110 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
james.maes	16bd4f768f	Merge pull request 'perf: skip tag lookups for tables with no tag assignments (forward-port of #67)' (#68) from fix/read-path-statements-develop into develop	2026-11-10 22:11:02 +00:00
james.maesandClaude Fable 5.2	04d04adc6e	perf(security): authorize each distinct parent once per child result page ... ParentRowCustomizer and AttachmentPreviewCustomizer read the parent of every child row to decide whether the row is visible, so rows that share a parent repeated the same read: 61 tasks on two parents ran 61 parent reads (TagLookupStatementBudgetTest), now 2. The answer is kept only for the one postQuery call, where the session and transaction cannot change. Also correct the README: the Tags field is selectable. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>	2026-11-10 15:58:11 -06:00
james.maesandClaude Fable 5.2	de922d7e86	perf(tags): no assignment lookup for tables that have no tags ... Core selects the Tags virtual field for every query or get that names no fields, so every such read of a TAGS-enabled table ran one collaboration_record_tag statement, including reads made one record at a time and tables that were never tagged. Measured with 600 added parents (TagLookupStatementBudgetTest): a 602-row list ran 2 statements, 600 gets ran 600 and 100 single-record updates ran 100; now 1 probe, then 0. TaggedTables remembers, per node and per table, whether any assignment exists (core Memoization registered with ClusterCaches, 30 s by default, withTagLookupCacheSeconds, 0 disables). Writing an assignment clears it at once and after commit and publishes memo-invalidate. The probe ignores row locks and returns no data, so the answer is the same for every session. Tables with assignments are read as before: one lookup per 500 parents. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>	2026-11-10 15:58:11 -06:00
james.maes	c6b077321c	Merge pull request 'fix(upload-journal): accept Amazon Aurora PostgreSQL (full_page_writes=off)' (#65) from fix/aurora-full-page-writes-develop into develop	2026-11-10 15:27:45 +00:00
james.maesandClaude Fable 5.2	2837f4152a	fix(upload-journal): accept Amazon Aurora PostgreSQL (full_page_writes=off) ... The PostgreSQL 17 journal qualification refused every call on Amazon Aurora PostgreSQL, which reports full_page_writes=off and does not expose the parameter in its cluster parameter group. - The settings query gains an eleventh column, to_regprocedure('aurora_version()') IS NOT NULL, which is true on Aurora, false on stock PostgreSQL 17 and never raises. - full_page_writes=off is accepted only when that column is true. Stock PostgreSQL with full_page_writes=off is still refused, and synchronous_commit, fsync, recovery, isolation, time zone and lock timeout are unchanged on Aurora. - The decision is the pure function tornPagesPrevented, covered by an always-on unit test. The opt-in real PostgreSQL 17 suite simulates Aurora with a function named aurora_version() and full_page_writes=off on the server. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>	2026-11-09 20:26:14 -06:00
james.maes	f0f1db7642	Merge pull request 'build(deps): align with Gitea registry (parent 2.2.1, QQQ 4.1.0-RC.26)' (#63) from deps/align-gitea into develop	2026-11-07 18:03:38 +00:00
james.maesandClaude Opus 5.5	c8150bc9cf	build(deps): align with Gitea registry (parent 2.2.1, QQQ 4.1.0-RC.26) ... pom.xml: parent -> 2.2.1 pom.xml: qqq.version: 4.1.0-RC.22 -> 4.1.0-RC.26 pom.xml: qbit-object-storage: 0.1.0-RC.4 -> 0.1.0-RC.10 pom.xml: qbit-redis.version: 0.1.0-RC.1 -> 0.2.0-RC.1 pom.xml: qbit-email.version: 0.1.0-RC.2 -> 0.1.0-RC.11 pom.xml: revision -> 1.2.0-SNAPSHOT Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-11-07 17:55:19 +00:00
james.maes	1c6ae14f59	fix(security): check parent-table read permission with a real GetInput (#62)	2026-11-06 19:30:40 +00:00
James Maes	9248f3f32c	test: keep develop's CollaborationApiHttpTest unchanged	2026-11-06 13:16:55 -06:00
James MaesandClaude Fable 5.1	468dc42751	fix(security): check parent-table read permission with a real GetInput ... ParentRecordAccess.canReadTable passed a bare AbstractActionInput to PermissionsHelper.hasTablePermission, so the READ check did not run as an actual table read. Pass a GetInput for the table instead, which is what a host's table permission rules evaluate. Ported from the GG Forge parent-read qualification. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>	2026-11-06 13:16:06 -06:00
