83 Commits
Author SHA1 Message Date
tliu93 355c8eb8b8 fix: add paho-mqtt and pymodbus to dev-requirements lock (CI install was missing M5 deps)
pytest / test (push) Successful in 5m13s
frontend / frontend (push) Successful in 1m48s
2026-06-23 20:06:13 +02:00
tliu93 1ffdaebbbd M6: add Tibber API + NL energy pricing + DSMR reference doc
frontend / frontend (push) Successful in 1m54s
pytest / test (push) Failing after 32s
2026-06-23 19:56:02 +02:00
tliu93 c64693b8e5 M6: add design doc — generic pricing layer (manual/tibber) + DSMR cost engine 2026-06-23 19:53:26 +02:00
tliu93 c89d083942 M5: add read-only per-meter readings tabs to Records page; English-only UI strings 2026-06-22 22:16:55 +02:00
tliu93 6090b98ab0 M5: enable SQLite foreign_keys pragma for DB-level FK enforcement 2026-06-22 20:45:47 +02:00
tliu93 489cd7df3d M5: default Modbus polling to off (opt-in) to avoid unnecessary DB writes 2026-06-22 20:28:12 +02:00
tliu93 35cdc7f22a M5: use DB-merged runtime settings for MQTT/discovery; add MQTT test button; move Expose panel into config accordion 2026-06-22 20:11:54 +02:00
tliu93 75d685f04d M5: roll Energy chart window for live auto-refresh (English label); render boolean config fields as switches 2026-06-22 19:40:18 +02:00
tliu93 935e68846c M5: Energy view — per-metric decimal formatting and chart auto-refresh toggle 2026-06-22 19:18:29 +02:00
tliu93 4767a7af60 M5-T13: document Modbus/Energy/MQTT, mark M5 done, finalize OpenAPI 2026-06-22 16:10:11 +02:00
tliu93 45d87f36f7 M5-T12: add expose toggle API and Home Assistant Expose settings panel 2026-06-22 15:50:49 +02:00
tliu93 35b95f5c88 M5-T11: add HA discovery and state publishing wired to polling 2026-06-22 15:32:43 +02:00
tliu93 9d05f5ea62 M5-T10: add paho MQTT client with lifespan connect/reconnect and mqtt/test endpoint 2026-06-22 15:06:30 +02:00
tliu93 360cddc05b M5-T09: add exposed_entities table and ExposableEntity provider framework 2026-06-22 14:50:40 +02:00
tliu93 14b846549e M5-T08: add MQTT, HA Discovery, and Modbus polling config fields 2026-06-22 14:27:21 +02:00
tliu93 68165f0e01 M5-T07: add latest-reading cards and Recharts trend charts; readings return most-recent rows 2026-06-22 14:17:59 +02:00
tliu93 2dc469274a M5-T06: add Energy device management UI and sidebar entry 2026-06-22 13:52:33 +02:00
tliu93 702a1cec00 M5-T05: add Modbus JSON API (device CRUD, readings, metrics, test) 2026-06-22 13:37:48 +02:00
tliu93 49d15d8ffe M5-T04: add Modbus poll service and APScheduler polling job 2026-06-22 13:24:11 +02:00
tliu93 c89cea4953 M5-T03: add Modbus TCP driver, YAML profile framework, sdm120, modbus_cli 2026-06-22 13:03:50 +02:00
tliu93 d9c82e39fb M5-T02: add modbus_device and modbus_reading tables and models 2026-06-22 12:45:50 +02:00
tliu93 eac7860b51 M5-T01B: render ConfigPage sections as Mantine Accordion 2026-06-22 12:34:04 +02:00
tliu93 da7d0f9d62 M5-T01: refactor AppLayout from top bar to sidebar navigation 2026-06-22 12:26:52 +02:00
tliu93 e49a6ad963 M5(design): generic modbus_device + JSON readings; YAML profiles; Accordion config; manual CLI/MQTT testing
frontend / frontend (push) Successful in 1m26s
pytest / test (push) Successful in 3m52s
Rework the M5 design doc data model and add manual-testing surface:

- Two-layer model: rename energy_meters/energy_readings -> generic
  modbus_device + modbus_reading; readings become a JSON payload blob
  instead of a wide fixed-column table (aggregation via SQLite
  json_extract, generated-column index as a future lever).
- Protocol knowledge moves to read-only YAML profiles (data+functions,
  no OOP inheritance); deployment/config (friendly_name, unit_id, host)
  stays on the device row; multiple devices share one profile.
- UUID = stable internal identity, also anchors HA discovery unique_id
  (Z2M rename model); device metadata for HA derived from the profile.
- Naming scheme C: modbus_* / /api/modbus/devices at the data+API layer,
  'Energy' as the first domain view in the frontend.
- New task M5-T01B: ConfigPage Accordion sectioning (avoids a second
  in-page side nav conflicting with the T01 app sidebar).
- Manual testing: read-only modbus_cli 'probe' subcommand (FC03/04 only,
  no register writes); mqtt/test now publishes a test message viewable
  in MQTT Explorer; no MQTT CLI (manual verification via UI + HA).
2026-06-22 11:55:39 +02:00
tliu93 d70b985eb0 fix(ci): recompile dev-requirements.txt to include pyotp (pytest CI install)
pytest / test (push) Successful in 3m51s
frontend / frontend (push) Successful in 1m26s
2026-06-22 11:08:06 +02:00
tliu93 e3c47b2762 docs(CLAUDE): switch to direct-on-main workflow for this single-user repo
frontend / frontend (push) Successful in 1m35s
pytest / test (push) Failing after 31s
2026-06-22 10:37:34 +02:00
tliu93 dbf17f3592 docs(CLAUDE): pin default sub-agent models (Implementer/Fixer=Sonnet, Reviewer=Opus) with manual-override clause 2026-06-22 09:50:46 +02:00
tliu93 ae75e4582d M4-T09: document login hardening, finalize OpenAPI and roadmap 2026-06-21 23:01:50 +02:00
tliu93 ee1264b66b M4-T08: add two-step login and TOTP settings panel (qrcode.react) 2026-06-21 22:42:41 +02:00
tliu93 51da8c5716 M4-T07: add admin_cli disable-totp and reissue-totp escape commands 2026-06-21 22:27:07 +02:00
tliu93 81bd32f613 M4-T06: add TOTP second factor to login (totp_code + recovery codes) 2026-06-21 22:08:55 +02:00
tliu93 ee3031013e M4-T05: add TOTP service and setup/enable/disable/status API 2026-06-21 21:55:17 +02:00
tliu93 5026967cee M4-T04: add AuthUser TOTP columns and auth_recovery_code table 2026-06-21 21:38:22 +02:00
tliu93 5481fbc99c M4-T03: add admin_cli escape hatch (reset-password, unlock, list-admin) 2026-06-21 21:28:41 +02:00
tliu93 e480a84a44 M4-T02: add exponential login back-off service and wire into login endpoint 2026-06-21 21:20:28 +02:00
tliu93 64d3882c16 M4-T01: add auth_login_throttle table + LoginThrottle model 2026-06-21 21:03:20 +02:00
tliu93 6b67b8e0f8 modify compose dev port 2026-06-21 20:49:24 +02:00
tliu93 0cb94d85ec docs(design): plan M4 login-hardening (first) + M5 IoT/energy
- Add docs/references/SDM120-Modbus-Protocol.md extracted from the SDM120 PDF
- M4 login hardening: brute-force exponential backoff, CLI escape hatch
  (reset-password/unlock/disable-totp), optional TOTP 2FA
- M5 IoT/energy: Modbus-TCP SDM120 polling + energy tables, MQTT/HA
  Discovery expose framework, frontend sidebar + energy view
- Add manual acceptance walkthroughs (M4 lock/unlock; M5 energy_cli read)
- Index both milestones in docs/design/README.md
2026-06-21 20:46:12 +02:00
tliu93 9da88db221 docs(roadmap): graduate next-phase work out of Future Ideas
frontend / frontend (push) Successful in 1m18s
pytest / test (push) Successful in 1m32s
Promote four now-decided directions from Future Ideas into a new
"下一阶段:已确定要做" section (roadmap altitude, not yet broken into task cards):
TOTP second factor for the public dashboard, frontend optimization (scope TBD),
MQTT / IoT integration, and a settings-page long-lived API token (PAT-style,
related to but distinct from M3's mobile OAuth). Future Ideas is now an empty,
purpose-stated bucket for not-yet-decided ideas.
2026-06-13 17:38:50 +02:00
tliu93 1a3aaea933 fix(docker): pin frontend-build to $BUILDPLATFORM to avoid QEMU V8 crash
frontend / frontend (push) Successful in 1m19s
pytest / test (push) Successful in 1m31s
docker-image / build-and-push (push) Successful in 3m55s
Multi-arch image builds (linux/amd64,linux/arm64) emulate the foreign arch via
QEMU. The unpinned `FROM node:22-slim` frontend-build stage was built per
target arch, so Node ran under emulation and V8's baseline JIT crashed (SIGTRAP
/ exit 133 on `npm ci`) on the CI runner's QEMU.

Pin the stage to --platform=$BUILDPLATFORM so the static, arch-independent SPA
dist/ is built once on the native build host and COPYd into each arch's runtime
stage. Node never runs emulated; both arch images are still produced.
2026-06-13 17:21:05 +02:00
tliu93 bf7fd71a21 Merge pull request 'Feature/m2 frontend v2' (#8) from feature/m2-frontend-v2 into main
pytest / test (push) Successful in 1m29s
frontend / frontend (push) Successful in 1m15s
docker-image / build-and-push (push) Failing after 59s
Reviewed-on: #8
2026-06-13 17:00:19 +02:00
tliu93 962ba26c7c docs(roadmap): add Future Ideas — TOTP 2FA for the public dashboard
frontend / frontend (push) Successful in 1m15s
pytest / test (push) Successful in 1m30s
frontend / frontend (pull_request) Successful in 1m16s
pytest / test (pull_request) Successful in 1m30s
Record TOTP (RFC 6238) as a deferred hardening idea for the now public-facing
Web dashboard: second factor on the single-admin login, with CLI-only password
reset and a CLI TOTP reset/recovery path that works even if the recovery codes
are lost (no lock-out dead end). Not M2.5, not scheduled — parked under a new
Future Ideas section.
2026-06-13 15:29:20 +02:00
tliu93 da236643f2 M2: frontend walkthrough fixes + explicit dev compose stack
frontend / frontend (push) Successful in 2m0s
pytest / test (push) Successful in 1m32s
Post-M2 self-walkthrough polish, batched into one commit.

Map / heat:
- fix heat-layer white-screen crash after login (add layer to map before
  setLatLngs; an off-map leaflet.heat layer has a null _map and throws)
- normalize each heat layer to the densest pixel cell visible in the CURRENT
  viewport (maxZoom:0 so intensity factor f=1) and recompute on moveend/zoomend,
  so sparse poo data reaches red and stays normalized at any zoom level
- dark CARTO basemap tiles when the color scheme is dark

UI:
- dark-mode toggle in the top-right, beside the settings gear
- switch top-right nav (records / theme / settings / logout) to Feather icons
  with hover tooltips
- home: Grafana-style quick time-range presets + back/forward shift buttons,
  placed between the From/To pickers and Apply; fix Select/tooltip z-index
  (Leaflet stacking) and the shift-button height alignment

API client:
- stop flooding GET /api/session with 401s: the session probe and the login
  endpoint own their 401s (no global redirect), which fixes the logout hang and
  the spinning login page

Compose:
- rename docker-compose.override.yml -> docker-compose.dev.yml as an explicit,
  non-auto-layered dev stack (8001, -dev container names, prod-copy ./data DB);
  update tests/test_deployment.py (read dev.yml, tolerate the !override tag) and
  the README "Docker Compose" section

Tests:
- pixel-grid peak counter, time-range presets, heat-layer ordering regression,
  and 401-redirect regression
2026-06-13 15:20:50 +02:00
tliu93 bd09523e94 M2-T13: docs wrap-up + retire frontend constraints + dependency cleanup
- README: add 前端 v2 (React SPA) section (dev/build/codegen/hosting/gates),
  update directory listing, drop stale Jinja descriptions
- architecture-overview: retire '不引入前后端分离' constraint; reflect SPA + JSON API
- roadmap: mark M2 done
- remove orphaned jinja2 dependency (recompile requirements*.txt; no other churn)
- delete empty tests/test_auth.py stub; drop dead _extract_csrf_token in test_api_data
- verified image still builds and app imports with the slimmer deps
2026-06-13 15:20:50 +02:00
tliu93 53f1245d83 docs(m2): mark M2-T12 done 2026-06-13 15:20:50 +02:00
tliu93 51f712f602 M2-T12: multi-stage Dockerfile (node build -> python runtime) + frontend CI
- Dockerfile: node:22-slim stage runs npm ci + npm run build; python runtime
  stage COPY --from copies dist to /app/frontend/dist (matches SPA_DIST_DIR);
  runtime image has no node
- .dockerignore: exclude frontend/node_modules and frontend/dist from context
- .github/workflows/frontend.yml: npm ci + codegen-sync + lint/typecheck/test/build
- tests/test_deployment.py: skip COPY --from sources in the context-existence
  check; assert the multi-stage frontend build wiring
- verified with a real docker build (image serves SPA, no node at runtime)
2026-06-13 15:20:50 +02:00
tliu93 f8b1e5fc71 docs(m2): mark M2-T11 done 2026-06-13 15:20:50 +02:00
tliu93 a9830c42d8 M2-T11: serve React SPA from FastAPI and remove Jinja pages
- app/main.py serves the SPA build (SPA_DIST_DIR, default frontend/dist):
  mounts /assets and a GET catch-all returning index.html for client routes;
  catch-all 404s on /api/*, never swallows /docs, /openapi.json, /static, assets,
  ingestion/ticktick/status; skips SPA serving when dist absent (backend-only CI)
- delete app/api/routes/pages.py, app/api/routes/auth.py, app/templates/
  (all replaced by /api/* + SPA; auth service layer kept)
- remove/replace Jinja page tests (JSON coverage already in test_api_*);
  add tests/test_spa_hosting.py for the fallback contract
- regenerate openapi/ (Jinja paths gone) and frontend schema.d.ts
2026-06-13 15:20:50 +02:00
tliu93 8aa7316b26 docs(m2): mark M2-T09 done 2026-06-13 15:20:50 +02:00
tliu93 32d93bba2a M2-T09: build data visualization UI (heatmap map as home page)
- self-contained RecordsMap (only module importing leaflet/react-leaflet/
  leaflet.heat/leaflet.markercluster); OSM tiles, swappable behind clean props
- heatmap layers for location + poo (primary); time-range selector fetches
  only the window (locations server-filtered; poo client-filtered)
- toggleable scatter layer with marker clustering; point-select reuses T10's
  edit/delete modals + hooks; query-key prefixes refresh map on mutation
- pure map logic isolated + unit-tested; leaflet mocked in component tests
- responsive layout; typed client only
2026-06-13 15:20:50 +02:00
tliu93 0d988a9b28 docs(m2): mark M2-T10 done 2026-06-13 15:20:50 +02:00
tliu93 ef7ea6b971 M2-T10: build records management UI (paginated lists + single-record CRUD)
- reusable src/records/ module: useUpdate/useDelete Poo+Location hooks
  (encodeURIComponent PK, prefix-based query invalidation), EditPooModal,
  EditLocationModal, ConfirmDeleteModal — exported for the map (T09) to reuse
- RecordsPage (/records): paginated poo + location tables (page size 100),
  edit + delete-with-confirm, refresh on success
- query keys ['poo']/['locations'] so map and list invalidations cross-cut
- typed client only; vitest tests
2026-06-13 15:20:50 +02:00
tliu93 6cc6382515 docs(m2): mark M2-T08 done 2026-06-13 15:20:50 +02:00
tliu93 ef2bd3c9c5 M2-T08: build config UI (replaces Jinja config page)
- GET /api/config renders sections; secret fields shown as empty password inputs
- save handles full-field submission semantics: always send non-secret values,
  send secret only when user typed a new value (blank secret keeps old)
- SMTP test button reflects tri-state (success / config-error 400 / failed 502)
  by reading ApiError.body.result
- typed client only; responsive Mantine layout; vitest tests
2026-06-13 15:20:50 +02:00
tliu93 cc2c02a2e2 docs(m2): mark M2-T07 done 2026-06-13 15:20:50 +02:00
tliu93 b2e26f0b17 M2-T07: build auth UI (login, session bootstrap, forced password change, logout)
- real Mantine login form -> POST /api/auth/login; 401 inline error; redirect when already authed
- ProtectedRoute: loading state, preserves intended destination, gates force_password_change
- ChangePasswordPage forced-change gate -> POST /api/auth/password
- logout control in AppLayout nav -> POST /api/auth/logout
- typed client only; vitest tests for the login flow
2026-06-13 15:20:50 +02:00
tliu93 8975acc48b docs(m2): mark M2-T06 done 2026-06-13 15:20:50 +02:00
tliu93 6cfeb2b865 M2-T06: scaffold React SPA frontend with typed OpenAPI client
- Vite + React 18 + TypeScript + Mantine + TanStack Query + react-router-dom
- typed client: openapi-typescript -> src/api/schema.d.ts (committed), openapi-fetch
- fetch wrapper middleware: cookies, X-CSRF-Token on writes, 401 -> /login,
  non-401 errors carry parsed JSON body
- SessionProvider/useSession (GET /api/session), ProtectedRoute skeleton
- app shell (Mantine + router) with placeholder login/home/config pages + gear nav
- dev proxy to FastAPI; vitest smoke test; frontend README
- npm scripts: dev/build/preview/lint/typecheck/test/codegen
2026-06-13 15:20:50 +02:00
tliu93 dba9e28540 docs(m2): mark M2-T05 done 2026-06-13 15:20:50 +02:00
tliu93 2bc5d6ea9a M2-T05: add SMTP test action API (POST /api/config/smtp/test)
- reuses send_smtp_test_email; tri-state result success(200)/config-error(400)/failed(502)
- session + CSRF protected; never echoes SMTP secrets
- SmtpTestResponse schema; regenerate openapi/
- extend tests/test_api_config.py (3 states + 401 + missing-CSRF 403)
2026-06-13 15:20:50 +02:00
tliu93 3ec663e138 docs(m2): mark M2-T04 done 2026-06-12 23:35:56 +02:00
tliu93 048414c5cb M2-T04: add single-row record CRUD API (patch/delete)
- PATCH/DELETE /api/locations/{person}/{datetime} and /api/poo/{timestamp}
- update only non-PK fields (PK immutable); 404 on missing PK
- delete scoped to exact full PK with rowcount guard (0->404, 1->ok);
  no batch/truncate/drop path
- session + CSRF protected; bare ingestion endpoints untouched
- service helpers in app/services/location.py and poo.py; regenerate openapi/
- tests/test_api_record_crud.py
2026-06-12 23:33:08 +02:00
tliu93 9ce3f2a0b8 docs(m2): mark M2-T03 done 2026-06-12 23:27:02 +02:00
tliu93 0fba7cfe11 M2-T03: add read-only data JSON API
- GET /api/locations (inclusive time window start/end, pagination, cap 5000)
- GET /api/poo (pagination, cap 1000, newest first)
- GET /api/public-ip (current state + recent history, cap 1000)
- all session-protected, read-only, bounded (no full-table export)
- typed response schemas; register router; regenerate openapi/
- tests/test_api_data.py
2026-06-12 23:24:17 +02:00
tliu93 d8303eaa3d docs(m2): mark M2-T02 done 2026-06-12 23:18:43 +02:00
tliu93 8da1f13e60 M2-T02: add session/auth JSON API for the SPA
- GET /api/session (user + csrf_token, 401 when unauthenticated)
- POST /api/auth/login (sets HttpOnly session cookie; 401 on bad creds; no CSRF)
- POST /api/auth/logout (session+CSRF; revokes session, clears cookie; 204)
- POST /api/auth/password (session+CSRF; reuses change_password; 400 on failure; 204)
- reuses app/services/auth.py and shared require_session/require_csrf deps
- register router in app/main.py; regenerate openapi/
- tests/test_api_session.py
2026-06-12 23:15:56 +02:00
tliu93 de77019ce3 docs(m2): mark M2-T01 done 2026-06-12 23:11:38 +02:00
tliu93 c2b1b7b751 M2-T01: add config JSON API (GET/PUT /api/config)
- new app/api/routes/api/ package with shared require_session (401) and
  require_csrf (presence-only X-CSRF-Token, 403) dependencies
- GET /api/config returns masked config sections; PUT /api/config reuses
  save_config_updates (blank secret keeps old; invalid -> 422, no write)
- session-protected; PUT also CSRF-protected
- register router in app/main.py; regenerate openapi/
- tests/test_api_config.py
2026-06-12 23:08:14 +02:00
tliu93 3628ac51e5 chore(m2): green the ruff baseline before M2 orchestration
- ignore E402 in scripts/*.py (deliberate sys.path bootstrap before app imports)
- drop unused pathlib.Path import in tests/test_auth.py

Establishes a clean ruff gate so each M2 task can be verified green at its boundary.
2026-06-12 22:56:21 +02:00
tliu93 1756192270 docs: record future-ideas backlog and refine CLAUDE.md workflow rules
pytest / test (push) Successful in 46s
- Add docs/future-ideas.md: unscheduled backlog (more data sources/types,
  long-term storage, Home Assistant data persistence, MQTT client, near-term PWA).
- CLAUDE.md: codify M1 lessons — reviewer blind-review discipline,
  build-context integrity checks when deleting/moving files,
  fixup-vs-standalone-commit boundary, and a pre-release walkthrough
  (run app + docker build + manual smoke) before tagging.
2026-06-12 22:50:47 +02:00
tliu93 66ec9979cc docs(m2): lock M2 frontend design decisions
pytest / test (push) Failing after 11m46s
Record the decisions reached in planning into docs/design/m2-frontend-v2.md:
component library = Mantine; map = Leaflet (react-leaflet + leaflet.heat +
markercluster, isolated behind a component seam for a future MapLibre swap);
OpenAPI typed client committed + CI-checked; CSRF simplified to SameSite=Lax +
a custom write header (no per-session token); heatmap-first map as the home
view with a required time-range picker and an auxiliary paginated list;
record CRUD edits non-PK fields and deletes single rows (no UI create); bare
ingestion endpoints stay until M3; trips optional. Wireframes intentionally
skipped for this milestone.
2026-06-12 22:40:57 +02:00
tliu93 c1a5d7a425 fix(docker): stop COPYing removed alembic_location/alembic_poo into the image
pytest / test (push) Successful in 50s
docker-image / build-and-push (push) Successful in 3m52s
M1-T04 deleted the alembic_location / alembic_poo chains and their .ini files,
but the Dockerfile still COPYed those four paths, so the release image build
failed at 'COPY alembic_poo.ini ./' (path not found). Drop the four stale COPY
lines (only alembic_app remains). Add test_dockerfile_copy_sources_exist, which
asserts every Dockerfile COPY source exists in the build context, so this class
of breakage fails pytest instead of only surfacing in the image CI.

Verified with a real local 'docker build' (succeeds) and pytest (98 passed).
2026-06-12 20:48:34 +02:00
tliu93 1e0b235cef Merge pull request 'Feature/m1 db consolidation' (#7) from feature/m1-db-consolidation into main
pytest / test (push) Has been cancelled
docker-image / build-and-push (push) Has been cancelled
Reviewed-on: #7
2026-06-12 20:33:34 +02:00
tliu93 a337b06c94 M1-rework: rename leftover pk_cols param in reconciliation test stubs
pytest / test (push) Successful in 48s
pytest / test (pull_request) Failing after 10m49s
Round-2 audit nit (review-notes/M1-full-review-2.md): two _always_fail
monkeypatch stubs still named their (ignored) third positional parameter
pk_cols after _reconcile switched to a full columns list. Rename to columns
for consistency. Test-only, no behavior change.

pytest 97 passed; ruff clean.
2026-06-12 19:11:13 +02:00
tliu93 1cbe6c46d2 M1-rework: harden legacy-migration reconciliation to full-row equality
Audit finding (review-notes/M1-full-review-1.md, FINDING 1): _reconcile only
checked primary-key presence, so a source row skipped by INSERT OR IGNORE due
to a value difference against a pre-existing same-PK target row would
false-pass. Compare ALL columns with SQLite's NULL-safe IS operator instead,
so reconciliation is a true full-row guarantee (idempotent re-runs still pass
because the rows match column-for-column). Add tests for the value-mismatch
abort and for idempotency under full-row reconciliation. Remove the now-unused
pk_cols parameter.

pytest 97 passed; ruff clean (pre-existing only); data-safety grep still empty.
2026-06-12 19:05:56 +02:00
tliu93 2f634006d2 M1-T07: align docs to single-DB reality and re-export OpenAPI
Rewrite README (single app.db + one alembic_app chain, legacy data moved
once via scripts.migrate_legacy_data, accurate test list) and remove the
Grafana Provisioning section. Update architecture-overview to the unified
data layer (one Base, app-DB engine with WAL) and retire the
alembic_location / alembic_poo sections. Mark M1 done in the roadmap.
Re-export openapi/, which catches the spec up to the already-existing
/config/smtp/test and /public-ip/check endpoints (purely additive; M1's
DB-session dependency swap produced no schema change).

pytest 95 passed; ruff clean (pre-existing only); OpenAPI export idempotent.
2026-06-12 17:13:28 +02:00
tliu93 dc624bb7e5 M1-T06: remove Grafana from compose and delete provisioning
Drop the grafana service and the homeautomation_grafana_storage volume
declaration from docker-compose.yml (migration and app services unchanged),
and delete the grafana/ provisioning + dashboards directory. Visualization
moves to the M2 React frontend; the named volume's actual data is retired
manually as an ops step, never by automation.

docker compose config -q passes; pytest 95 passed; ruff clean.
2026-06-12 17:02:49 +02:00
tliu93 af8c602988 M1-T05: drop location/poo database config from Settings and tests
Remove the dead location_database_url / poo_database_url fields and the
location_sqlite_path / poo_sqlite_path computed properties from Settings;
drop them from the config-page payload and from .env.example. Update the
test hardcodes (test_config, test_public_ip, test_smtp) and reduce the
conftest test_database_urls fixture to the single app DB. The one-time
migration script keeps reading legacy URLs from env/CLI, independent of
Settings.

pytest 95 passed; ruff clean (pre-existing only).
2026-06-12 16:57:54 +02:00
tliu93 0d898e09f2 M1-T04: converge startup chain onto the single app DB
run_all_migrations() now adopts/initializes only the app DB and returns
{'app': ...}. app/main.py drops the location/poo readiness checks
(ensure_location_db_ready / ensure_poo_db_ready) and their imports;
ensure_runtime_dirs only provisions the app DB path; lifespan still
fail-closes on a missing/unmanaged app DB. Delete the retired
location/poo adopt scripts and the alembic_location / alembic_poo
chains. Update tests to single-DB expectations and drop the obsolete
location/poo adoption + readiness tests.

pytest 95 passed; ruff clean (pre-existing only); a fresh app DB
initialized via scripts.run_migrations contains location + poo_records.
2026-06-12 16:50:05 +02:00
tliu93 3d3c2bcc57 M1-T03: unify data layer, models, deps and routes onto single app DB
Collapse the three data layers into one. app/db.py now exposes a single
Base, a cached engine bound to app_database_url with SQLite WAL enabled, and
get_engine/get_session_local/reset_db_caches/get_db_session. Delete
app/auth_db.py, app/poo_db.py and app/models/base.py. All models (auth,
config, public_ip, location, poo) inherit the one Base and register on a
single metadata. Dependencies converge to a single get_db; all routes use it.

Also update the alembic env.py files (app/location/poo) and tests that
imported the removed modules so the suite stays green, and drop the obsolete
test_legacy_style_location_db test whose flow (app reading a separate location
DB) no longer exists. Location/poo Alembic chains, adopt scripts and adoption
tests remain for M1-T04; config fields remain for M1-T05.

pytest 109 passed; ruff clean (pre-existing only); WAL verified; single
Base.metadata holds all seven tables.
2026-06-12 16:35:07 +02:00
tliu93 bc8dd062d5 M1-T02: add idempotent legacy data migration script
scripts/migrate_legacy_data.py copies rows from the legacy locationRecorder.db
/ pooRecorder.db into the unified app DB's location / poo_records tables using
ATTACH + INSERT OR IGNORE (idempotent via PK-conflict skip; explicit columns,
never SELECT *). After copy it reconciles every source row against the target
and raises / exits non-zero on any shortfall. Missing legacy files are a safe
no-op (skipped); --dry-run writes nothing. Not part of the Alembic chain; run
manually once at cut-over. Never deletes or overwrites any file.

