Commit Graph
7 Commits
Author SHA1 Message Date
Simon AxelssonandClaude Sonnet 5 9740e20818 feat(data-os): TS/Python quality bridge + provenance fields + fault-injection tests
CI (SIAX Cloud) / security (push) Successful in 23s
Security Scan (SIAX Cloud) / security (push) Successful in 22s
CI (SIAX Cloud) / integration (push) Successful in 29s
CI (SIAX Cloud) / quality (push) Successful in 45s
CI (SIAX Cloud) / security (pull_request) Successful in 27s
CI (SIAX Cloud) / integration (pull_request) Successful in 31s
Security Scan (SIAX Cloud) / security (pull_request) Successful in 29s
CI (SIAX Cloud) / quality (pull_request) Successful in 48s
Continues P010 (branch renamed from feat/p010-source-registry-persistence):
closes the three remaining BEFORE_STATE.md gaps (§2 TS/Python split-brain,
§3 missing provenance fields, §4 untested failure modes), all additive to
the production pipeline — the Python QA filter's delivery decision is
unchanged by any of this.

TS/Python bridge (§2):
  - New route surface app/api/0pt/quality/* (rules/runs/issues CRUD +
    POST /api/0pt/quality/bridge-run) over the shared QualityService
    singleton (src/services/quality.ts), fixing the same "no shared
    instance" defect the Source Registry had.
  - src/quality/quality-bridge.ts auto-provisions a deterministic,
    tenant-scoped baseline rule set on first use so the Python engine never
    pre-registers rules through the CRUD surface.
  - services/api/workers/quality_bridge.py: the Python-side HTTP client.
    Feature-flagged off by default (OPT_QUALITY_BRIDGE_ENABLED), fails open
    on any error (never raises into run_job, never changes `delivered`).
    Advisory only — real per-row gating against the TS engine isn't
    implementable yet (ExpectationEngine returns dataset-level pass/fail
    counts + samples, not a per-row boolean array); documented, not faked.
  - Live-verified end-to-end: `next dev` (AUTH_ENFORCE=false) + the real
    Python pipeline hitting the real brreg.no API, round-tripping through
    the real bridge route, receiving a real QualityService score.

Provenance (§3): sources/connectors.py's Provenance dataclass gains
retrieved_at, transform, connector_version, producer, quality,
entity_resolution, artifact_ref (all previously absent from the real
dataflow entirely) via a make_provenance() factory used by every connector.
resolve.py stamps entity_resolution on merged records' provenance (never on
unmerged ones). pipeline.py stamps artifact_ref at delivery and quality
once the bridge succeeds.

Failure tests (§4), real fault injection against real code (services/api,
new pytest + respx suite, not vitest mocks):
  - dedup-determinism + duplicate-input + out-of-order: resolve.py under
    shuffled/interleaved/exhaustive-permutation input, plus an explicit,
    verified test of the one case that IS legitimately order-sensitive
    (non-transitive similarity in a greedy clusterer) instead of hiding it.
  - schema-evolution: respx-mocked upstream responses with added/removed/
    alien-shaped fields through the real brreg connector + pipeline.
  - restart: connector exceptions mid-job (including one from respx itself)
    never crash run_job.
  - backfill: the real property that holds given no persistence layer
    exists (replay is compute-idempotent) — NOT a fabricated dedup-store
    test.
  - CDC-interruption and true backfill/delivery-store idempotency:
    explicitly documented as not implementable against this codebase
    (no CDC pipeline, no delivery store exist anywhere) via skip markers
    with reasons, in tests/test_known_gaps.py, rather than silently
    dropped or faked.

Testing: 256/256 vitest (249 pre-existing + 7 new bridge route/service
tests) and 28/28 pytest (2 documented skips) green locally. tsc -p
tsconfig.check.json has zero new errors (14 pre-existing errors, all in
the untracked/unaudited src/fleet, src/graph, src/metrics, src/projections
paths this pass does not touch, per BEFORE_STATE.md).