james.maes	73681b3a2e	Merge pull request 'ci: pin shared workflow to v1' (#60) from ci/pin-v1-develop into develop	2026-11-06 16:04:53 +00:00
james.maes	483dc271fe	Merge pull request 'fix(custom-fields): reject over-long STRING values and reload help after publication' (#61) from fix/custom-field-length-and-help-20261106 into develop	2026-11-06 15:32:01 +00:00
james.maes	813894f8c6	Merge pull request 'Integrate attachments, upload journal, presence and Redis presence for KOF22' (#59) from kof22/collab-v2 into develop v1.0.0	2026-11-06 14:52:51 +00:00
james.maesandClaude Fable 5.1	ef625ca300	fix(custom-fields): reject over-long STRING values and reload help after publication ... Bounded STRING custom fields now carry QQQ's native ValueTooLongBehavior.ERROR, so a write longer than maxLength is rejected with a field error and the stored value is preserved. The QQQ default is PASS_THROUGH, which let over-long values through unchecked. After the runtime metadata overlay refreshes (immediately, or in the after-commit callback of the originating transaction), persisted help content is reloaded so sections and fields created by publishing a definition receive their stored helpContent rows without a restart. Two CustomFieldsTest regressions cover invalid-write preservation and stored help on a newly published section. Carried locally by GG Forge as collaboration-custom-fields.patch (supersedes collaboration-field-length.patch). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Signed-off-by: James Maes <james@kof22.com>	2026-11-06 08:39:48 -06:00
james.maesandClaude Opus 5.5	6060b737c5	ci: pin shared workflow to v1 ... Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-11-06 14:06:32 +00:00
james.maesandClaude Sonnet 5.5	024cfc21a9	release: prepare Collaboration 1.0 (Core RC.22 BOM, Redis RC.1, current status docs) ... Replace the default-active Core snapshot profile with the released Core 4.1.0-RC.22 BOM, pin qbit-redis to its first release candidate, and replace the snapshot-era status text with the current 1.0 status. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>	2026-11-05 17:25:50 -06:00
James MaesandClaude Sonnet 5.5	63245b14e0	feat(collaboration): integrate presence, Redis presence, virtual-field and fixture work on the attachment base ... One reviewable integration on current develop 7d61f29e. It starts from feat/record-files-standalone (attachment lifecycle, upload journal, retention SPI; descends from develop) and adds the owner stack content from feature/collab, feature/collab-int-20261029, feature/field-values-1196, feature/presence-recovery-1300, test/presence-capacity-1302 and the fix/attachment-transactions-1289 / fix/mywork-query-bounds-1291 work, plus KOF22 adaptations: CollaborationQBitConfig.attachmentLifecycle, CollaborationRecords.insertActivity/query(transaction), ActivityWriter source feature routing, and the Redis/fixture tests. Build: develop CI files are preserved. pom.xml only adds the Redis API guardrail allow-list and an optional qbit-redis dependency; the core and qbit-redis versions stay on the canonical SNAPSHOT lines with markers to repin after the core and redis integration PRs are published (order CORE -> REDIS -> COLLAB). No PRIVATE or feature-snapshot coordinate is introduced. Not yet compiled: it needs the core integration artifact (virtual-field providers). Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>	2026-11-05 17:19:49 -06:00
James MaesandGPT-6 Codex	fb7c6316bf	docs: record successful Record Files consumer verification ... Co-Authored-By: GPT-6 Codex <noreply@openai.com>	2026-11-05 17:19:48 -06:00
James MaesandGPT-6 Codex	9fbe20c9d6	fix: handle upload recovery metadata and align Core dependency ... Verify all 269 Collaboration tests, including PostgreSQL attachment recovery and MySQL/PostgreSQL journals, against the compatible Core snapshot. Co-Authored-By: GPT-6 Codex <noreply@openai.com>	2026-11-05 17:19:47 -06:00
3 people	1c05422901	chore: preserve paused QBit work and document remaining issues ... Save the current implementation and README, status, and known-issue documentation at the user-requested stop. No completed release or deployment is claimed. Co-Authored-By: Claude <noreply@anthropic.com> Co-Authored-By: GPT-6 Codex <noreply@openai.com>	2026-11-05 17:19:47 -06:00