Validated end-to-end on copies of the real production DBs: dry-run reported
75103 location + 874 poo rows and wrote nothing; the real run copied all rows
with reconciliation passing; a second run copied 0 (idempotent).
2026-06-12 16:13:55 +02:00
tliu93 427a491380 M1-T01: add app-chain revision creating location + poo_records tables
Add alembic_app revision 20260611_06_merge_location_poo_tables that builds
empty location and poo_records tables (REAL float columns, matching the
production schema and the adopt scripts' EXPECTED_*_TABLE_INFO constants).
Update APP_BASELINE_REVISION to the new head. Schema-only; data migration
is handled separately by scripts/migrate_legacy_data.py (M1-T02).
2026-06-12 16:02:46 +02:00
tliu93 b359bbe3bf docs: add next-phase roadmap, milestone design docs, and CLAUDE.md
pytest / test (push) Successful in 54s
- roadmap.md: M1 (DB consolidation) -> M2 (React SPA) -> M3 (token/mobile)
- docs/design/: agent-pipeline design docs with atomic tasks for M1-M3
- CLAUDE.md: workflow, doc map, commit conventions, review-notes briefing flow
- .gitignore: ignore local review-notes/
2026-06-12 15:37:17 +02:00
204 changed files with 47968 additions and 3832 deletions
+3
View File
@@ -8,3 +8,6 @@ data
openapi
src
# Frontend host build artifacts — built inside the node stage, not needed from context
frontend/node_modules
frontend/dist
+17 -2
View File
@@ -4,8 +4,6 @@ APP_NAME=Home Automation Backend (Python)
APP_ENV=production
APP_HOSTNAME=home-automation.example.com
APP_DATABASE_URL=sqlite:////app/data/app.db
LOCATION_DATABASE_URL=sqlite:////app/data/locationRecorder.db
POO_DATABASE_URL=sqlite:////app/data/pooRecorder.db
AUTH_BOOTSTRAP_USERNAME=admin
AUTH_BOOTSTRAP_PASSWORD=change-me
@@ -30,3 +28,20 @@ TICKTICK_CLIENT_ID=
TICKTICK_CLIENT_SECRET=
TICKTICK_TOKEN=
HOME_ASSISTANT_ACTION_TASK_PROJECT_ID=
# Optional: Modbus polling (global kill-switch; default false — opt-in).
# MODBUS_POLLING_ENABLED=false
# Optional: MQTT broker connection.
# Leave MQTT_ENABLED=false (or unset) when MQTT is not needed.
MQTT_ENABLED=false
MQTT_BROKER_HOST=
MQTT_BROKER_PORT=1883
MQTT_USERNAME=
MQTT_PASSWORD=
MQTT_TLS_ENABLED=false
# Optional: Home Assistant MQTT Discovery.
# Requires MQTT_ENABLED=true and a running MQTT broker.
HA_DISCOVERY_ENABLED=false
HA_DISCOVERY_PREFIX=homeassistant
+49
View File
@@ -0,0 +1,49 @@
name: frontend
on:
push:
branches:
- "**"
pull_request:
workflow_dispatch:
jobs:
frontend:
runs-on: ubuntu-latest
steps:
- name: Check out repository
uses: actions/checkout@v4
- name: Set up Node
uses: actions/setup-node@v4
with:
node-version: "22"
cache: npm
cache-dependency-path: frontend/package-lock.json
- name: Install dependencies
working-directory: frontend
run: npm ci
- name: Check codegen is in sync
working-directory: frontend
run: |
npm run codegen
git diff --exit-code src/api/schema.d.ts
- name: Lint
working-directory: frontend
run: npm run lint
- name: Type-check
working-directory: frontend
run: npm run typecheck
- name: Test
working-directory: frontend
run: npm run test
- name: Build
working-directory: frontend
run: npm run build
+1
View File
@@ -5,3 +5,4 @@
__pycache__/
*.pyc
data/
review-notes/
+181
View File
@@ -0,0 +1,181 @@
# CLAUDE.md — Home Automation Backend
本文件每次会话自动加载。它定义本项目的**工作流程、文档位置、commit 规范**。请在动手前先读完。
## 项目速览
- 个人用 home-automation 后端:**FastAPI + SQLite + SQLAlchemy + Alembic**,服务端模板(JinjaM2 将换成 React SPA)。
- 单 admin 鉴权(Argon2 + server-side session cookie),runtime config 落 `app_config` 表。
- 模块:public IPv4 monitor、SMTP 通知、location recorder、poo recorder、Home Assistant in/out、TickTick OAuth。
- 已发布 `v1.0.3`。下一阶段方向:**M1 单库化 → M2 React 前端 → M3 token/移动端(远期,M2 后再说)**。
- **当前现实**:在 M1 完成前仍是**三个独立 SQLite 库**app / location / poo),三套 DeclarativeBase、三条 Alembic 链。不要假设已经单库——以代码现状为准。
- 明确不做:Notion 模块。
## 文档地图与「开工前必读」
文档都在 `docs/`
| 路径 | 作用 |
| --- | --- |
| `docs/roadmap.md` | 全局规划与里程碑总览 |
| `docs/design/README.md` | **协作契约**:任务卡格式、原子任务定义、校验闸门、数据安全红线 |
| `docs/design/m1-db-consolidation.md` | M1 原子任务(含真实代码现状盘点 + 人工 runbook) |
| `docs/design/m2-frontend-v2.md` | M2 原子任务 + API 契约 + 前端校验闸门 |
| `docs/design/m3-token-mobile.md` | M3(远期,暂缓) |
| `docs/*.md`auth / public-ip-monitor / location-recorder …) | 各模块说明,按需读 |
**开工时读取顺序**
1. `docs/design/README.md`(每轮都读,它是流程与验收的共同契约)。
2. 本轮对应的 milestone 文档(如 `docs/design/m1-db-consolidation.md`),定位要做的任务卡。
3. 任务卡 `Files` 列出的源文件 + 该模块的 `docs/*.md`(按需)。
4. `docs/roadmap.md` 仅在需要全局视角时读。
## 工作流程
### 实现模式(由用户的提示词决定)
- **默认逐步**:给一个 milestone 文档,按其中原子任务**一步一步**实现。
- **(a) 只实现一步**:用户说"只实现一步 / 这一个任务"时,**只做那一个任务卡**,跑完校验闸门后停下,等用户确认,不要顺手往下做。
- **(b) 完成整个 milestone**:仅当用户在提示词里**显式要求启用 sub-agent**时,才起 implementer / reviewer / fixer sub-agent(模型用下方**『默认模型档位』**,用户人工指定则覆盖),按任务依赖顺序跑完整条链。
- **Sub-agent 纪律**:只在用户显式要求时才 spawn sub-agent;单步/小改动在主线内联完成。起 sub-agent 时按下方**『默认模型档位』**选择模型(用户人工指定则以人工指定为准),用 Agent 工具的 `model` 字段落实。
### 默认模型档位(实现模式 sub-agent;可被人工指定覆盖)
起 implementer / reviewer / fixer sub-agent 时,**默认**用下列模型档位,无需用户每次人工指定:
| 角色 | 默认模型 | 推理档位 |
| --- | --- | --- |
| **Implementer** | **Sonnet** | high reasoning effort |
| **Fixer**(返工) | **Sonnet** | high reasoning effort |
| **Reviewer** | **Opus** | extra-high reasoning effort |
- **人工指定覆盖**:若用户在提示词里**显式指定了其他模型**(针对任一角色),则**以用户人工指定为准**,覆盖上述默认。
- 用 Agent 工具的 `model` 字段落实模型选择;该字段当前仅支持 `sonnet` / `opus` / `haiku` / `fable`
- **推理档位说明**:Agent 工具未暴露独立的 reasoning-effort 旋钮,"high / extra-high reasoning" 通过 spawn prompt 里的显式指令传达(要求 implementer/fixer 动手前充分推理边界条件;要求 reviewer 以对抗性外部审计心态最高强度复核)。若所在 harness 提供真正的 effort 设置,则一并按此档位设置。
### 角色(Orchestrator → Implementer → Reviewer → Fixer
- 我(主线)= **Orchestrator**:挑依赖已满足的下一个任务、派发、转述结果、维护任务 `Status`
- **Implementer**(默认 **Sonnet**high reasoning;见上方『默认模型档位』):一次一个任务,严格按任务卡,不扩范围。
- **Reviewer**(默认 **Opus**extra-high reasoning):实现完成后起 Reviewer sub-agent,按任务卡 `Acceptance criteria` + `Reviewer checklist` 复核、**独立重跑校验闸门**,驱动返工直到本轮 PASS。
- **Fixer**(默认 **Sonnet**high reasoning):按 reviewer 的编号返工清单返工;**每轮返工起一个干净的 Fixer**(与首次实现的 Implementer 分开冷启动),先读对应 `review-notes/<task>-review-<n>.md` 再改。
#### Reviewer 盲审纪律(M1 教训)
M1 里 review **从未触发过一次 rework**,根因是 orchestrator 把自己的结论 / 辩护喂给了 reviewer,造成 context bleed、review 沦为橡皮图章。所以:
- reviewer 必须**冷启动(Clear-Agent)、最小化喂料**——spawn prompt 只给:① 任务卡(`Acceptance criteria` + `Reviewer checklist`)、② 对应的 `review-notes/<task>-impl|rework-<n>.md` 路径、③ 要审的 diff / commit 范围。
- **不要**在 prompt 里塞 orchestrator 自己的判断、"我觉得没问题"、对实现选择的辩护,或上一轮 reviewer 的倾向性结论。让它**独立得出结论、独立重跑校验闸门**。
- 事后另起的整库**独立盲审**(如对抗复审)同理:Clear-Agent、最小上下文,把它当"**外部审计**"而非"确认自己没错"。
### 校验闸门(每个任务结束都要全绿)
根目录、激活 `.venv` 后:
```bash
pytest # 权威闸门(CI 跑的就是它)
ruff check . # line-length=100
python scripts/export_openapi.py && git diff --exit-code openapi/ # 改了路由/schema 才需要,且产物须入库
```
前端任务(M2)在 `frontend/` 下另跑 `npm run lint && npm run typecheck && npm run test && npm run build`(详见 m2 文档 §8)。
**不过闸门就不算完成**,不得跳过、不得留红给下一轮。
### 构建上下文完整性(M1 Dockerfile 教训)
`docker build` **不在 pytest/ruff 闸门里**——M1 删了 `alembic_location/poo` 后忘了同步 `Dockerfile``COPY`,单元闸门全绿却把坏掉的镜像构建一路漏到 release tag。所以:
- 任务**删除 / 移动 / 重命名文件或目录**时,必须 grep 构建清单是否还在引用它们:`Dockerfile`(尤其 `COPY` 源)、`docker/``*.ini`、CI workflow、`requirements*.txt` 等。
- 已有回归测试 `tests/test_deployment.py::test_dockerfile_copy_sources_exist` 守"Dockerfile `COPY` 源必须存在于构建上下文";新增 / 改动 `COPY` 时确保它仍覆盖得到。
- Reviewer 审"删 / 移文件"类任务时,**必须顺带核对构建清单引用**,把它当 acceptance 的一部分。
## 每轮简报(`review-notes/`
每轮工作都要在 `review-notes/` 下产出**中文简报**。该目录**已在 `.gitignore` 忽略**,纯本地、不入库——它是 agent 之间和与人之间的交接载体,不是仓库产物。
- **实现 / 返工简报**:每轮实现完成后(无论首次实现还是返工),写一份。文件名建议 `<task-id>-impl-<n>.md` / `<task-id>-rework-<n>.md`(如 `M1-T03-impl-1.md``M1-T03-rework-1.md`)。至少包含:
1. **本轮修改的具体内容**(改了哪些文件、做了什么、为什么)。
2. **自动化测试结果**`pytest` / `ruff` / 前端闸门的实际输出或结论,通过/失败逐项写清)。
3. **若需人工 walkthrough**:写明具体步骤(怎么启动、点哪里、预期看到什么);若无需人工验证,明确写"无需人工 walkthrough"。
- **review 简报**:每轮 review 后写一份,文件名建议 `<task-id>-review-<n>.md`(如 `M1-T03-review-1.md`)。至少包含:评审结论(`PASS` 或带编号的返工清单)、对照任务卡 `Acceptance criteria` + `Reviewer checklist` 的逐条核对、reviewer 独立重跑校验闸门的结果。
**用途**:① reviewer 审核时参考对应的实现简报;② implementer 返工时参考对应的 review 简报;③ 人类(用户)通读这些简报确认有无问题。简报之间用文件名里的 `<task-id>` 与轮次 `<n>` 对应起来。
### Orchestrator 派发契约(让简报真正被读到)
**关键**:sub-agent 冷启动、不继承主线上下文,**不会因为本文件提到简报就自动去读**对应文件。简报能流转,靠的是 orchestrator(主线)在**每次 spawn 时把路径显式写进 prompt**,而不是被动约定。所以派发时必须做到:
- **显式告诉它「先读哪个简报」**:
- 派 implementer 做**首次实现** → 传任务卡位置(milestone 文档路径 + task id);无前置简报。
- 派 implementer 做**返工** → 必须传对应的 `review-notes/<task>-review-<n>.md` 路径,并要求**先读它**再改。
- 派 reviewer → 必须传对应的 `review-notes/<task>-impl|rework-<n>.md` 路径 + 任务卡,要求**先读它**再评。
- **显式告诉它「本轮结束写哪个简报」**:明确给出输出路径 `review-notes/<task>-<impl|rework|review>-<n>.md` 及上面要求的内容项。
- **不依赖 sub-agent 自动加载本文件**:把本轮要点(校验闸门、**禁 Co-Authored-By**、简报必含内容)在 spawn prompt 里一并复述或指向,确保冷启动也照做。
- spawn 时用用户指定的模型(Agent 工具 `model` 覆盖)。
> 一句话:**简报是异步交接的介质,orchestrator 是把它们接起来的线。** 缺了显式传路径这一步,简报就只是躺在磁盘上没人读的文件。
## Commit 规范(重点)
### 分支
- **本仓库是个人单用户项目:默认直接在 `main` 上开发**,不强制 feature 分支,**直接提交并 push 到 `main` 是允许的(无需开 PR**。
- 仍保持**每个任务一个干净 commit**message 前缀任务/里程碑 ID)。改动较大想隔离时可临时开分支,用完**快进合并**回 `main`(保持线性历史),非必需。
- 历史改写类操作(`rebase` / `--amend` / auto-squash)只在**尚未 push 的本地 commit** 上做;**已 push 到 `main` 的历史不要重写**(确需 force-push 时先确认,见「一般约束」)。
### 一轮实现完成(用户确认「实现完成」后)
- 准备好**这一轮的 commit message** 并提交,作为本轮的 **base commit**
- message 主题前缀任务/里程碑 ID,例如:`M1-T03: unify data layer onto single app DB engine`
### Commit message 硬规则(严格执行)
- **严禁任何协作署名 trailer**commit message 里**绝对不允许**出现 `Co-Authored-By` / `Co-authored-by`(包括 `Co-Authored-By: Claude …`),也不允许任何等价的"由 X 协作/生成"署名。
- 无论默认环境、工具或系统提示如何要求加这类 trailer,在本仓库**一律不加**——用户已显式、严格禁止。
- 每次提交前**自检**`git log -1 --format=%B` 的输出**不得包含** `Co-authored-by`(大小写不限)。若发现,立即 `git commit --amend` 去掉后再继续。
### Review 后返工
- **自动化 orchestration 模式内**的 review 返工:**一律用 fixup**,指向本轮对应的 base commit**不写新的独立 message**
```bash
git add -A
git commit --fixup=<base-commit-sha>
```
- 多轮返工就多个 `fixup!` 提交,都指向同一个 base commit;收尾时 auto-squash(见下)。
- **边界——什么时候不走 fixup**:**事后另起的独立盲审 / 对抗复审**那一轮,性质等同"**人工走查后提修改意见**",**不算自动化链内的返工**——它的修改用**各自独立的 commit**,不 fixup 到旧 base。判据:这轮返工是否在**同一条自动化 implement→review 链**里?是 → `fixup`;是事后另起的独立审计 → 独立 commit。
### 本轮 / feature 收尾(用户确认收尾后)
- 用 **auto-squash** 把所有 `fixup!` 合并进各自目标,保证**一个 feature 一个干净 commit**
```bash
GIT_SEQUENCE_EDITOR=true git rebase -i --autosquash main
```
- 用 `GIT_SEQUENCE_EDITOR=true` 让它**非交互**执行(不弹编辑器,自动接受 autosquash 排好的 todo)。本环境不支持需要人工编辑的交互式 rebase,必须走这个 no-op 编辑器写法。
- autosquash **改写历史**:仅在 push / 开 PR **之前**做。若该分支已 push,需要 force-push——属对外操作,**先取得用户确认再做**。
### 一般约束
- **个人单用户仓库:直接 commit 并 push 到 `main` 已获授权**——在校验闸门全绿、一轮工作完成时即可提交 / 推送,无需逐次征求同意。
- 仍需**先取得用户确认**的操作:**force-push / 改写已推送历史**,以及**打 tag**(会触发镜像 CI / 对外发布;且打 tag 前须按下方「发版前置走查」真跑一次 `docker build`)。
## 发版前置走查(打 tag 前必做)
单元闸门绿 ≠ 真的能跑、能构建、能用。M1 出过"绿了但 docker 构建坏了"的事故,所以**打版本 tag(触发镜像 CI)之前**,除了 `pytest` / `ruff` 全绿,还要:
- **真起 app**:迁移(`python -m scripts.run_migrations`)→ `uvicorn app.main:app ...`,确认能正常启动、关键路由不 500。
- **真跑镜像构建**:本地 `docker build`(多阶段就跑完整条),确认构建通过、`COPY` 源都在。
- **关键功能人工瞄一眼**:尤其前端 / 可视化类(M2 的热力图、首页地图)——自动闸门判断不了"渲染对不对、UX 顺不顺",这部分**靠看跑起来的 app,不靠读代码**。
- 上述任一不过 → **不打 tag**。tag 一旦 push 会触发 docker 镜像 CI / 对外发布,属对外操作,**先确认**。
## 数据安全红线(不可违反)
- 任何脚本 / migration **都不得删除或覆盖用户数据文件**(旧 `.db`、备份、volume)。删除只能是人工、事后、保留归档的独立步骤(见 `docs/design/m1-db-consolidation.md` §6 runbook)。
- 涉及历史数据的迁移**先在备份副本上演练**;迁移脚本必须幂等且搬完对账行数。
- Review 时只要发现"删文件 / drop 有数据的表 / truncate"出现在自动化任务里,直接判返工。
## 常用命令
```bash
# 环境
python -m venv .venv && source .venv/bin/activate && pip install -r dev-requirements.txt
# 迁移(初始化/适配 DB
python -m scripts.run_migrations
# 起服务
uvicorn app.main:app --reload --host 0.0.0.0 --port 8000
# 测试 / lint / OpenAPI 导出
pytest
ruff check .
python scripts/export_openapi.py
```
+20 -4
View File
@@ -1,3 +1,20 @@
# Stage 1: build the React SPA.
# Pin to the native build host ($BUILDPLATFORM) so the Node/V8 build never runs
# under QEMU during multi-arch builds — emulated Node crashes V8's baseline JIT
# (SIGTRAP / exit 133 on `npm ci`). The dist/ output is static JS/CSS, i.e.
# architecture-independent, so building it once and COPYing it into each
# target-arch runtime stage is both correct and avoids the emulator entirely.
FROM --platform=$BUILDPLATFORM node:22-slim AS frontend-build
WORKDIR /frontend
COPY frontend/package.json frontend/package-lock.json ./
RUN npm ci
COPY frontend/ ./
RUN npm run build
# Stage 2: python runtime (no node)
FROM python:3.12-slim
ENV PYTHONDONTWRITEBYTECODE=1 \
@@ -11,15 +28,14 @@ RUN pip install --no-cache-dir -r requirements.txt
COPY app ./app
COPY alembic_app ./alembic_app
COPY alembic_app.ini ./
COPY alembic_location ./alembic_location
COPY alembic_location.ini ./
COPY alembic_poo ./alembic_poo
COPY alembic_poo.ini ./
COPY scripts ./scripts
COPY docker ./docker
COPY README.md ./
RUN mkdir -p /app/data
# Copy the built SPA dist from the frontend-build stage
COPY --from=frontend-build /frontend/dist ./frontend/dist
EXPOSE 8000
ENTRYPOINT ["/app/docker/entrypoint.sh"]
+254 -109
View File
@@ -4,16 +4,19 @@
当前系统已经包含:
- FastAPI Web 应用与服务端模板页面
- SQLite + SQLAlchemy + Alembic 的库结构
- username/password + server-side session 鉴权
- FastAPI Web 应用React SPA 前端 + JSON API
- SQLite + SQLAlchemy + Alembic 的库结构
- username/password + server-side session 鉴权(含登录加固,见下文)
- runtime config 页面与 app DB 持久化
- public IPv4 monitor、历史持久化与定时检查
- SMTP 配置、测试发信与 public IPv4 changed 邮件通知
- location recorder
- poo recorder
- Home Assistant inbound / outbound integration
- Home Assistant inbound / outbound integrationREST 通道)
- TickTick OAuth 与 action task 集成
- **Modbus 设备采集**:通过 YAML profile(首个:SDM120 电表)按设备周期轮询 Modbus-TCP 网关,解码工程量并落通用读数表(`modbus_device` + `modbus_reading`
- **MQTT + Home Assistant Discovery**:以可勾选方式把 Modbus 设备/工程量注册为 HA device/entity(含 binary_sensor online),state 周期发布;配置变更可重连重发
- **前端侧边栏 + Energy 视图**:侧边导航替换顶栏;Energy 页管理 Modbus 设备、展示最新读数与 Recharts 走势图;Config 页 Accordion 分区展开;Expose 设置勾选 HA 可暴露实体
- pytest 测试与 OpenAPI 导出脚本
- Docker / Compose 部署入口
@@ -23,44 +26,40 @@
## 当前配置现实
当前系统仍然是三个独立的 SQLite 数据库文件,而不是单一数据库
当前系统使用单一 SQLite 数据库文件`app.db`),所有数据表都在其中
- `app` 级共享数据使用自己的 DB 文件
- `location` 模块使用自己的 DB 文件
- `poo` 模块使用自己的 DB 文件
- auth(单个 admin 用户、server-side session
- runtime config 持久化(`app_config` 表)
- public IPv4 当前状态与变化历史
- location 记录(`location` 表)
- poo 记录(`poo_records` 表)
- Modbus 设备定义(`modbus_device` 表)
- Modbus 通用读数(`modbus_reading` 表,JSON payload
- HA 实体暴露开关(`exposed_entity_toggle` 表)
当前阶段明确不借这次重构把这些 DB 合并。配置层已经显式反映这一点
配置层只保留一个数据库环境变量
- `APP_DATABASE_URL`
- `LOCATION_DATABASE_URL`
- `POO_DATABASE_URL`
目前 auth、`location``poo` 都已经接到各自独立的数据库文件。
`app.db` 不会在应用启动时自动创建,需要先运行:
其中 `app` 级共享 DB 当前主要用于:
```bash
python -m scripts.run_migrations
```
- 单个 admin 用户
- server-side session
- runtime config 持久化
- public IPv4 当前状态与变化历史
这部分现在也使用 Alembic 管理:
- `app db` 不会在应用启动时自动创建
- 需要先运行 `python scripts/app_db_adopt.py`
- 这个脚本会创建新 DB 并建好 schema
该命令会通过 Alembic 将 `app.db` 初始化或升级到最新 head(含全部表,包括 M5 新增的 `modbus_device``modbus_reading``exposed_entity_toggle`)。
## 当前目录
主要目录如下:
- `app/`: FastAPI 应用代码
- `alembic_app/`: App DB 的 Alembic migration 环境
- `alembic_location/`: Location DB 的 Alembic migration 环境
- `alembic_poo/`: Poo DB 的 Alembic migration 环境
- `app/`: FastAPI 应用代码(包含 JSON API、业务服务、数据模型)
- `frontend/`: React SPA 前端(Vite + React + TypeScript + Mantine
- `alembic_app/`: App DB 的 Alembic migration 环境(管理所有表,含 M5 新增的 `modbus_device``modbus_reading``exposed_entity_toggle`
- `tests/`: pytest 测试
- `docs/`: 当前系统说明文档
- `scripts/`: 辅助脚本,例如 OpenAPI 导出
- `openapi/`: OpenAPI schema 静态产物(`openapi.json` / `openapi.yaml`),纳入版本控制
## 依赖管理
@@ -121,31 +120,81 @@ uvicorn app.main:app --reload --host 0.0.0.0 --port 8000
启动后可访问:
- 应用首页:`http://localhost:8000/`
- 应用首页React SPA`http://localhost:8000/`
- 健康检查:`http://localhost:8000/status`
- Swagger UI`http://localhost:8000/docs`
- ReDoc`http://localhost:8000/redoc`
## 前端 v2React SPA
M2 用 React SPA 取代了原有 Jinja 服务端模板,由 FastAPI 同源托管(同一容器、同一 origin)。
### 技术栈
- **Vite + React + TypeScript + Mantine**(组件库)
- **TanStack Query**(数据请求/缓存)
- **Leaflet / react-leaflet**(地图与热力图)
- **Recharts**Energy 视图走势图,M5 引入)
- **openapi-typescript + openapi-fetch**(类型化 API client,由 `openapi/openapi.json` 生成)
### 本地开发(前端)
前端开发服务器会把 `/api``/location``/poo``/public-ip``/homeassistant``/ticktick``/status` 等路径代理到后端 FastAPI`:8000`)。
```bash
cd frontend
npm install
npm run dev # 启动 Vite dev server(默认 :5173),代理后端
```
### 构建
```bash
cd frontend
npm run build # 产出 frontend/dist
```
FastAPI 启动时若 `frontend/dist/index.html` 存在,则自动挂载该目录,并对非 `/api` 路径做 SPA fallback(返回 `index.html`)。该路径可通过环境变量 `SPA_DIST_DIR` 覆盖(默认值为 `frontend/dist`,与多阶段 Dockerfile 中 `COPY``/app/frontend/dist` 一致)。
### 类型化 API Client
前端 API client 由后端 OpenAPI schema 自动生成:
```bash
cd frontend
npm run codegen # 从 ../openapi/openapi.json 生成 src/api/schema.d.ts
```
生成物(`src/api/schema.d.ts`)已提交入库,CI 会校验它与 `openapi/openapi.json` 保持同步。
### 前端校验闸门
```bash
cd frontend
npm run lint # ESLint
npm run typecheck # TypeScript 类型检查
npm run test # Vitest 单元测试
npm run build # 构建,确认产出 dist
```
## 数据库与 Alembic
当前默认使用 SQLite,并区分三个数据库文件:
当前使用单一 SQLite 数据库文件:
- App DB`sqlite:///./data/app.db`
- Location DB`sqlite:///./data/locationRecorder.db`
- Poo DB`sqlite:///./data/pooRecorder.db`
- 数据目录:`./data/`
初始化 migration 环境后,可继续添加模型并生成迁移
所有模型(auth / config / public_ip / location / poo / modbus / expose)共用同一个 `Base`,均通过单一 Alembic 链管理
当前 `app``location` `poo` 都已经有各自独立的 Alembic 链路。
- App Alembic 环境:`alembic_app.ini` + `alembic_app/`
- Location Alembic 环境:`alembic_location.ini` + `alembic_location/`
- Poo Alembic 环境:`alembic_poo.ini` + `alembic_poo/`
- Alembic 环境:`alembic_app.ini` + `alembic_app/`
- 统一 migration job`python -m scripts.run_migrations`
- App DB 初始化:`python scripts/app_db_adopt.py`
- Location DB 接管 / 初始化:`python scripts/location_db_adopt.py`
- Poo DB 接管 / 初始化:`python scripts/poo_db_adopt.py`
- App DB 接管 / 初始化:`python scripts/app_db_adopt.py`
历史 location / poo 数据(旧版本遗留的独立 DB 文件)已通过以下脚本一次性迁移至 `app.db`(幂等,不删除旧文件):
```bash
python -m scripts.migrate_legacy_data
```
## 基础鉴权
@@ -153,9 +202,9 @@ uvicorn app.main:app --reload --host 0.0.0.0 --port 8000
- 认证模型:`username/password`
- 会话模型:server-side session + cookie
- 当前主要受保护页面:`/config`
- 当前公开页面:`/login`
- 当前公开 API现有业务 API 暂未在这一轮统一收口到 auth 下
- 当前受保护入口:React SPA`/` 等客户端路由)调用 `/api/*` JSON 端点
- 当前公开页面:`/login`SPA 登录页)
- 当前公开 API裸 ingestion 端点(`/location/record``/poo/record` 等设备调用端点)暂未收口到 session 保护(M3 再做)
安全实现的当前边界:
@@ -163,7 +212,7 @@ uvicorn app.main:app --reload --host 0.0.0.0 --port 8000
- session cookie 使用 `HttpOnly`
- `Secure` 默认随 `APP_ENV` 切换:非 development 时默认开启
- `SameSite=Lax`
- 登录表单和登出表单都有基础 CSRF 防护
- 写请求(POST/PUT/PATCH/DELETE)需携带 `X-CSRF-Token` headerSameSite=Lax + 自定义 header 纵深防御,无需 per-session token 值比对)
首次启动时,如果 `APP_DATABASE_URL` 对应的 auth DB 里还没有用户,应用会使用:
@@ -177,12 +226,145 @@ uvicorn app.main:app --reload --host 0.0.0.0 --port 8000
首次登录后会被要求立即修改密码。这个 bootstrap 只用于首个用户落库,不是后续的完整配置管理方案。
当前前端主要有两条页面路径
React SPA 主要页面路由(客户端路由,均由 FastAPI fallback 到 `index.html`
- `/login`
- `/config`
- `/login`:登录页
- `/`:首页(地图热力图主视图)
- `/config`:配置页(取代原 Jinja `/config`
- `/records`:记录管理列表页
无论是本地 `host:port` 还是反向代理后的域名访问,登录成功后都使用相对路径跳转到 `/config`
无论是本地 `host:port` 还是反向代理后的域名访问,登录成功后进入 SPA 首页(`/`
## M4 登录加固
M4 在基础鉴权之上叠加了三层防御,详细说明见 [`docs/auth.md`](./docs/auth.md)。
### 防爆破 / 指数退避
登录失败超过 3 次后进入指数退避(`wait = min(900s, 1s × 2^(failures-3))`),期间请求返回 `429 Too Many Requests`(含 `Retry-After` 响应头);成功登录后自动清零。退避按 **client IP****username** 双键取较大值,不会因此永久锁定账号(只是延迟,不是封号)。
- 全局开关:`AUTH_LOGIN_THROTTLE_ENABLED`CONFIG_FIELDS,默认 `true`
- 反代后需要 `AUTH_TRUST_FORWARDED_FOR=true` 才会读 `X-Forwarded-For``.env` 部署级配置,默认 `false`
### CLI 逃生通道
拿到服务器 CLI 权限时,可以在**不依赖任何已存凭据**(无需密码、恢复码)的情况下重置密码、解锁退避、关停 TOTP:
```bash
# 重置密码(不加 --password 则交互式输入,不回显)
python -m scripts.admin_cli reset-password admin
# 解锁退避(被 429 挡住时使用)
python -m scripts.admin_cli unlock --all # 清所有退避行
python -m scripts.admin_cli unlock --ip 1.2.3.4 # 按 IP 清
python -m scripts.admin_cli unlock --username admin # 按 username 清
# TOTP 相关(需要先启用,见下文)
python -m scripts.admin_cli disable-totp admin # 关停 TOTP(零凭据,逃生口)
python -m scripts.admin_cli reissue-totp admin # 重新发放 TOTP secret(打印新 URI
# 查看用户列表
python -m scripts.admin_cli list-admin
```
在 Docker 容器内执行时:
```bash
docker compose exec app python -m scripts.admin_cli <command>
```
### 可选 TOTP 二次验证
admin 可在 React SPA 设置页(`/config`)自选启用 RFC 6238 TOTP
1. 设置页点「启用 TOTP」→ 后端生成 `otpauth://` URI,前端渲染二维码(`qrcode.react`
2. 用 Authenticator App(如 Google Authenticator、Authy)扫码
3. 输入当前 6 位动态码确认 → TOTP 启用
4. 妥善保存一次性展示的 10 个恢复码(格式 `xxxx-xxxx`
启用后,登录需要两步:密码 → 6 位动态码(或恢复码,一次性)。不启用则维持纯密码登录,行为不变。
恢复码丢失时,可用 CLI 逃生:`python -m scripts.admin_cli disable-totp admin`,随后即可纯密码登录。
TOTP issuer 标签(显示在 Authenticator 里)通过 `AUTH_TOTP_ISSUER` 环境变量配置(`.env` 部署级),默认回退 `app_name`
## M5 Modbus 设备采集 / Energy / MQTT + HA Discovery
M5 给后端接入家庭 IoT 生态,新增通用 Modbus 采集链路(首个领域:能耗)、MQTT + Home Assistant Discovery 发布,以及前端侧边栏与 Energy 视图。
### 依赖
后端新增:
- `pymodbus`Modbus-TCP 客户端(轮询电表等 slave 设备)
- `paho-mqtt`MQTT 客户端(HA Discovery 与 state 发布)
- `pyyaml`YAML profile 加载(设备协议声明式描述)
前端新增:
- `recharts`Energy 视图走势图
### Modbus 设备采集
采集链路采用两层分离:**YAML profile**(协议知识,随代码走)+ **`modbus_device` 数据库行**(部署/可配置信息)+ **`modbus_reading` 通用读数表**JSON payload 遥测)。
- **profile**(如 `sdm120.yaml`)描述:读哪些寄存器(FC04 块读)、每个量的 key/unit/device_class/ha_component。纯协议知识,不含 unit_id / friendly_name 等部署项。
- **`modbus_device` 行**friendly_name、网关 host/port、Modbus slave `unit_id`(电表 Meter ID,设备面板可改故落 DB)、选用哪个 profile、采样周期、是否启用。
- **`modbus_reading` 行**device_id FK、recorded_at、payloadJSON,如 `{"voltage": 230.2, "current": 1.3, ...}`)。
- 多设备可共享同一 profile(如两块 SDM120 共用 `sdm120` profile,各自独立 unit_id 和 friendly_name)。
- APScheduler 后台 job 周期轮询所有 `enabled` 设备,更新 `last_poll_at` / `last_poll_ok`。全局开关 `MODBUS_POLLING_ENABLED`CONFIG_FIELDS)。
**手工命令行试读**(不依赖 DB,最快验证网关连通性):
```bash
# 按 profile 解码读一次(验证整套解码链路)
python -m scripts.modbus_cli read --host <网关IP> --port 502 --unit 1 --profile sdm120
# 手工指定请求内容(first-contact 验证,不依赖 profile
python -m scripts.modbus_cli probe --host <网关IP> --port 502 --unit 1 --fc 4 --address 0x0000 --count 2 --decode float32
```
CLI 工具为受控手工验证而设(设备需接市电),仅暴露读功能码(FC03/04),无写寄存器子命令。
### MQTT + Home Assistant Discovery
后端作为 MQTT 发布方,把 Modbus 设备/工程量以 HA Discovery 协议自动注册为 device/entity
- 每个 Modbus 设备 = 一个 HA device`unique_id` 锚定于设备 `uuid`,不随改名变)
- 各工程量 = sensor entitydevice_class/unit 取自 YAML profile);另有 binary_sensor `online`(取 `last_poll_ok`
- 每次轮询成功后推 state topic;连接成功或勾选变更时重发 retained discovery config
- `POST /api/config/mqtt/test`:试连 broker 并发布一条测试消息(可用 MQTT Explorer 验证链路)
**Config 页配置流程**
1.`/config` 的「MQTT」section 填写 broker host/port/username/password,启用 `MQTT_ENABLED`
2. 点「发送测试消息」确认 broker 链路通
3. 启用 `HA_DISCOVERY_ENABLED`
4. 在「Home Assistant Expose」面板勾选要暴露的实体,点「重新发布 discovery」
5. 在 Home Assistant 确认对应 device/entity 出现
### API 端点(M5 新增)
| 端点 | 用途 |
| --- | --- |
| `GET /api/modbus/devices` | 列出 Modbus 设备 |
| `POST /api/modbus/devices` | 新建设备 |
| `GET /api/modbus/devices/{uuid}` | 单个设备 |
| `PATCH /api/modbus/devices/{uuid}` | 修改设备(含 enable/disable|
| `DELETE /api/modbus/devices/{uuid}` | 删除设备;有读数时 409 |
| `GET /api/modbus/devices/{uuid}/metrics` | 该设备 profile 的量目录(key/unit/device_class|
| `GET /api/modbus/devices/{uuid}/latest` | 最新一条读数 payload |
| `GET /api/modbus/devices/{uuid}/readings` | 时间范围读数(start/end/limit),供走势图 |
| `POST /api/modbus/devices/{uuid}/test` | 即时试读(不落库) |
| `GET /api/modbus/profiles` | 列出可用 profile 名 + 描述 |
| `GET /api/expose` | 可暴露实体目录 + 勾选状态 + MQTT/Discovery 状态 |
| `PUT /api/expose` | 设置逐 key 暴露开关 |
| `POST /api/expose/republish` | 手动重发 discovery |
| `POST /api/config/mqtt/test` | 试连 broker 并发布测试消息 |
### 前端视图(M5 新增)
- **侧边栏**:把顶栏改为侧边导航(Home / Records / Energy / Config + 主题切换 + 注销),当前路由高亮,移动端可折叠。
- **`/energy`Energy 视图)**:设备 CRUD(新建/编辑/删除,删除有二次确认;有读数时引导改用禁用);最新读数卡片(字段标签/单位取自 profile metrics);时间序列走势图(Recharts,支持电压/电流/功率/电能,带时间范围选择)。
- **Config 页 Accordion**:各大 config section 可独立折叠/展开;「Home Assistant Expose」面板按设备分组勾选可暴露实体、显示 MQTT/Discovery 连接状态、「重新发布 discovery」按钮。
- SPA 路由新增 `/energy`
## Config 持久化
@@ -197,7 +379,7 @@ uvicorn app.main:app --reload --host 0.0.0.0 --port 8000
这意味着:
- location / poo / app DB 地址仍然属于 bootstrap 范畴
- app DB 地址`APP_DATABASE_URL`仍然属于 bootstrap 范畴
- 运行时可编辑配置主要通过 `app_config` 表持久化
- token / secret 这类运行时必须可取回的配置,目前允许明文存储在 config 表中
- 登录密码仍然单独使用 Argon2 哈希,不走 config 表明文存储
@@ -209,6 +391,9 @@ uvicorn app.main:app --reload --host 0.0.0.0 --port 8000
- SMTP 基础配置
- TickTick OAuth 配置
- Home Assistant 配置
- MQTT broker 配置(`MQTT_ENABLED``MQTT_BROKER_HOST/PORT/USERNAME/PASSWORD``MQTT_TLS_ENABLED`
- Home Assistant Discovery 配置(`HA_DISCOVERY_ENABLED``HA_DISCOVERY_PREFIX`
- Modbus 采集配置(`MODBUS_POLLING_ENABLED`
其中 SMTP password 与其他 secret 字段一致:
@@ -241,8 +426,8 @@ uvicorn app.main:app --reload --host 0.0.0.0 --port 8000
当前系统已经提供最小可用的 SMTP 能力:
- SMTP 配置可在 `/config` 页面填写并保存到 `app_config`
- 可通过 config 页面发送测试邮件
- SMTP 配置可在 React SPA `/config` 页面填写并保存到 `app_config`(通过 `PUT /api/config`
- 可通过 config 页面发送测试邮件`POST /api/config/smtp/test`
- 邮件 `From` 头支持显示名,例如 `Home Automation <sender@example.com>`
当前 SMTP 配置项包括:
@@ -294,18 +479,20 @@ python scripts/export_openapi.py
当前 Compose 分成两层:
- `docker-compose.yml`:默认使用 registry image,适合部署 / 生产拉取
- `docker-compose.override.yml`仅为本地开发追加 `build: .`
- `docker-compose.yml`:默认使用 registry image,适合部署 / 生产拉取(暴露 8881
- `docker-compose.dev.yml`:本地开发显式叠加层——追加 `build: .`、独立 project /
容器名(`-dev` 后缀)、暴露 8001,并把 DB 指向挂载的 `./data` 副本,可与生产栈在同一台机器上并存
本地开发启动方式:
本地开发启动方式(显式叠加 dev 层)
```bash
docker compose up -d --build
docker compose -f docker-compose.yml -f docker-compose.dev.yml up -d --build
```
上面的命令会自动叠加 `docker-compose.override.yml`,因此本地仍然会按当前工作目录重新 build。
dev 层刻意不沿用 `docker-compose.override.yml` 这种会被 `docker compose up` 自动叠加的文件名,
因此默认的 `docker compose up` 只用生产基础文件,不会把开发端口 / 配置误带到生产。
如果要按生产方式直接从 registry 拉取并启动,显式只使用基础 compose 文件:
如果要按生产方式直接从 registry 拉取并启动,使用基础 compose 文件:
```bash
docker compose -f docker-compose.yml pull
@@ -318,55 +505,6 @@ docker compose -f docker-compose.yml up -d
docker compose logs -f app
```
## Grafana Provisioning
当前仓库支持通过 Grafana provisioning 自动加载 SQLite datasource 和 repo 内的 dashboard 导出文件。
需要保留的文件路径如下:
- `grafana/provisioning/datasources/locationrecorder.yaml`
- `grafana/provisioning/datasources/poorecorder.yaml`
- `grafana/provisioning/dashboards/provider.yaml`
- `grafana/dashboards/locationrecorder.json`
- `grafana/dashboards/poorecorder.json`
这些文件的职责分别是:
- `grafana/provisioning/datasources/locationrecorder.yaml`:声明 `locationrecorder` SQLite datasource,并指向 `/data/home-automation/locationRecorder.db`
- `grafana/provisioning/datasources/poorecorder.yaml`:声明 `poorecorder` SQLite datasource,并指向 `/data/home-automation/pooRecorder.db`
- `grafana/provisioning/dashboards/provider.yaml`:告诉 Grafana 从 `/var/lib/grafana/dashboards` 扫描并加载 dashboard JSON
- `grafana/dashboards/locationrecorder.json`location recorder dashboard 导出文件,内容本身不需要在 compose 中改写
- `grafana/dashboards/poorecorder.json`poo recorder dashboard 导出文件,内容本身不需要在 compose 中改写
当前 `docker-compose.yml` 中,Grafana service 需要挂载以下目录:
- `./grafana/provisioning -> /etc/grafana/provisioning:ro`
- `./grafana/dashboards -> /var/lib/grafana/dashboards:ro`
同时保留现有 named volume `homeautomation_grafana_storage:/var/lib/grafana` 作为 Grafana 运行态数据存储。
一键启动前,至少需要以下文件已经存在:
- `grafana/provisioning/datasources/locationrecorder.yaml`
- `grafana/provisioning/datasources/poorecorder.yaml`
- `grafana/provisioning/dashboards/provider.yaml`
- `grafana/dashboards/locationrecorder.json`
- `grafana/dashboards/poorecorder.json`
启动方式:
```bash
docker compose up -d
```
启动后会发生的事情:
- Grafana 容器会安装 `frser-sqlite-datasource` 插件
- Grafana 会读取 `/etc/grafana/provisioning/datasources/` 下的 datasource YAML
- Grafana 会读取 `/etc/grafana/provisioning/dashboards/provider.yaml`
- Grafana 会从 `/var/lib/grafana/dashboards/` 自动导入两个 dashboard JSON
- 现有 Grafana named volume 继续负责保存 Grafana 运行态数据,不会覆盖 repo 内的 dashboard 与 provisioning 文件
## Container Image CI
项目提供了一个 release image workflow
@@ -411,9 +549,16 @@ pytest
当前测试包含:
- app 基本启动测试
- `/status` endpoint 测试
- 登录 / session 基础流程测试
- app 启动与 `/status` 检查
- 登录 / session / 鉴权流程
- runtime config 读写
- public IPv4 monitor
- SMTP 配置与测试发信
- location / poo recorder 端点
- Home Assistant inbound 集成
- TickTick OAuth
- 部署与迁移(`run_migrations`
- legacy 数据迁移脚本(`migrate_legacy_data`
## OpenAPI 导出
+8 -3
View File
@@ -3,11 +3,16 @@ from logging.config import fileConfig
from alembic import context
from sqlalchemy import engine_from_config, pool
from app.auth_db import AuthBase
from app.config import get_settings
from app.db import Base
from app.models.config import AppConfigEntry # noqa: F401
from app.models.auth import AuthSession, AuthUser # noqa: F401
from app.models.auth import AuthSession, AuthUser, RecoveryCode # noqa: F401
from app.models.auth_throttle import LoginThrottle # noqa: F401
from app.models.public_ip import PublicIPHistory, PublicIPState # noqa: F401
from app.models.location import Location # noqa: F401
from app.models.poo import PooRecord # noqa: F401
from app.models.modbus import ModbusDevice, ModbusReading # noqa: F401
from app.models.expose import ExposedEntityToggle # noqa: F401
config = context.config
@@ -19,7 +24,7 @@ configured_url = config.get_main_option("sqlalchemy.url")
if not configured_url or configured_url == "sqlite:///./data/app.db":
config.set_main_option("sqlalchemy.url", settings.app_database_url)
target_metadata = AuthBase.metadata
target_metadata = Base.metadata
def run_migrations_offline() -> None:
@@ -0,0 +1,43 @@
"""merge location and poo_records tables into app chain
Revision ID: 20260611_06_merge_location_poo_tables
Revises: 20260429_05_public_ip_monitor
Create Date: 2026-06-11 00:00:01.000000
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
revision: str = "20260611_06_merge_location_poo_tables"
down_revision: Union[str, None] = "20260429_05_public_ip_monitor"
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
op.create_table(
"location",
sa.Column("person", sa.Text(), nullable=False),
sa.Column("datetime", sa.Text(), nullable=False),
sa.Column("latitude", sa.REAL(), nullable=False),
sa.Column("longitude", sa.REAL(), nullable=False),
sa.Column("altitude", sa.REAL(), nullable=True),
sa.PrimaryKeyConstraint("person", "datetime"),
)
op.create_table(
"poo_records",
sa.Column("timestamp", sa.Text(), nullable=False),
sa.Column("status", sa.Text(), nullable=False),
sa.Column("latitude", sa.REAL(), nullable=False),
sa.Column("longitude", sa.REAL(), nullable=False),
sa.PrimaryKeyConstraint("timestamp"),
)
def downgrade() -> None:
op.drop_table("poo_records")
op.drop_table("location")
@@ -0,0 +1,43 @@
"""add auth_login_throttle table for exponential back-off throttling
Revision ID: 20260621_07_auth_login_throttle
Revises: 20260611_06_merge_location_poo_tables
Create Date: 2026-06-21 00:00:00.000000
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
revision: str = "20260621_07_auth_login_throttle"
down_revision: Union[str, None] = "20260611_06_merge_location_poo_tables"
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
op.create_table(
"auth_login_throttle",
sa.Column("id", sa.Integer(), autoincrement=True, nullable=False),
sa.Column("key", sa.String(length=255), nullable=False),
sa.Column("scope", sa.String(length=16), nullable=False),
sa.Column("failures", sa.Integer(), nullable=False),
sa.Column("first_failed_at", sa.DateTime(timezone=True), nullable=False),
sa.Column("last_failed_at", sa.DateTime(timezone=True), nullable=False),
sa.Column("next_allowed_at", sa.DateTime(timezone=True), nullable=True),
sa.PrimaryKeyConstraint("id"),
sa.UniqueConstraint("scope", "key", name="uq_auth_login_throttle_scope_key"),
)
op.create_index(
"ix_auth_login_throttle_scope_key",
"auth_login_throttle",
["scope", "key"],
unique=False,
)
def downgrade() -> None:
op.drop_index("ix_auth_login_throttle_scope_key", table_name="auth_login_throttle")
op.drop_table("auth_login_throttle")
+61
View File
@@ -0,0 +1,61 @@
"""add TOTP fields to auth_users and create auth_recovery_code table
Revision ID: 20260621_08_totp
Revises: 20260621_07_auth_login_throttle
Create Date: 2026-06-21 00:00:00.000000
"""
from typing import Sequence, Union
import sqlalchemy as sa
from alembic import op
revision: str = "20260621_08_totp"
down_revision: Union[str, None] = "20260621_07_auth_login_throttle"
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# Add totp_secret (nullable) to auth_users — existing rows get NULL, which is correct.
op.add_column("auth_users", sa.Column("totp_secret", sa.String(length=64), nullable=True))
# Add totp_enabled (NOT NULL) with server_default="0" so existing rows default to false.
op.add_column(
"auth_users",
sa.Column(
"totp_enabled",
sa.Boolean(),
nullable=False,
server_default="0",
),
)
# Create auth_recovery_code table for one-time TOTP recovery codes.
op.create_table(
"auth_recovery_code",
sa.Column("id", sa.Integer(), autoincrement=True, nullable=False),
sa.Column("user_id", sa.Integer(), nullable=False),
sa.Column("code_hash", sa.String(length=255), nullable=False),
sa.Column("used_at", sa.DateTime(timezone=True), nullable=True),
sa.ForeignKeyConstraint(["user_id"], ["auth_users.id"]),
sa.PrimaryKeyConstraint("id"),
)
op.create_index(
op.f("ix_auth_recovery_code_user_id"),
"auth_recovery_code",
["user_id"],
unique=False,
)
def downgrade() -> None:
# Drop auth_recovery_code table first (has FK to auth_users).
op.drop_index(op.f("ix_auth_recovery_code_user_id"), table_name="auth_recovery_code")
op.drop_table("auth_recovery_code")
# Drop the two TOTP columns from auth_users.
# Use batch_alter_table for SQLite compatibility (alembic's portable column-drop path).
with op.batch_alter_table("auth_users") as batch_op:
batch_op.drop_column("totp_enabled")
batch_op.drop_column("totp_secret")
@@ -0,0 +1,81 @@
"""add modbus_device and modbus_reading tables
Revision ID: 20260622_09_modbus_tables
Revises: 20260621_08_totp
Create Date: 2026-06-22 00:00:00.000000
"""
from typing import Sequence, Union
import sqlalchemy as sa
from alembic import op
revision: str = "20260622_09_modbus_tables"
down_revision: Union[str, None] = "20260621_08_totp"
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# modbus_device — deployment/configurable metadata for each polled device.
op.create_table(
"modbus_device",
sa.Column("id", sa.Integer(), autoincrement=True, nullable=False),
sa.Column("uuid", sa.String(length=36), nullable=False),
sa.Column("friendly_name", sa.String(length=255), nullable=False),
sa.Column("transport", sa.String(length=16), nullable=False),
sa.Column("host", sa.String(length=255), nullable=False),
sa.Column("port", sa.Integer(), nullable=False),
sa.Column("unit_id", sa.Integer(), nullable=False),
sa.Column("profile", sa.String(length=64), nullable=False),
sa.Column("poll_interval_s", sa.Integer(), nullable=False),
sa.Column("enabled", sa.Boolean(), nullable=False),
sa.Column("last_poll_at", sa.DateTime(timezone=True), nullable=True),
sa.Column("last_poll_ok", sa.Boolean(), nullable=True),
sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
sa.PrimaryKeyConstraint("id"),
sa.UniqueConstraint("uuid", name="uq_modbus_device_uuid"),
)
# modbus_reading — generic telemetry, one row per device per poll cycle.
op.create_table(
"modbus_reading",
sa.Column("id", sa.Integer(), autoincrement=True, nullable=False),
sa.Column("device_id", sa.Integer(), nullable=False),
sa.Column("recorded_at", sa.DateTime(timezone=True), nullable=False),
sa.Column("payload", sa.JSON(), nullable=False),
sa.ForeignKeyConstraint(
["device_id"],
["modbus_device.id"],
ondelete="RESTRICT",
),
sa.PrimaryKeyConstraint("id"),
)
# Individual index on recorded_at (from the ORM-level index=True).
op.create_index(
"ix_modbus_reading_recorded_at",
"modbus_reading",
["recorded_at"],
unique=False,
)
# Composite index for efficient time-range queries per device.
op.create_index(
"ix_modbus_reading_device_recorded",
"modbus_reading",
["device_id", "recorded_at"],
unique=False,
)
def downgrade() -> None:
# Drop the reading table first (it has a FK referencing modbus_device).
op.drop_index("ix_modbus_reading_device_recorded", table_name="modbus_reading")
op.drop_index("ix_modbus_reading_recorded_at", table_name="modbus_reading")
op.drop_table("modbus_reading")
# Drop the device table.
op.drop_table("modbus_device")
@@ -0,0 +1,44 @@
"""add exposed_entity_toggle table
Revision ID: 20260622_10_exposed_entities
Revises: 20260622_09_modbus_tables
Create Date: 2026-06-22 00:00:00.000000
"""
from typing import Sequence, Union
import sqlalchemy as sa
from alembic import op
revision: str = "20260622_10_exposed_entities"
down_revision: Union[str, None] = "20260622_09_modbus_tables"
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# exposed_entity_toggle — per-entity on/off switch for MQTT / HA Discovery.
op.create_table(
"exposed_entity_toggle",
sa.Column("id", sa.Integer(), autoincrement=True, nullable=False),
sa.Column("key", sa.String(length=255), nullable=False),
sa.Column("enabled", sa.Boolean(), nullable=False),
sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
sa.PrimaryKeyConstraint("id"),
sa.UniqueConstraint("key", name="uq_exposed_entity_toggle_key"),
)
# Index on key for fast single-key lookups (toggle by key).
op.create_index(
"ix_exposed_entity_toggle_key",
"exposed_entity_toggle",
["key"],
unique=True,
)
def downgrade() -> None:
# Drop index then table — only removes what this revision created.
op.drop_index("ix_exposed_entity_toggle_key", table_name="exposed_entity_toggle")
op.drop_table("exposed_entity_toggle")
-37
View File
@@ -1,37 +0,0 @@
[alembic]
script_location = alembic_location
prepend_sys_path = .
path_separator = os
sqlalchemy.url = sqlite:///./data/locationRecorder.db
[loggers]
keys = root,sqlalchemy,alembic
[handlers]
keys = console
[formatters]
keys = generic
[logger_root]
level = WARN
handlers = console
[logger_sqlalchemy]
level = WARN
handlers =
qualname = sqlalchemy.engine
[logger_alembic]
level = INFO
handlers =
qualname = alembic
[handler_console]
class = StreamHandler
args = (sys.stderr,)
level = NOTSET
formatter = generic
[formatter_generic]
format = %(levelname)-5.5s [%(name)s] %(message)s
-2
View File
@@ -1,2 +0,0 @@
This directory contains the Alembic migration environment for the Python rewrite skeleton.
-48
View File
@@ -1,48 +0,0 @@
from logging.config import fileConfig
from alembic import context
from sqlalchemy import engine_from_config, pool
from app.config import get_settings
from app.models import Location # noqa: F401
from app.models.base import Base
config = context.config
if config.config_file_name is not None:
fileConfig(config.config_file_name)
settings = get_settings()
configured_url = config.get_main_option("sqlalchemy.url")
if not configured_url or configured_url == "sqlite:///./data/locationRecorder.db":
config.set_main_option("sqlalchemy.url", settings.location_database_url)
target_metadata = Base.metadata
def run_migrations_offline() -> None:
url = config.get_main_option("sqlalchemy.url")
context.configure(url=url, target_metadata=target_metadata, literal_binds=True)
with context.begin_transaction():
context.run_migrations()
def run_migrations_online() -> None:
connectable = engine_from_config(
config.get_section(config.config_ini_section, {}),
prefix="sqlalchemy.",
poolclass=pool.NullPool,
)
with connectable.connect() as connection:
context.configure(connection=connection, target_metadata=target_metadata)
with context.begin_transaction():
context.run_migrations()
if context.is_offline_mode():
run_migrations_offline()
else:
run_migrations_online()
-26
View File
@@ -1,26 +0,0 @@
"""${message}
Revision ID: ${up_revision}
Revises: ${down_revision | comma,n}
Create Date: ${create_date}
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
# revision identifiers, used by Alembic.
revision: str = ${repr(up_revision)}
down_revision: Union[str, None] = ${repr(down_revision)}
branch_labels: Union[str, Sequence[str], None] = ${repr(branch_labels)}
depends_on: Union[str, Sequence[str], None] = ${repr(depends_on)}
def upgrade() -> None:
${upgrades if upgrades else "pass"}
def downgrade() -> None:
${downgrades if downgrades else "pass"}
-1
View File
@@ -1 +0,0 @@
@@ -1,33 +0,0 @@
"""location baseline
Revision ID: 20260419_01_location_baseline
Revises:
Create Date: 2026-04-19 00:00:00.000000
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
revision: str = "20260419_01_location_baseline"
down_revision: Union[str, None] = None
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
op.create_table(
"location",
sa.Column("person", sa.Text(), nullable=False),
sa.Column("datetime", sa.Text(), nullable=False),
sa.Column("latitude", sa.Float(), nullable=False),
sa.Column("longitude", sa.Float(), nullable=False),
sa.Column("altitude", sa.Float(), nullable=True),
sa.PrimaryKeyConstraint("person", "datetime"),
)
def downgrade() -> None:
op.drop_table("location")
-37
View File
@@ -1,37 +0,0 @@
[alembic]
script_location = alembic_poo
prepend_sys_path = .
path_separator = os
sqlalchemy.url = sqlite:///./data/pooRecorder.db
[loggers]
keys = root,sqlalchemy,alembic
[handlers]
keys = console
[formatters]
keys = generic
[logger_root]
level = WARN
handlers = console
[logger_sqlalchemy]
level = WARN
handlers =
qualname = sqlalchemy.engine
[logger_alembic]
level = INFO
handlers = console
qualname = alembic
[handler_console]
class = StreamHandler
args = (sys.stderr,)
level = NOTSET
formatter = generic
[formatter_generic]
format = %(levelname)-5.5s [%(name)s] %(message)s
-48
View File
@@ -1,48 +0,0 @@
from logging.config import fileConfig
from alembic import context
from sqlalchemy import engine_from_config, pool
from app.config import get_settings
from app.models.poo import PooRecord # noqa: F401
from app.poo_db import PooBase
config = context.config
if config.config_file_name is not None:
fileConfig(config.config_file_name)
settings = get_settings()
configured_url = config.get_main_option("sqlalchemy.url")
if not configured_url or configured_url == "sqlite:///./data/pooRecorder.db":
config.set_main_option("sqlalchemy.url", settings.poo_database_url)
target_metadata = PooBase.metadata
def run_migrations_offline() -> None:
url = config.get_main_option("sqlalchemy.url")
context.configure(url=url, target_metadata=target_metadata, literal_binds=True)
with context.begin_transaction():
context.run_migrations()
def run_migrations_online() -> None:
connectable = engine_from_config(
config.get_section(config.config_ini_section, {}),
prefix="sqlalchemy.",
poolclass=pool.NullPool,
)
with connectable.connect() as connection:
context.configure(connection=connection, target_metadata=target_metadata)
with context.begin_transaction():
context.run_migrations()
if context.is_offline_mode():
run_migrations_offline()
else:
run_migrations_online()
@@ -1,32 +0,0 @@
"""poo baseline
Revision ID: 20260420_01_poo_baseline
Revises:
Create Date: 2026-04-20 00:00:00.000000
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
revision: str = "20260420_01_poo_baseline"
down_revision: Union[str, None] = None
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
op.create_table(
"poo_records",
sa.Column("timestamp", sa.Text(), nullable=False),
sa.Column("status", sa.Text(), nullable=False),
sa.Column("latitude", sa.Float(), nullable=False),
sa.Column("longitude", sa.Float(), nullable=False),
sa.PrimaryKeyConstraint("timestamp"),
)
def downgrade() -> None:
op.drop_table("poo_records")
View File
+306
View File
@@ -0,0 +1,306 @@
from __future__ import annotations
import logging
from fastapi import APIRouter, Depends, HTTPException, status
from fastapi.responses import JSONResponse
from sqlalchemy.orm import Session
from app.api.routes.api.deps import require_csrf, require_session
from app.config import Settings, get_settings
from app.dependencies import get_app_settings, get_db
from app.integrations.mqtt import MQTT_SETTINGS_KEYS, mqtt_manager
from app.schemas.config import (
ConfigField,
ConfigResponse,
ConfigSection,
ConfigUpdateRequest,
ConfigUpdateResponse,
MqttTestResponse,
SmtpTestResponse,
)
from app.services.auth import AuthenticatedSession
from app.services.config_page import ConfigSaveError, build_config_sections, save_config_updates
from app.services.email import EmailConfigurationError, EmailDeliveryError, send_smtp_test_email
logger = logging.getLogger(__name__)
router = APIRouter(prefix="/api", tags=["api-config"])
def _sections_from_raw(sections_raw: list[dict]) -> list[ConfigSection]:
result = []
for section in sections_raw:
fields = [ConfigField(**f) for f in section["fields"]]
result.append(ConfigSection(name=section["name"], fields=fields))
return result
@router.get("/config", response_model=ConfigResponse)
def get_config(
db: Session = Depends(get_db),
settings: Settings = Depends(get_app_settings),
_auth: AuthenticatedSession = Depends(require_session),
) -> ConfigResponse:
"""Return all configuration sections. Secret field values are masked (empty string)."""
sections_raw = build_config_sections(db, settings)
return ConfigResponse(sections=_sections_from_raw(sections_raw))
@router.put("/config", response_model=ConfigUpdateResponse)
def put_config(
body: ConfigUpdateRequest,
db: Session = Depends(get_db),
settings: Settings = Depends(get_app_settings),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> ConfigUpdateResponse:
"""
Save configuration updates.
- Blank secret value keeps the existing stored value (no change).
- Invalid values return 422 and nothing is written to the database.
- If MQTT-related settings changed, the MQTT client reconnects automatically.
"""
# Detect whether any MQTT-related key is being submitted (non-secret change
# or non-blank secret change) so we know to reconnect after saving.
mqtt_keys_submitted = any(k.lower() in MQTT_SETTINGS_KEYS for k in body.updates)
try:
save_config_updates(db, body.updates, settings)
except ConfigSaveError as exc:
logger.warning("Rejected config update via API: %s", exc)
raise HTTPException(
status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
detail="invalid config submission",
) from exc
# Re-read settings after save (save_config_updates clears the settings cache).
# Use build_runtime_settings so the reconnect picks up DB-stored values, not
# just the bootstrap env (otherwise a broker configured via the UI is ignored).
from app.services.config_page import build_runtime_settings
refreshed_settings = build_runtime_settings(db, get_settings())
# Reconnect MQTT client if any MQTT setting was updated.
if mqtt_keys_submitted:
logger.info("MQTT settings changed — triggering reconnect.")
mqtt_manager.reconnect(refreshed_settings)
sections_raw = build_config_sections(db, refreshed_settings)
return ConfigUpdateResponse(sections=_sections_from_raw(sections_raw))
@router.post(
"/config/smtp/test",
responses={
200: {"model": SmtpTestResponse},
400: {"model": SmtpTestResponse},
502: {"model": SmtpTestResponse},
},
)
def post_smtp_test(
settings: Settings = Depends(get_app_settings),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> JSONResponse:
"""
Send a test SMTP email using the current runtime settings.
Returns a structured result indicating success or the category of failure.
Three possible outcomes:
- 200 { "result": "success", "message": ... }
- 400 { "result": "config-error", "message": ... } (EmailConfigurationError)
- 502 { "result": "failed", "message": ... } (EmailDeliveryError)
SMTP credentials are never echoed in the response.
"""
try:
send_smtp_test_email(settings)
except EmailConfigurationError as exc:
logger.warning("SMTP test rejected due to configuration: %s", exc)
return JSONResponse(
status_code=status.HTTP_400_BAD_REQUEST,
content={"result": "config-error", "message": str(exc)},
)
except EmailDeliveryError as exc:
logger.warning("SMTP test delivery failed: %s", exc)
return JSONResponse(
status_code=status.HTTP_502_BAD_GATEWAY,
content={"result": "failed", "message": str(exc)},
)
return JSONResponse(
status_code=status.HTTP_200_OK,
content={"result": "success", "message": "Test email sent successfully."},
)
# ---------------------------------------------------------------------------
# POST /api/config/mqtt/test — M5-T10
# ---------------------------------------------------------------------------
class _MqttConfigurationError(ValueError):
"""Raised when MQTT settings are incomplete or disabled."""
class _MqttConnectionError(RuntimeError):
"""Raised when MQTT broker connection or publish fails."""
@router.post(
"/config/mqtt/test",
responses={
200: {"model": MqttTestResponse},
400: {"model": MqttTestResponse},
502: {"model": MqttTestResponse},
},
)
def post_mqtt_test(
settings: Settings = Depends(get_app_settings),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> JSONResponse:
"""
Test MQTT broker connectivity by attempting to connect and publishing a
test message to ``<ha_discovery_prefix>/home-automation/test``.
The message is visible in MQTT Explorer (or any subscriber) so users can
confirm the full broker publish path is working.
Three possible outcomes:
- 200 { "result": "success", "message": ... }
- 400 { "result": "config-error", "message": ... } (not configured)
- 502 { "result": "failed", "message": ... } (connection/publish error)
MQTT credentials are never echoed in the response.
"""
try:
_run_mqtt_test(settings)
except _MqttConfigurationError as exc:
logger.warning("MQTT test rejected due to configuration: %s", exc)
return JSONResponse(
status_code=status.HTTP_400_BAD_REQUEST,
content={"result": "config-error", "message": str(exc)},
)
except _MqttConnectionError as exc:
logger.warning("MQTT test connection/publish failed: %s", exc)
return JSONResponse(
status_code=status.HTTP_502_BAD_GATEWAY,
content={"result": "failed", "message": str(exc)},
)
return JSONResponse(
status_code=status.HTTP_200_OK,
content={
"result": "success",
"message": (
f"Test message published to "
f"{settings.ha_discovery_prefix}/home-automation/test."
),
},
)
def _run_mqtt_test(settings: Settings) -> None:
"""Attempt a transient MQTT connection and publish a test message.
Raises
------
_MqttConfigurationError
When MQTT is not enabled or broker host is not configured.
_MqttConnectionError
When the broker is unreachable or the publish fails.
"""
import socket
import paho.mqtt.client as mqtt
if not settings.mqtt_broker_host:
raise _MqttConfigurationError("MQTT broker host is not configured.")
test_topic = f"{settings.ha_discovery_prefix}/home-automation/test"
test_payload = '{"source": "home-automation", "event": "mqtt_test"}'
connected_event = __import__("threading").Event()
published_event = __import__("threading").Event()
connect_error: list[str] = []
client = mqtt.Client(
callback_api_version=mqtt.CallbackAPIVersion.VERSION2,
client_id="home-automation-test",
)
def _on_connect(
_client: mqtt.Client,
_userdata: object,
_flags: mqtt.ConnectFlags,
reason_code: mqtt.ReasonCode,
_properties: mqtt.Properties | None,
) -> None:
if reason_code.is_failure:
connect_error.append(f"Broker refused connection: {reason_code}")
connected_event.set()
def _on_publish(
_client: mqtt.Client,
_userdata: object,
_mid: int,
_reason_code: mqtt.ReasonCode,
_properties: mqtt.Properties | None,
) -> None:
published_event.set()
client.on_connect = _on_connect
client.on_publish = _on_publish
if settings.mqtt_tls_enabled:
try:
client.tls_set()
except Exception as exc:
raise _MqttConnectionError(f"TLS setup failed: {exc}") from exc
if settings.mqtt_username:
client.username_pw_set(
username=settings.mqtt_username,
password=settings.mqtt_password or None,
)
client.loop_start()
try:
try:
client.connect(
host=settings.mqtt_broker_host,
port=settings.mqtt_broker_port,
keepalive=10,
)
except (OSError, socket.error) as exc:
# Sanitise: ensure password never leaks into the error message.
msg = _sanitize_mqtt_error(str(exc), settings.mqtt_password)
raise _MqttConnectionError(f"Cannot reach broker: {msg}") from exc
# Wait up to 5 s for connection acknowledgement.
if not connected_event.wait(timeout=5):
raise _MqttConnectionError("Broker connection timed out (5 s).")
if connect_error:
raise _MqttConnectionError(connect_error[0])
# Publish test message.
client.publish(test_topic, payload=test_payload, qos=1, retain=False)
# Wait up to 5 s for the publish ACK (QoS 1).
if not published_event.wait(timeout=5):
raise _MqttConnectionError("Publish timed out (5 s) — broker reachable but no ACK.")
finally:
try:
client.disconnect()
except Exception:
pass
client.loop_stop()
def _sanitize_mqtt_error(message: str, password: str | None) -> str:
"""Replace *password* in *message* with ``[redacted]``."""
if password:
return message.replace(password, "[redacted]")
return message
+275
View File
@@ -0,0 +1,275 @@
from __future__ import annotations
from fastapi import APIRouter, Body, Depends, HTTPException, Query, status
from sqlalchemy import desc, select
from sqlalchemy.orm import Session
from app.api.routes.api.deps import require_csrf, require_session
from app.dependencies import get_db
from app.models.location import Location
from app.models.poo import PooRecord
from app.models.public_ip import PublicIPHistory, PublicIPState
from app.schemas.data import (
LocationRecord,
LocationUpdateRequest,
LocationsResponse,
PooRecord as PooRecordSchema,
PooResponse,
PooUpdateRequest,
PublicIPHistorySchema,
PublicIPResponse,
PublicIPStateSchema,
)
from app.services.auth import AuthenticatedSession
from app.services.location import delete_location, update_location
from app.services.poo import delete_poo_record, update_poo_record
router = APIRouter(prefix="/api", tags=["api-data"])
@router.get("/locations", response_model=LocationsResponse)
def get_locations(
limit: int = Query(default=1000, ge=1, le=5000),
offset: int = Query(default=0, ge=0),
start: str | None = Query(default=None),
end: str | None = Query(default=None),
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
) -> LocationsResponse:
"""
Return location records with optional time-window filtering and pagination.
- ``start`` / ``end`` are ISO8601 strings; filtering is **inclusive** on both bounds.
- Results are ordered by ``datetime`` ascending.
- ``limit`` is capped at 5000 to prevent full-table exports.
"""
stmt = select(Location)
if start is not None:
stmt = stmt.where(Location.datetime >= start)
if end is not None:
stmt = stmt.where(Location.datetime <= end)
stmt = stmt.order_by(Location.datetime).offset(offset).limit(limit)
rows = db.execute(stmt).scalars().all()
items = [
LocationRecord(
person=row.person,
datetime=row.datetime,
latitude=row.latitude,
longitude=row.longitude,
altitude=row.altitude,
)
for row in rows
]
return LocationsResponse(items=items, limit=limit, offset=offset)
@router.get("/poo", response_model=PooResponse)
def get_poo(
limit: int = Query(default=100, ge=1, le=1000),
offset: int = Query(default=0, ge=0),
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
) -> PooResponse:
"""
Return poo records ordered by timestamp descending (most recent first).
``limit`` is capped at 1000 to prevent full-table exports.
"""
stmt = (
select(PooRecord)
.order_by(desc(PooRecord.timestamp))
.offset(offset)
.limit(limit)
)
rows = db.execute(stmt).scalars().all()
items = [
PooRecordSchema(
timestamp=row.timestamp,
status=row.status,
latitude=row.latitude,
longitude=row.longitude,
)
for row in rows
]
return PooResponse(items=items, limit=limit, offset=offset)
@router.get("/public-ip", response_model=PublicIPResponse)
def get_public_ip(
limit: int = Query(default=100, ge=1, le=1000),
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
) -> PublicIPResponse:
"""
Return the current public IP state and recent history.
- ``state`` is ``null`` if no IP check has been performed yet.
- ``history`` is ordered by ``observed_at`` descending (most recent first).
- ``limit`` applies to the history list and is capped at 1000.
"""
state_row = db.execute(
select(PublicIPState).where(PublicIPState.id == 1).limit(1)
).scalar_one_or_none()
history_rows = db.execute(
select(PublicIPHistory).order_by(desc(PublicIPHistory.observed_at)).limit(limit)
).scalars().all()
state = PublicIPStateSchema.model_validate(state_row) if state_row is not None else None
history = [PublicIPHistorySchema.model_validate(row) for row in history_rows]
return PublicIPResponse(state=state, history=history)
# ---------------------------------------------------------------------------
# PATCH /api/locations/{person}/{datetime}
# ---------------------------------------------------------------------------
@router.patch("/locations/{person}/{datetime}", response_model=LocationRecord)
def patch_location(
person: str,
datetime: str,
body: LocationUpdateRequest = Body(default=LocationUpdateRequest()),
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> LocationRecord:
"""
Update the non-PK fields of a single location record.
- ``person`` and ``datetime`` identify the row (composite PK) and are immutable.
- Only ``latitude``, ``longitude``, and ``altitude`` may be updated.
- Omitted body fields are left unchanged.
- Returns **404** if the PK does not exist.
"""
row = update_location(
db,
person,
datetime,
latitude=body.latitude,
longitude=body.longitude,
altitude=body.altitude,
)
if row is None:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="location record not found",
)
return LocationRecord(
person=row.person,
datetime=row.datetime,
latitude=row.latitude,
longitude=row.longitude,
altitude=row.altitude,
)
# ---------------------------------------------------------------------------
# DELETE /api/locations/{person}/{datetime}
# ---------------------------------------------------------------------------
@router.delete(
"/locations/{person}/{datetime}",
status_code=status.HTTP_204_NO_CONTENT,
response_model=None,
)
def delete_location_record(
person: str,
datetime: str,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> None:
"""
Delete the single location record identified by its composite PK.
- Exactly one row is deleted; **404** if the PK does not exist.
- No batch delete / truncate path is available.
"""
deleted = delete_location(db, person, datetime)
if not deleted:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="location record not found",
)
# ---------------------------------------------------------------------------
# PATCH /api/poo/{timestamp}
# ---------------------------------------------------------------------------
@router.patch("/poo/{timestamp}", response_model=PooRecordSchema)
def patch_poo(
timestamp: str,
body: PooUpdateRequest = Body(default=PooUpdateRequest()),
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> PooRecordSchema:
"""
Update the non-PK fields of a single poo record.
- ``timestamp`` is the PK and is immutable.
- Only ``status``, ``latitude``, and ``longitude`` may be updated.
- Omitted body fields are left unchanged.
- Returns **404** if the PK does not exist.
"""
row = update_poo_record(
db,
timestamp,
status=body.status,
latitude=body.latitude,
longitude=body.longitude,
)
if row is None:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="poo record not found",
)
return PooRecordSchema(
timestamp=row.timestamp,
status=row.status,
latitude=row.latitude,
longitude=row.longitude,
)
# ---------------------------------------------------------------------------
# DELETE /api/poo/{timestamp}
# ---------------------------------------------------------------------------
@router.delete(
"/poo/{timestamp}",
status_code=status.HTTP_204_NO_CONTENT,
response_model=None,
)
def delete_poo(
timestamp: str,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> None:
"""
Delete the single poo record identified by its PK.
- Exactly one row is deleted; **404** if the PK does not exist.
- No batch delete / truncate path is available.
"""
deleted = delete_poo_record(db, timestamp)
if not deleted:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail="poo record not found",
)
+28
View File
@@ -0,0 +1,28 @@
from __future__ import annotations
from fastapi import Depends, Header, HTTPException, status
from app.dependencies import get_current_auth_session
from app.services.auth import AuthenticatedSession
def require_session(
auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> AuthenticatedSession:
if auth is None:
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail="authentication required",
)
return auth
def require_csrf(
_auth: AuthenticatedSession = Depends(require_session),
x_csrf_token: str | None = Header(default=None, alias="X-CSRF-Token"),
) -> None:
if not x_csrf_token:
raise HTTPException(
status_code=status.HTTP_403_FORBIDDEN,
detail="missing CSRF token",
)
+221
View File
@@ -0,0 +1,221 @@
"""Expose API routes (M5-T12).
Three endpoints:
GET /api/expose — Return catalog of exposable entities, per-key
toggle state, and MQTT/Discovery connection status.
PUT /api/expose — Set per-key toggle state (map key → bool);
triggers a HA Discovery re-publish on success.
POST /api/expose/republish — Manually trigger a full HA Discovery re-publish.
Auth model:
- GET: session required (no CSRF — read-only).
- PUT, POST: session + CSRF required.
"""
from __future__ import annotations
import logging
from datetime import UTC, datetime
from fastapi import APIRouter, Depends, HTTPException, status
from sqlalchemy.orm import Session
from app.api.routes.api.deps import require_csrf, require_session
from app.config import get_settings
from app.dependencies import get_db
from app.services.config_page import build_runtime_settings
from app.integrations.expose import CatalogEntry, build_catalog
from app.integrations.mqtt import mqtt_manager
from app.models.expose import ExposedEntityToggle
from app.schemas.expose import (
CatalogEntrySchema,
DeviceInfoSchema,
ExposeResponse,
ExposeUpdateRequest,
ExposeUpdateResponse,
ExposableEntitySchema,
MqttStatusSchema,
RepublishResponse,
)
from app.services.auth import AuthenticatedSession
logger = logging.getLogger(__name__)
router = APIRouter(prefix="/api", tags=["api-expose"])
# ---------------------------------------------------------------------------
# Internal helpers
# ---------------------------------------------------------------------------
def _get_mqtt_status(db: Session) -> MqttStatusSchema:
"""Build MQTT/Discovery status from live runtime state (DB-merged settings)."""
settings = build_runtime_settings(db, get_settings())
return MqttStatusSchema(
mqtt_configured=mqtt_manager.is_configured(settings),
mqtt_connected=mqtt_manager.is_connected,
discovery_enabled=bool(settings.ha_discovery_enabled),
)
def _entry_to_schema(entry: CatalogEntry) -> CatalogEntrySchema:
"""Convert a CatalogEntry to its Pydantic schema form.
Excludes ``value_getter`` because it is a non-serialisable callable.
"""
entity = entry.entity
return CatalogEntrySchema(
entity=ExposableEntitySchema(
key=entity.key,
component=entity.component,
device=DeviceInfoSchema(
identifiers=list(entity.device.identifiers),
name=entity.device.name,
),
device_class=entity.device_class,
unit=entity.unit,
name=entity.name,
state_class=entity.state_class,
),
enabled=entry.enabled,
)
def _build_response_data(
session: Session,
) -> tuple[list[CatalogEntrySchema], MqttStatusSchema]:
"""Return (catalog_entries, mqtt_status) for building GET / PUT responses."""
catalog = build_catalog(session)
catalog_schema = [_entry_to_schema(e) for e in catalog]
mqtt_status = _get_mqtt_status(session)
return catalog_schema, mqtt_status
# ---------------------------------------------------------------------------
# GET /api/expose
# ---------------------------------------------------------------------------
@router.get("/expose", response_model=ExposeResponse)
def get_expose(
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
) -> ExposeResponse:
"""Return the full exposable-entity catalog with toggle states and MQTT status.
The catalog is computed dynamically from registered providers (e.g. the
Modbus provider enumerates all enabled devices and their metric entities).
Toggle states come from the ``exposed_entity_toggle`` table; entities with
no row default to ``enabled=False``.
"""
catalog_schema, mqtt_status = _build_response_data(db)
return ExposeResponse(catalog=catalog_schema, mqtt_status=mqtt_status)
# ---------------------------------------------------------------------------
# PUT /api/expose
# ---------------------------------------------------------------------------
@router.put("/expose", response_model=ExposeUpdateResponse)
def put_expose(
body: ExposeUpdateRequest,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> ExposeUpdateResponse:
"""Set per-entity toggle state.
Accepts a map of ``{key: bool}`` and upserts rows in the
``exposed_entity_toggle`` table. Only keys present in ``body.toggles``
are touched; other entities' toggles are left unchanged.
After writing the toggles, triggers a HA Discovery re-publish so any
changes (enabled ↔ disabled) are reflected in Home Assistant immediately.
"""
now = datetime.now(UTC)
for key, enabled in body.toggles.items():
existing = (
db.query(ExposedEntityToggle)
.filter(ExposedEntityToggle.key == key)
.first()
)
if existing is not None:
existing.enabled = enabled
existing.updated_at = now
else:
db.add(
ExposedEntityToggle(
key=key,
enabled=enabled,
updated_at=now,
)
)
db.commit()
# Trigger discovery re-publish after toggle change.
_trigger_republish(db)
catalog_schema, mqtt_status = _build_response_data(db)
return ExposeUpdateResponse(catalog=catalog_schema, mqtt_status=mqtt_status)
# ---------------------------------------------------------------------------
# POST /api/expose/republish
# ---------------------------------------------------------------------------
@router.post("/expose/republish", response_model=RepublishResponse)
def post_expose_republish(
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> RepublishResponse:
"""Manually trigger a full HA Discovery re-publish.
Calls ``publish_discovery(session)`` from the HA discovery service (M5-T11).
Returns a status indicating whether the publish was attempted (or skipped
because MQTT / discovery is not enabled / connected).
"""
settings = build_runtime_settings(db, get_settings())
if not (settings.mqtt_enabled and settings.ha_discovery_enabled and mqtt_manager.is_connected):
return RepublishResponse(
ok=False,
message="MQTT / HA Discovery not enabled or broker not connected.",
)
try:
_trigger_republish(db)
return RepublishResponse(
ok=True,
message="HA Discovery re-published successfully.",
)
except Exception as exc:
logger.exception("post_expose_republish: unexpected error during publish")
raise HTTPException(
status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
detail="Discovery re-publish failed.",
) from exc
# ---------------------------------------------------------------------------
# Internal: trigger discovery re-publish
# ---------------------------------------------------------------------------
def _trigger_republish(session: Session) -> None:
"""Call publish_discovery from the HA discovery service.
No-op if MQTT / discovery is not enabled or broker is not connected
(publish_discovery guards internally). All errors are swallowed to avoid
breaking the API response.
"""
try:
from app.services.ha_discovery import publish_discovery
publish_discovery(session)
except Exception:
logger.exception("_trigger_republish: publish_discovery raised an error")
+442
View File
@@ -0,0 +1,442 @@
"""Modbus device CRUD, readings, metrics, and test-read API (M5-T05).
All endpoints are under /api/modbus, require an authenticated session, and
write endpoints (POST/PATCH/DELETE + test-read) additionally require a
non-empty X-CSRF-Token header.
Deletion safety (red-line):
DELETE /api/modbus/devices/{uuid} explicitly queries the application
layer for existing modbus_reading rows before deleting. If any exist
the endpoint returns 409 and suggests disabling the device instead.
We do NOT rely on the SQLite FK RESTRICT constraint because SQLite does
not enforce foreign-key constraints at runtime unless
PRAGMA foreign_keys=ON is set per connection (and it is not in this
project — see M5-T02 review note OBS-1).
"""
from __future__ import annotations
import logging
from datetime import UTC, datetime
from typing import Any
from fastapi import APIRouter, Depends, HTTPException, Query, status
from sqlalchemy import func, select
from sqlalchemy.orm import Session
from app.api.routes.api.deps import require_csrf, require_session
from app.dependencies import get_db
from app.integrations.modbus import driver as modbus_driver
from app.integrations.modbus.driver import ModbusDriverError
from app.integrations.modbus.profiles import (
ProfileNotFoundError,
decode as decode_profile,
list_profiles,
load_profile,
)
from app.models.modbus import ModbusDevice, ModbusReading
from app.schemas.modbus import (
MetricInfo,
ModbusDeviceCreate,
ModbusDeviceListResponse,
ModbusDeviceResponse,
ModbusDeviceUpdate,
ModbusLatestResponse,
ModbusMetricsResponse,
ModbusProfilesResponse,
ModbusReadingResponse,
ModbusReadingsResponse,
ModbusTestReadResponse,
ProfileSummary,
)
from app.services.auth import AuthenticatedSession
logger = logging.getLogger(__name__)
router = APIRouter(prefix="/api/modbus", tags=["api-modbus"])
# Maximum number of readings that can be returned per request.
_READINGS_LIMIT_MAX = 5000
# ---------------------------------------------------------------------------
# Helpers
# ---------------------------------------------------------------------------
def _get_device_or_404(db: Session, uuid: str) -> ModbusDevice:
"""Return the device with the given UUID or raise 404."""
device = db.execute(
select(ModbusDevice).where(ModbusDevice.uuid == uuid)
).scalar_one_or_none()
if device is None:
raise HTTPException(
status_code=status.HTTP_404_NOT_FOUND,
detail=f"Device '{uuid}' not found.",
)
return device
def _label_from_key(key: str) -> str:
"""Derive a human-readable label from a metric key.