No PR opened, per instructions.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-09-01 21:02:14 +02:00
Simon AxelssonandClaude Sonnet 5 07b13bd833 feat(data-os): Source Registry Postgres persistence + full HTTP route surface
Replaces the in-memory (per-instance, not even per-process-shared) Maps in
SourceService/SourceRegistryService with real Postgres persistence, and adds
the register/query/describe/health/rights/provenance/freshness HTTP routes
that never existed (grep for SourceRegistryService in app/ previously
returned nothing).

- sql/002_source_registry.sql: opt_sources, opt_source_rights,
  opt_source_provenance, opt_source_health_checks, opt_source_freshness.
  Numbered 002 (not 006) because 004's opt_source_rate_policies already
  forward-references opt_sources, which no sql/ file defined until now —
  fixes 004's pre-existing broken migration ordering (opt_datasets remains
  a separate, unrelated, still-open gap in 004, flagged not fixed).
- src/repositories/*: SourceRepository / SourceRegistryRepository seams,
  each with an in-memory implementation (business logic extracted verbatim)
  and a Postgres implementation, chosen via resolveStoreMode — production
  still fails hard with no DATABASE_URL (Master-DoD §3), but a configured
  production now actually works instead of always being blocked.
- SourceService / SourceRegistryService: storage moved behind the
  repository seam; getHealth/getFreshness are now async and tenant-scoped
  at the accessor itself (needed once they're reachable directly over HTTP,
  not only after an earlier caller's tenant check).
- src/services/source-registry.ts: shared singleton instances for the
  sources/* route group, fixing the "each route module owned its own
  private registry" defect (source-rate-policies and .../query previously
  couldn't see a source registered via POST /api/0pt/sources).
- New routes: POST/GET /api/0pt/sources, GET /api/0pt/sources/[id],
  GET+PUT .../rights, GET+PUT .../provenance, GET+POST .../health,
  GET .../freshness.
- Business logic unchanged: fail-closed purpose gate, cross-tenant
  NOT_FOUND (no existence oracle, §50), bounded 100-row health history.

Tested against a real Postgres container (not mocked) —
tests/p0/source-registry-postgres.realinfra.test.ts, env-gated on
SOURCE_REGISTRY_TEST_DATABASE_URL, skips honestly when unset. All 249
pre-existing P0 tests + 7 new realinfra tests pass; also smoke-tested the
full HTTP route surface end-to-end against next dev + live Postgres.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-09-01 20:42:41 +02:00
Simon AxelssonandClaude Fable 5 b46807da75 feat(data-os): P0 capability layer — source registry, search, quality, dataset versions, metadata, analytics, features, ER
- Source Registry (§10): rights/license/purpose metadata, provenance,
  freshness SLA, health checks, describe/query; purpose gate FAILS CLOSED
  (no rights record => deny); cross-tenant probes => NOT_FOUND.
- Search (§10/§42): OpenSearch REST adapter (no SDK) + SearchService with
  MANDATORY tenant + purpose/rights filtering, versioned index definitions,
  bulk, generation-per-rebuild (opt-{name}--g{n}) with atomic alias cutover,
  rollback, failed-rebuild generation discard. OpenSearch never canonical.
- Quality (§10): versioned rule registry (GE-JSON interchange), local
  engine for the GE core subset (never silently passes unsupported types),
  runs/results/issues/score/trend, whole-run delegation to the GE backend
  via N0D (0pt.quality.ge_run) - never fabricated results.
- Dataset (§10/§5/§53): immutable numbered versions holding canonical
  st0re:artifact refs only, idempotent createVersion, export registers the
  ST0RE artifact + AUD0 evidence + OpenLineage.
- Metadata (§10): OpenMetadata PUT /api/v1/tables projection adapter +
  projectAll rebuild path; projection never canonical.
- Analytics (§10): ClickHouse HTTP adapter, closed metric-query DSL with
  FORCED tenant predicate + readonly/max_execution_time/max_result_rows,
  TTL retention. Zero connection to b0ts/quarantined pipelines.
- Features (§10): versioned feature-view registry, real Feast
  get-online-features adapter, materialize/get-historical as N0D delegation.
- ER (§10): generic candidate API with deterministic blocking + transparent
  scoring + per-field explanations, product hard-rule hook (force both
  directions), review-band handoff, Splink batch via N0D contract.