James MaesandClaude Sonnet 5.5	89e77b67b9	feat(attachments): reconcile reviewed attachment lifecycle, upload journal and retention SPI ... Selective three-way reconciliation onto develop 7d61f29 (base c4eb972 = original ancestor; theirs = reviewed b6fb2382 third-party-src/qbit-collaboration). Attachments package (main and tests), upload journal (H2/MySQL/PostgreSQL), cleanup queue, recovery/adoption processes, attachment schema fields, tables, pagination, delivery schema, help text and docs. Upstream presence work, CI workflows and release convention are untouched. Redis presence, my-work bounds, notifier and custom-field changes in b6 are deliberately not included. Targets qqq 4.1.0-RC.21 BOM. Needs core QBackendTransaction after-rollback callbacks (not in RC.21): see docs/CORE-DEPENDENCY.md. Not compiled against RC.21. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>	2026-11-05 17:19:46 -06:00
james.maes	9c51710f09	Merge pull request 'ci: use shared ci-workflows Maven workflow' (#57) from ci/ci-workflows into develop	2026-11-05 23:15:25 +00:00
james.maesandClaude Opus 5.5	a0f9a172d0	ci: use shared ci-workflows Maven workflow ... Replace the platform-ci release callers with one caller of KofTwentyTwo/ci-workflows maven.yaml (qqq-orb port). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-11-05 22:03:19 +00:00
forge-recovery	7d61f29e92	ci: enroll shared Gitea lifecycle	2026-11-02 22:12:48 +00:00
forge-recovery	6e60585af1	ci: enroll shared Gitea lifecycle	2026-11-02 21:31:05 +00:00
forge-recoveryandCodex	594e755da4	ci: enroll shared Gitea release-candidate lifecycle ... Co-Authored-By: Codex <noreply@openai.com>	2026-11-02 20:14:28 +00:00
3 people	4679821cb4	ci: add the isolated postpublish job to the QRun release caller ... Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Co-Authored-By: Codex <noreply@openai.com>	2026-11-02 15:47:51 +00:00
forge-recovery	f8337989e8	ci: add the QRun release caller (#53)	2026-11-02 04:17:58 +00:00
james.maesandClaude Fable 5.1	6cf98bf1fd	Merge feat/per-session-presence: per-connection viewers and counts (extension 1.2.0) ... Fixes #54. Each merged viewer carries connections; GET /presence?perConnection=true adds one lease per connection with opaque ids; presence.changed carries leaseCount and leases. Security: the raw session uuid is never stored, replicated or returned. The heartbeat uses an HMAC-SHA256 derived sessionId keyed by QQQ_PRESENCE_SESSION_SECRET (random per-process secret when unset, so set it on every pod). The public leases() accessor and Lease record are removed. 162 tests green, including a test that the uuid is absent from every bus message and snapshot. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>	2026-11-01 17:27:25 -06:00
james.maesandClaude Fable 5.1	0ae1e62b7b	fix(presence): never store or publish the session uuid ... The session uuid is a bearer credential (clients resume sessions with it as a cookie or bearer token). The first cut kept it on the lease, replicated it on the internal presence bus and exposed it through a public leases() accessor. Derive an opaque session id instead: "s" + the first 16 hex digits of HMAC-SHA256(secret, sessionUuid), with QQQ_PRESENCE_SESSION_SECRET (shared across pods) or a random per-process secret. The raw uuid is dropped at heartbeat; only the opaque id is stored, replicated and returned (per-connection viewers and presence.changed leases). The leases() accessor and Lease record are removed. Tests assert the raw uuid never appears in bus messages, snapshots or HTTP responses. Refs #54 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>	2026-11-01 16:35:11 -06:00
james.maesandClaude Fable 5.1	47cbc7c72d	feat(presence): per-connection viewers and connection counts ... Two terminals or tabs of one user collapsed into a single viewer, so clients could not tell "Admin, twice" from "Admin". The registry already kept one lease per connection; expose it additively: - merged viewers carry connections (live leases behind the viewer) - snapshots and presence.changed carry leaseCount - GET /presence?perConnection=true adds leases, one viewer per lease with an opaque connection id, mode, field and lease start; presence.changed always carries leases (subscribers cannot opt in) - leases record the server-side session uuid, never sent to clients - counts() still counts users; connectionCounts() counts leases and the Who's here card shows both Extension collaboration 1.2.0; regenerated committed document. Fixes #54 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>	2026-11-01 16:31:37 -06:00