Example: ``"active_power"`` → ``"Active Power"``
"""
return key.replace("_", " ").title()
# ---------------------------------------------------------------------------
# GET /api/modbus/profiles
# ---------------------------------------------------------------------------
@router.get("/profiles", response_model=ModbusProfilesResponse)
def get_profiles(
_auth: AuthenticatedSession = Depends(require_session),
) -> ModbusProfilesResponse:
"""List all available Modbus YAML profiles (name + description).
Intended for the front-end's device-creation profile drop-down.
"""
pairs = list_profiles()
return ModbusProfilesResponse(
profiles=[ProfileSummary(name=name, description=desc) for name, desc in pairs]
)
# ---------------------------------------------------------------------------
# Device CRUD
# ---------------------------------------------------------------------------
@router.get("/devices", response_model=ModbusDeviceListResponse)
def list_devices(
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
) -> ModbusDeviceListResponse:
"""Return all Modbus devices (no pagination — device counts stay small)."""
devices = db.execute(select(ModbusDevice)).scalars().all()
items = [ModbusDeviceResponse.model_validate(d) for d in devices]
return ModbusDeviceListResponse(items=items, total=len(items))
@router.post(
"/devices",
response_model=ModbusDeviceResponse,
status_code=status.HTTP_201_CREATED,
)
def create_device(
body: ModbusDeviceCreate,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> ModbusDeviceResponse:
"""Create a new Modbus device.
- Validates that the referenced ``profile`` exists; returns 422 if not.
- Returns 201 with the created device on success.
"""
# Validate profile exists (422 if not)
try:
load_profile(body.profile)
except ProfileNotFoundError:
raise HTTPException(
status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
detail=f"Profile '{body.profile}' does not exist.",
)
now = datetime.now(UTC)
device = ModbusDevice(
friendly_name=body.friendly_name,
transport=body.transport,
host=body.host,
port=body.port,
unit_id=body.unit_id,
profile=body.profile,
poll_interval_s=body.poll_interval_s,
enabled=body.enabled,
created_at=now,
updated_at=now,
)
db.add(device)
db.commit()
db.refresh(device)
logger.info("Created Modbus device %r (uuid=%s)", device.friendly_name, device.uuid)
return ModbusDeviceResponse.model_validate(device)
@router.get("/devices/{uuid}", response_model=ModbusDeviceResponse)
def get_device(
uuid: str,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
) -> ModbusDeviceResponse:
"""Return a single Modbus device by UUID."""
device = _get_device_or_404(db, uuid)
return ModbusDeviceResponse.model_validate(device)
@router.patch("/devices/{uuid}", response_model=ModbusDeviceResponse)
def patch_device(
uuid: str,
body: ModbusDeviceUpdate,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> ModbusDeviceResponse:
"""Partially update a Modbus device (including enable/disable).
Only fields explicitly provided in the request body are updated.
Providing ``profile`` triggers a profile-existence check (422 if unknown).
"""
device = _get_device_or_404(db, uuid)
update_data = body.model_dump(exclude_none=True)
# If profile is being changed, validate it exists.
if "profile" in update_data:
try:
load_profile(update_data["profile"])
except ProfileNotFoundError:
raise HTTPException(
status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
detail=f"Profile '{update_data['profile']}' does not exist.",
)
for field, value in update_data.items():
setattr(device, field, value)
device.updated_at = datetime.now(UTC)
db.commit()
db.refresh(device)
logger.info("Updated Modbus device %r (uuid=%s)", device.friendly_name, device.uuid)
return ModbusDeviceResponse.model_validate(device)
@router.delete(
"/devices/{uuid}",
status_code=status.HTTP_204_NO_CONTENT,
response_model=None,
)
def delete_device(
uuid: str,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> None:
"""Delete a Modbus device.
Returns 409 Conflict if the device has any associated readings; use
``enabled=false`` to disable it instead.
**Application-layer safety**: the check is performed via an explicit
SELECT COUNT query — not by relying on SQLite's FK RESTRICT constraint,
which is not enforced at runtime in this project (no PRAGMA foreign_keys=ON).
"""
device = _get_device_or_404(db, uuid)
# Application-layer guard: refuse deletion if any readings exist.
# We do NOT rely on SQLite FK RESTRICT — it is not enforced at runtime
# without PRAGMA foreign_keys=ON, which is not set in this project.
reading_count: int = db.execute(
select(func.count(ModbusReading.id)).where(ModbusReading.device_id == device.id)
).scalar_one()
if reading_count > 0:
raise HTTPException(
status_code=status.HTTP_409_CONFLICT,
detail=(
f"Device '{uuid}' has {reading_count} associated reading(s) and cannot be "
"deleted. Disable the device (set enabled=false) instead to stop polling "
"without losing historical data."
),
)
logger.info("Deleting Modbus device %r (uuid=%s)", device.friendly_name, uuid)
db.delete(device)
db.commit()
# ---------------------------------------------------------------------------
# Readings endpoints
# ---------------------------------------------------------------------------
@router.get("/devices/{uuid}/latest", response_model=ModbusLatestResponse)
def get_latest_reading(
uuid: str,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
) -> ModbusLatestResponse:
"""Return the most recent reading for a device.
If no readings exist yet, returns ``{"found": false, "recorded_at": null,
"payload": null}`` (200, not 404) so the front-end can distinguish
"device exists but has no data" from "device not found".
"""
device = _get_device_or_404(db, uuid)
row = db.execute(
select(ModbusReading)
.where(ModbusReading.device_id == device.id)
.order_by(ModbusReading.recorded_at.desc())
.limit(1)
).scalar_one_or_none()
if row is None:
return ModbusLatestResponse(found=False, recorded_at=None, payload=None)
return ModbusLatestResponse(
found=True,
recorded_at=row.recorded_at,
payload=row.payload,
)
@router.get("/devices/{uuid}/readings", response_model=ModbusReadingsResponse)
def get_readings(
uuid: str,
start: datetime | None = Query(default=None),
end: datetime | None = Query(default=None),
limit: int = Query(default=500, ge=1, le=_READINGS_LIMIT_MAX),
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
) -> ModbusReadingsResponse:
"""Return time-range readings for a device.
When the window contains more rows than ``limit``, the **most recent** N rows
are returned (``ORDER BY recorded_at DESC LIMIT n``), then reversed to
ascending order before being sent to the client. This ensures that for long
time-range requests (e.g. 6 h / 24 h) the caller always sees the latest data
rather than the oldest segment of the window.
When the window has fewer rows than ``limit`` the full window is returned in
ascending order — behaviour is identical to a plain ascending query.
The response schema is unchanged: items are always ``recorded_at`` ascending.
The query uses the ``(device_id, recorded_at)`` composite index for
efficient time-window scans.
Query parameters:
- ``start``: inclusive lower bound (ISO8601 datetime)
- ``end``: inclusive upper bound (ISO8601 datetime)
- ``limit``: max rows to return (default 500, max {_READINGS_LIMIT_MAX})
"""
device = _get_device_or_404(db, uuid)
stmt = (
select(ModbusReading)
.where(ModbusReading.device_id == device.id)
.order_by(ModbusReading.recorded_at.desc())
.limit(limit)
)
if start is not None:
stmt = stmt.where(ModbusReading.recorded_at >= start)
if end is not None:
stmt = stmt.where(ModbusReading.recorded_at <= end)
# Fetch most-recent N rows, then reverse to restore ascending order for the response.
rows = list(reversed(db.execute(stmt).scalars().all()))
items = [ModbusReadingResponse(recorded_at=r.recorded_at, payload=r.payload) for r in rows]
return ModbusReadingsResponse(items=items)
# ---------------------------------------------------------------------------
# Metrics endpoint
# ---------------------------------------------------------------------------
@router.get("/devices/{uuid}/metrics", response_model=ModbusMetricsResponse)
def get_metrics(
uuid: str,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
) -> ModbusMetricsResponse:
"""Return the metric catalogue for a device's profile.
Each entry has ``key``, ``label`` (derived from key if the profile has
none — underscores → spaces, title-cased), ``unit``, and ``device_class``.
This is the authoritative metadata source for front-end card labels and
chart axis labels.
"""
device = _get_device_or_404(db, uuid)
try:
profile = load_profile(device.profile)
except ProfileNotFoundError:
raise HTTPException(
status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
detail=f"Device profile '{device.profile}' could not be loaded.",
)
metrics = [
MetricInfo(
key=m.key,
label=_label_from_key(m.key),
unit=m.unit,
device_class=m.device_class,
)
for m in profile.metrics
]
return ModbusMetricsResponse(profile=profile.name, metrics=metrics)
# ---------------------------------------------------------------------------
# Test-read endpoint
# ---------------------------------------------------------------------------
@router.post("/devices/{uuid}/test", response_model=ModbusTestReadResponse)
def test_read(
uuid: str,
db: Session = Depends(get_db),
_auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> ModbusTestReadResponse:
"""Immediately read and decode the device once without persisting any data.
Useful for validating gateway connectivity and Modbus address settings
before relying on the background polling job.
This is a write-class endpoint (it initiates network I/O on demand) and
therefore requires a CSRF token. The response payload is never stored.
"""
device = _get_device_or_404(db, uuid)
try:
profile = load_profile(device.profile)
except ProfileNotFoundError:
raise HTTPException(
status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
detail=f"Device profile '{device.profile}' could not be loaded.",
)
try:
registers = modbus_driver.read_blocks(
device.host,
device.port,
device.unit_id,
[{"start": b.start, "count": b.count} for b in profile.blocks],
)
payload: dict[str, Any] = decode_profile(profile, registers)
return ModbusTestReadResponse(ok=True, payload=payload)
except ModbusDriverError as exc:
logger.warning(
"Test read failed for device %r (uuid=%s): %s",
device.friendly_name,
uuid,
exc,
)
return ModbusTestReadResponse(ok=False, error=str(exc))
except Exception as exc: # noqa: BLE001
logger.warning(
"Unexpected error during test read for device %r (uuid=%s): %s",
device.friendly_name,
uuid,
exc,
)
return ModbusTestReadResponse(ok=False, error=f"Unexpected error: {exc}")
+324
View File
@@ -0,0 +1,324 @@
from __future__ import annotations
import logging
from fastapi import APIRouter, Depends, HTTPException, Request, Response, status
from sqlalchemy.orm import Session
from app.api.routes.api.deps import require_csrf, require_session
from app.config import Settings
from app.dependencies import get_app_settings, get_db
from app.schemas.session import (
LoginRequest,
PasswordChangeRequest,
SessionResponse,
SessionUser,
)
from app.schemas.totp import (
TotpDisableRequest,
TotpEnableRequest,
TotpSetupResponse,
TotpStatusResponse,
)
from app.services.auth import (
AuthPasswordChangeError,
AuthenticatedSession,
authenticate_user,
change_password,
create_session,
revoke_session,
)
from app.services import login_throttle
from app.services import totp as totp_service
logger = logging.getLogger(__name__)
router = APIRouter(prefix="/api", tags=["api-session"])
def _build_session_response(auth: AuthenticatedSession) -> SessionResponse:
return SessionResponse(
user=SessionUser(
username=auth.user.username,
force_password_change=auth.user.force_password_change,
),
csrf_token=auth.session.csrf_token,
)
@router.get("/session", response_model=SessionResponse)
def get_session(
auth: AuthenticatedSession = Depends(require_session),
) -> SessionResponse:
"""Return the current session user and CSRF token. Returns 401 if not authenticated."""
return _build_session_response(auth)
def _get_client_ip(request: Request, *, trust_forwarded_for: bool) -> str:
"""Extract the client IP address from the request.
When ``trust_forwarded_for`` is True (reverse-proxy deployments) the
left-most value from the ``X-Forwarded-For`` header is used. Otherwise the
direct socket IP (``request.client.host``) is used.
"""
if trust_forwarded_for:
xff = request.headers.get("X-Forwarded-For", "")
if xff:
return xff.split(",")[0].strip()
if request.client is not None:
return request.client.host
return "unknown"
@router.post("/auth/login", response_model=SessionResponse)
def post_login(
body: LoginRequest,
request: Request,
response: Response,
db: Session = Depends(get_db),
settings: Settings = Depends(get_app_settings),
) -> SessionResponse:
"""
Authenticate with username and password.
On success, sets an HttpOnly session cookie and returns the session user + CSRF token.
On failure, returns 401 with no cookie set.
Repeated failures trigger exponential back-off (429 + Retry-After).
No X-CSRF-Token required (unauthenticated endpoint).
"""
client_ip = _get_client_ip(request, trust_forwarded_for=settings.auth_trust_forwarded_for)
# --- Throttle check (before any password verification) ---
if settings.auth_login_throttle_enabled:
wait_seconds = login_throttle.check_and_get_wait(
db, ip=client_ip, username=body.username
)
if wait_seconds > 0:
raise HTTPException(
status_code=status.HTTP_429_TOO_MANY_REQUESTS,
detail="too many failed login attempts; please try again later",
headers={"Retry-After": str(wait_seconds)},
)
# --- Password verification ---
user = authenticate_user(db, username=body.username, password=body.password)
if user is None:
if settings.auth_login_throttle_enabled:
login_throttle.register_failure(db, ip=client_ip, username=body.username)
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail="invalid username or password",
)
# --- TOTP second-factor check (only when TOTP is enabled for the user) ---
if user.totp_enabled:
if not body.totp_code:
# Password correct but no TOTP code supplied: signal the front-end to
# prompt for the second factor. Do NOT issue a session.
# Deliberate: this is a normal two-step protocol step from a legitimate
# user — we do NOT register a throttle failure here. The attacker must
# already have the correct password to reach this branch, and counting
# each normal first-step as a failure would risk locking out the real
# admin on every login attempt.
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail={"totp_required": True},
)
# TOTP code supplied — verify against time-based code or a recovery code.
totp_ok = totp_service.verify_totp_code(user, body.totp_code)
if not totp_ok:
totp_ok = totp_service.verify_recovery_code(db, user=user, code=body.totp_code)
if not totp_ok:
# Second-factor failure is an active attack signal — register failure
# so that repeated wrong TOTP/recovery-code guesses trigger back-off.
if settings.auth_login_throttle_enabled:
login_throttle.register_failure(db, ip=client_ip, username=body.username)
raise HTTPException(
status_code=status.HTTP_401_UNAUTHORIZED,
detail="invalid username or password",
)
# --- Success: clear back-off state and issue session ---
if settings.auth_login_throttle_enabled:
login_throttle.clear(db, ip=client_ip, username=body.username)
auth_session, raw_token = create_session(db, user=user, settings=settings)
logger.info("Created API authenticated session for user '%s'", user.username)
response.set_cookie(
key=settings.auth_session_cookie_name,
value=raw_token,
max_age=settings.auth_session_ttl_hours * 3600,
httponly=True,
secure=settings.auth_cookie_secure,
samesite="lax",
path="/",
)
auth = AuthenticatedSession(user=user, session=auth_session)
return _build_session_response(auth)
@router.post("/auth/logout")
def post_logout(
response: Response,
db: Session = Depends(get_db),
settings: Settings = Depends(get_app_settings),
auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> Response:
"""
Revoke the current session and clear the session cookie.
Requires authentication and X-CSRF-Token header.
Returns 204 No Content.
"""
revoke_session(db, auth_session=auth.session)
logger.info("Revoked API authenticated session for user '%s'", auth.user.username)
no_content = Response(status_code=status.HTTP_204_NO_CONTENT)
no_content.delete_cookie(settings.auth_session_cookie_name, path="/")
return no_content
@router.post("/auth/password")
def post_change_password(
body: PasswordChangeRequest,
db: Session = Depends(get_db),
auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> Response:
"""
Change the current user's password.
Requires authentication and X-CSRF-Token header.
On AuthPasswordChangeError returns 400 with a generic message.
On success, force_password_change becomes False (handled by the service).
Returns 204 No Content.
"""
try:
change_password(
db,
user=auth.user,
current_password=body.current_password,
new_password=body.new_password,
confirm_password=body.confirm_password,
)
except AuthPasswordChangeError as exc:
logger.info(
"Rejected password change for user '%s': %s",
auth.user.username,
exc,
)
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="password change failed",
) from exc
logger.info("Password updated for user '%s'", auth.user.username)
return Response(status_code=status.HTTP_204_NO_CONTENT)
# ---------------------------------------------------------------------------
# TOTP endpoints (M4-T05)
# ---------------------------------------------------------------------------
@router.post("/auth/totp/setup", response_model=TotpSetupResponse)
def post_totp_setup(
db: Session = Depends(get_db),
settings: Settings = Depends(get_app_settings),
auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> TotpSetupResponse:
"""
Generate a new pending TOTP secret, otpauth URI, and one-time recovery codes.
The secret is stored in the DB but TOTP is NOT yet enabled (totp_enabled stays
False until the user confirms with POST /api/auth/totp/enable).
Recovery codes are returned here as plaintext exactly once; their Argon2 hashes
are persisted immediately so enable only needs to flip the enabled flag.
Repeating this call replaces any prior pending secret and regenerates codes.
Requires: session cookie + X-CSRF-Token.
"""
secret, otpauth_uri, recovery_codes = totp_service.setup(
db,
user=auth.user,
issuer=settings.effective_totp_issuer,
)
return TotpSetupResponse(
secret=secret,
otpauth_uri=otpauth_uri,
recovery_codes=recovery_codes,
)
@router.post("/auth/totp/enable", status_code=status.HTTP_204_NO_CONTENT)
def post_totp_enable(
body: TotpEnableRequest,
db: Session = Depends(get_db),
auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> Response:
"""
Enable TOTP by confirming with the current 6-digit code from the authenticator app.
Requires a prior call to POST /api/auth/totp/setup (so that a pending secret
exists). On success, totp_enabled becomes True.
Returns 400 if the code is wrong or there is no pending secret.
Requires: session cookie + X-CSRF-Token.
"""
ok = totp_service.enable(db, user=auth.user, code=body.code)
if not ok:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="invalid TOTP code or no pending setup",
)
return Response(status_code=status.HTTP_204_NO_CONTENT)
@router.post("/auth/totp/disable", status_code=status.HTTP_204_NO_CONTENT)
def post_totp_disable(
body: TotpDisableRequest,
db: Session = Depends(get_db),
auth: AuthenticatedSession = Depends(require_session),
_csrf: None = Depends(require_csrf),
) -> Response:
"""
Disable TOTP. The caller must provide exactly one of:
- ``password``: the user's current login password, OR
- ``code``: the current 6-digit TOTP code.
On success: totp_enabled=False, totp_secret cleared, all recovery codes deleted.
Returns 400 if neither credential matches or neither is provided.
Requires: session cookie + X-CSRF-Token.
"""
ok = totp_service.disable(
db,
user=auth.user,
password=body.password,
code=body.code,
)
if not ok:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="invalid credential; provide a valid password or TOTP code",
)
return Response(status_code=status.HTTP_204_NO_CONTENT)
@router.get("/auth/totp", response_model=TotpStatusResponse)
def get_totp_status(
auth: AuthenticatedSession = Depends(require_session),
) -> TotpStatusResponse:
"""
Return the current TOTP status for the authenticated user.
Response contains only ``{"enabled": bool}``.
Secret and recovery codes are NEVER returned here.
Requires: session cookie only (no CSRF — read-only).
"""
return TotpStatusResponse(enabled=auth.user.totp_enabled)
-234
View File
@@ -1,234 +0,0 @@
import logging
from pathlib import Path
from fastapi import APIRouter, Depends, Form, Request, status
from fastapi.responses import HTMLResponse, RedirectResponse, Response
from fastapi.templating import Jinja2Templates
from sqlalchemy.orm import Session
from app.config import Settings
from app.dependencies import get_app_settings, get_auth_db, get_current_auth_session
from app.services.auth import (
AuthenticatedSession,
authenticate_user,
change_password,
create_session,
AuthPasswordChangeError,
issue_login_csrf_token,
revoke_session,
validate_csrf_token,
)
from app.services.config_page import build_config_sections, is_ticktick_oauth_ready
logger = logging.getLogger(__name__)
templates = Jinja2Templates(directory=str(Path(__file__).resolve().parents[2] / "templates"))
router = APIRouter(tags=["auth"])
LOGIN_CSRF_COOKIE_NAME = "login_csrf"
@router.get("/login", response_class=HTMLResponse)
def login_page(
request: Request,
settings: Settings = Depends(get_app_settings),
current_auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> Response:
if current_auth is not None:
return RedirectResponse(url="/config", status_code=status.HTTP_303_SEE_OTHER)
csrf_token = issue_login_csrf_token()
response = templates.TemplateResponse(
request,
"login.html",
{
"app_name": settings.app_name,
"app_env": settings.app_env,
"csrf_token": csrf_token,
"error_message": None,
},
)
_set_login_csrf_cookie(response, settings=settings, token=csrf_token)
return response
@router.post("/login", response_class=HTMLResponse)
def login_submit(
request: Request,
username: str = Form(),
password: str = Form(),
csrf_token: str = Form(),
session: Session = Depends(get_auth_db),
settings: Settings = Depends(get_app_settings),
) -> Response:
cookie_csrf_token = request.cookies.get(LOGIN_CSRF_COOKIE_NAME)
if not validate_csrf_token(expected=cookie_csrf_token, actual=csrf_token):
logger.warning("Rejected login attempt due to CSRF validation failure")
return _render_login_error(
request,
settings=settings,
status_code=status.HTTP_400_BAD_REQUEST,
error_message="invalid login request",
)
user = authenticate_user(session, username=username, password=password)
if user is None:
return _render_login_error(
request,
settings=settings,
status_code=status.HTTP_401_UNAUTHORIZED,
error_message="invalid username or password",
)
auth_session, raw_token = create_session(session, user=user, settings=settings)
response = RedirectResponse(url="/config", status_code=status.HTTP_303_SEE_OTHER)
response.delete_cookie(LOGIN_CSRF_COOKIE_NAME, path="/login")
response.set_cookie(
key=settings.auth_session_cookie_name,
value=raw_token,
max_age=settings.auth_session_ttl_hours * 3600,
httponly=True,
secure=settings.auth_cookie_secure,
samesite="lax",
path="/",
)
logger.info("Created authenticated session for user '%s'", user.username)
return response
@router.post("/config/change-password", response_class=HTMLResponse)
def change_password_submit(
request: Request,
current_password: str = Form(),
new_password: str = Form(),
confirm_password: str = Form(),
csrf_token: str = Form(),
session: Session = Depends(get_auth_db),
settings: Settings = Depends(get_app_settings),
current_auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> Response:
if current_auth is None:
return RedirectResponse(url="/login", status_code=status.HTTP_303_SEE_OTHER)
if not validate_csrf_token(expected=current_auth.session.csrf_token, actual=csrf_token):
logger.warning("Rejected password change attempt due to CSRF validation failure")
return _render_config_page(
request,
settings=settings,
auth_db_session=session,
current_auth=current_auth,
status_code=status.HTTP_400_BAD_REQUEST,
password_change_error="invalid password change request",
)
try:
change_password(
session,
user=current_auth.user,
current_password=current_password,
new_password=new_password,
confirm_password=confirm_password,
)
except AuthPasswordChangeError as exc:
logger.info(
"Rejected password change for user '%s': %s",
current_auth.user.username,
exc,
)
return _render_config_page(
request,
settings=settings,
auth_db_session=session,
current_auth=current_auth,
status_code=status.HTTP_400_BAD_REQUEST,
password_change_error="password change failed",
)
logger.info("Password updated for user '%s'", current_auth.user.username)
return RedirectResponse(url="/config", status_code=status.HTTP_303_SEE_OTHER)
@router.post("/logout")
def logout(
request: Request,
csrf_token: str = Form(),
session: Session = Depends(get_auth_db),
settings: Settings = Depends(get_app_settings),
current_auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> RedirectResponse:
if current_auth is not None and validate_csrf_token(
expected=current_auth.session.csrf_token, actual=csrf_token
):
revoke_session(session, auth_session=current_auth.session)
logger.info("Revoked authenticated session for user '%s'", current_auth.user.username)
else:
logger.warning("Rejected logout request due to missing session or invalid CSRF token")
response = RedirectResponse(url="/login", status_code=status.HTTP_303_SEE_OTHER)
response.delete_cookie(settings.auth_session_cookie_name, path="/")
return response
def _render_login_error(
request: Request,
*,
settings: Settings,
status_code: int,
error_message: str,
) -> HTMLResponse:
csrf_token = issue_login_csrf_token()
response = templates.TemplateResponse(
request,
"login.html",
{
"app_name": settings.app_name,
"app_env": settings.app_env,
"csrf_token": csrf_token,
"error_message": error_message,
},
status_code=status_code,
)
_set_login_csrf_cookie(response, settings=settings, token=csrf_token)
return response
def _set_login_csrf_cookie(response: HTMLResponse, *, settings: Settings, token: str) -> None:
response.set_cookie(
key=LOGIN_CSRF_COOKIE_NAME,
value=token,
max_age=1800,
httponly=True,
secure=settings.auth_cookie_secure,
samesite="lax",
path="/login",
)
def _render_config_page(
request: Request,
*,
settings: Settings,
auth_db_session: Session,
current_auth: AuthenticatedSession,
status_code: int,
password_change_error: str | None,
) -> HTMLResponse:
return templates.TemplateResponse(
request,
"config.html",
{
"app_name": settings.app_name,
"app_env": settings.app_env,
"current_username": current_auth.user.username,
"csrf_token": current_auth.session.csrf_token,
"force_password_change": current_auth.user.force_password_change,
"password_change_error": password_change_error,
"config_error": None,
"config_saved": False,
"config_sections": build_config_sections(auth_db_session, settings),
"ticktick_oauth_ready": is_ticktick_oauth_ready(settings),
"ticktick_redirect_uri": settings.ticktick_redirect_uri,
"ticktick_oauth_notice": None,
"ticktick_oauth_error": None,
},
status_code=status_code,
)
+1 -3
View File
@@ -11,7 +11,6 @@ from app.dependencies import (
get_app_settings,
get_db,
get_homeassistant_client,
get_poo_db,
get_ticktick_client,
)
from app.integrations.homeassistant import (
@@ -36,7 +35,6 @@ INTERNAL_SERVER_ERROR_MESSAGE = "internal server error"
async def publish_from_homeassistant(
request: Request,
db: Session = Depends(get_db),
poo_db: Session = Depends(get_poo_db),
settings: Settings = Depends(get_app_settings),
homeassistant_client: HomeAssistantClient = Depends(get_homeassistant_client),
ticktick_client: TickTickClient = Depends(get_ticktick_client),
@@ -49,7 +47,7 @@ async def publish_from_homeassistant(
db,
envelope,
ticktick_client=ticktick_client,
poo_session=poo_db,
poo_session=db,
settings=settings,
homeassistant_client=homeassistant_client,
)
-240
View File
@@ -1,240 +0,0 @@
import logging
from pathlib import Path
from fastapi import APIRouter, Depends, Request, status
from fastapi.responses import HTMLResponse, RedirectResponse, Response
from fastapi.templating import Jinja2Templates
from app.config import Settings, get_settings
from app.dependencies import get_app_settings, get_auth_db, get_current_auth_session
from app.services.auth import AuthenticatedSession
from app.services.config_page import (
ConfigSaveError,
build_config_sections,
is_ticktick_oauth_ready,
save_config_updates,
)
from app.services.email import EmailConfigurationError, EmailDeliveryError, is_smtp_ready, send_smtp_test_email
from sqlalchemy.orm import Session
templates = Jinja2Templates(directory=str(Path(__file__).resolve().parents[2] / "templates"))
router = APIRouter(tags=["pages"])
logger = logging.getLogger(__name__)
def _ticktick_oauth_notice(status_value: str | None) -> tuple[str | None, str | None]:
if status_value == "success":
return "TickTick authorization completed successfully.", None
if status_value == "invalid-state":
return None, "TickTick authorization failed due to invalid OAuth state. Start the flow again."
if status_value == "invalid-callback":
return None, "TickTick authorization callback was missing required parameters."
if status_value == "failed":
return None, "TickTick authorization failed. Check server logs for the provider response and verify TickTick app credentials and redirect URI."
return None, None
def _smtp_test_notice(status_value: str | None) -> tuple[str | None, str | None]:
if status_value == "success":
return "SMTP test email sent successfully.", None
if status_value == "config-error":
return None, "SMTP test failed. Check required SMTP settings before sending a test email."
if status_value == "failed":
return None, "SMTP test failed. Check saved SMTP settings and server reachability."
return None, None
def _build_config_context(
*,
auth_db_session: Session,
settings: Settings,
current_auth: AuthenticatedSession,
config_saved: bool,
config_error: str | None,
password_change_error: str | None,
ticktick_oauth_notice: str | None,
ticktick_oauth_error: str | None,
smtp_test_notice: str | None,
smtp_test_error: str | None,
) -> dict[str, object]:
return {
"app_name": settings.app_name,
"app_env": settings.app_env,
"current_username": current_auth.user.username,
"csrf_token": current_auth.session.csrf_token,
"force_password_change": current_auth.user.force_password_change,
"password_change_error": password_change_error,
"config_error": config_error,
"config_saved": config_saved,
"config_sections": build_config_sections(auth_db_session, settings),
"ticktick_oauth_ready": is_ticktick_oauth_ready(settings),
"ticktick_redirect_uri": settings.ticktick_redirect_uri,
"ticktick_oauth_notice": ticktick_oauth_notice,
"ticktick_oauth_error": ticktick_oauth_error,
"smtp_test_ready": is_smtp_ready(settings),
"smtp_test_notice": smtp_test_notice,
"smtp_test_error": smtp_test_error,
}
@router.get("/", response_class=HTMLResponse)
def home(
request: Request,
current_auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> RedirectResponse:
if current_auth is None:
return RedirectResponse(url="/login", status_code=status.HTTP_303_SEE_OTHER)
return RedirectResponse(url="/config", status_code=status.HTTP_303_SEE_OTHER)
@router.get("/admin", response_class=HTMLResponse)
def admin_redirect(
request: Request,
current_auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> RedirectResponse:
if current_auth is None:
return RedirectResponse(url="/login", status_code=status.HTTP_303_SEE_OTHER)
return RedirectResponse(url="/config", status_code=status.HTTP_303_SEE_OTHER)
@router.get("/config", response_class=HTMLResponse)
def config_page(
request: Request,
auth_db_session: Session = Depends(get_auth_db),
settings: Settings = Depends(get_app_settings),
current_auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> Response:
if current_auth is None:
return RedirectResponse(url="/login", status_code=status.HTTP_303_SEE_OTHER)
ticktick_oauth_notice, ticktick_oauth_error = _ticktick_oauth_notice(
request.query_params.get("ticktick_oauth")
)
smtp_test_notice, smtp_test_error = _smtp_test_notice(request.query_params.get("smtp_test"))
context = _build_config_context(
auth_db_session=auth_db_session,
settings=settings,
current_auth=current_auth,
config_saved=request.query_params.get("saved") == "1",
config_error=None,
password_change_error=None,
ticktick_oauth_notice=ticktick_oauth_notice,
ticktick_oauth_error=ticktick_oauth_error,
smtp_test_notice=smtp_test_notice,
smtp_test_error=smtp_test_error,
)
return templates.TemplateResponse(request, "config.html", context)
@router.post("/config", response_class=HTMLResponse)
async def config_submit(
request: Request,
auth_db_session: Session = Depends(get_auth_db),
settings: Settings = Depends(get_app_settings),
current_auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> Response:
if current_auth is None:
return RedirectResponse(url="/login", status_code=status.HTTP_303_SEE_OTHER)
form = await request.form()
csrf_token = form.get("csrf_token")
if csrf_token != current_auth.session.csrf_token:
logger.warning("Rejected config update due to CSRF validation failure")
context = _build_config_context(
auth_db_session=auth_db_session,
settings=settings,
current_auth=current_auth,
config_saved=False,
config_error="invalid config update request",
password_change_error=None,
ticktick_oauth_notice=None,
ticktick_oauth_error=None,
smtp_test_notice=None,
smtp_test_error=None,
)
return templates.TemplateResponse(
request,
"config.html",
context,
status_code=status.HTTP_400_BAD_REQUEST,
)
try:
save_config_updates(auth_db_session, dict(form), settings)
except ConfigSaveError:
logger.warning("Rejected config update due to invalid submitted values")
refreshed_settings = get_settings()
context = _build_config_context(
auth_db_session=auth_db_session,
settings=refreshed_settings,
current_auth=current_auth,
config_saved=False,
config_error="invalid config submission",
password_change_error=None,
ticktick_oauth_notice=None,
ticktick_oauth_error=None,
smtp_test_notice=None,
smtp_test_error=None,
)
return templates.TemplateResponse(
request,
"config.html",
context,
status_code=status.HTTP_400_BAD_REQUEST,
)
return RedirectResponse(url="/config?saved=1", status_code=status.HTTP_303_SEE_OTHER)
@router.post("/config/smtp/test", response_class=HTMLResponse)
async def smtp_test_submit(
request: Request,
auth_db_session: Session = Depends(get_auth_db),
settings: Settings = Depends(get_app_settings),