93 new unit/contract tests; suite 248 passed / 1 skipped.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-31 16:49:32 +02:00
Simon AxelssonandClaude Fable 5 b7038d7675 feat(platform): canonical SIAX context/errors + real-contract clients, no prod mock fallback
- src/platform/: SiaxRequestContext + propagation headers + 12 canonical
  error codes (documented mirror of cl0ud packages/contract), SiaxError,
  resource refs (st0re:artifact:...), and the §3/§10 dependency-mode guard:
  production runtime + unconfigured dependency/store => hard
  DEPENDENCY_UNAVAILABLE at construction, never silent mock/in-memory.
- src/clients/: all four platform clients rewritten against the REAL wire
  contracts verified on sibling main branches this session:
  * cl0ud: POST /api/policy/evaluate ({data} envelope, allow+requiresApproval)
    - the old /v1/cl0ud/policies/evaluate path never existed on CL0UD
  * st0re: camelCase ArtifactInput on /v1/st0re/artifacts/register incl.
    X-Tenant-Id/X-Actor-Id + X-Idempotency-Key - the old snake_case body
    would have been a 400 against the live registry
  * n0d: required_role + resource_requirements + body idempotency_key +
    CloudEvents callback_url per n0d createJobSchema; adds getJob/getJobResult
  * aud0: /v1/aud0/openlineage is AUD0's only lineage ingest -
    /v1/aud0/lineage/events never existed; legacy shape converts to a
    spec-shaped OpenLineage RunEvent (deterministic runId dedup)
- src/lineage/: OpenLineage subset types + LineageEmitter (never-throw,
  §16 evidence semantics)
- shared SiaxHttp transport: timeouts, typed SiaxError mapping, context
  propagation, no raw upstream bodies in errors

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-08-31 16:35:01 +02:00
adminandClaude Sonnet 5 bda21e87eb Live-verify Iceberg write path against ST0RE + close field/contract gaps
Re-runs this session's task spec against 0pt's actual repo state, which
turned out to already have a substantial, tested implementation from a
prior pass (35b6ec6): dataset<->Iceberg-table mapping, source rate policy,
and an ST0RE lakehouse client that was honestly marked "not live-verified".

This pass found api.st0re.siax.io IS reachable from this environment and
used it -- first via curl to safely discover the real contract, then
through 0pt's own updated code -- to drive a real namespace -> table ->
snapshot write against production ST0RE (task item 3's "write ONE first
test dataset" proof). That surfaced and fixed three real bugs the
mock-only pass could not have caught:

- ST0RE requires X-Tenant-Id/X-Actor-Id headers on every lakehouse route;
  the previous HTTP client sent none. Fixed via a threaded
  St0reRequestContext.
- IcebergTableInput.datasetId is a foreign key into ST0RE's OWN dataset
  table, not a free-form string -- sending 0pt's dataset_id 400s live with
  a Postgres FK violation. Fixed by no longer sending it; 0pt's
  dataset<->table linkage lives entirely in DatasetIcebergMapping.
- ST0RE enforces a real unique constraint on lakehouse namespaces
  (tenant_id, project_id, name) with no upsert route. Fixed with a
  client-side lookup-or-create (listNamespaces + findOrRegisterNamespace).

Also closed a field-level gap against the task's literal schema spec:
0pt.dataset_iceberg_mapping.schema.v1 now carries namespace_id (a real
reference to ST0RE's LakehouseNamespace.namespaceId, not guessed) and
mapped_at, both additive alongside the existing namespace/created_at
fields (sql/005, JSON schema, types, service).

tests/p0/iceberg-write-service.realinfra.test.ts replaces the prior
it.todo with a real SEAM-pattern live test: attempts the real HTTP call,
skips honestly if ST0RE is unreachable, never fabricates success. It ran
successfully against production ST0RE this session (real namespace/table/
snapshot UUIDs logged in docs/ICEBERG_WRITE_BOUNDARY.md).

