+
- **For a write predicate's author (`@objectstack/service-settings`).** `SettingsWritePredicateInput.batch` holds every key the write sets in the namespace, the asked key among them, with the value it would store (`null` clears). `setMany` always passes it, frozen. A rule over several keys judges the state after the whole write from it instead of one key against the rest as they stand. Existing predicates need no change.
Lines changed: 20 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,20 @@
1+
---
2+
'@objectstack/plugin-auth': patch
3+
'@objectstack/service-settings': patch
4+
---
5+
6+
fix(plugin-auth)!: clearing the allowed IP ranges applies at once, and a console save that leaves out the saving address is refused
7+
8+
Clause-②: yes (narrowing) — a settings console save (`PUT /api/settings/auth`) of an `allowed_ip_ranges` list none of whose entries matches the saving request's own address is refused, where it was stored before; and `@objectstack/service-settings` widens its published surface: three new exported types (`SettingsWriteOptions`, `SettingsRequestTransport`, `SettingsWritePredicateOptions`), `setMany`'s fourth parameter accepts `SettingsWriteOptions`, `registerWritePredicate` takes an optional fourth `options`, and `SettingsWritePredicateInput` gains an optional `transport`.
9+
Metadata-protocol: none
10+
Skills-docs: none — `skills/**` teaches no `allowed_ip_ranges`; `content/docs/permissions/authentication.mdx` describes the setting and does not yet state the save refusal or the recovery, which ship here in the refusal text and this entry (the page is outside this change's surface and is named for a follow-up).
11+
12+
<!-- adr-0087: not-required (no-migration-prescription) no metadata an author writes changes and no stored row is rewritten: a console save of a setting value is refused when it would lock out the saving request -->
13+
14+
**BREAKING** for an administrator who saves, through the settings console, an allowed-IP list (ADR-0069 D5) that does not include the address they are connected from.
15+
16+
- **Clearing applies at once.** The console sends an emptied field as `null`, which resolves with source `default`, and the settings binding used to skip a default-sourced `allowed_ip_ranges`: the last stored list kept refusing sign-in (`403 IP_NOT_ALLOWED`) until a restart while the field showed empty. A cleared or restored-default value now hands the gate back to the deployment's own declaration in the same pass: no list when it declared none, the `allowedIpRanges` the plugin options or an `auth:configure` handler declared otherwise.
17+
- **The save is refused.** A list no entry of which matches the saving request's address is refused with `400 SETTINGS_VALIDATION`, `invalid_value` at `allowed_ip_ranges`, the address in `constraint.clientAddress`, and a message in the caller's language that names it. The address is the one the gate judges at sign-in: the transport's peer address, or the forwarded one when the stack declares `server.trustProxy`. When the server cannot establish the address, every non-empty list is refused, because the gate would refuse that request too. Clearing the list is never refused, and an in-process write (boot, seed, migration) carries no request and is not judged.
18+
- **Recovery when already locked out.** Start the server with `OS_AUTH_ALLOWED_IP_RANGES` set to an empty value. The settings env override wins over the stored list and opens the gate; remove it once the list is fixed.
19+
- **`@objectstack/service-settings`.** `setMany`'s options (`SettingsWriteOptions`) and a write predicate's input gain an optional `transport` (`SettingsRequestTransport`: the transport's peer address and request headers), which the settings routes set on a save so a write predicate can judge the change against the caller's own connection. A predicate sees only the headers it names at registration (`registerWritePredicate(…, { requestHeaders })`, `SettingsWritePredicateOptions`), so a credential header never reaches one that did not ask for it. The transport identifies and authorizes nothing and is never stored.
20+
- **Who moves.** An administrator adds their own address or range to the list, or saves it from an address inside it.
Lines changed: 15 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,15 @@
1+
---
2+
'@objectstack/cloud-connection': patch
3+
'@objectstack/cli': patch
4+
---
5+
6+
fix(cloud-connection,cli): the install-local ledger follows `OS_HOME`, so `os dev --fresh` and a second instance under another `OS_HOME` start without an earlier run's installs
7+
8+
Clause-②: yes (widening)
9+
Metadata-protocol: none
10+
11+
- **Now.** With no `storageDir`, the install-local ledger (`LocalManifestSource`, and so `MarketplaceInstallLocalPlugin`) lives at `$OS_HOME/installed-packages` when `OS_HOME` is set, and at `CWD/.objectstack/installed-packages` when it is not. Before, it was always the working directory, so `os dev --fresh` (a fresh `OS_HOME` and database) rehydrated a package an earlier run installed, and so did a second instance under another `OS_HOME`.
12+
- **Unchanged.** `os dev` without `OS_HOME` keeps the ledger in the project, beside its database. `os start` in project mode sets `OS_HOME` to `CWD/.objectstack`, which is the same directory as before. An explicit `storageDir` is used as given.
13+
- **Existing installs.** When the ledger moved (an `OS_HOME` other than `CWD/.objectstack`) and its new directory does not exist yet, the boot copies the `.json` files of the working-directory ledger into it once, and logs a warning naming both directories. The old files are left in place and are not read again; remove them. A copy that fails is logged as an error, and the next boot tries again. Under `os dev --fresh` nothing is copied.
14+
- **`os doctor`** reads the same directory the runtime does, `OS_HOME` included.
15+
- **The widening.** `@objectstack/cloud-connection` exports `resolveInstalledPackagesDir({ env?, cwd? })` (returns `{ dir, legacyDir }`), `INSTALLED_PACKAGES_HOME_SUBDIR`, and `LocalManifestSource.legacyDir` / `adoptLegacyEntries()`. `os doctor` takes the ledger directory from `resolveInstalledPackagesDir`; a `@objectstack/cloud-connection` without it is reported as a reader that does not declare the ledger directory. Align the two packages' versions.

0 commit comments

Comments
 (0)