current_auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> Response:
if current_auth is None:
return RedirectResponse(url="/login", status_code=status.HTTP_303_SEE_OTHER)
form = await request.form()
csrf_token = form.get("csrf_token")
if csrf_token != current_auth.session.csrf_token:
logger.warning("Rejected SMTP test due to CSRF validation failure")
context = _build_config_context(
auth_db_session=auth_db_session,
settings=settings,
current_auth=current_auth,
config_saved=False,
config_error=None,
password_change_error=None,
ticktick_oauth_notice=None,
ticktick_oauth_error=None,
smtp_test_notice=None,
smtp_test_error="invalid SMTP test request",
)
return templates.TemplateResponse(
request,
"config.html",
context,
status_code=status.HTTP_400_BAD_REQUEST,
)
try:
send_smtp_test_email(settings)
except EmailConfigurationError as exc:
logger.warning("SMTP test email rejected due to configuration: %s", exc)
return RedirectResponse(
url="/config?smtp_test=config-error",
status_code=status.HTTP_303_SEE_OTHER,
)
except EmailDeliveryError as exc:
logger.warning("SMTP test email failed: %s", exc)
return RedirectResponse(
url="/config?smtp_test=failed",
status_code=status.HTTP_303_SEE_OTHER,
)
return RedirectResponse(
url="/config?smtp_test=success",
status_code=status.HTTP_303_SEE_OTHER,
)
+3 -3
View File
@@ -7,7 +7,7 @@ from pydantic import ValidationError
from sqlalchemy.orm import Session
from app.config import Settings
from app.dependencies import get_app_settings, get_homeassistant_client, get_poo_db
from app.dependencies import get_app_settings, get_homeassistant_client, get_db
from app.integrations.homeassistant import HomeAssistantClient
from app.schemas.poo import PooRecordRequest
from app.services.poo import publish_latest_poo_status, record_poo
@@ -21,7 +21,7 @@ INTERNAL_SERVER_ERROR_MESSAGE = "internal server error"
@router.post("/poo/record")
async def create_poo_record(
request: Request,
db: Session = Depends(get_poo_db),
db: Session = Depends(get_db),
settings: Settings = Depends(get_app_settings),
homeassistant_client: HomeAssistantClient = Depends(get_homeassistant_client),
) -> Response:
@@ -56,7 +56,7 @@ async def create_poo_record(
@router.get("/poo/latest")
def notify_latest_poo(
db: Session = Depends(get_poo_db),
db: Session = Depends(get_db),
settings: Settings = Depends(get_app_settings),
homeassistant_client: HomeAssistantClient = Depends(get_homeassistant_client),
) -> Response:
+2 -2
View File
@@ -1,7 +1,7 @@
from fastapi import APIRouter, Depends, HTTPException, status
from sqlalchemy.orm import Session
from app.dependencies import get_auth_db, get_current_auth_session
from app.dependencies import get_db, get_current_auth_session
from app.schemas.public_ip import PublicIPCheckResponse
from app.config import get_settings
from app.services.auth import AuthenticatedSession
@@ -12,7 +12,7 @@ router = APIRouter(tags=["public-ip"])
@router.get("/public-ip/check", response_model=PublicIPCheckResponse)
def run_public_ip_check(
session: Session = Depends(get_auth_db),
session: Session = Depends(get_db),
current_auth: AuthenticatedSession | None = Depends(get_current_auth_session),
) -> PublicIPCheckResponse:
if current_auth is None:
+2 -2
View File
@@ -7,7 +7,7 @@ from sqlalchemy.orm import Session
from app.config import Settings
from app.dependencies import (
get_app_settings,
get_auth_db,
get_db,
get_current_auth_session,
get_ticktick_client,
)
@@ -39,7 +39,7 @@ def start_ticktick_auth(
@router.get("/ticktick/auth/code")
def handle_ticktick_auth_code(
request: Request,
auth_db_session: Session = Depends(get_auth_db),
auth_db_session: Session = Depends(get_db),
settings: Settings = Depends(get_app_settings),
ticktick_client: TickTickClient = Depends(get_ticktick_client),
) -> Response:
-53
View File
@@ -1,53 +0,0 @@
from collections.abc import Generator
from functools import lru_cache
from sqlalchemy import create_engine
from sqlalchemy.orm import DeclarativeBase, Session, sessionmaker
from app.config import get_settings
class AuthBase(DeclarativeBase):
pass
def _build_connect_args(database_url: str) -> dict[str, object]:
connect_args: dict[str, object] = {}
if database_url.startswith("sqlite"):
connect_args["check_same_thread"] = False
return connect_args
@lru_cache
def _get_auth_engine(database_url: str):
return create_engine(database_url, connect_args=_build_connect_args(database_url))
@lru_cache
def _get_auth_session_local(database_url: str):
engine = _get_auth_engine(database_url)
return sessionmaker(bind=engine, autoflush=False, autocommit=False, class_=Session)
def get_auth_engine():
settings = get_settings()
return _get_auth_engine(settings.app_database_url)
def get_auth_session_local():
settings = get_settings()
return _get_auth_session_local(settings.app_database_url)
def reset_auth_db_caches() -> None:
_get_auth_session_local.cache_clear()
_get_auth_engine.cache_clear()
def get_auth_db_session() -> Generator[Session, None, None]:
session_local = get_auth_session_local()
session = session_local()
try:
yield session
finally:
session.close()
+26 -13
View File
@@ -12,9 +12,6 @@ class Settings(BaseSettings):
app_hostname: str = "localhost:8000"
app_database_url: str = "sqlite:///./data/app.db"
location_database_url: str = "sqlite:///./data/locationRecorder.db"
poo_database_url: str = "sqlite:///./data/pooRecorder.db"
ticktick_client_id: str = ""
ticktick_client_secret: str = ""
ticktick_token: str = ""
@@ -40,6 +37,26 @@ class Settings(BaseSettings):
auth_session_cookie_name: str = "home_automation_session"
auth_session_ttl_hours: int = 12
auth_cookie_secure_override: bool | None = True
auth_login_throttle_enabled: bool = True
auth_trust_forwarded_for: bool = False
auth_totp_issuer: str = "" # defaults to app_name when empty
# Modbus polling — global kill-switch (CONFIG_FIELDS registered in T08).
# Off by default: polling is opt-in so a fresh deploy writes no modbus_reading
# rows until the admin explicitly enables it (matches mqtt/discovery defaults).
modbus_polling_enabled: bool = False
# MQTT broker connection (T08 wires into CONFIG_FIELDS/UI; T10 builds the client).
mqtt_enabled: bool = False
mqtt_broker_host: str = ""
mqtt_broker_port: int = 1883
mqtt_username: str = ""
mqtt_password: str = ""
mqtt_tls_enabled: bool = False
# Home Assistant MQTT Discovery (T08 wires into CONFIG_FIELDS/UI; T11 does publishing).
ha_discovery_enabled: bool = False
ha_discovery_prefix: str = "homeassistant"
model_config = SettingsConfigDict(
env_file=".env",
@@ -77,21 +94,11 @@ class Settings(BaseSettings):
raw_path = database_url[len(prefix) :]
return Path(raw_path)
@computed_field
@property
def location_sqlite_path(self) -> Path | None:
return self._sqlite_path_from_url(self.location_database_url)
@computed_field
@property
def app_sqlite_path(self) -> Path | None:
return self._sqlite_path_from_url(self.app_database_url)
@computed_field
@property
def poo_sqlite_path(self) -> Path | None:
return self._sqlite_path_from_url(self.poo_database_url)
@computed_field
@property
def auth_cookie_secure(self) -> bool:
@@ -99,6 +106,12 @@ class Settings(BaseSettings):
return self.auth_cookie_secure_override
return not self.is_development
@computed_field
@property
def effective_totp_issuer(self) -> str:
"""The issuer label shown in Authenticator apps. Falls back to app_name."""
return self.auth_totp_issuer.strip() or self.app_name
@lru_cache
def get_settings() -> Settings:
+42 -8
View File
@@ -1,6 +1,8 @@
from collections.abc import Generator
from functools import lru_cache
from sqlalchemy import create_engine
from sqlalchemy import create_engine, event
from sqlalchemy.engine import Engine
from sqlalchemy.orm import DeclarativeBase, Session, sessionmaker
from app.config import get_settings
@@ -10,18 +12,50 @@ class Base(DeclarativeBase):
pass
settings = get_settings()
connect_args: dict[str, object] = {}
if settings.location_database_url.startswith("sqlite"):
def _build_connect_args(database_url: str) -> dict[str, object]:
connect_args: dict[str, object] = {}
if database_url.startswith("sqlite"):
connect_args["check_same_thread"] = False
return connect_args
engine = create_engine(settings.location_database_url, connect_args=connect_args)
SessionLocal = sessionmaker(bind=engine, autoflush=False, autocommit=False, class_=Session)
@lru_cache
def _get_engine(database_url: str) -> Engine:
engine = create_engine(database_url, connect_args=_build_connect_args(database_url))
if database_url.startswith("sqlite"):
@event.listens_for(engine, "connect")
def _enable_sqlite_wal(dbapi_connection, _connection_record):
cursor = dbapi_connection.cursor()
cursor.execute("PRAGMA journal_mode=WAL")
cursor.execute("PRAGMA foreign_keys=ON")
cursor.close()
return engine
@lru_cache
def _get_session_local(database_url: str) -> sessionmaker:
engine = _get_engine(database_url)
return sessionmaker(bind=engine, autoflush=False, autocommit=False, class_=Session)
def get_engine() -> Engine:
return _get_engine(get_settings().app_database_url)
def get_session_local() -> sessionmaker:
return _get_session_local(get_settings().app_database_url)
def reset_db_caches() -> None:
_get_session_local.cache_clear()
_get_engine.cache_clear()
def get_db_session() -> Generator[Session, None, None]:
session = SessionLocal()
session_local = get_session_local()
session = session_local()
try:
yield session
finally:
+3 -13
View File
@@ -3,30 +3,20 @@ from collections.abc import Generator
from fastapi import Depends, Request
from sqlalchemy.orm import Session
from app.auth_db import get_auth_db_session
from app.config import Settings, get_settings
from app.db import get_db_session
from app.integrations.homeassistant import HomeAssistantClient
from app.integrations.ticktick import TickTickClient
from app.poo_db import get_poo_db_session
from app.services.auth import AuthenticatedSession, get_authenticated_session
from app.services.config_page import build_runtime_settings
def get_auth_db() -> Generator[Session, None, None]:
yield from get_auth_db_session()
def get_app_settings(session: Session = Depends(get_auth_db)) -> Settings:
return build_runtime_settings(session, get_settings())
def get_db() -> Generator[Session, None, None]:
yield from get_db_session()
def get_poo_db() -> Generator[Session, None, None]:
yield from get_poo_db_session()
def get_app_settings(session: Session = Depends(get_db)) -> Settings:
return build_runtime_settings(session, get_settings())
def get_homeassistant_client(settings: Settings = Depends(get_app_settings)) -> HomeAssistantClient:
@@ -39,7 +29,7 @@ def get_ticktick_client(settings: Settings = Depends(get_app_settings)) -> TickT
def get_current_auth_session(
request: Request,
session: Session = Depends(get_auth_db),
session: Session = Depends(get_db),
settings: Settings = Depends(get_app_settings),
) -> AuthenticatedSession | None:
raw_token = request.cookies.get(settings.auth_session_cookie_name)
+350
View File
@@ -0,0 +1,350 @@
"""Expose framework: ExposableEntity, provider protocol, and build_catalog.
This module defines:
- ``ExposableEntity`` — a dataclass describing one publishable MQTT/HA entity.
- Provider protocol — a simple callable ``(session) -> list[ExposableEntity]``.
- A provider registry and registration helper.
- ``build_catalog(session)`` — merges all registered providers' entities with
per-key toggle state from the ``exposed_entity_toggle`` table.
Design notes
------------
- **No over-engineering**: providers are plain callables, not abstract base
classes. This project is personal / single-user; keep it flat.
- **Key stability**: entity keys MUST be derived from device uuid + metric key
(e.g. ``"modbus.<uuid>.voltage"``), never from auto-increment IDs. Stable
keys mean the toggle table survives device removal/re-addition without drift.
- **Metadata from profile**: the modbus provider reads ``device_class``,
``unit``, and ``component`` directly from the YAML profile's ``metrics[]``
rather than maintaining a separate mapping.
- **value_getter**: a callable ``() -> object | None`` or ``None`` if not yet
implemented (T11 will wire real getters). Presence of the field lets T11
call it without changing the dataclass interface.
"""
from __future__ import annotations
import logging
from dataclasses import dataclass, field
from typing import Any, Callable, Optional, Protocol
from sqlalchemy.orm import Session
logger = logging.getLogger(__name__)
# ---------------------------------------------------------------------------
# ExposableEntity
# ---------------------------------------------------------------------------
@dataclass
class DeviceInfo:
"""Grouping metadata that maps an entity to a logical HA device.
``identifiers`` corresponds to ``device.identifiers`` in HA Discovery
config — a stable tuple used to group entities under one device card.
``name`` is the human-readable device name (friendly_name).
"""
identifiers: tuple[str, ...]
"""Stable identifiers for HA device grouping (e.g. ``("modbus", "<uuid>")``).
These must not change over the device lifetime; they anchor the HA device
card even when friendly_name changes.
"""
name: str
"""Human-readable device name (may change; triggers HA discovery re-publish)."""
@dataclass
class ExposableEntity:
"""One publishable entity in the MQTT / HA Discovery expose framework.
Fields
------
key
Stable unique identifier for this entity. Used as the toggle table
key and as part of the HA Discovery ``unique_id`` derivation.
Convention: ``"<provider>.<device_uuid>.<metric_key>"``,
e.g. ``"modbus.abc123.voltage"`` or ``"modbus.abc123.online"``.
component
HA MQTT component type: ``"sensor"``, ``"binary_sensor"``, ``"switch"``, etc.
device
Grouping info (DeviceInfo) that determines which HA device card this
entity belongs to.
device_class
HA ``device_class`` string (e.g. ``"voltage"``, ``"power"``, ``"None"``).
Empty string or ``None`` means no device_class (e.g. for the online sensor).
unit
Physical unit string (e.g. ``"V"``, ``"A"``, ``"kWh"``).
Empty string for dimensionless.
name
Human-readable entity label (friendly display name, may include device
context, e.g. ``"SDM120 Voltage"``).
value_getter
Optional callable ``(session) -> object | None`` that returns the current
entity value given an open SQLAlchemy session. ``None`` means "not yet
wired". T11 calls this to obtain the current state before publishing.
Signature: ``Callable[[Session], Any]``.
state_class
HA ``state_class`` string (e.g. ``"measurement"``, ``"total_increasing"``).
``None`` means not set.
"""
key: str
component: str
device: DeviceInfo
device_class: Optional[str]
unit: str
name: str
value_getter: Optional[Callable[["Session"], Any]] = field(default=None, repr=False)
state_class: Optional[str] = None
# ---------------------------------------------------------------------------
# Provider protocol
# ---------------------------------------------------------------------------
class EntityProvider(Protocol):
"""Protocol for entity provider callables.
A provider is any callable that accepts a ``Session`` and returns a list
of ``ExposableEntity`` objects. Using a Protocol (not ABC) keeps the
implementation lightweight — any matching callable qualifies.
"""
def __call__(self, session: Session) -> list[ExposableEntity]: ...
# ---------------------------------------------------------------------------
# Provider registry
# ---------------------------------------------------------------------------
_REGISTRY: list[EntityProvider] = []
def register_provider(provider: EntityProvider) -> EntityProvider:
"""Register an entity provider.
Can be used as a decorator::
@register_provider
def my_provider(session: Session) -> list[ExposableEntity]:
...
Or called directly::
register_provider(my_provider)
Returns the provider unchanged (for decorator compatibility).
"""
_REGISTRY.append(provider)
return provider
def get_providers() -> list[EntityProvider]:
"""Return a snapshot of the currently registered providers."""
return list(_REGISTRY)
# ---------------------------------------------------------------------------
# Catalog builder
# ---------------------------------------------------------------------------
@dataclass
class CatalogEntry:
"""An entity from the catalog, enriched with its current toggle state."""
entity: ExposableEntity
enabled: bool
"""True if this entity is currently enabled in the toggle table."""
def build_catalog(session: Session) -> list[CatalogEntry]:
"""Enumerate all entities from registered providers and attach toggle state.
For each entity key, the toggle state is looked up in ``exposed_entity_toggle``.
Entities with no toggle row default to ``enabled=False``.
Parameters
----------
session:
Active SQLAlchemy session for DB access.
Returns
-------
list[CatalogEntry]
All entities from all registered providers, each paired with its
current ``enabled`` state.
"""
from app.models.expose import ExposedEntityToggle # local import to avoid circular
# Collect all entities from all providers.
all_entities: list[ExposableEntity] = []
for provider in _REGISTRY:
try:
entities = provider(session)
all_entities.extend(entities)
except Exception:
logger.exception("Provider %r raised an error; skipping its entities", provider)
if not all_entities:
return []
# Load all toggle rows in one query for efficiency.
keys = [e.key for e in all_entities]
toggle_rows = session.query(ExposedEntityToggle).filter(
ExposedEntityToggle.key.in_(keys)
).all()
toggle_map: dict[str, bool] = {row.key: row.enabled for row in toggle_rows}
return [
CatalogEntry(entity=entity, enabled=toggle_map.get(entity.key, False))
for entity in all_entities
]
# ---------------------------------------------------------------------------
# Modbus provider
# ---------------------------------------------------------------------------
def _modbus_provider(session: Session) -> list[ExposableEntity]:
"""Enumerate ExposableEntity objects for all enabled Modbus devices.
For each enabled ``ModbusDevice``:
- Loads its YAML profile.
- Produces one ``sensor`` entity per metric in ``profile.metrics``
(device_class / unit / component taken from the metric spec).
- Produces one ``binary_sensor`` entity named "online" (derived from
``device.last_poll_ok``).
Entity keys follow the pattern ``"modbus.<uuid>.<metric_key>"``, which is
stable across device renames and DB rebuilds.
"""
from app.models.modbus import ModbusDevice # local import
from app.integrations.modbus.profiles import (
load_profile,
ProfileNotFoundError,
ProfileValidationError,
)
devices: list[ModbusDevice] = session.query(ModbusDevice).filter(
ModbusDevice.enabled.is_(True)
).all()
entities: list[ExposableEntity] = []
for device in devices:
device_info = DeviceInfo(
identifiers=("modbus", device.uuid),
name=device.friendly_name,
)
# Load the profile to get metric metadata.
try:
profile = load_profile(device.profile)
except (ProfileNotFoundError, ProfileValidationError) as exc:
logger.warning(
"Skipping device %r (uuid=%s): cannot load profile %r: %s",
device.friendly_name,
device.uuid,
device.profile,
exc,
)
continue
# One sensor entity per profile metric.
for metric in profile.metrics:
entity_key = f"modbus.{device.uuid}.{metric.key}"
# Capture device id and metric key for the value_getter closure.
# The getter queries the most recent ModbusReading for this device
# and extracts the metric value from its payload.
_device_id = device.id
_metric_key = metric.key
def _make_value_getter(
dev_id: int, m_key: str
) -> Callable[["Session"], Any]:
"""Return a getter that fetches the latest reading value from DB."""
def _getter(sess: "Session") -> Any:
from app.models.modbus import ModbusReading
from sqlalchemy import desc
reading = (
sess.query(ModbusReading)
.filter(ModbusReading.device_id == dev_id)
.order_by(desc(ModbusReading.recorded_at))
.first()
)
if reading is None:
return None
return reading.payload.get(m_key)
return _getter
entities.append(
ExposableEntity(
key=entity_key,
component=metric.ha_component,
device=device_info,
device_class=metric.device_class or None,
unit=metric.unit,
name=f"{device.friendly_name} {metric.key.replace('_', ' ').title()}",
value_getter=_make_value_getter(_device_id, _metric_key),
state_class=metric.state_class,
)
)
# One binary_sensor entity for device "online" status.
online_key = f"modbus.{device.uuid}.online"
_dev_id_online = device.id
def _make_online_getter(dev_id: int) -> Callable[["Session"], Any]:
"""Return a getter that reports the device online status from DB."""
def _getter(sess: "Session") -> Any:
from app.models.modbus import ModbusDevice
dev = sess.query(ModbusDevice).filter(
ModbusDevice.id == dev_id
).first()
if dev is None:
return None
# Map last_poll_ok to HA binary_sensor payload strings.
if dev.last_poll_ok is True:
return "ON"
return "OFF"
return _getter
entities.append(
ExposableEntity(
key=online_key,
component="binary_sensor",
device=device_info,
device_class="connectivity",
unit="",
name=f"{device.friendly_name} Online",
value_getter=_make_online_getter(_dev_id_online),
state_class=None,
)
)
return entities
# Register the modbus provider at module load time.
register_provider(_modbus_provider)
+1
View File
@@ -0,0 +1 @@
"""Modbus integration package — driver, profile loader, and decoder."""
+203
View File
@@ -0,0 +1,203 @@
"""Modbus TCP driver — thin wrapper around pymodbus.
This module provides a single public function ``read_blocks`` that performs
one or more FC04 (Read Input Registers) block reads against a Modbus TCP
gateway and returns a flat ``dict[register_address -> 16-bit_value]`` map.
Design decisions
----------------
- **Only read function codes** (FC03/FC04) are exposed. There is no write path.
- float32 decoding uses ``struct.unpack('>f', ...)`` directly rather than the
pymodbus ``BinaryPayloadDecoder`` helper, which has had API churn across 3.x
releases. Big-endian word order + big-endian byte order means the 4 raw bytes
are already in standard network (big-endian) order.
- ``ModbusTcpClient`` is used in synchronous mode (``connect()`` / ``close()``).
The client is created fresh per call; this keeps the driver stateless and
avoids threading concerns at the cost of one TCP handshake per read cycle.
APScheduler jobs call this from a thread pool, so stateless is safer.
- pymodbus 3.13.x uses ``device_id=`` as the slave-address keyword argument
(renamed from ``slave=`` in earlier 3.x releases).
Exceptions
----------
``ModbusDriverError``
Base exception for all driver-level errors.
``ModbusConnectionError``
Raised when the TCP connection to the gateway cannot be established or is
lost during the read.
``ModbusResponseError``
Raised when the gateway responds with a Modbus exception frame or when the
returned register count does not match the requested count.
"""
from __future__ import annotations
import logging
import struct
from typing import TYPE_CHECKING
from pymodbus.client import ModbusTcpClient
from pymodbus.exceptions import ConnectionException, ModbusException
if TYPE_CHECKING:
from collections.abc import Sequence
logger = logging.getLogger(__name__)
# ---------------------------------------------------------------------------
# Custom exceptions
# ---------------------------------------------------------------------------
class ModbusDriverError(RuntimeError):
"""Base class for all Modbus driver errors."""
class ModbusConnectionError(ModbusDriverError):
"""Cannot connect to (or lost connection with) the Modbus TCP gateway."""
class ModbusResponseError(ModbusDriverError):
"""Modbus gateway returned an exception frame or an unexpected response."""
# ---------------------------------------------------------------------------
# Block definition type alias
# ---------------------------------------------------------------------------
#: A single read block: ``{"start": int, "count": int}``.
Block = dict[str, int]
# ---------------------------------------------------------------------------
# Public API
# ---------------------------------------------------------------------------
def read_blocks(
host: str,
port: int,
unit_id: int,
blocks: Sequence[Block],
*,
timeout: float = 3.0,
) -> dict[int, int]:
"""Read one or more contiguous register blocks via FC04 (input registers).
Parameters
----------
host:
Hostname or IP address of the Modbus TCP gateway.
port:
TCP port (typically 502).
unit_id:
Modbus slave / unit address (the device's Meter ID for SDM120).
blocks:
Sequence of ``{"start": int, "count": int}`` dicts describing the
contiguous register ranges to read. ``count`` is the number of
16-bit registers (not bytes).
timeout:
TCP connect/read timeout in seconds (default 3 s).
Returns
-------
dict[int, int]
Flat mapping of ``{register_address: 16-bit_register_value}`` for
every register returned across all blocks.
Raises
------
ModbusConnectionError
If the TCP connection to the gateway fails.
ModbusResponseError
If the gateway returns a Modbus exception frame or an unexpected
number of registers.
"""
client = ModbusTcpClient(host, port=port, timeout=timeout)
try:
connected = client.connect()
if not connected:
raise ModbusConnectionError(
f"Could not connect to Modbus gateway at {host}:{port}"
)
registers: dict[int, int] = {}
for block in blocks:
start: int = block["start"]
count: int = block["count"]
_read_block(client, unit_id, start, count, registers)
return registers
except ConnectionException as exc:
raise ModbusConnectionError(
f"Connection to Modbus gateway {host}:{port} failed: {exc}"
) from exc
except ModbusException as exc:
raise ModbusResponseError(f"Modbus protocol error: {exc}") from exc
finally:
client.close()
def _read_block(
client: ModbusTcpClient,
unit_id: int,
start: int,
count: int,
result: dict[int, int],
) -> None:
"""Read one block and merge into *result*. Raises on any error."""
try:
response = client.read_input_registers(start, count=count, device_id=unit_id)
except ConnectionException as exc:
raise ModbusConnectionError(
f"Lost connection while reading registers 0x{start:04X}+{count}: {exc}"
) from exc
if response.isError():
raise ModbusResponseError(
f"Modbus exception reading registers 0x{start:04X}+{count}: {response}"
)
regs = response.registers
if len(regs) != count:
raise ModbusResponseError(
f"Expected {count} registers from 0x{start:04X}, got {len(regs)}"
)
for offset, value in enumerate(regs):
result[start + offset] = value
# ---------------------------------------------------------------------------
# Decoding helpers
# ---------------------------------------------------------------------------
def registers_to_float(hi_reg: int, lo_reg: int) -> float:
"""Decode two 16-bit registers to a big-endian IEEE-754 float32.
The SDM120 (and most Modbus float devices) use big-endian word order:
the high 16-bit register comes first, then the low register. Combined
with big-endian byte order within each register, the raw 4-byte sequence
is standard network-byte-order (big-endian) float32.
Parameters
----------
hi_reg:
The first (higher-address-range, most-significant) 16-bit register.
lo_reg:
The second (lower-address-range, least-significant) 16-bit register.
Returns
-------
float
The decoded IEEE-754 single-precision floating-point value.
Example
-------
>>> registers_to_float(0x4366, 0x3334)
230.20001220703125
"""
raw = struct.pack(">HH", hi_reg, lo_reg)
return struct.unpack(">f", raw)[0]
+231
View File
@@ -0,0 +1,231 @@
"""Modbus profile loader, validator, and decoder.
A *profile* is a YAML file that describes the protocol-level knowledge for
a particular Modbus device model:
- Which registers to read (``blocks`` for efficient bulk reads).
- How to decode each quantity (``metrics``: register address, type, unit, …).
- Home Assistant metadata per metric (``device_class``, ``ha_component``, …).
Profiles live in ``app/integrations/modbus/profiles/<name>.yaml`` and are
located at runtime relative to *this file* (not CWD), so they work whether
the package is installed as a wheel, run in-process, or invoked via
``python -m scripts.modbus_cli``.
Design notes
------------
- **Purely data + functions** — no abstract base classes or inheritance.
- ``ModbusProfile`` and ``MetricSpec`` are Pydantic models: they validate on
construction and raise ``pydantic.ValidationError`` for malformed YAML.
- ``decode`` uses ``driver.registers_to_float`` for float32 quantities.
Unsupported types are silently skipped with a warning (future-proofing for
int16/uint16 etc.).
"""
from __future__ import annotations
import logging
from pathlib import Path
from typing import Optional
import yaml
from pydantic import BaseModel, ValidationError
from app.integrations.modbus.driver import registers_to_float
logger = logging.getLogger(__name__)
# Directory containing the YAML profiles, located relative to *this* file.
_PROFILES_DIR = Path(__file__).parent / "profiles"
# ---------------------------------------------------------------------------
# Pydantic models for profile validation
# ---------------------------------------------------------------------------
class MetricSpec(BaseModel):
"""Specification for a single measurable quantity within a profile."""
key: str
"""Stable identifier used as the key in decoded payloads (e.g. ``"voltage"``)."""
address: int
"""Starting Modbus register address (0-based, e.g. ``0x0000`` for voltage)."""
type: str
"""Encoding type — currently only ``"float32"`` is supported."""
unit: str
"""Physical unit string (e.g. ``"V"``, ``"A"``, ``"kWh"``). Empty string for dimensionless."""
device_class: str
"""Home Assistant device class (e.g. ``"voltage"``, ``"power"``, ``"energy"``)."""
ha_component: str
"""Home Assistant component type (e.g. ``"sensor"``)."""
state_class: Optional[str] = None
"""Home Assistant state class, if applicable (e.g. ``"total_increasing"``)."""
class BlockSpec(BaseModel):
"""A contiguous register block for bulk reading."""
start: int
"""First register address in the block."""
count: int
"""Number of 16-bit registers to read."""
class ModbusProfile(BaseModel):
"""Complete description of a Modbus device's protocol-level knowledge."""
name: str
"""Short identifier matching the YAML file name (e.g. ``"sdm120"``)."""
description: str
"""Human-readable description of the device."""
function_code: int
"""Modbus function code for reading measurements (4 = FC04 input registers)."""
word_order: str
"""Word (register) order: ``"big"`` means high register first."""
byte_order: str
"""Byte order within each register: ``"big"`` means standard network order."""
blocks: list[BlockSpec]
"""Register blocks to read in bulk, reducing Modbus transaction count."""
metrics: list[MetricSpec]
"""List of individual measurable quantities and their decoding rules."""
# ---------------------------------------------------------------------------
# Public API
# ---------------------------------------------------------------------------
class ProfileNotFoundError(FileNotFoundError):
"""Raised when the requested profile YAML file does not exist."""
class ProfileValidationError(ValueError):
"""Raised when a profile YAML file fails Pydantic validation."""
def load_profile(name: str) -> ModbusProfile:
"""Load and validate a Modbus device profile by name.
The profile file is expected at
``app/integrations/modbus/profiles/<name>.yaml``.
Parameters
----------
name:
Profile name without extension (e.g. ``"sdm120"``).
Returns
-------
ModbusProfile
Validated profile model.
Raises
------
ProfileNotFoundError
If ``profiles/<name>.yaml`` does not exist.
ProfileValidationError
If the YAML is syntactically valid but the content fails schema
validation (missing required fields, wrong types, etc.).
"""
path = _PROFILES_DIR / f"{name}.yaml"
if not path.exists():
raise ProfileNotFoundError(
f"Modbus profile '{name}' not found (looked for {path})"
)
with path.open("r", encoding="utf-8") as fh:
raw = yaml.safe_load(fh)
try:
profile = ModbusProfile.model_validate(raw)
except ValidationError as exc:
raise ProfileValidationError(
f"Profile '{name}' failed validation: {exc}"
) from exc
return profile
def list_profiles() -> list[tuple[str, str]]:
"""Return ``(name, description)`` pairs for all available profiles.
Scans ``app/integrations/modbus/profiles/*.yaml``, loads each, and
returns a sorted list of ``(name, description)`` tuples. Profiles that
fail to load are skipped with a warning.
"""
results: list[tuple[str, str]] = []
if not _PROFILES_DIR.exists():
return results
for path in sorted(_PROFILES_DIR.glob("*.yaml")):
name = path.stem
try:
profile = load_profile(name)
results.append((profile.name, profile.description))
except (ProfileNotFoundError, ProfileValidationError, Exception) as exc:
logger.warning("Skipping malformed profile '%s': %s", name, exc)
return results
def decode(profile: ModbusProfile, registers: dict[int, int]) -> dict[str, float]:
"""Decode raw register values into a mapping of ``{key: float_value}``.
For each metric in *profile*, the function looks up the two registers at
``address`` and ``address + 1`` in *registers* and decodes them as a
big-endian float32 (high register first).
Metrics whose registers are absent from *registers* are skipped (not
an error — this can happen if a block was not requested or a device does
not populate that register). Metrics with unsupported ``type`` values
are also skipped.
Parameters
----------
profile:
Validated ``ModbusProfile`` describing the device.
registers:
Mapping of ``{register_address: 16-bit_value}`` as returned by
``driver.read_blocks``.
Returns
-------
dict[str, float]
Decoded engineering values keyed by each metric's ``key`` field.
"""
result: dict[str, float] = {}
for metric in profile.metrics:
if metric.type == "float32":
hi_addr = metric.address
lo_addr = metric.address + 1
if hi_addr not in registers or lo_addr not in registers:
logger.debug(
"Skipping metric '%s': registers 0x%04X/0x%04X not available",
metric.key,
hi_addr,
lo_addr,
)
continue
value = registers_to_float(registers[hi_addr], registers[lo_addr])
result[metric.key] = value
else:
logger.warning(
"Metric '%s' has unsupported type '%s', skipping",
metric.key,
metric.type,
)
return result
@@ -0,0 +1,74 @@
name: sdm120
description: Eastron SDM120 single-phase energy meter
function_code: 4 # input registers (FC04)
word_order: big # high register first (big-endian word order)
byte_order: big # high byte first within each register
blocks:
# Registers 3000130095 (0x00000x005F): voltage through max export demand
- { start: 0x0000, count: 0x0060 }
# Registers 3034330346 (0x01560x0159): total active/reactive energy
- { start: 0x0156, count: 0x0004 }
metrics:
# Core measurements — addresses from SDM120 Modbus Protocol §4 (FC04 input registers)
# Each float32 occupies two consecutive 16-bit registers; address = Hex start in §4.
- key: voltage
address: 0x0000 # register 30001 — Voltage (V)
type: float32
unit: "V"
device_class: voltage
ha_component: sensor
- key: current
address: 0x0006 # register 30007 — Current (A)
type: float32
unit: "A"