james.maesandClaude Fable 5.1	1ae6508793	fix(presence): always send viewers, empty after the last leave (#53) ... The presence payload omitted viewers once the last viewer left, so clients kept showing the previous list. Viewers are now always serialized, as an empty list after the last leave. Fixes #52 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>	2026-11-01 04:25:29 -06:00
james.maesandCodex	03407ae79a	ci: adopt managed orb 0.8 patch track (#43) ... Co-authored-by: Codex <noreply@openai.com>	2026-10-21 15:26:43 -05:00
james.maesandCodex	9d6767e731	ci: adopt shared orb 0.7.18 (#42) ... Co-authored-by: Codex <noreply@openai.com>	2026-10-21 14:43:27 -05:00
james.maesandCodex	aa6a4a3b58	ci: adopt shared orb 0.7.17 (#41) ... Co-authored-by: Codex <noreply@openai.com>	2026-10-21 14:15:33 -05:00
james.maesandCodex	d9a46afbf9	ci: use the shared branch-aware CodeQL worker (#39) ... Co-authored-by: Codex <noreply@openai.com>	2026-10-21 10:35:35 -05:00
james.maesandCodex	b5308a8bab	ci: adopt central orb 0.7.12 and stable gates (#37) ... Co-authored-by: Codex <noreply@openai.com>	2026-10-21 08:50:12 -05:00
james.maesandCodex	0caaa3717e	ci: centralize rc release completion in qqq orb (#36) ... Co-authored-by: Codex <noreply@openai.com>	2026-10-21 02:23:46 -05:00
james.maesandCodex	464db76bc3	ci: adopt shared orb 0.7.11 (#35) ... Co-authored-by: Codex <noreply@openai.com>	2026-10-21 00:54:30 -05:00
james.maes	19a06166f5	Merge pull request #34 from QRun-IO/feature/central-ci ... ci: adopt shared QQQ orb 0.7.4	2026-10-20 19:32:48 -05:00
james.maesandCodex	61060e2793	ci: adopt central QQQ orb 0.7.4 ... Co-Authored-By: Codex <noreply@openai.com>	2026-10-20 19:26:47 -05:00
james.maes	4e88ed8303	Merge pull request #31 from QRun-IO/ci/release-rc-secrets-inherit ... ci(release-rc): pass secrets so the shared workflow can sign the RC tag	2026-10-16 08:03:57 -05:00
james.maesandClaude Opus 5.5	7a7b804106	ci(release-rc): pass secrets to the shared workflow so it can sign the RC tag ... QRun-IO/.github#2 adds a tag-and-release job that signs vX with the QRUN_TAG_SIGNING_KEY secret after Maven Central confirms the artifact. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-15 16:19:25 -05:00
james.maes	4c95151a86	Merge pull request #29 from QRun-IO/feature/ui-api-1.6 ... feat(ui-api): extension for QQQ UI API 1.6.0 (operationIds, BasicErrorResponseV1, QQQ 4.1.0-RC.15)	2026-10-15 16:12:53 -05:00
james.maesandClaude Opus 5.5	41bbf2c428	chore: target QQQ 4.1.0-RC.15 (UI API 1.6.0) ... Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-15 12:27:04 -05:00
james.maesandClaude Opus 5.5	2f6be765f2	feat(ui-api): extension for QQQ UI API 1.6.0 (operationIds, BasicErrorResponseV1, QQQ 4.1.0-RC.14) ... Regenerated against UI API 1.6.0: every operation has an operationId and the standard error responses (BasicErrorResponseV1: error, code, message, fieldErrors, traceId); the extension's own error schema is kept but deprecated; hand-written error bodies go through QJavalinUtils so they carry code/message/traceId. Extension MINOR bumped. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-15 11:43:37 -05:00
james.maes	92dd901409	Merge pull request #28 from QRun-IO/fix/producer-version-from-build ... fix: qbit metadata version comes from the build	2026-10-15 01:42:56 -05:00
james.maesandClaude Opus 5.5	6a4ea584f5	fix: the qbit metadata version comes from the build, not a hardcoded constant ... The producer read a literal version, so released artifacts reported a stale or -SNAPSHOT version. qbit-collaboration.properties (version=${project.version}) is Maven-filtered and the producer loads it; tests assert producer == bundle == artifact version. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-15 01:27:45 -05:00
james.maes	03b3bbaca2	Merge pull request #27 from QRun-IO/fix/help-bundle-version-test ... test(help): bundle version test compares with the artifact version	2026-10-14 23:18:42 -05:00
james.maesandClaude Opus 5.5	31ce528064	test(help): compare the help bundle version with the artifact version, not a hardcoded SNAPSHOT ... The bundle version is now Maven-filtered, so a literal SNAPSHOT in the test fails every release build. Surefire passes the artifact version in qbit.artifactVersion. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-14 23:07:29 -05:00