Two ST0RE-side cross-tenant read bugs were found and flagged (not fixed
here -- out of scope for this repo): GET .../namespaces ignores its own
tenant header and query filters; GET .../tables/:id doesn't check the
caller's tenant against the table's owner.

npm test: 155 passed, 0 failed, 0 todo (up from 154 passed + 1 todo).
npx tsc --noEmit: 1 pre-existing, unrelated error in
quality-report-service.ts (present before this pass, part of the wider
uncommitted P0 data-platform build already in this repo) -- not touched.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-07-13 00:18:10 +02:00
adminandClaude Sonnet 5 35b6ec63bf Add 0pt Iceberg write-path ownership: dataset<->table mapping + source rate policy
Real, tested implementation of docs/11-lakehouse-write-plan.md's items 2-8
(the "future implementation session" that doc named): 0pt is now the
canonical writer of Iceberg tables through ST0RE's real, shipped
/v1/st0re/lakehouse/* table registry (sax3l/st0re commit 9ea0abb), never
persisting lakehouse table/catalog/snapshot state itself.

- src/types + src/services for 0pt.dataset_iceberg_mapping.schema.v1
  (dataset -> ST0RE tableId reference) and 0pt.source_rate_policy.schema.v1
  (the sourceRatePolicy ownership repair, moved out of N0D per ADR-0030).
- src/clients/st0re-lakehouse-client.ts: new client matching ST0RE's actual
  camelCase route contract, with the same mock/HTTP boundary convention as
  this repo's existing cl0ud/n0d/aud0/st0re clients. Honest caveat in its
  module doc comment: not verified against a live ST0RE instance in this
  pass (no reachable ST0RE_URL); every test exercises the deterministic mock.
- src/services/iceberg-write-service.ts: registers tables/snapshots through
  ST0RE, requests N0D jobs only for heavy writes (never executes them
  in-process), emits all four AUD0 event names
  (dataset_iceberg_table_mapped / iceberg_write_requested /
  iceberg_snapshot_created / source_rate_policy_created).
- sql/004, packages/schemas/*.json, app/api/0pt/lakehouse + sources/rate-policies
  routes, tests/p0/*.test.ts (154 passed + 1 honest it.todo for the live-ST0RE
  case, up from 142 baseline, 0 failures).
- Reconciled a real collision: an earlier untracked draft schema/test pair
  (tests/p0/enterprise-data-os-expansion.test.ts) asserted a different,
  now-superseded source_rate_policy shape; updated that one test block only.
- docs/ICEBERG_WRITE_BOUNDARY.md states the ownership boundary plainly and
  is explicit throughout about what's real-and-tested vs. contract-only.
  docs/11-lakehouse-write-plan.md gets a status-update section, not a rewrite.
- Data quality rules (quality-check-service.ts, quality-report-service.ts)
  were not touched -- confirmed still 0pt-owned.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-07-12 19:46:47 +02:00
adminandClaude Sonnet 5 c89731115c feat: add 0pt.query_source, 0pt.document_ocr, 0pt.write_external_record capabilities
Three new, tenant-agnostic P0 capabilities:

- 0pt.query_source: bounded, timeout/row-capped ad-hoc SQL query execution
  against registered database sources, via a Fabric/OneLake adapter
  (Azure AD credential-based mssql connection), with AUD0 audit events
  and an app/api/0pt/sources/[sourceId]/query endpoint.

- 0pt.document_ocr (document capture + extraction): mailbox-based document
  intake via Microsoft Graph, OCR/field extraction via Azure Document
  Intelligence, with capture/extraction API endpoints and services that
  never fabricate extracted values.

- 0pt.write_external_record (external writeback): a generic writeback-adapter
  contract plus a concrete Odoo bills adapter that matches partners via
  Fabric, backing an external-writeback-service for pushing records to
  downstream systems of record.

Known interim gaps:
- src/adapters/local-blob-store.ts is an interim ST0RE substitute pending
  the real ST0RE artifact-store integration.
- mssql is used by the Fabric/OneLake adapter but deliberately not yet
  added to package.json (pending `pnpm add mssql` once the concurrent
  session's package.json/dependency changes settle).

A separate, concurrently-active session has unrelated in-flight changes
in this same working tree (data-platform catalog/lineage/feature-store,
dataset quality, lakehouse/Iceberg docs, deploy scaffolding, package.json/
pnpm-lock.yaml updates, etc.) — none of that is touched or included here.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-07-08 19:32:43 +02:00