device_class: current
ha_component: sensor
- key: active_power
address: 0x000C # register 30013 — Active power (W)
type: float32
unit: "W"
device_class: power
ha_component: sensor
- key: power_factor
address: 0x001E # register 30031 — Power factor (dimensionless)
type: float32
unit: ""
device_class: power_factor
ha_component: sensor
- key: frequency
address: 0x0046 # register 30071 — Frequency (Hz)
type: float32
unit: "Hz"
device_class: frequency
ha_component: sensor
- key: import_energy
address: 0x0048 # register 30073 — Import active energy (kWh)
type: float32
unit: "kWh"
device_class: energy
state_class: total_increasing
ha_component: sensor
- key: export_energy
address: 0x004A # register 30075 — Export active energy (kWh)
type: float32
unit: "kWh"
device_class: energy
state_class: total_increasing
ha_component: sensor
- key: total_energy
address: 0x0156 # register 30343 — Total active energy (kWh)
type: float32
unit: "kWh"
device_class: energy
state_class: total_increasing
ha_component: sensor
+257
View File
@@ -0,0 +1,257 @@
"""MQTT client integration (paho-mqtt 2.x).
Provides :class:`MqttManager`: a long-lived MQTT client that runs paho's
loop in a background thread. Designed to be started in FastAPI's lifespan
and shared across the application as a module-level singleton (``mqtt_manager``).
Key design decisions
--------------------
- **paho 2.x API**: ``Client`` requires ``CallbackAPIVersion.VERSION2`` as the
first positional argument. VERSION2 on_connect callback receives
``(client, userdata, connect_flags, reason_code, properties)`` — not the
older RC int.
- **Background thread**: ``loop_start()`` spawns a daemon thread; ``loop_stop()``
joins it cleanly on shutdown.
- **Never crash the process**: all paho operations are wrapped in try/except;
connection failure is logged but does not propagate to the caller.
- **Password safety**: the MQTT password is never passed to ``logger`` calls.
- **Reconnect**: ``reconnect(settings)`` tears down the old client and
establishes a fresh connection with the new settings. Callers (e.g. PUT
/api/config) call this after saving MQTT-related config values.
"""
from __future__ import annotations
import logging
import threading
from typing import TYPE_CHECKING
import paho.mqtt.client as mqtt
if TYPE_CHECKING:
from app.config import Settings
logger = logging.getLogger(__name__)
# MQTT settings keys that, when changed, should trigger a reconnect.
MQTT_SETTINGS_KEYS = {
"mqtt_enabled",
"mqtt_broker_host",
"mqtt_broker_port",
"mqtt_username",
"mqtt_password",
"mqtt_tls_enabled",
}
def _is_configured(settings: Settings) -> bool:
"""Return True if MQTT is enabled *and* the broker host is set."""
return bool(settings.mqtt_enabled and settings.mqtt_broker_host)
class MqttManager:
"""Long-lived MQTT client wrapper.
Lifecycle
---------
1. ``connect(settings)`` — start background loop + connect to broker.
2. ``publish(...)`` — publish messages while connected.
3. ``disconnect()`` — graceful teardown (called on app shutdown).
4. ``reconnect(settings)`` — disconnect then re-connect with new settings
(called after saving updated MQTT configuration).
When MQTT is not enabled or not configured, all methods are no-ops.
Connection failures are caught and logged; they do not raise.
"""
def __init__(self) -> None:
self._client: mqtt.Client | None = None
self._lock = threading.Lock()
self._connected = False
# ------------------------------------------------------------------
# Public properties
# ------------------------------------------------------------------
@property
def is_connected(self) -> bool:
"""True if the underlying paho client is currently connected."""
return self._connected
# ------------------------------------------------------------------
# Public interface
# ------------------------------------------------------------------
def is_configured(self, settings: Settings) -> bool:
"""Return True if MQTT is enabled and broker host is configured."""
return _is_configured(settings)
def connect(self, settings: Settings) -> None:
"""Connect to the MQTT broker and start the background loop thread.
No-op if MQTT is not enabled or broker host is not set.
Connection errors are logged but do not raise.
"""
if not _is_configured(settings):
logger.debug("MQTT not configured or not enabled — skipping connect.")
return
with self._lock:
self._start_client(settings)
def disconnect(self) -> None:
"""Disconnect from the broker and stop the background loop thread.
No-op if no client is active.
"""
with self._lock:
self._stop_client()
def reconnect(self, settings: Settings) -> None:
"""Disconnect the current client (if any) and reconnect with *settings*.
Call this after saving updated MQTT configuration values.
No-op if MQTT is not enabled or broker host is not set.
"""
with self._lock:
self._stop_client()
self.connect(settings)
def publish(
self,
topic: str,
payload: str | bytes | None,
*,
retain: bool = False,
qos: int = 0,
) -> None:
"""Publish a message to *topic*.
If the client is not connected the call is silently skipped.
Errors are logged but do not raise.
"""
with self._lock:
client = self._client
if client is None or not self._connected:
logger.debug("MQTT publish skipped — not connected (topic=%s).", topic)
return
try:
client.publish(topic, payload=payload, qos=qos, retain=retain)
except Exception:
logger.exception("MQTT publish error (topic=%s).", topic)
# ------------------------------------------------------------------
# Internal helpers — must be called with self._lock held
# ------------------------------------------------------------------
def _start_client(self, settings: Settings) -> None:
"""Build a fresh paho Client, configure it, and call loop_start + connect."""
client = mqtt.Client(
callback_api_version=mqtt.CallbackAPIVersion.VERSION2,
client_id="home-automation",
)
# Callbacks — VERSION2 on_connect signature:
# (client, userdata, connect_flags, reason_code, properties)
def _on_connect(
_client: mqtt.Client,
_userdata: object,
_flags: mqtt.ConnectFlags,
reason_code: mqtt.ReasonCode,
_properties: mqtt.Properties | None,
) -> None:
if reason_code.is_failure:
logger.warning("MQTT connection refused: %s", reason_code)
self._connected = False
else:
logger.info("MQTT connected (reason_code=%s).", reason_code)
self._connected = True
# VERSION2 on_disconnect signature:
# (client, userdata, disconnect_flags, reason_code, properties)
def _on_disconnect(
_client: mqtt.Client,
_userdata: object,
_disconnect_flags: mqtt.DisconnectFlags,
reason_code: mqtt.ReasonCode,
_properties: mqtt.Properties | None,
) -> None:
self._connected = False
if reason_code.is_failure:
logger.warning("MQTT disconnected unexpectedly (reason_code=%s).", reason_code)
else:
logger.info("MQTT disconnected cleanly.")
client.on_connect = _on_connect
client.on_disconnect = _on_disconnect
# TLS
if settings.mqtt_tls_enabled:
try:
client.tls_set()
except Exception:
logger.exception("MQTT TLS setup failed.")
return
# Credentials — never log the password
if settings.mqtt_username:
client.username_pw_set(
username=settings.mqtt_username,
password=settings.mqtt_password or None,
)
# Start background loop thread *before* connect so paho can handle
# the connection handshake asynchronously.
client.loop_start()
try:
client.connect(
host=settings.mqtt_broker_host,
port=settings.mqtt_broker_port,
keepalive=60,
)
except Exception:
# Log without leaking the password
logger.exception(
"MQTT connect failed (host=%s, port=%s). Stopping loop.",
settings.mqtt_broker_host,
settings.mqtt_broker_port,
)
try:
client.loop_stop()
except Exception:
pass
return
self._client = client
logger.info(
"MQTT client started (host=%s, port=%s).",
settings.mqtt_broker_host,
settings.mqtt_broker_port,
)
def _stop_client(self) -> None:
"""Disconnect and stop the background loop. Idempotent."""
client = self._client
if client is None:
return
self._client = None
self._connected = False
try:
client.disconnect()
except Exception:
logger.debug("MQTT disconnect raised (ignoring).", exc_info=True)
try:
client.loop_stop()
except Exception:
logger.debug("MQTT loop_stop raised (ignoring).", exc_info=True)
logger.info("MQTT client stopped.")
# ---------------------------------------------------------------------------
# Module-level singleton — shared across lifespan and route handlers
# ---------------------------------------------------------------------------
mqtt_manager = MqttManager()
+144 -38
View File
@@ -1,32 +1,51 @@
import logging
import os
from contextlib import asynccontextmanager
from pathlib import Path
from fastapi import FastAPI
from fastapi import FastAPI, HTTPException, Request
from fastapi.responses import FileResponse
from fastapi.staticfiles import StaticFiles
from apscheduler.schedulers.background import BackgroundScheduler
from apscheduler.triggers.interval import IntervalTrigger
from sqlalchemy.orm import Session
from app import models # noqa: F401
from app.api.routes.auth import router as auth_router
from app.api.routes import pages, status
import app.auth_db as auth_db
from app.api.routes.api.config import router as api_config_router
from app.api.routes.api.data import router as api_data_router
from app.api.routes.api.expose import router as api_expose_router
from app.api.routes.api.modbus import router as api_modbus_router
from app.api.routes.api.session import router as api_session_router
from app.api.routes import status
from app.db import get_session_local
from app.api.routes.homeassistant import router as homeassistant_router
from app.api.routes.location import router as location_router
from app.api.routes.poo import router as poo_router
from app.api.routes.public_ip import router as public_ip_router
from app.api.routes.ticktick import router as ticktick_router
from app.config import get_settings
from app.integrations.mqtt import mqtt_manager
from app.services.auth import AuthBootstrapError, initialize_auth_schema
from app.services.config_page import seed_missing_config_from_bootstrap, sync_app_hostname_from_bootstrap
from app.services.config_page import build_runtime_settings, seed_missing_config_from_bootstrap, sync_app_hostname_from_bootstrap
from app.services.public_ip import check_public_ipv4_and_notify
from app.services.modbus_poll import poll_all_enabled_devices, BASE_POLL_TICK_SECONDS
from app.services.ha_discovery import publish_discovery, publish_states
from scripts.app_db_adopt import AppDatabaseAdoptionError, validate_app_runtime_db
from scripts.location_db_adopt import LocationDatabaseAdoptionError, validate_location_runtime_db
from scripts.poo_db_adopt import PooDatabaseAdoptionError, validate_poo_runtime_db
logger = logging.getLogger(__name__)
_REPO_ROOT = Path(__file__).resolve().parents[1]
def _get_spa_dist_dir() -> Path:
env_val = os.environ.get("SPA_DIST_DIR")
if env_val:
return Path(env_val)
return _REPO_ROOT / "frontend" / "dist"
def _run_scheduled_public_ip_check() -> None:
session_local = auth_db.get_auth_session_local()
session_local = get_session_local()
session: Session = session_local()
try:
check_public_ipv4_and_notify(session, bootstrap_settings=get_settings())
@@ -34,8 +53,50 @@ def _run_scheduled_public_ip_check() -> None:
session.close()
def _run_scheduled_modbus_poll() -> None:
session_local = get_session_local()
session: Session = session_local()
try:
poll_all_enabled_devices(session, bootstrap_settings=get_settings())
finally:
session.close()
def _run_scheduled_ha_state_publish() -> None:
"""Periodic job: publish discovery configs + state + availability for all enabled exposed entities.
Runs every 60 seconds. When the MQTT broker is connected:
- Publishes (or re-publishes) all HA Discovery configs (retained, idempotent).
- Publishes the current state / availability for all enabled entities.
This also serves as the reliable "publish discovery after connect" mechanism:
because paho connects asynchronously, a synchronous call immediately after
``mqtt_manager.connect()`` would fire before the TCP handshake completes and
be a no-op. Instead, this periodic job picks it up within 60 seconds of the
broker becoming available — retained payloads make repeated publishes harmless.
Additionally, if MQTT is configured in DB but the manager is not yet connected
(e.g. MQTT was enabled via UI after startup), this job attempts to connect so
the user does not need to restart the server.
"""
session_local = get_session_local()
session: Session = session_local()
try:
runtime_settings = build_runtime_settings(session, get_settings())
# Reconnect if MQTT is configured (in DB) but not yet connected.
if mqtt_manager.is_configured(runtime_settings) and not mqtt_manager.is_connected:
logger.info("_run_scheduled_ha_state_publish: MQTT configured but not connected — attempting connect.")
mqtt_manager.connect(runtime_settings)
publish_discovery(session)
publish_states(session)
except Exception:
logger.exception("_run_scheduled_ha_state_publish: unexpected error")
finally:
session.close()
def ensure_auth_db_ready() -> None:
session_local = auth_db.get_auth_session_local()
session_local = get_session_local()
session: Session = session_local()
try:
validate_app_runtime_db(get_settings().app_database_url)
@@ -50,41 +111,16 @@ def ensure_auth_db_ready() -> None:
session.close()
def ensure_location_db_ready() -> None:
settings = get_settings()
if settings.location_sqlite_path is None:
return
try:
validate_location_runtime_db(settings.location_database_url)
except LocationDatabaseAdoptionError as exc:
raise RuntimeError(str(exc)) from exc
def ensure_poo_db_ready() -> None:
settings = get_settings()
if settings.poo_sqlite_path is None:
return
try:
validate_poo_runtime_db(settings.poo_database_url)
except PooDatabaseAdoptionError as exc:
raise RuntimeError(str(exc)) from exc
def ensure_runtime_dirs() -> None:
settings = get_settings()
for path in (settings.app_sqlite_path, settings.location_sqlite_path, settings.poo_sqlite_path):
if path is not None:
path.parent.mkdir(parents=True, exist_ok=True)
if settings.app_sqlite_path is not None:
settings.app_sqlite_path.parent.mkdir(parents=True, exist_ok=True)
@asynccontextmanager
async def lifespan(_: FastAPI):
ensure_runtime_dirs()
ensure_auth_db_ready()
ensure_location_db_ready()
ensure_poo_db_ready()
scheduler = BackgroundScheduler(timezone="UTC")
scheduler.add_job(
_run_scheduled_public_ip_check,
@@ -94,8 +130,42 @@ async def lifespan(_: FastAPI):
max_instances=1,
coalesce=True,
)
scheduler.add_job(
_run_scheduled_modbus_poll,
trigger=IntervalTrigger(seconds=BASE_POLL_TICK_SECONDS),
id="modbus-poll",
replace_existing=True,
max_instances=1,
coalesce=True,
)
# Periodic HA state / availability publish (60-second fallback sweep).
scheduler.add_job(
_run_scheduled_ha_state_publish,
trigger=IntervalTrigger(seconds=60),
id="ha-state-publish",
replace_existing=True,
max_instances=1,
coalesce=True,
)
scheduler.start()
# MQTT: connect using DB-merged runtime settings so broker configured via UI
# is picked up on restart (not just from env/bootstrap settings).
# Discovery will be published by the first run of _run_scheduled_ha_state_publish
# (within 60 s of startup), after the async paho handshake completes.
_startup_session_local = get_session_local()
_startup_session: Session = _startup_session_local()
try:
_startup_runtime_settings = build_runtime_settings(_startup_session, get_settings())
finally:
_startup_session.close()
mqtt_manager.connect(_startup_runtime_settings)
yield
# MQTT: clean shutdown before the process exits.
mqtt_manager.disconnect()
scheduler.shutdown(wait=False)
@@ -116,13 +186,49 @@ def create_app() -> FastAPI:
app.mount("/static", StaticFiles(directory=static_dir), name="static")
app.include_router(status.router)
app.include_router(auth_router)
app.include_router(pages.router)
app.include_router(api_config_router)
app.include_router(api_data_router)
app.include_router(api_expose_router)
app.include_router(api_modbus_router)
app.include_router(api_session_router)
app.include_router(homeassistant_router)
app.include_router(location_router)
app.include_router(poo_router)
app.include_router(public_ip_router)
app.include_router(ticktick_router)
# SPA hosting: mount frontend/dist if it exists and has index.html.
# If the SPA dist is absent (e.g. backend-only CI), skip SPA serving entirely
# so that pytest stays green with only the API routes registered.
spa_dist = _get_spa_dist_dir()
spa_index = spa_dist / "index.html"
if spa_dist.is_dir() and spa_index.is_file():
spa_assets = spa_dist / "assets"
if spa_assets.is_dir():
app.mount("/assets", StaticFiles(directory=spa_assets), name="spa-assets")
# Resolve the dist root once so the containment check is fast and consistent.
_spa_root = spa_dist.resolve()
@app.get("/{full_path:path}", include_in_schema=False)
async def spa_fallback(full_path: str, request: Request) -> FileResponse: # noqa: RUF029
# Explicit 404 for unmatched /api/* — never return index.html for API paths.
if full_path.startswith("api/"):
raise HTTPException(status_code=404, detail="not found")
# Resolve candidate to an absolute path and verify it stays within the SPA
# dist root. Without this check, URL-encoded ".." sequences (e.g. "..%2f")
# bypass Starlette's path parameter handling and allow arbitrary file reads.
candidate = (spa_dist / full_path).resolve()
if candidate.is_file() and candidate.is_relative_to(_spa_root):
return FileResponse(candidate)
# For any path outside the dist root, or for SPA client routes that don't
# correspond to a real file, return index.html so the SPA router handles it.
return FileResponse(spa_index)
else:
logger.warning(
"SPA dist not found at %s — SPA hosting disabled (API-only mode).", spa_dist
)
return app
+2
View File
@@ -3,6 +3,7 @@
from app.models.auth import AuthSession, AuthUser
from app.models.config import AppConfigEntry
from app.models.location import Location
from app.models.poo import PooRecord
from app.models.public_ip import PublicIPHistory, PublicIPState
__all__ = [
@@ -10,6 +11,7 @@ __all__ = [
"AuthSession",
"AuthUser",
"Location",
"PooRecord",
"PublicIPHistory",
"PublicIPState",
]
+25 -3
View File
@@ -3,10 +3,10 @@ from datetime import datetime
from sqlalchemy import Boolean, DateTime, ForeignKey, Integer, String
from sqlalchemy.orm import Mapped, mapped_column, relationship
from app.auth_db import AuthBase
from app.db import Base
class AuthUser(AuthBase):
class AuthUser(Base):
__tablename__ = "auth_users"
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
@@ -15,11 +15,15 @@ class AuthUser(AuthBase):
is_active: Mapped[bool] = mapped_column(Boolean, nullable=False, default=True)
force_password_change: Mapped[bool] = mapped_column(Boolean, nullable=False, default=True)
created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), nullable=False)
# TOTP fields (Phase B, M4-T04) — nullable/false by default so existing users are unaffected
totp_secret: Mapped[str | None] = mapped_column(String(64), nullable=True)
totp_enabled: Mapped[bool] = mapped_column(Boolean, nullable=False, default=False)
sessions: Mapped[list["AuthSession"]] = relationship(back_populates="user")
recovery_codes: Mapped[list["RecoveryCode"]] = relationship(back_populates="user")
class AuthSession(AuthBase):
class AuthSession(Base):
__tablename__ = "auth_sessions"
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
@@ -31,3 +35,21 @@ class AuthSession(AuthBase):
revoked_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)
user: Mapped[AuthUser] = relationship(back_populates="sessions")
class RecoveryCode(Base):
"""One-time TOTP recovery codes for AuthUser.
``code_hash`` stores the Argon2 hash of the plaintext code (plaintext is
returned only at setup time and never stored). ``used_at`` is NULL while
the code is still valid; set to the consumption timestamp when consumed.
"""
__tablename__ = "auth_recovery_code"
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
user_id: Mapped[int] = mapped_column(ForeignKey("auth_users.id"), nullable=False, index=True)
code_hash: Mapped[str] = mapped_column(String(255), nullable=False)
used_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)
user: Mapped[AuthUser] = relationship(back_populates="recovery_codes")
+31
View File
@@ -0,0 +1,31 @@
from datetime import datetime
from sqlalchemy import DateTime, Index, Integer, String, UniqueConstraint
from sqlalchemy.orm import Mapped, mapped_column
from app.db import Base
class LoginThrottle(Base):
"""Tracks per-key login failure state for exponential back-off throttling.
``scope`` is either ``'ip'`` or ``'user'``; ``key`` is the IP address or
username string respectively. The pair ``(scope, key)`` is unique — one
row per tracked entity. A row is deleted on successful login (clear).
"""
__tablename__ = "auth_login_throttle"
__table_args__ = (
UniqueConstraint("scope", "key", name="uq_auth_login_throttle_scope_key"),
Index("ix_auth_login_throttle_scope_key", "scope", "key"),
)
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
key: Mapped[str] = mapped_column(String(255), nullable=False)
scope: Mapped[str] = mapped_column(String(16), nullable=False) # 'ip' | 'user'
failures: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
first_failed_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), nullable=False)
last_failed_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), nullable=False)
next_allowed_at: Mapped[datetime | None] = mapped_column(
DateTime(timezone=True), nullable=True
)
-4
View File
@@ -1,4 +0,0 @@
from app.db import Base
__all__ = ["Base"]
+2 -2
View File
@@ -3,10 +3,10 @@ from datetime import datetime
from sqlalchemy import DateTime, Integer, String
from sqlalchemy.orm import Mapped, mapped_column
from app.auth_db import AuthBase
from app.db import Base
class AppConfigEntry(AuthBase):
class AppConfigEntry(Base):
__tablename__ = "app_config"
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
+49
View File
@@ -0,0 +1,49 @@
"""SQLAlchemy model for the exposed-entity toggle table.
``ExposedEntityToggle`` stores the per-entity enable/disable state for the
MQTT / HA Discovery expose framework. The *catalog* of what *can* be
exposed is computed dynamically by provider functions (see
``app/integrations/expose.py``); this table only records which entries the
user has explicitly enabled.
Key design decisions
--------------------
- ``key`` is a stable string identifier derived from device uuid + metric key
(e.g. ``"modbus.<uuid>.voltage"``), so it does not drift if rows are
deleted and re-inserted.
- Default state for an entity not yet in this table is **disabled** (false).
``build_catalog`` in expose.py treats a missing row as ``enabled=False``.
- Only one row per entity key (``key`` is UNIQUE).
"""
from __future__ import annotations
from datetime import datetime
from sqlalchemy import Boolean, DateTime, Integer, String
from sqlalchemy.orm import Mapped, mapped_column
from app.db import Base
class ExposedEntityToggle(Base):
"""Per-entity on/off switch for MQTT / HA Discovery publishing.
Rows are created on demand (first toggle); entities with no row are
treated as disabled.
"""
__tablename__ = "exposed_entity_toggle"
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
# Stable entity key, e.g. "modbus.<uuid>.voltage" or "modbus.<uuid>.online".
key: Mapped[str] = mapped_column(String(255), unique=True, nullable=False)
# Whether this entity should be published via MQTT / HA Discovery.
enabled: Mapped[bool] = mapped_column(Boolean, nullable=False, default=False)
# Last time this row was created or modified.
updated_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False
)
+114
View File
@@ -0,0 +1,114 @@
"""SQLAlchemy models for Modbus device management and telemetry.
Two tables:
- modbus_device: deployment/configurable metadata for each polled device.
- modbus_reading: generic telemetry rows (one per device per poll cycle).
"""
from __future__ import annotations
import uuid as _uuid
from datetime import datetime
from sqlalchemy import Boolean, DateTime, ForeignKey, Index, Integer, String
from sqlalchemy.orm import Mapped, mapped_column, relationship
from sqlalchemy.types import JSON
from app.db import Base
def _uuid4_str() -> str:
return str(_uuid.uuid4())
class ModbusDevice(Base):
"""Deployment-layer record for a Modbus slave device reachable via a TCP gateway.
Protocol knowledge (register map, decoding rules) lives in the YAML profile
referenced by ``profile``. Per-device variables (network address, slave ID,
display name, poll rate) live here.
"""
__tablename__ = "modbus_device"
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
# Stable internal identity — used as the API path key and HA Discovery unique_id anchor.
uuid: Mapped[str] = mapped_column(
String(36), unique=True, nullable=False, default=_uuid4_str
)
# Human-readable label (may be changed; changing it triggers re-publish of HA Discovery).
friendly_name: Mapped[str] = mapped_column(String(255), nullable=False)
# Transport protocol — only "tcp" is supported for now.
transport: Mapped[str] = mapped_column(String(16), nullable=False, default="tcp")
# TCP gateway address.
host: Mapped[str] = mapped_column(String(255), nullable=False)
port: Mapped[int] = mapped_column(Integer, nullable=False, default=502)
# Modbus slave address (set on the device panel; different meters on the same gateway
# must have distinct unit_ids).
unit_id: Mapped[int] = mapped_column(Integer, nullable=False, default=1)
# Which YAML profile to use for register mapping and decoding (e.g. "sdm120").
profile: Mapped[str] = mapped_column(String(64), nullable=False)
# Polling interval in seconds.
poll_interval_s: Mapped[int] = mapped_column(Integer, nullable=False, default=5)
# Whether this device is included in the periodic poll sweep.
enabled: Mapped[bool] = mapped_column(Boolean, nullable=False, default=True)
# Poll-status fields (updated by the poll service after each attempt).
last_poll_at: Mapped[datetime | None] = mapped_column(
DateTime(timezone=True), nullable=True
)
last_poll_ok: Mapped[bool | None] = mapped_column(Boolean, nullable=True)
# Audit timestamps.
created_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False
)
updated_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False
)
# Relationship to readings (back-reference; not loaded eagerly).
readings: Mapped[list["ModbusReading"]] = relationship(
back_populates="device", cascade="save-update, merge"
)
class ModbusReading(Base):
"""Generic telemetry row: one device, one poll instant, all decoded metrics as JSON.
``payload`` contains the full dict of engineering values produced by the YAML profile
decoder, e.g. ``{"voltage": 230.2, "current": 1.3, ...}``. The profile is the key
to interpreting which keys exist and what units they carry.
"""
__tablename__ = "modbus_reading"
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
# FK to the device that produced this reading.
# ON DELETE RESTRICT: prevents accidental deletion of a device that has historical data.
device_id: Mapped[int] = mapped_column(
ForeignKey("modbus_device.id", ondelete="RESTRICT"), nullable=False
)
# The UTC timestamp of when the sample was taken — real indexed column, NOT embedded in payload.
recorded_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False, index=True
)
# Profile-decoded engineering values as a JSON object.
payload: Mapped[dict] = mapped_column(JSON, nullable=False)
device: Mapped["ModbusDevice"] = relationship(back_populates="readings")
# Composite index for efficient time-range queries scoped to a single device.
Index("ix_modbus_reading_device_recorded", ModbusReading.device_id, ModbusReading.recorded_at)
+2 -2
View File
@@ -1,10 +1,10 @@
from sqlalchemy import Float, String
from sqlalchemy.orm import Mapped, mapped_column
from app.poo_db import PooBase
from app.db import Base
class PooRecord(PooBase):
class PooRecord(Base):
__tablename__ = "poo_records"
timestamp: Mapped[str] = mapped_column(String, primary_key=True)
+3 -3
View File
@@ -3,10 +3,10 @@ from datetime import datetime
from sqlalchemy import DateTime, Integer, String
from sqlalchemy.orm import Mapped, mapped_column
from app.auth_db import AuthBase
from app.db import Base
class PublicIPState(AuthBase):
class PublicIPState(Base):
__tablename__ = "public_ip_state"
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
@@ -20,7 +20,7 @@ class PublicIPState(AuthBase):
last_provider: Mapped[str | None] = mapped_column(String(64), nullable=True)
class PublicIPHistory(AuthBase):
class PublicIPHistory(Base):
__tablename__ = "public_ip_history"
id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
-28
View File
@@ -1,28 +0,0 @@
from collections.abc import Generator
from sqlalchemy import create_engine
from sqlalchemy.orm import DeclarativeBase, Session, sessionmaker
from app.config import get_settings
class PooBase(DeclarativeBase):
pass
settings = get_settings()
connect_args: dict[str, object] = {}
if settings.poo_database_url.startswith("sqlite"):
connect_args["check_same_thread"] = False
poo_engine = create_engine(settings.poo_database_url, connect_args=connect_args)
PooSessionLocal = sessionmaker(bind=poo_engine, autoflush=False, autocommit=False, class_=Session)
def get_poo_db_session() -> Generator[Session, None, None]:
session = PooSessionLocal()
try:
yield session
finally:
session.close()
+47
View File
@@ -0,0 +1,47 @@
from __future__ import annotations
from typing import Literal
from pydantic import BaseModel
class ConfigField(BaseModel):
env_name: str
label: str
value: str
secret: bool
input_type: str
configured: bool
class ConfigSection(BaseModel):
name: str
fields: list[ConfigField]
class ConfigResponse(BaseModel):
sections: list[ConfigSection]
class ConfigUpdateRequest(BaseModel):
"""Flat mapping of env_name → value, mirroring the existing form semantics."""
updates: dict[str, str]
class ConfigUpdateResponse(BaseModel):
sections: list[ConfigSection]
class SmtpTestResponse(BaseModel):
"""Response from POST /api/config/smtp/test."""
result: Literal["success", "config-error", "failed"]
message: str
class MqttTestResponse(BaseModel):
"""Response from POST /api/config/mqtt/test."""
result: Literal["success", "config-error", "failed"]
message: str
+92
View File
@@ -0,0 +1,92 @@
from __future__ import annotations
from datetime import datetime
from pydantic import BaseModel
# ---------------------------------------------------------------------------
# Location
# ---------------------------------------------------------------------------
class LocationRecord(BaseModel):
person: str
datetime: str
latitude: float
longitude: float
altitude: float | None
class LocationsResponse(BaseModel):
items: list[LocationRecord]
limit: int
offset: int
class LocationUpdateRequest(BaseModel):
"""PATCH body for a location record — all fields optional; PK fields excluded."""
latitude: float | None = None
longitude: float | None = None
altitude: float | None = None
# ---------------------------------------------------------------------------
# Poo
# ---------------------------------------------------------------------------
class PooRecord(BaseModel):
timestamp: str
status: str
latitude: float
longitude: float
class PooResponse(BaseModel):
items: list[PooRecord]
limit: int
offset: int
class PooUpdateRequest(BaseModel):
"""PATCH body for a poo record — all fields optional; PK field excluded."""
status: str | None = None
latitude: float | None = None
longitude: float | None = None
# ---------------------------------------------------------------------------
# Public IP
# ---------------------------------------------------------------------------
class PublicIPStateSchema(BaseModel):
id: int
current_ipv4: str
previous_ipv4: str | None
first_seen_at: datetime
last_checked_at: datetime
last_changed_at: datetime | None
last_check_status: str
last_check_error: str | None
last_provider: str | None
model_config = {"from_attributes": True}
class PublicIPHistorySchema(BaseModel):
id: int
ipv4: str
observed_at: datetime
change_type: str
provider: str | None
model_config = {"from_attributes": True}
class PublicIPResponse(BaseModel):
state: PublicIPStateSchema | None
history: list[PublicIPHistorySchema]
+95
View File
@@ -0,0 +1,95 @@
"""Pydantic schemas for the Expose API (M5-T12).
Three endpoints:
GET /api/expose — catalog + toggle state + MQTT/Discovery status
PUT /api/expose — set toggles (map key → bool)
POST /api/expose/republish — trigger discovery re-publish
"""
from __future__ import annotations
from pydantic import BaseModel
# ---------------------------------------------------------------------------
# Nested schemas
# ---------------------------------------------------------------------------
class DeviceInfoSchema(BaseModel):
"""HA device grouping info for an exposable entity."""
identifiers: list[str]
name: str
class ExposableEntitySchema(BaseModel):
"""One exposable entity in the catalog.
``value_getter`` is intentionally excluded — it is a non-serialisable
callable and is only used internally by the HA Discovery service.
"""
key: str
component: str
device: DeviceInfoSchema
device_class: str | None
unit: str
name: str
state_class: str | None = None
class CatalogEntrySchema(BaseModel):
"""An entity from the catalog with its current toggle state."""
entity: ExposableEntitySchema
enabled: bool
class MqttStatusSchema(BaseModel):
"""Connection status for MQTT and HA Discovery."""
mqtt_configured: bool
mqtt_connected: bool
discovery_enabled: bool
# ---------------------------------------------------------------------------
# Response schemas
# ---------------------------------------------------------------------------
class ExposeResponse(BaseModel):
"""Response for GET /api/expose."""
catalog: list[CatalogEntrySchema]
mqtt_status: MqttStatusSchema
class ExposeUpdateResponse(BaseModel):
"""Response for PUT /api/expose (returns updated catalog + status)."""
catalog: list[CatalogEntrySchema]
mqtt_status: MqttStatusSchema
class RepublishResponse(BaseModel):
"""Response for POST /api/expose/republish."""
ok: bool
message: str
# ---------------------------------------------------------------------------
# Request schemas
# ---------------------------------------------------------------------------
class ExposeUpdateRequest(BaseModel):
"""Request body for PUT /api/expose.
``toggles`` is a map from entity key to desired enabled state (bool).
Only keys present in the map are updated; absent keys are untouched.
"""
toggles: dict[str, bool]
+154
View File
@@ -0,0 +1,154 @@
"""Pydantic schemas for the Modbus device CRUD + readings + metrics API (M5-T05)."""
from __future__ import annotations
from datetime import datetime
from typing import Any
from pydantic import BaseModel, Field
# ---------------------------------------------------------------------------
# Device schemas
# ---------------------------------------------------------------------------
class ModbusDeviceCreate(BaseModel):
"""Request body for POST /api/modbus/devices."""
friendly_name: str = Field(..., min_length=1, max_length=255)
transport: str = Field(default="tcp", max_length=16)
host: str = Field(..., min_length=1, max_length=255)
port: int = Field(default=502, ge=1, le=65535)
unit_id: int = Field(default=1, ge=0, le=247)
profile: str = Field(..., min_length=1, max_length=64)
poll_interval_s: int = Field(default=5, ge=1, le=3600)
enabled: bool = Field(default=True)
class ModbusDeviceUpdate(BaseModel):
"""Request body for PATCH /api/modbus/devices/{uuid} — all fields optional."""
friendly_name: str | None = Field(default=None, min_length=1, max_length=255)
transport: str | None = Field(default=None, max_length=16)
host: str | None = Field(default=None, min_length=1, max_length=255)
port: int | None = Field(default=None, ge=1, le=65535)
unit_id: int | None = Field(default=None, ge=0, le=247)
profile: str | None = Field(default=None, min_length=1, max_length=64)
poll_interval_s: int | None = Field(default=None, ge=1, le=3600)
enabled: bool | None = None
class ModbusDeviceResponse(BaseModel):
"""Response schema for a single Modbus device."""
uuid: str
friendly_name: str
transport: str
host: str
port: int
unit_id: int
profile: str
poll_interval_s: int
enabled: bool
last_poll_at: datetime | None
last_poll_ok: bool | None
created_at: datetime
updated_at: datetime
model_config = {"from_attributes": True}
class ModbusDeviceListResponse(BaseModel):
"""Response schema for listing Modbus devices."""
items: list[ModbusDeviceResponse]
total: int
# ---------------------------------------------------------------------------
# Reading schemas
# ---------------------------------------------------------------------------
class ModbusReadingResponse(BaseModel):
"""A single reading row: timestamp + decoded payload."""
recorded_at: datetime
payload: dict[str, Any]
model_config = {"from_attributes": True}
class ModbusReadingsResponse(BaseModel):
"""Response schema for the readings time-range endpoint."""
items: list[ModbusReadingResponse]
class ModbusLatestResponse(BaseModel):
"""Response for the /latest endpoint.
``found`` is False and ``recorded_at``/``payload`` are None when the device
has no readings yet. Callers should check ``found`` before using the values.
"""
found: bool
recorded_at: datetime | None
payload: dict[str, Any] | None
# ---------------------------------------------------------------------------
# Metrics / profile schemas
# ---------------------------------------------------------------------------
class MetricInfo(BaseModel):
"""Metadata for a single measurable quantity in a device's profile."""
key: str
label: str
unit: str
device_class: str
class ModbusMetricsResponse(BaseModel):
"""Response schema for GET /api/modbus/devices/{uuid}/metrics."""
profile: str
metrics: list[MetricInfo]
# ---------------------------------------------------------------------------
# Profile list schema
# ---------------------------------------------------------------------------
class ProfileSummary(BaseModel):
"""One entry in the GET /api/modbus/profiles response."""
name: str
description: str
class ModbusProfilesResponse(BaseModel):
"""Response schema for GET /api/modbus/profiles."""
profiles: list[ProfileSummary]
# ---------------------------------------------------------------------------
# Test-read schema
# ---------------------------------------------------------------------------
class ModbusTestReadResponse(BaseModel):
"""Response for POST /api/modbus/devices/{uuid}/test.
On success ``ok=True`` and ``payload`` contains the decoded values.
On failure ``ok=False`` and ``error`` describes the problem.
"""
ok: bool
payload: dict[str, Any] | None = None
error: str | None = None
+25
View File
@@ -0,0 +1,25 @@
from __future__ import annotations
from pydantic import BaseModel
class SessionUser(BaseModel):
username: str
force_password_change: bool
class SessionResponse(BaseModel):
user: SessionUser
csrf_token: str
class LoginRequest(BaseModel):
username: str
password: str
totp_code: str | None = None
class PasswordChangeRequest(BaseModel):
current_password: str
new_password: str
confirm_password: str
+59
View File
@@ -0,0 +1,59 @@
"""Pydantic schemas for TOTP setup / enable / disable / status endpoints (M4-T05)."""
from __future__ import annotations
from pydantic import BaseModel
# ---------------------------------------------------------------------------
# Setup (POST /api/auth/totp/setup) — response
# ---------------------------------------------------------------------------
class TotpSetupResponse(BaseModel):
"""Returned once after a setup call.
``secret`` and ``recovery_codes`` are **one-time plaintext values**.
They are never returned again by any subsequent API call.
The frontend must display and instruct the user to save them before confirming.
"""
secret: str
otpauth_uri: str
recovery_codes: list[str]
# ---------------------------------------------------------------------------
# Enable (POST /api/auth/totp/enable) — request
# ---------------------------------------------------------------------------
class TotpEnableRequest(BaseModel):
"""The user confirms setup by providing the 6-digit TOTP code."""
code: str
# ---------------------------------------------------------------------------
# Disable (POST /api/auth/totp/disable) — request
# ---------------------------------------------------------------------------
class TotpDisableRequest(BaseModel):
"""Disable TOTP by proving identity.
Exactly one of ``password`` or ``code`` must be provided.
"""
password: str | None = None
code: str | None = None
# ---------------------------------------------------------------------------
# Status (GET /api/auth/totp) — response
# ---------------------------------------------------------------------------
class TotpStatusResponse(BaseModel):
"""Minimal status response — never exposes secret or recovery codes."""
enabled: bool
+45 -9
View File
@@ -7,7 +7,7 @@ from typing import Any
from sqlalchemy import select
from sqlalchemy.orm import Session
from app.auth_db import reset_auth_db_caches
from app.db import reset_db_caches
from app.config import Settings, get_settings
from app.models.config import AppConfigEntry
@@ -25,9 +25,9 @@ class ConfigField:
CONFIG_FIELDS: tuple[ConfigField, ...] = (
ConfigField("System", "APP_NAME", "app_name", "App Name"),
ConfigField("System", "APP_ENV", "app_env", "App Env"),
ConfigField("System", "APP_DEBUG", "app_debug", "App Debug"),
ConfigField("System", "APP_DEBUG", "app_debug", "App Debug", input_type="checkbox"),
ConfigField("System", "APP_HOSTNAME", "app_hostname", "App Hostname"),
ConfigField("SMTP", "SMTP_ENABLED", "smtp_enabled", "SMTP Enabled"),
ConfigField("SMTP", "SMTP_ENABLED", "smtp_enabled", "SMTP Enabled", input_type="checkbox"),
ConfigField("SMTP", "SMTP_HOST", "smtp_host", "SMTP Host"),
ConfigField("SMTP", "SMTP_PORT", "smtp_port", "SMTP Port"),
ConfigField("SMTP", "SMTP_USERNAME", "smtp_username", "SMTP Username"),
@@ -35,7 +35,7 @@ CONFIG_FIELDS: tuple[ConfigField, ...] = (
ConfigField("SMTP", "SMTP_FROM_NAME", "smtp_from_name", "SMTP From Name"),
ConfigField("SMTP", "SMTP_FROM_ADDRESS", "smtp_from_address", "SMTP From Address"),
ConfigField("SMTP", "SMTP_TO_ADDRESS", "smtp_to_address", "SMTP To Address"),
ConfigField("SMTP", "SMTP_USE_STARTTLS", "smtp_use_starttls", "SMTP Use STARTTLS"),
ConfigField("SMTP", "SMTP_USE_STARTTLS", "smtp_use_starttls", "SMTP Use STARTTLS", input_type="checkbox"),
ConfigField(
"Authentication",
"AUTH_SESSION_COOKIE_NAME",
@@ -49,6 +49,13 @@ CONFIG_FIELDS: tuple[ConfigField, ...] = (
"auth_cookie_secure_override",
"Cookie Secure Override",
),
ConfigField(
"Authentication",
"AUTH_LOGIN_THROTTLE_ENABLED",
"auth_login_throttle_enabled",
"Login Throttle Enabled",
input_type="checkbox",
),
ConfigField("Poo", "POO_WEBHOOK_ID", "poo_webhook_id", "Poo Webhook ID", secret=True),
ConfigField(
"Poo",
@@ -96,6 +103,26 @@ CONFIG_FIELDS: tuple[ConfigField, ...] = (
"home_assistant_action_task_project_id",
"Home Assistant Action Task Project ID",
),
ConfigField("MQTT", "MQTT_ENABLED", "mqtt_enabled", "MQTT Enabled", input_type="checkbox"),
ConfigField("MQTT", "MQTT_BROKER_HOST", "mqtt_broker_host", "MQTT Broker Host"),
ConfigField("MQTT", "MQTT_BROKER_PORT", "mqtt_broker_port", "MQTT Broker Port", input_type="number"),
ConfigField("MQTT", "MQTT_USERNAME", "mqtt_username", "MQTT Username"),
ConfigField("MQTT", "MQTT_PASSWORD", "mqtt_password", "MQTT Password", secret=True),
ConfigField("MQTT", "MQTT_TLS_ENABLED", "mqtt_tls_enabled", "MQTT TLS Enabled", input_type="checkbox"),
ConfigField(
"Home Assistant Discovery",
"HA_DISCOVERY_ENABLED",
"ha_discovery_enabled",
"HA Discovery Enabled",
input_type="checkbox",
),
ConfigField(
"Home Assistant Discovery",
"HA_DISCOVERY_PREFIX",
"ha_discovery_prefix",
"HA Discovery Prefix",
),
ConfigField("Modbus", "MODBUS_POLLING_ENABLED", "modbus_polling_enabled", "Modbus Polling Enabled", input_type="checkbox"),
)
@@ -127,7 +154,7 @@ def sync_app_hostname_from_bootstrap(session: Session, bootstrap_settings: Setti
current_values["APP_HOSTNAME"] = bootstrap_hostname
_persist_config_values(session, current_values)
get_settings.cache_clear()
reset_auth_db_caches()
reset_db_caches()
def build_runtime_settings(session: Session, bootstrap_settings: Settings) -> Settings:
@@ -184,7 +211,7 @@ def save_config_updates(session: Session, form_data: dict[str, str], bootstrap_s
_validate_config_values(merged_values, bootstrap_settings)
_persist_config_values(session, merged_values)
get_settings.cache_clear()
reset_auth_db_caches()
reset_db_caches()
def save_config_value(
@@ -199,7 +226,7 @@ def save_config_value(
_validate_config_values(current_values, bootstrap_settings)
_persist_config_values(session, current_values)
get_settings.cache_clear()
reset_auth_db_caches()
reset_db_caches()
def is_ticktick_oauth_ready(settings: Settings) -> bool:
@@ -260,8 +287,6 @@ def _settings_payload(settings: Settings) -> dict[str, Any]:
"app_debug": settings.app_debug,
"app_hostname": settings.app_hostname,
"app_database_url": settings.app_database_url,
"location_database_url": settings.location_database_url,
"poo_database_url": settings.poo_database_url,
"ticktick_client_id": settings.ticktick_client_id,
"ticktick_client_secret": settings.ticktick_client_secret,
"ticktick_token": settings.ticktick_token,
@@ -286,4 +311,15 @@ def _settings_payload(settings: Settings) -> dict[str, Any]:
"auth_session_cookie_name": settings.auth_session_cookie_name,
"auth_session_ttl_hours": settings.auth_session_ttl_hours,
"auth_cookie_secure_override": settings.auth_cookie_secure_override,
"auth_login_throttle_enabled": settings.auth_login_throttle_enabled,
"auth_trust_forwarded_for": settings.auth_trust_forwarded_for,
"modbus_polling_enabled": settings.modbus_polling_enabled,
"mqtt_enabled": settings.mqtt_enabled,
"mqtt_broker_host": settings.mqtt_broker_host,
"mqtt_broker_port": settings.mqtt_broker_port,
"mqtt_username": settings.mqtt_username,
"mqtt_password": settings.mqtt_password,
"mqtt_tls_enabled": settings.mqtt_tls_enabled,
"ha_discovery_enabled": settings.ha_discovery_enabled,
"ha_discovery_prefix": settings.ha_discovery_prefix,
}
+408
View File
@@ -0,0 +1,408 @@
"""HA MQTT Discovery service — build and publish discovery configs + state.
This module handles:
1. Building HA MQTT Discovery payloads for each ``ExposableEntity``.
2. Publishing discovery configs (retained) for enabled entities.
3. Clearing (empty-payload) discovery configs for disabled entities.
4. Publishing current entity state values.
5. A per-device helper for use in the Modbus poll loop.
Design
------
- **No-op when MQTT / discovery is not configured**: every public function
checks ``ha_discovery_enabled`` and ``mqtt_manager.is_connected`` before
doing any work. Callers never need to guard these conditions.
- **Stable unique_id**: derived from device ``uuid`` + metric ``key``, never
from mutable fields like ``friendly_name`` or auto-increment DB ids.
- **Discovery topic format**: ``<prefix>/<component>/<node_id>/<object_id>/config``
where ``node_id`` is the device uuid (slugified to be safe) and
``object_id`` is a uuid-prefixed stable string.
- **State topic format**: ``<prefix>/<component>/<node_id>/<object_id>/state``
- **Availability topic**: ``<prefix>/modbus/<node_id>/availability`` (shared
across all entities of the same device; publishes "online"/"offline").
- **best-effort**: functions catch all exceptions internally so that callers
(e.g. the Modbus poll loop) never crash due to MQTT publish errors.
"""
from __future__ import annotations
import json
import logging
from typing import Any
from sqlalchemy.orm import Session
from app.integrations.expose import ExposableEntity, build_catalog
from app.integrations.mqtt import mqtt_manager
from app.services.config_page import build_runtime_settings
logger = logging.getLogger(__name__)
# ---------------------------------------------------------------------------
# Internal helpers
# ---------------------------------------------------------------------------
def _should_publish(settings: Any) -> bool:
"""Return True only if MQTT and HA Discovery are both enabled and connected."""
return bool(
settings.mqtt_enabled
and settings.ha_discovery_enabled
and mqtt_manager.is_connected
)
def _node_id(device_uuid: str) -> str:
"""Stable MQTT node_id derived from device uuid (replace hyphens for safety)."""
return device_uuid.replace("-", "_")
def _object_id(entity: ExposableEntity) -> str:
"""Stable MQTT object_id derived from entity key (dots + hyphens → underscores)."""
return entity.key.replace(".", "_").replace("-", "_")
def _discovery_topic(entity: ExposableEntity, prefix: str) -> str:
"""Build the HA Discovery config topic for *entity*.
Format: ``<prefix>/<component>/<node_id>/<object_id>/config``
"""
node = _node_id(entity.device.identifiers[1]) # device uuid
obj = _object_id(entity)
return f"{prefix}/{entity.component}/{node}/{obj}/config"
def _state_topic(entity: ExposableEntity, prefix: str) -> str:
"""Build the state publish topic for *entity*.
Format: ``<prefix>/<component>/<node_id>/<object_id>/state``
"""
node = _node_id(entity.device.identifiers[1])
obj = _object_id(entity)
return f"{prefix}/{entity.component}/{node}/{obj}/state"
def _availability_topic(device_uuid: str, prefix: str) -> str:
"""Shared availability topic for all entities of a device.
Format: ``<prefix>/modbus/<node_id>/availability``
"""
node = _node_id(device_uuid)
return f"{prefix}/modbus/{node}/availability"
def _unique_id(entity: ExposableEntity) -> str:
"""Stable unique_id — device uuid + metric key (never from mutable fields)."""
device_uuid = entity.device.identifiers[1]
# entity.key is already "modbus.<uuid>.<metric_key>" — use it as the seed
return f"{device_uuid}_{entity.key.replace('.', '_')}"
# ---------------------------------------------------------------------------
# Public: build discovery payload
# ---------------------------------------------------------------------------
def build_discovery_payload(
entity: ExposableEntity,
prefix: str,
) -> tuple[str, dict[str, Any]]:
"""Build the HA MQTT Discovery config topic and payload dict for *entity*.
Parameters
----------
entity:
An ``ExposableEntity`` (from ``build_catalog``).
prefix:
The discovery prefix (e.g. ``"homeassistant"``).
Returns
-------
tuple[str, dict]
``(topic, config_dict)`` — the config topic and the HA Discovery
payload (not yet JSON-serialised).
"""
device_uuid = entity.device.identifiers[1]
avail_topic = _availability_topic(device_uuid, prefix)
state_t = _state_topic(entity, prefix)
topic = _discovery_topic(entity, prefix)
config: dict[str, Any] = {
"unique_id": _unique_id(entity),
"name": entity.name,
"state_topic": state_t,
"availability": [{"topic": avail_topic}],
"availability_mode": "all",
"device": {
"identifiers": list(entity.device.identifiers),
"name": entity.device.name,
},
}
# Component-specific fields
if entity.component == "binary_sensor":
# "online" binary sensor: payload ON/OFF
config["payload_on"] = "ON"
config["payload_off"] = "OFF"
if entity.device_class:
config["device_class"] = entity.device_class
else:
# sensor (and others)
if entity.device_class:
config["device_class"] = entity.device_class
if entity.unit:
config["unit_of_measurement"] = entity.unit
if entity.state_class:
config["state_class"] = entity.state_class
return topic, config
# ---------------------------------------------------------------------------
# Public: publish discovery
# ---------------------------------------------------------------------------
def publish_discovery(session: Session) -> None:
"""Publish HA Discovery configs for all entities in the catalog.
- Enabled entities: publish retained JSON config payload.
- Disabled entities: publish empty (b"") payload to clear any previously
retained config from HA.
No-op if MQTT / discovery is not enabled or the client is not connected.
All MQTT errors are caught internally — this function never raises.
"""
from app.config import get_settings
settings = build_runtime_settings(session, get_settings())
if not _should_publish(settings):
logger.debug("publish_discovery: skipped (MQTT/Discovery not enabled or not connected)")
return
prefix = settings.ha_discovery_prefix
try:
catalog = build_catalog(session)
except Exception:
logger.exception("publish_discovery: failed to build catalog; aborting")
return
for entry in catalog:
entity = entry.entity
try:
topic, config = build_discovery_payload(entity, prefix)
if entry.enabled:
payload = json.dumps(config)
mqtt_manager.publish(topic, payload, retain=True)
logger.debug(
"publish_discovery: published config for %r%s", entity.key, topic
)
else:
# Clear retained config for disabled entities.
mqtt_manager.publish(topic, b"", retain=True)
logger.debug(
"publish_discovery: cleared config for disabled entity %r%s",
entity.key,
topic,
)
except Exception:
logger.exception(
"publish_discovery: error processing entity %r; continuing", entity.key
)
# ---------------------------------------------------------------------------
# Public: publish states
# ---------------------------------------------------------------------------
def publish_states(session: Session) -> None:
"""Publish current state values for all enabled entities.
Calls each entity's ``value_getter`` to obtain the current value and
publishes it to the entity's state topic. Also publishes availability
(online/offline) for each device.
No-op if MQTT / discovery is not enabled or the client is not connected.
All errors are caught internally — this function never raises.
"""
from app.config import get_settings
settings = build_runtime_settings(session, get_settings())
if not _should_publish(settings):
logger.debug("publish_states: skipped (MQTT/Discovery not enabled or not connected)")
return
prefix = settings.ha_discovery_prefix
try:
catalog = build_catalog(session)
except Exception:
logger.exception("publish_states: failed to build catalog; aborting")
return
# Collect enabled entries and publish
for entry in catalog:
if not entry.enabled:
continue
entity = entry.entity
try:
_publish_entity_state(entity, prefix, session)
except Exception:
logger.exception(
"publish_states: error publishing state for %r; continuing", entity.key
)
def _publish_entity_state(
entity: ExposableEntity,
prefix: str,
session: Session,
) -> None:
"""Publish one entity's current state value to its state topic.
Also publishes the availability topic for ``binary_sensor`` "online" entities.
"""
state_t = _state_topic(entity, prefix)
device_uuid = entity.device.identifiers[1]
if entity.component == "binary_sensor" and "online" in entity.key:
# The online sensor represents device availability.
# Ask the value_getter for the current ON/OFF string, then also publish
# the shared availability topic (online/offline).
raw_value = None
if entity.value_getter is not None:
try:
raw_value = entity.value_getter(session)
except Exception:
logger.exception("value_getter raised for online entity %r", entity.key)
# Default to offline when no reading is available.
online = (raw_value == "ON")
avail_payload = "online" if online else "offline"
avail_topic = _availability_topic(device_uuid, prefix)
mqtt_manager.publish(avail_topic, avail_payload, retain=False)
# The state of the binary_sensor itself
state_payload = "ON" if online else "OFF"
mqtt_manager.publish(state_t, state_payload, retain=False)
else:
# Regular sensor: call value_getter(session) if available.
value = None
if entity.value_getter is not None:
try:
value = entity.value_getter(session)
except Exception:
logger.exception("value_getter raised for entity %r", entity.key)
if value is None:
logger.debug("publish_states: no value for %r — skipping state publish", entity.key)
return
mqtt_manager.publish(state_t, str(value), retain=False)
# ---------------------------------------------------------------------------
# Public: per-device state push (called by modbus_poll after each poll)
# ---------------------------------------------------------------------------
def publish_device_state(session: Session, device: Any) -> None:
"""Publish state + availability for all enabled entities of *device*.
Called by ``modbus_poll.poll_device`` after a successful (or failed) poll.
Publishes:
- The current availability topic ("online" / "offline") for the device.
- The state topic for each **enabled** entity of the device.
No-op if MQTT / discovery is not enabled or the client is not connected.
All errors are caught internally — never raises.
"""
from app.config import get_settings
settings = build_runtime_settings(session, get_settings())
if not _should_publish(settings):
return
prefix = settings.ha_discovery_prefix
device_uuid = device.uuid
online = bool(device.last_poll_ok)
# Publish availability topic first.
try:
avail_topic = _availability_topic(device_uuid, prefix)
avail_payload = "online" if online else "offline"
mqtt_manager.publish(avail_topic, avail_payload, retain=False)
except Exception:
logger.exception(
"publish_device_state: failed to publish availability for device uuid=%s", device_uuid
)
if not online:
# Device is offline — no point pushing stale state values.
return
# Publish state for each enabled entity of this device.
try:
catalog = build_catalog(session)
except Exception:
logger.exception(
"publish_device_state: failed to build catalog for device uuid=%s", device_uuid
)
return
for entry in catalog:
if not entry.enabled:
continue
entity = entry.entity
# Only process entities belonging to this device.
if entity.device.identifiers[1] != device_uuid:
continue
# Skip the online binary_sensor itself (availability handled above).
if entity.component == "binary_sensor" and "online" in entity.key:
# Publish the binary_sensor state too.
try:
state_t = _state_topic(entity, prefix)
mqtt_manager.publish(state_t, "ON", retain=False)
except Exception:
logger.exception(
"publish_device_state: error publishing online sensor state for %r",
entity.key,
)
continue
# Regular sensor — call value_getter(session).
try:
value = None
if entity.value_getter is not None:
value = entity.value_getter(session)
if value is None:
continue
state_t = _state_topic(entity, prefix)
mqtt_manager.publish(state_t, str(value), retain=False)
except Exception:
logger.exception(
"publish_device_state: error publishing state for entity %r", entity.key
)
def publish_device_offline(session: Session, device_uuid: str) -> None:
"""Publish an "offline" availability payload for *device_uuid*.
Called by ``modbus_poll.poll_device`` when a poll fails, to immediately
reflect the device going offline in HA (before the next full state sweep).
No-op if MQTT / discovery is not enabled or the client is not connected.
Never raises.
"""
from app.config import get_settings
settings = build_runtime_settings(session, get_settings())
if not _should_publish(settings):
return
try:
prefix = settings.ha_discovery_prefix
avail_topic = _availability_topic(device_uuid, prefix)
mqtt_manager.publish(avail_topic, "offline", retain=False)
except Exception:
logger.exception(
"publish_device_offline: failed for device uuid=%s", device_uuid
)
+56 -1
View File
@@ -1,6 +1,6 @@
from datetime import datetime, timezone
from sqlalchemy import insert
from sqlalchemy import delete, insert, select
from sqlalchemy.orm import Session
from app.models.location import Location
@@ -40,3 +40,58 @@ def record_location(session: Session, payload: LocationRecordRequest) -> None:
)
session.execute(stmt)
session.commit()
def update_location(
session: Session,
person: str,
datetime_pk: str,
*,
latitude: float | None,
longitude: float | None,
altitude: float | None,
) -> Location | None:
"""Update non-PK fields of a single location row.
Returns the updated ORM object, or ``None`` if the PK does not exist.
The caller must not pass PK fields — they are immutable.
Only fields with a non-``None`` value are written; ``altitude`` being
``None`` in the request means "leave unchanged", not "clear to NULL".
"""
row = session.execute(
select(Location).where(
Location.person == person,
Location.datetime == datetime_pk,
)
).scalar_one_or_none()
if row is None:
return None
if latitude is not None:
row.latitude = latitude
if longitude is not None:
row.longitude = longitude
if altitude is not None:
row.altitude = altitude
session.commit()
session.refresh(row)
return row
def delete_location(session: Session, person: str, datetime_pk: str) -> bool:
"""Delete the single location row identified by its full composite PK.
Returns ``True`` if exactly one row was deleted, ``False`` if the PK did
not exist (caller should raise 404). The DELETE is scoped to the exact PK
— no batch/truncate path exists.
"""
result = session.execute(
delete(Location).where(
Location.person == person,
Location.datetime == datetime_pk,
)
)
session.commit()
return result.rowcount == 1
+179
View File
@@ -0,0 +1,179 @@
"""Exponential back-off throttle service for the login endpoint.
Design
------
Failures are tracked per (scope, key) in the ``auth_login_throttle`` table:
* scope ``'ip'`` → key is the client IP address
* scope ``'user'`` → key is the username
On each login attempt the endpoint checks **both** keys and takes the **maximum**
remaining wait (in seconds). A wait > 0 means the caller is still inside a
back-off window and should receive 429.
Exponential formula
-------------------
The first ``N_FREE`` failures are free (no delay). After that::
wait = min(BACKOFF_CAP_S, BACKOFF_BASE_S * 2 ** (failures - N_FREE))
Constants (module-level so they can be tuned without touching the algorithm):
* ``N_FREE`` = 3 — failures before any delay starts
* ``BACKOFF_BASE_S`` = 1 — base wait in seconds (1st delayed attempt → 2 s)
* ``BACKOFF_CAP_S`` = 900 — hard cap (15 min)
Example delay schedule:
failures 1, 2, 3 → 0 s (free)
failures 4 → 2 s
failures 5 → 4 s
failures 6 → 8 s
failures 7 → 16 s
failures 13+ → 900 s (capped)
Timezone notes
--------------
The model uses ``DateTime(timezone=True)``. SQLite stores timestamps without
timezone info, so when rows are read back the column value may be timezone-naive.
We normalise every stored value to UTC-aware via ``_as_utc`` before comparing
against ``datetime.now(UTC)``.
"""
from __future__ import annotations
import math
from datetime import UTC, datetime, timedelta
from sqlalchemy import delete, select
from sqlalchemy.orm import Session
from app.models.auth_throttle import LoginThrottle
# ---------------------------------------------------------------------------
# Tunable constants
# ---------------------------------------------------------------------------
N_FREE: int = 3 # failures before any backoff starts
BACKOFF_BASE_S: int = 1 # seconds — wait after the first delayed failure
BACKOFF_CAP_S: int = 900 # seconds — maximum wait (15 min)
# Internal scope labels (keep in sync with model constraint)
_SCOPE_IP = "ip"
_SCOPE_USER = "user"
# ---------------------------------------------------------------------------
# Public API
# ---------------------------------------------------------------------------
def check_and_get_wait(session: Session, *, ip: str, username: str) -> int:
"""Return the number of seconds the caller must still wait, or 0 if allowed.
Checks both the IP row and the username row; returns the *larger* remaining
wait so that both keys must have their back-off window expire before the
caller is permitted to try again.
"""
now = datetime.now(UTC)
wait_ip = _remaining_wait(session, scope=_SCOPE_IP, key=ip, now=now)
wait_user = _remaining_wait(session, scope=_SCOPE_USER, key=username, now=now)
return max(wait_ip, wait_user)
def register_failure(session: Session, *, ip: str, username: str) -> None:
"""Record one login failure for both the IP and the username keys.
Updates (or creates) the throttle row for each key, increments the failure
counter, and recomputes ``next_allowed_at`` using the exponential formula.
"""
now = datetime.now(UTC)
_upsert_failure(session, scope=_SCOPE_IP, key=ip, now=now)
_upsert_failure(session, scope=_SCOPE_USER, key=username, now=now)
session.commit()
def clear(session: Session, *, ip: str, username: str) -> None:
"""Delete the throttle rows for the given IP and username (successful login).
It is safe to call this even if no rows exist for the given keys.
"""
session.execute(
delete(LoginThrottle).where(
(LoginThrottle.scope == _SCOPE_IP) & (LoginThrottle.key == ip)
| (LoginThrottle.scope == _SCOPE_USER) & (LoginThrottle.key == username)
)
)
session.commit()
# ---------------------------------------------------------------------------
# Helpers
# ---------------------------------------------------------------------------
def _remaining_wait(session: Session, *, scope: str, key: str, now: datetime) -> int:
"""Return the number of seconds that ``key`` in ``scope`` must still wait.
Returns 0 if the key has no throttle row or if the back-off window has
already expired.
"""
row = session.scalar(
select(LoginThrottle)
.where(LoginThrottle.scope == scope, LoginThrottle.key == key)
.limit(1)
)
if row is None or row.next_allowed_at is None:
return 0
next_allowed = _as_utc(row.next_allowed_at)
if next_allowed <= now:
return 0
remaining = (next_allowed - now).total_seconds()
# Round up so callers are never told "0 seconds" while still blocked.
return math.ceil(remaining)
def _compute_next_allowed_at(failures: int, now: datetime) -> datetime | None:
"""Compute the earliest time the key should be allowed to try again.
Returns ``None`` for the first ``N_FREE`` failures (no delay).
"""
if failures <= N_FREE:
return None
wait_s = min(BACKOFF_CAP_S, BACKOFF_BASE_S * (2 ** (failures - N_FREE)))
return now + timedelta(seconds=wait_s)
def _upsert_failure(session: Session, *, scope: str, key: str, now: datetime) -> None:
"""Create or update the throttle row for (scope, key) with one more failure."""
row = session.scalar(
select(LoginThrottle)
.where(LoginThrottle.scope == scope, LoginThrottle.key == key)
.limit(1)
)
if row is None:
new_failures = 1
row = LoginThrottle(
scope=scope,
key=key,
failures=new_failures,
first_failed_at=now,
last_failed_at=now,
next_allowed_at=_compute_next_allowed_at(new_failures, now),
)
session.add(row)
else:
row.failures += 1
row.last_failed_at = now
row.next_allowed_at = _compute_next_allowed_at(row.failures, now)
def _as_utc(value: datetime | None) -> datetime | None:
"""Normalise a possibly-naive datetime to UTC-aware."""
if value is None:
return None
if value.tzinfo is None:
return value.replace(tzinfo=UTC)
return value.astimezone(UTC)
+209
View File
@@ -0,0 +1,209 @@
"""Modbus polling service — periodic read-and-store for enabled Modbus devices.
Design notes
------------
- ``poll_device`` loads the device's YAML profile, calls the driver to bulk-read
all register blocks in a single TCP connection, decodes the raw registers into
an engineering-value dict via the profile, and persists one ``ModbusReading``
row. It also stamps ``last_poll_at`` / ``last_poll_ok`` on the device.
- All exceptions are caught and logged inside ``poll_device``; the function
never re-raises. This prevents a single broken device from aborting the
sweep for all other devices.
- ``poll_all_enabled_devices`` respects per-device ``poll_interval_s``: it skips
a device whose ``last_poll_at`` is recent enough. This lets the APScheduler
job run on a short base tick (e.g. 5 s) while each device self-regulates its
own cadence.
- The global ``modbus_polling_enabled`` flag from ``Settings`` (merged with any
DB overrides via ``build_runtime_settings``) gates the entire sweep as a
no-op. T08 will expose this flag in CONFIG_FIELDS/UI; for now it is
controlled by the ``MODBUS_POLLING_ENABLED`` env var (or the default True).
"""
from __future__ import annotations
import logging
from datetime import UTC, datetime
from sqlalchemy import select
from sqlalchemy.orm import Session
from app.config import Settings
from app.integrations.modbus import driver, profiles
from app.models.modbus import ModbusDevice, ModbusReading
from app.services.config_page import build_runtime_settings
from app.services.ha_discovery import publish_device_offline, publish_device_state
logger = logging.getLogger(__name__)
# Base polling tick in seconds (used by APScheduler; each device checks its own interval).
BASE_POLL_TICK_SECONDS = 5
def poll_device(session: Session, device: ModbusDevice) -> ModbusReading | None:
"""Read one enabled Modbus device, decode, and persist a reading row.
Parameters
----------
session:
Open SQLAlchemy session. The caller is responsible for session
lifecycle (open / commit / close).
device:
``ModbusDevice`` ORM instance to poll.
Returns
-------
ModbusReading | None
The newly-persisted reading on success, or ``None`` on any failure.
Failure is logged and the device's ``last_poll_ok`` is set to ``False``.
"""
now = datetime.now(UTC)
try:
profile = profiles.load_profile(device.profile)
registers = driver.read_blocks(
device.host,
device.port,
device.unit_id,
[{"start": b.start, "count": b.count} for b in profile.blocks],
)
payload = profiles.decode(profile, registers)
reading = ModbusReading(
device_id=device.id,
recorded_at=now,
payload=payload,
)
session.add(reading)
device.last_poll_at = now
device.last_poll_ok = True
session.commit()
logger.debug(
"Polled device %r (id=%d): %d metrics recorded at %s",
device.friendly_name,
device.id,
len(payload),
now.isoformat(),
)
# Best-effort: push MQTT state after successful poll.
# Must never let MQTT errors crash the poll loop.
try:
publish_device_state(session, device)
except Exception: # noqa: BLE001
logger.exception(
"poll_device: MQTT state publish failed for device %r (id=%d); "
"continuing (poll itself succeeded)",
device.friendly_name,
device.id,
)
return reading
except Exception: # noqa: BLE001
logger.exception(
"Failed to poll device %r (id=%d, host=%s:%d, unit_id=%d)",
device.friendly_name,
device.id,
device.host,
device.port,
device.unit_id,
)
# Rollback first — if session.add(reading) already ran in the success
# branch before commit() failed, that pending ModbusReading must be
# discarded. Without rollback it would remain staged and could be
# flushed/committed by a subsequent operation (even for a different
# device), producing a spurious reading row for a poll that failed.
# Rollback also expires all ORM objects; device fields are re-set below.
try:
session.rollback()
except Exception: # noqa: BLE001
logger.exception(
"session.rollback() failed for device id=%d; session may be unusable", device.id
)
try:
device.last_poll_at = now
device.last_poll_ok = False
session.commit()
except Exception: # noqa: BLE001
logger.exception(
"Also failed to persist last_poll_ok=False for device id=%d", device.id
)
# Best-effort: publish "offline" availability after failed poll.
try:
publish_device_offline(session, device.uuid)
except Exception: # noqa: BLE001
logger.exception(
"poll_device: MQTT offline publish failed for device %r (id=%d)",
device.friendly_name,
device.id,
)
return None
def poll_all_enabled_devices(
session: Session,
*,
bootstrap_settings: Settings,
) -> None:
"""Sweep all enabled Modbus devices and poll each one whose interval has elapsed.
Global kill-switch: if ``build_runtime_settings(...).modbus_polling_enabled``
is ``False``, this function returns immediately without touching any device.
Per-device interval: a device is skipped if its ``last_poll_at`` is not yet
``poll_interval_s`` seconds in the past. This allows the APScheduler job to
run on a short fixed tick while each device self-regulates its own cadence.
Exceptions from individual devices are swallowed inside ``poll_device``;
this function itself will not raise.
Parameters
----------
session:
Open SQLAlchemy session.
bootstrap_settings:
The bootstrap ``Settings`` instance (from ``get_settings()``), used as
the base for ``build_runtime_settings`` to pick up any DB overrides.
"""
try:
runtime_settings = build_runtime_settings(session, bootstrap_settings)
except Exception: # noqa: BLE001
logger.exception("Failed to build runtime settings for Modbus poll; aborting sweep")
return
if not runtime_settings.modbus_polling_enabled:
logger.debug("Modbus polling is disabled (modbus_polling_enabled=False); skipping sweep")
return
try:
devices = session.execute(
select(ModbusDevice).where(ModbusDevice.enabled == True) # noqa: E712
).scalars().all()
except Exception: # noqa: BLE001
logger.exception("Failed to query enabled Modbus devices; aborting sweep")
return
now = datetime.now(UTC)
for device in devices:
# Respect per-device poll interval: skip if last poll was too recent.
if device.last_poll_at is not None:
# SQLite may return a naive datetime despite DateTime(timezone=True) —
# treat naive values as UTC so the comparison is safe.
last_poll = device.last_poll_at
if last_poll.tzinfo is None:
last_poll = last_poll.replace(tzinfo=UTC)
elapsed = (now - last_poll).total_seconds()
if elapsed < device.poll_interval_s:
logger.debug(
"Skipping device %r (id=%d): %.1fs elapsed < %ds interval",
device.friendly_name,
device.id,
elapsed,
device.poll_interval_s,
)
continue
poll_device(session, device)
+48 -1
View File
@@ -4,7 +4,7 @@ from dataclasses import dataclass
from datetime import datetime, timezone
import logging
from sqlalchemy import desc, insert, select
from sqlalchemy import delete, desc, insert, select
from sqlalchemy.orm import Session
from app.config import Settings
@@ -74,6 +74,53 @@ def record_poo(
logger.warning("Failed to trigger poo webhook on Home Assistant: %s", exc)
def update_poo_record(
session: Session,
timestamp_pk: str,
*,
status: str | None,
latitude: float | None,
longitude: float | None,
) -> PooRecord | None:
"""Update non-PK fields of a single poo record row.
Returns the updated ORM object, or ``None`` if the PK does not exist.
The ``timestamp`` PK is immutable and must not be passed as an update field.
Only fields with a non-``None`` value are written.
"""
row = session.execute(
select(PooRecord).where(PooRecord.timestamp == timestamp_pk)
).scalar_one_or_none()
if row is None:
return None
if status is not None:
row.status = status
if latitude is not None:
row.latitude = latitude
if longitude is not None:
row.longitude = longitude
session.commit()
session.refresh(row)
return row
def delete_poo_record(session: Session, timestamp_pk: str) -> bool:
"""Delete the single poo record row identified by its PK.
Returns ``True`` if exactly one row was deleted, ``False`` if the PK did
not exist (caller should raise 404). The DELETE is scoped to the exact PK
— no batch/truncate path exists.
"""
result = session.execute(
delete(PooRecord).where(PooRecord.timestamp == timestamp_pk)
)
session.commit()
return result.rowcount == 1
def get_latest_poo_record(session: Session) -> LatestPooRecord | None:
stmt = select(PooRecord).order_by(desc(PooRecord.timestamp)).limit(1)
record = session.execute(stmt).scalar_one_or_none()
+234
View File
@@ -0,0 +1,234 @@
"""TOTP service: secret management, recovery-code lifecycle, setup/enable/disable (M4-T05).
State-machine summary
---------------------
DISABLED (default)
totp_secret=None, totp_enabled=False, no recovery codes
PENDING (after setup, before enable)
totp_secret=<base32>, totp_enabled=False, recovery codes stored as hashes
- Re-calling setup replaces the pending secret and regenerates recovery codes.
ENABLED (after enable)
totp_secret=<base32>, totp_enabled=True, recovery codes stored as hashes
After disable:
totp_secret=None, totp_enabled=False, recovery codes deleted → back to DISABLED
Recovery-code timing
--------------------
1. setup → generate 10 plaintext codes, persist their Argon2 hashes immediately,
return plaintext to caller (ONLY time).
2. enable → verify 6-digit TOTP code; if ok, flip totp_enabled=True.
Recovery codes are already in the DB from step 1.
3. disable → clear secret, clear enabled flag, delete all recovery codes.
Security note: ``totp_secret`` is stored as plaintext (same posture as other
secrets in this project: protected by file-system permissions on the SQLite
database). Recovery codes are stored as Argon2 hashes only.
"""
from __future__ import annotations
import logging
import secrets
import string
import pyotp
from sqlalchemy import delete, select
from sqlalchemy.orm import Session
from app.models.auth import AuthUser, RecoveryCode
from app.services.auth import hash_password, verify_password
logger = logging.getLogger(__name__)
# Number of recovery codes to generate per setup
_RECOVERY_CODE_COUNT = 10
# Alphabet for each 4-character segment: lowercase letters + digits, no ambiguous chars
_CODE_ALPHABET = string.ascii_lowercase + string.digits
# ---------------------------------------------------------------------------
# Internal helpers
# ---------------------------------------------------------------------------
def _generate_recovery_code() -> str:
"""Return a single recovery code in ``xxxx-xxxx`` format."""
segment = lambda: "".join(secrets.choice(_CODE_ALPHABET) for _ in range(4)) # noqa: E731
return f"{segment()}-{segment()}"
def _verify_totp_code(secret: str, code: str) -> bool:
"""Verify a 6-digit TOTP code against the given base-32 secret (±1 window)."""
return pyotp.TOTP(secret).verify(code, valid_window=1)
def _delete_recovery_codes(db: Session, *, user_id: int) -> None:
"""Delete ALL recovery codes for a user (called on setup re-run and disable)."""
db.execute(delete(RecoveryCode).where(RecoveryCode.user_id == user_id))
# ---------------------------------------------------------------------------
# Public API
# ---------------------------------------------------------------------------
def setup(
db: Session,
*,
user: AuthUser,
issuer: str,
) -> tuple[str, str, list[str]]:
"""Generate a new pending TOTP secret and recovery codes.
The secret is stored in ``user.totp_secret`` (``totp_enabled`` stays
``False``). Any pre-existing pending secret and recovery codes are
replaced atomically.
Returns
-------
(secret, otpauth_uri, plaintext_recovery_codes)
``plaintext_recovery_codes`` are returned **once** here and MUST NOT be
stored or returned elsewhere.
"""
# Generate new TOTP secret
new_secret = pyotp.random_base32()
otpauth_uri = pyotp.TOTP(new_secret).provisioning_uri(
name=user.username,
issuer_name=issuer,
)
# Generate plaintext recovery codes
plaintext_codes = [_generate_recovery_code() for _ in range(_RECOVERY_CODE_COUNT)]
# --- Persist atomically ---
# 1. Delete any old pending recovery codes (idempotent on re-setup)
assert user.id is not None # mypy guard; always set after DB insertion
_delete_recovery_codes(db, user_id=user.id)
# 2. Update the secret (pending — totp_enabled stays False)
user.totp_secret = new_secret
# 3. Persist new recovery codes as Argon2 hashes
for plaintext in plaintext_codes:
db.add(RecoveryCode(user_id=user.id, code_hash=hash_password(plaintext)))
db.commit()
db.refresh(user)
logger.info("TOTP setup (pending) for user '%s'; %d recovery codes generated.", user.username, _RECOVERY_CODE_COUNT)
return new_secret, otpauth_uri, plaintext_codes
def enable(db: Session, *, user: AuthUser, code: str) -> bool:
"""Enable TOTP after the user proves they can generate the correct code.
Parameters
----------
code:
The current 6-digit TOTP code from the user's authenticator app.
Returns
-------
True on success; False if the code is invalid or no secret is pending.
"""
if not user.totp_secret:
logger.info("TOTP enable rejected for '%s': no pending secret.", user.username)
return False
if not _verify_totp_code(user.totp_secret, code):
logger.info("TOTP enable rejected for '%s': wrong code.", user.username)
return False
user.totp_enabled = True
db.commit()
db.refresh(user)
logger.info("TOTP enabled for user '%s'.", user.username)
return True
def disable(
db: Session,
*,
user: AuthUser,
password: str | None = None,
code: str | None = None,
) -> bool:
"""Disable TOTP. Caller must supply exactly one of ``password`` or ``code``.
On success: ``totp_enabled=False``, ``totp_secret=None``, all recovery
codes deleted.
Returns
-------
True on success; False if neither credential is valid.
"""
if not password and not code:
logger.info("TOTP disable rejected for '%s': no credential provided.", user.username)
return False
if password:
if not verify_password(password, user.password_hash):
logger.info("TOTP disable rejected for '%s': wrong password.", user.username)
return False
elif code:
if not user.totp_secret or not _verify_totp_code(user.totp_secret, code):
logger.info("TOTP disable rejected for '%s': wrong TOTP code.", user.username)
return False
# Clear TOTP state
assert user.id is not None
_delete_recovery_codes(db, user_id=user.id)
user.totp_enabled = False
user.totp_secret = None
db.commit()
db.refresh(user)
logger.info("TOTP disabled for user '%s'.", user.username)
return True
def verify_totp_code(user: AuthUser, code: str) -> bool:
"""Verify a 6-digit TOTP code for a user.
Returns True if the code is valid (±1 time window), False otherwise.
The user must have a ``totp_secret`` set; returns False if not.
This is the public entry-point for T06 login two-factor verification.
"""
if not user.totp_secret:
return False
return _verify_totp_code(user.totp_secret, code)
def verify_recovery_code(db: Session, *, user: AuthUser, code: str) -> bool:
"""Verify and consume a one-time recovery code.
Finds the first unused recovery code whose hash matches ``code``, marks it
as consumed (sets ``used_at``), and returns ``True``. Returns ``False`` if
no matching unused code exists.
This function is provided for T06 (login two-factor) but lives here so the
TOTP service owns all recovery-code logic.
"""
from datetime import UTC, datetime
unused = db.execute(
select(RecoveryCode).where(
RecoveryCode.user_id == user.id,
RecoveryCode.used_at.is_(None),
)
).scalars().all()
for rc in unused:
if verify_password(code, rc.code_hash):
rc.used_at = datetime.now(UTC)
db.commit()
logger.info(
"Recovery code consumed for user '%s' (id=%d).", user.username, rc.id
)
return True
return False
-16
View File
@@ -1,16 +0,0 @@
<!DOCTYPE html>
<html lang="zh-CN">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>{% block title %}{{ app_name }}{% endblock %}</title>
<link rel="icon" href="data:,">
<link rel="stylesheet" href="/static/styles.css">
</head>
<body>
<main class="shell">
{% block content %}{% endblock %}
</main>
</body>
</html>
-139
View File
@@ -1,139 +0,0 @@
{% extends "base.html" %}
{% block title %}Config · {{ app_name }}{% endblock %}
{% block content %}
<section class="panel">
<p class="eyebrow">Configuration</p>
<h1>Config</h1>
{% if force_password_change %}
<div class="alert">
首次登录后需要先修改密码。完成后再继续长期使用当前配置页面。
</div>
{% endif %}
{% if password_change_error %}
<div class="alert">{{ password_change_error }}</div>
{% endif %}
{% if config_error %}
<div class="alert">{{ config_error }}</div>
{% endif %}
{% if config_saved %}
<div class="notice">config saved to the app database. Some changes may require an app restart.</div>
{% endif %}
{% if ticktick_oauth_error %}
<div class="alert">{{ ticktick_oauth_error }}</div>
{% endif %}
{% if ticktick_oauth_notice %}
<div class="notice">{{ ticktick_oauth_notice }}</div>
{% endif %}
{% if smtp_test_error %}
<div class="alert">{{ smtp_test_error }}</div>
{% endif %}
{% if smtp_test_notice %}
<div class="notice">{{ smtp_test_notice }}</div>
{% endif %}
<div class="meta single-column">
<div>
<dt>当前用户</dt>
<dd>admin</dd>
</div>
</div>
<section class="config-block">
<h2>Change Password</h2>
<form class="auth-form" method="post" action="/config/change-password">
<input type="hidden" name="csrf_token" value="{{ csrf_token }}">
<label>
<span>Current Password</span>
<input type="password" name="current_password" autocomplete="current-password" required>
</label>
<label>
<span>New Password</span>
<input type="password" name="new_password" autocomplete="new-password" required>
</label>
<label>
<span>Confirm New Password</span>
<input type="password" name="confirm_password" autocomplete="new-password" required>
</label>
<button type="submit">修改密码</button>
</form>
</section>
<section class="config-block">
<h2>Config</h2>
<form class="config-form" method="post" action="/config">
<input type="hidden" name="csrf_token" value="{{ csrf_token }}">
{% for section in config_sections %}
<fieldset class="config-section">
<legend>{{ section.name }}</legend>
{% for field in section.fields %}
<label>
<span>{{ field.label }}</span>
{% if field.secret %}
<input type="{{ field.input_type }}" name="{{ field.env_name }}" value="" placeholder="leave blank to keep current value">
<small>{% if field.configured %}configured{% else %}not configured{% endif %}</small>
{% else %}
<input type="{{ field.input_type }}" name="{{ field.env_name }}" value="{{ field.value }}">
{% endif %}
</label>
{% endfor %}
{% if section.name == "TickTick" %}
<div class="integration-action-row">
<div>
<p class="integration-action-title">TickTick OAuth</p>
<p class="integration-action-copy">Redirect URI: {{ ticktick_redirect_uri or "configure APP_HOSTNAME to generate the callback URI" }}</p>
{% if ticktick_oauth_ready %}
<p class="integration-action-copy">Use the saved TickTick client settings to start the authorization flow.</p>
{% else %}
<p class="integration-action-copy">Fill in App Hostname, TickTick Client ID, and TickTick Client Secret before starting OAuth.</p>
{% endif %}
</div>
{% if ticktick_oauth_ready %}
<a class="button-link" href="/ticktick/auth/start">Authorize TickTick</a>
{% else %}
<span class="button-link disabled" aria-disabled="true">Authorize TickTick</span>
{% endif %}
</div>
{% endif %}
{% if section.name == "SMTP" %}
<div class="integration-action-row">
<div>
<p class="integration-action-title">SMTP Test Email</p>
<p class="integration-action-copy">Save the SMTP settings first, then send a simple plaintext test email to the configured recipient.</p>
</div>
{% if smtp_test_ready %}
<button type="submit" formaction="/config/smtp/test" formmethod="post">Send SMTP Test</button>
{% else %}
<span class="button-link disabled" aria-disabled="true">Send SMTP Test</span>
{% endif %}
</div>
{% endif %}
</fieldset>
{% endfor %}
<button type="submit">Save Config</button>
</form>
</section>
<form class="logout-form" method="post" action="/logout">
<input type="hidden" name="csrf_token" value="{{ csrf_token }}">
<button type="submit">登出</button>
</form>
</section>
{% endblock %}
-36
View File
@@ -1,36 +0,0 @@
{% extends "base.html" %}
{% block title %}{{ app_name }}{% endblock %}
{% block content %}
<section class="panel">
<p class="eyebrow">Python Rewrite Skeleton</p>
<h1>{{ app_name }}</h1>
<p class="lead">
这是当前 Go 后端的 Python 重构基础骨架。此阶段仅提供应用入口、配置、数据库、
测试、模板和容器化基础,不包含业务逻辑迁移。
</p>
<dl class="meta">
<div>
<dt>运行环境</dt>
<dd>{{ app_env }}</dd>
</div>
<div>
<dt>健康检查</dt>
<dd><a href="/status">/status</a></dd>
</div>
<div>
<dt>OpenAPI</dt>
<dd><a href="/docs">/docs</a></dd>
</div>
<div>
<dt>登录</dt>
<dd><a href="/login">/login</a></dd>
</div>
<div>
<dt>Notion</dt>
<dd>{{ notion_status }}</dd>
</div>
</dl>
</section>
{% endblock %}
-33
View File
@@ -1,33 +0,0 @@
{% extends "base.html" %}
{% block title %}登录 · {{ app_name }}{% endblock %}
{% block content %}
<section class="panel auth-panel">
<p class="eyebrow">Authentication</p>
<h1>登录</h1>
<p class="lead">
登录成功后会进入受保护的 config 页面。
</p>
{% if error_message %}
<div class="alert">{{ error_message }}</div>
{% endif %}
<form class="auth-form" method="post" action="/login">
<input type="hidden" name="csrf_token" value="{{ csrf_token }}">
<label>
<span>Username</span>
<input type="text" name="username" autocomplete="username" required>
</label>
<label>
<span>Password</span>
<input type="password" name="password" autocomplete="current-password" required>
</label>
<button type="submit">登录</button>
</form>
</section>
{% endblock %}
+7 -5
View File
@@ -53,19 +53,17 @@ idna==3.11
# httpx
iniconfig==2.3.0
# via pytest
jinja2==3.1.6
# via -r requirements.in
mako==1.3.11
# via alembic
markupsafe==3.0.3
# via
# jinja2
# mako
# via mako
packaging==26.1
# via
# build
# pytest
# wheel
paho-mqtt==2.1.0
# via -r requirements.in
pip-tools==7.5.3
# via -r dev-requirements.in
pluggy==1.6.0
@@ -82,6 +80,10 @@ pydantic-settings==2.13.1
# via -r requirements.in
pygments==2.20.0
# via pytest
pymodbus==3.13.1
# via -r requirements.in
pyotp==2.10.0
# via -r requirements.in
pyproject-hooks==1.2.0
# via
# build
+28
View File
@@ -0,0 +1,28 @@
# Local dev override — use explicitly:
# docker compose -f docker-compose.yml -f docker-compose.dev.yml up --build
# Isolated from the production stack so both can run on this host at once:
# - distinct compose project name (separate network/grouping)
# - distinct container names (-dev suffix; Docker rejects duplicate names)
# - distinct image tag (local build doesn't clobber the prod :latest tag)
name: home-automation-dev
services:
migration:
build: .
image: home-automation:dev
container_name: home-automation-migration-dev
environment:
# In-container path for the mounted ./data volume (./data -> /app/data).
# Overrides the host-absolute APP_DATABASE_URL in .env for local compose runs.
APP_DATABASE_URL: "sqlite:////app/data/app.db"
app:
build: .
image: home-automation:dev
container_name: home-automation-app-dev
# Publish on 8002 for dev. `!override` REPLACES the base ports list instead of
# appending to it, so the dev stack does NOT also bind the production 8881.
ports: !override
- "127.0.0.1:8002:8000"
environment:
APP_DATABASE_URL: "sqlite:////app/data/app.db"
-6
View File
@@ -1,6 +0,0 @@
services:
migration:
build: .
app:
build: .
-19
View File
@@ -25,22 +25,3 @@ services:
- ./data:/app/data
- ./.env:/app/.env:ro
grafana:
image: grafana/grafana:latest
container_name: home-automation-grafana
depends_on:
- app
restart: unless-stopped
ports:
- "10.238.75.70:8882:3000"
environment:
GF_PLUGINS_PREINSTALL: frser-sqlite-datasource
volumes:
- ./data:/data/home-automation:ro
- ./grafana/provisioning:/etc/grafana/provisioning:ro
- ./grafana/dashboards:/var/lib/grafana/dashboards:ro
- homeautomation_grafana_storage:/var/lib/grafana
volumes:
homeautomation_grafana_storage:
name: homeautomation_grafana_storage
+104 -26
View File
@@ -19,51 +19,42 @@
- `main.py`
- FastAPI app factory
- lifespan
- lifespanAPScheduler 启停、MQTT 客户端起停、连接后触发 HA Discovery 发布)
- 基础路由注册
- `config.py`
- 环境变量驱动的 settings
- `auth_db.py`
- app 级共享 auth 数据库
- 环境变量驱动的 settings(含 M5 新增的 MQTT/HA Discovery/Modbus 配置项)
- `db.py`
- SQLAlchemy engine / session / Base
- 统一数据层:一个 `Base`、一个绑定 `app_database_url` 的 cached engineSQLite WAL)、`get_engine` / `get_session_local` / `reset_db_caches` / `get_db_session`
- `dependencies.py`
- 通用依赖注入
- `api/`
- HTTP routes
- 当前已迁入 `/login``/logout``/admin`
- 当前已迁入 `GET /public-ip/check`
- 当前已迁入 `POST /homeassistant/publish` 第一版入口
- 当前已迁入 `POST /poo/record``GET /poo/latest`
- `api/routes/api/`JSON API`/api/*` 前缀),供 React SPA 调用:会话/鉴权、配置读写、数据查询、记录 CRUD、Modbus 设备 CRUD + readings + metrics + test`/api/modbus/*`)、Expose 勾选 + 重发 discovery`/api/expose`)、MQTT 测试连接(`/api/config/mqtt/test`
- 裸 ingestion 端点:`GET /public-ip/check``POST /homeassistant/publish``POST /poo/record``GET /poo/latest`、TickTick OAuth 等
- `models/`
- SQLAlchemy models
- 当前 `auth``location``poo` 使用各自独立的数据库 base
- 所有模型(auth / config / public_ip / location / poo / modbus / expose)共用同一个 `Base`,均落在单一 `app.db`
- M5 新增:`ModbusDevice`(设备部署层)、`ModbusReading`(通用遥测,JSON payload)、`ExposedEntityToggle`HA 实体暴露开关)
- `schemas/`
- Pydantic schemas
- Pydantic schemasM5 新增 `modbus.py``expose.py`
- `services/`
- 业务服务层
- 当前已迁入 config page 的 DB 持久化逻辑
- 当前已迁入 public IPv4 检查、状态持久化与变化通知逻辑
- 当前已迁入 SMTP 发信与测试发信逻辑
- M5 新增:`modbus_poll.py`(采集 service,逐设备 poll + 落库 + 推 MQTT state)、`ha_discovery.py`(构建 HA Discovery payload、发布 retained config、发布 state
- `integrations/`
- 外部系统适配层
- 当前已迁入 Home Assistant outbound adapter
- `templates/`
- Jinja2 模板
- Home Assistant outbound adapterREST 通道,原有)
- M5 新增:`modbus/`pymodbus 薄封装:`driver.py` 块读 + float32 解码;`profiles.py` YAML profile 加载/校验/解码;`profiles/sdm120.yaml` SDM120 协议声明)
- M5 新增:`mqtt.py`paho-mqtt 长连接 `MqttManager`:lifespan 起/停、配置变更重连、`publish(topic, payload, retain)`
- M5 新增:`expose.py`(通用 expose 框架:`ExposableEntity`、provider 注册表、`build_catalog`Modbus provider 从 YAML profile 派生 sensor/binary_sensor 实体目录)
- `static/`
- 极简静态资源
### `alembic_location/`
Location DB 的 migration 基础设施。
### `alembic_app/`
App DB 的 migration 基础设施
### `alembic_poo/`
Poo DB 的 migration 基础设施。
App DB 的唯一 Alembic migration 链,同时管理 `location` / `poo_records` 表。M1 将三个独立 DB 合并进 `app.db` 后,`alembic_location/``alembic_poo/` 已退役,全部由此链统一管理
### `tests/`
@@ -73,17 +64,104 @@ pytest 测试目录。后续可以在这里自然扩展:
- mock tests
- integration tests
### `frontend/`
React SPA 前端(M2 引入)。Vite + React + TypeScript + Mantine,由 FastAPI 同源托管。
- `src/`React 源码
- `src/api/`:由 `openapi/openapi.json` 生成的类型化 client`schema.d.ts`+ fetch 封装
- `dist/``npm run build` 产物,由 FastAPI 的 `SPA_DIST_DIR` 挂载并对非 `/api` 路径做 fallback
### `scripts/`
辅助脚本目录。当前包含 OpenAPI 导出脚本。
辅助脚本目录。当前包含
- `export_openapi.py`:导出 OpenAPI schema 静态产物
- `run_migrations.py`:运行 Alembic migration
- `app_db_adopt.py`App DB 接管 / 初始化
- `migrate_legacy_data.py`:一次性历史数据搬迁脚本
- `admin_cli.py`Admin CLI 逃生通道(M4),见下方"登录加固"说明
- `modbus_cli.py`M5):Modbus 手工试读 CLI`read`/`probe` 两个只读子命令),不依赖 DB,供受控手工验证链路连通性
### `openapi/`
OpenAPI schema 静态产物(`openapi.json` / `openapi.yaml`),由 `python scripts/export_openapi.py` 生成,纳入版本控制。前端 codegen 以此为契约源。
## 登录加固(M4
M4 在基础 Argon2 + server-side session 鉴权之上叠加了三层防御:
**防爆破 / 指数退避**`app/services/login_throttle.py` 按 client IP 与 username 双键记失败计数,失败超过 3 次后指数增长等待时间(最长 15 分钟),`POST /api/auth/login` 在退避窗口内直接返回 `429 + Retry-After`,不执行 Argon2 验证。成功登录后清零。退避是延迟而非永久封号;全局开关 `AUTH_LOGIN_THROTTLE_ENABLED`CONFIG_FIELDS,默认开);反代后需设 `AUTH_TRUST_FORWARDED_FOR=true``.env` 部署级,默认 false)。
**CLI 逃生通道**`scripts/admin_cli.py`(入口 `python -m scripts.admin_cli`)直连本地 DB,**无需 HTTP 服务运行、无需任何已存凭据**,支持:重置密码(`reset-password`)、解锁退避(`unlock`)、关停 TOTP`disable-totp`,零凭据最终逃生)、重新发放 TOTP secret`reissue-totp`)、查看用户列表(`list-admin`)。CLI 只动 auth 行,不触碰用户数据表。
**可选 TOTP 二次验证**:admin 可在设置页自选启用 RFC 6238 TOTP。启用后登录为两步(密码 → 6 位动态码或一次性恢复码);不启用维持纯密码。TOTP secret 明文存库(与其他 secret 一致,靠文件权限保护);恢复码以 Argon2 哈希存储,使用后消费(一次性)。后端使用 `pyotp`,二维码由前端 `qrcode.react` 渲染。issuer 标签由 `AUTH_TOTP_ISSUER``.env` 部署级)配置,默认回退 `app_name`
详细说明:[`docs/auth.md`](./auth.md)
## M5 — 通用 Modbus 采集链路与 MQTT 通道
### Modbus 采集链路
```
YAML profile(协议知识) modbus_device 行(部署信息,DB
- 寄存器块/地址/类型 - host / port(网关 IP
- 每量:key/unit/device_class - unit_idModbus slave 地址)
- ha_component - friendly_name / profile / poll_interval_s / enabled
│ │
└────────────┬───────────────────────┘
│ APScheduler job(轮询所有 enabled 设备)
│ driver.py: ModbusTcpClient → FC04 块读 → 大端 float32 解码
│ profiles.py: decode(profile, registers) → dict[key → value]
modbus_reading 行
device_id · recorded_at · payload(JSON)
{"voltage": 230.2, "current": 1.3, "active_power": 295.0, ...}
```
**关键设计决策**
- 命名分层:存储/采集/API 全部通用 `modbus_*``/api/modbus/devices`);面向用户的领域视图叫 **Energy**(第一个)。接入新设备型号只需新增 YAML profile,不改表/不改 API。
- 读数为 JSON `payload`(无固定列),SQLite `json_extract` 在 DB 端做 AVG/GROUP BY(被 `(device_id, recorded_at)` 索引圈住)。
- FK `ON DELETE RESTRICT`:有读数的设备拒删,引导改为 `enabled=false`
- 设备 `uuid`(uuid4 内部生成)是**稳定身份锚点**API 路径键、HA Discovery `unique_id` 来源,不随 friendly_name 改变。
### 第二条 MQTT 通道(HA Discovery 发布)
与已有 REST 通道(`app/integrations/homeassistant.py``POST /homeassistant/publish`**并行、不冲突**
```
MQTT 通道(M5 新增):
paho-mqtt MqttManagerlifespan 长连接,配置变更可重连)
├─► Discovery configretained
│ topic: <prefix>/<component>/<node>/<object>/config
│ 内容:device 块(identifiers=uuid)、state_topic、unique_iduuid+key)、
│ namefriendly_name)、device_class、unit_of_measurement、availability
│ 时机:连接成功时 / 目录或勾选变更时(全量重发);
│ 取消勾选时发空 payload(清除 entity
└─► State / Availability(非 retained
时机:每次轮询成功后推该设备所有 enabled entity 的最新值;
周期兜底 job 重推所有 enabled entity + online topic
expose 框架:
provider 动态产出 ExposableEntity 目录(元数据从 YAML profile 派生)
ExposedEntityToggle 表:只存逐 key 开关(default=disabled
build_catalog(session) → 目录 + 勾选状态(合并所有 provider)
```
**HA entity 身份模型(Z2M 语义)**
- `unique_id` = `f"{device.uuid}_{metric.key}"`(稳定,改名不变)
- `name` = `friendly_name`(改名重发 discovery,HA 显示名跟着变、历史不丢)
- 每设备除各 sensor entity 外,另有 `binary_sensor` `online`(取 `last_poll_ok`)——这是"不止 sensor"的体现
## 当前约束
- 当前只搭骨架,不迁业务逻辑
- 当前数据库继续使用 SQLite
- 当前不引入前后端分离
- ~~当前不引入前后端分离~~ **已退役(M2**:现为 React SPA + JSON `/api` 层,由 FastAPI 同源托管
- 当前不设计 Notion 模块
- 当前通知能力仍保持极小范围,不引入独立通知中心或多渠道抽象
- Modbus 当前**仅 TCP**Waveshare RTU↔TCP 网关),**只读**FC03/04),不写设备寄存器
## 关于 Notion
+191 -82
View File
@@ -1,120 +1,229 @@
# 基础鉴权说明
# 鉴权说明
本文档说明当前 Python 重构项目里已经落地的第一版鉴权基座
这一轮只解决:
- 登录页
- 登录 / 登出流程
- server-side session
- 一个最小受保护页面
这一轮明确不解决:
- 完整 config persistence
- 完整 config CRUD
- 多用户权限系统
- OAuth / SSO / RBAC
本文档说明当前已落地的鉴权基座(基础 session 鉴权 + M4 登录加固)
## 当前 auth 模型
- 认证方式:`username/password`
- 认证方式:`username/password`(可选启用 TOTP 二次验证)
- 会话方式:server-side session
- 客户端凭据:session cookie
- 页面形态:Jinja server-side template
## 当前持久化
## 持久化
当前新增一个共享 App DB
所有 auth 相关数据存放在单一 App DB(`APP_DATABASE_URL`,默认 `sqlite:///./data/app.db`)中
- `APP_DATABASE_URL`
- 默认值:`sqlite:///./data/app.db`
当前 auth 相关数据存放在这个 DB 中:
- `auth_users`
- `auth_sessions`
- `app_config`
当前没有把 auth 数据和 `location` / `poo` DB 混放。
当前这部分现在也走 Alembic 管理:
- Alembic 环境:`alembic_app.ini` + `alembic_app/`
- 初始化脚本:`python scripts/app_db_adopt.py`
当前没有 legacy app DB,所以这一版脚本只负责初始化新库,不负责 legacy adoption。
`app_config` 现在承接运行时配置持久化。
其中:
- `.env` 负责 bootstrap / fallback
- `app_config` 表负责运行时配置覆盖
- 登录密码仍然属于认证数据,使用 Argon2 哈希,不存进 `app_config`
- `auth_users`:用户表(含 TOTP 字段 `totp_secret` / `totp_enabled`
- `auth_sessions`session token 哈希与过期时间
- `auth_login_throttle`:登录失败退避状态(按 IP / username 双键)
- `auth_recovery_code`TOTP 恢复码哈希(一次性)
- `app_config`runtime 配置持久化
## 首次启动与 bootstrap
如果 auth DB 中还没有任何用户,应用启动时会要求
如果 auth DB 中还没有任何用户,应用启动时会使用
- `AUTH_BOOTSTRAP_USERNAME`
- `AUTH_BOOTSTRAP_PASSWORD`
创建首个 admin 用户。
当前默认 bootstrap 值就是:
创建首个 admin 用户。当前默认 bootstrap 值为:
- username: `admin`
- password: `admin`
首次登录后,系统会强制要求修改密码。
如果你希望在首次启动前就覆盖默认值,可以直接设置环境变量:
## 基础安全设计
- `AUTH_BOOTSTRAP_USERNAME`
- `AUTH_BOOTSTRAP_PASSWORD`
建议流程是:
1. 配好 `.env`
2. 运行 `python scripts/app_db_adopt.py`
3. 启动应用
4. 用 `admin / admin` 首次登录
5. 立即修改密码
## 安全设计
当前这版已经落实的基础安全点:
当前这版已经落实的安全点:
- 密码不明文存储,使用 Argon2 哈希
- session cookie 为 `HttpOnly`
- cookie 使用 `SameSite=Lax`
- `Secure` cookie 在非 `development` 环境默认开启
- 登录表单与登出表单都有基础 CSRF 校验
- 写请求(POST/PUT/PATCH/DELETE)需携带 `X-CSRF-Token` headerSameSite=Lax + 自定义 header 纵深防御,无需 per-session token 值比对)
- session token 为随机生成,服务端只持久化 token hash
- session 有过期时间与显式失效机制
- session 有过期时间(默认 12 小时)与显式失效机制
## 当前受保护范围
---
当前这轮只保护了页面入口:
## M4 登录加固
- `GET /config`
- `POST /config`
- `POST /config/change-password`
- `POST /logout`
M4 在基础鉴权之上叠加了三层防御:防爆破/指数退避、CLI 逃生通道、可选 TOTP 二次验证。
相关流程:
### 1. 防爆破 / 指数退避
- `GET /login`
- `POST /login`
#### 机制
登录访问 `/config` 时会被重定向到 `/login`
登录失败按指数增长延迟,目标是拖垮暴力枚举,同时不因此永久锁定账号
## 下一步不在本轮内
**退避是延迟(429 + Retry-After),不是永久封号**——单 admin 场景下永久锁会被攻击者反向用来故意打锁,所以退避只增加等待时间,CLI 是最终逃生口。
后续可以在这个基座上继续做:
#### 双键计算
- 配置页面接入
- config persistence
- 更细的受保护路由范围
- 用户初始化 / 密码轮换的更正式 runbook
每次登录请求同时按 **client IP****username** 各记一套失败计数,本次需等待时间 = 两者退避的**较大值**。
- 按 IP:堵单点暴力(攻击 IP 越敲越慢,合法用户换 IP 不受影响)
- 按 username:全局兜底(即便攻击者分布式换 IP 也有一层保护)
#### 指数公式
```
N_FREE = 3 — 前 3 次失败不延迟(免费次数)
BASE = 1 秒
CAP = 900 秒(15 分钟)
wait = min(CAP, BASE × 2^(failures - N_FREE)) failures > N_FREE 时生效)
```
延迟示意:
| 累计失败次数 | 等待时间 |
| --- | --- |
| 13 | 0 秒(免费) |
| 4 | 2 秒 |
| 5 | 4 秒 |
| 6 | 8 秒 |
| 7 | 16 秒 |
| … | … |
| 13+ | 900 秒(封顶) |
成功登录后,该 IP 和 username 的退避状态**立即清零**。
#### 登录端点行为(`POST /api/auth/login`
1. 先查退避(在窗口内直接 `429`**不验密码**,节省 Argon2 计算并防止枚举)
2. 验密码失败 → 记一次失败(IP + username 各记),返回 `401`
3. 密码正确但 TOTP 启用且缺少 `totp_code` → 返回 `401 {totp_required: true}`,**不记失败**(这是正常两步流程的第一步,不是攻击信号)
4. 密码正确但 TOTP 验证失败 → 记一次失败,返回 `401`
5. 全部通过 → 清零退避状态,发 session cookie
#### 配置项
| 配置项 | 类型 | 默认 | 说明 |
| --- | --- | --- | --- |
| `AUTH_LOGIN_THROTTLE_ENABLED` | boolCONFIG_FIELDS | `true` | 全局开关,关闭后整个退避机制 no-op |
| `AUTH_TRUST_FORWARDED_FOR` | bool`.env` 部署级) | `false` | `true` 时取 `X-Forwarded-For` 最左作为 client IP;**反代后必须显式开启**,否则反代 IP 全视为同一 IP,按 IP 退避失效 |
> **反代部署注意**`AUTH_TRUST_FORWARDED_FOR` 默认 `false`(直接用 socket IP)。部署在 nginx 等反代后面时,应设 `AUTH_TRUST_FORWARDED_FOR=true`,并确保反代正确设置 `X-Forwarded-For`
---
### 2. CLI 逃生通道
#### 设计原则
CLI 通过 **直连本地 DB**`get_session_local()`)工作,**无需 HTTP 服务器运行、无需任何已存凭据**(密码、恢复码均不需要)。拿到服务器 CLI 权限本身就意味着对系统有完全控制,因此这是可接受且必须存在的最终逃生口。
CLI 只动 auth 相关行(`auth_users` 的密码/TOTP 字段、`auth_login_throttle``auth_recovery_code`),**绝不**触碰用户数据表(`location``poo_records``public_ip_state` 等)。
入口:`python -m scripts.admin_cli <command>`
#### 子命令
| 命令 | 用途 | 逃生场景 |
| --- | --- | --- |
| `reset-password <username> [--password <pwd>]` | 重置密码(不带 `--password` 则交互式输入,不回显) | 忘记密码 |
| `unlock [--all \| --ip <ip> \| --username <u>]` | 清 `auth_login_throttle` 行 | 被退避锁住(429)时解锁 |
| `disable-totp <username>` | 关停 TOTP(清 secret + 删全部恢复码),**零凭据可执行** | 恢复码全丢也能进 |
| `reissue-totp <username>` | 生成新 TOTP secret 并打印 `otpauth://` URI | 设备丢失、需重新注册 Authenticator |
| `list-admin` | 列出所有用户与状态列 | 排障用 |
#### 使用示例
```bash
# 重置密码(不带 --password 时交互式 prompt,不回显)
python -m scripts.admin_cli reset-password admin
# 重置密码(非交互,脚本里用)
python -m scripts.admin_cli reset-password admin --password "newpassword"
# 解锁所有退避
python -m scripts.admin_cli unlock --all
# 解锁特定 IP
python -m scripts.admin_cli unlock --ip 1.2.3.4
# 解锁特定 username
python -m scripts.admin_cli unlock --username admin
# 关停 TOTP(零凭据,最终逃生口)
python -m scripts.admin_cli disable-totp admin
# 重新发放 TOTP secret(打印新 otpauth:// URI,扫码重新注册)
python -m scripts.admin_cli reissue-totp admin
# 查看用户列表
python -m scripts.admin_cli list-admin
```
在 Docker 容器内执行:
```bash
docker compose exec app python -m scripts.admin_cli <command>
```
#### `reissue-totp` 语义说明
- 对**已启用** TOTP 的用户:新 secret **立即在登录时生效**,旧 Authenticator 生成的码立即失效;**无需**再走 web `enable` 步骤。
- 现有恢复码**不被删除**——恢复码是独立的随机哈希值,与 TOTP secret 无密码学绑定,reissue 后恢复码依然有效(仍可用于登录)。
- 如果需要完整清理(secret + 所有恢复码),使用 `disable-totp`
---
### 3. 可选 TOTP 二次验证
#### 设计
- TOTP 遵循 RFC 6238,使用 `pyotp` 库。
- 二维码在**前端**由 `qrcode.react` 渲染(后端只返回 `otpauth://` URI,不引图像依赖)。
- `totp_secret` 明文存库(与项目其他 secret 处理一致,靠数据库文件权限保护)。
- 恢复码以 Argon2 哈希存库,一次性(使用后标记 `used_at`)。
#### 启用流程
1. **setup**`POST /api/auth/totp/setup`):生成 pending secret,返回 secret、`otpauth://` URI、10 个明文恢复码(**仅此一次**);此时 `totp_enabled` 仍为 `false`
2. **扫码**:用 Authenticator App 扫前端渲染的二维码(或手动输入 secret)。
3. **enable**`POST /api/auth/totp/enable`):输入当前 6 位码确认 → `totp_enabled` 变为 `true`
4. 妥善保存恢复码(不会再次展示)。
#### 停用流程
**web 停用**`POST /api/auth/totp/disable`):需提供当前密码或当前 6 位 TOTP 码。成功后清 secret、删全部恢复码,恢复纯密码登录。
**CLI 停用**(逃生口,恢复码全丢时):`python -m scripts.admin_cli disable-totp admin`,零凭据,立即生效。
#### 登录二步流程
1. 提交 `POST /api/auth/login {username, password}` → 密码正确但 TOTP 已启用 → `401 {totp_required: true}`(不发 session
2. 前端切到第二屏,提交 `POST /api/auth/login {username, password, totp_code}` → 通过 → 发 session cookie
`totp_code` 可以是 6 位 TOTP 动态码,也可以是 `xxxx-xxxx` 格式恢复码(命中即消费,不可复用)。
#### API 端点
| 端点 | 用途 |
| --- | --- |
| `POST /api/auth/totp/setup` | 生成 pending secret + URI + 恢复码(一次性明文返回) |
| `POST /api/auth/totp/enable` | 带当前 6 位码确认启用 |
| `POST /api/auth/totp/disable` | 带密码或当前码停用 |
| `GET /api/auth/totp` | 返回当前 TOTP 状态(`{enabled: bool}`),不返回 secret/恢复码 |
全部端点需要 session cookie`GET /api/auth/totp` 不需 CSRF;其余写端点需 `X-CSRF-Token`)。
#### 配置项
| 配置项 | 类型 | 默认 | 说明 |
| --- | --- | --- | --- |
| `AUTH_TOTP_ISSUER` | str`.env` 部署级) | 空(回退 `app_name` | 显示在 Authenticator App 里的 issuer 标签 |
---
## 受保护范围
当前 JSON API 端点(`/api/*`)需要 session cookie;写端点需额外携带 `X-CSRF-Token` header。
裸 ingestion 端点(`/location/record``/poo/record` 等设备调用端点)暂未收口到 session 保护(M3 计划引入 token 鉴权)。
## 下一步(不在当前范围)
- M3token 鉴权(供脚本 / 设备 / 移动端调用 API),ingestion 端点收口。
+129
View File
@@ -0,0 +1,129 @@
# 设计文档与多模型协作约定
本目录把 `docs/roadmap.md` 里的三个里程碑展开成**可被 coding agent 流水线执行**的详细设计文档。
- [`m1-db-consolidation.md`](./m1-db-consolidation.md) — 单库化地基
- [`m2-frontend-v2.md`](./m2-frontend-v2.md) — React SPA 前端 v2
- [`m3-token-mobile.md`](./m3-token-mobile.md) — token 鉴权与移动端(远期)
- [`m4-login-hardening.md`](./m4-login-hardening.md) — 登录加固(防爆破/指数退避 + CLI 逃生 + 可选 TOTP**先做**
- [`m5-iot-energy.md`](./m5-iot-energy.md) — IoT 集成与能耗采集(Modbus/Energy + MQTT/HA Discovery + 前端侧边栏)
本文件定义**所有任务共用的格式与协作规则**,三个里程碑文档不再重复这些约定。
---
## 1. 协作模型:Orchestrator → Implementer → Reviewer
设计目标是让三个不同档位的模型分工协作:
| 角色 | 模型档位 | 职责 |
| --- | --- | --- |
| **Orchestrator(编排者)** | 强模型 | 读里程碑文档,挑出依赖已满足的下一个原子任务,派发给 implementer;收到 reviewer 的 PASS 后推进下一个任务 |
| **Implementer(实现者)** | 便宜模型 | 一次只做**一个**原子任务,严格按任务卡执行,跑完本地校验后回报 |
| **Reviewer(评审者)** | 强模型 | 对照验收标准 + Reviewer checklist 独立复核、独立跑校验闸门,返回 `PASS` 或一份编号返工清单 |
### 循环
```
Orchestrator 选任务 T(其 Depends 全部为 done
Implementer 实现 T ──► 跑校验闸门 ──► 回报 diff + 校验输出
Reviewer 复核 T
├── PASS ─────────► Orchestrator 标记 T 为 done,进入下一个任务
└── REWORK[1..n] ─► 退回 Implementer,按编号逐条修,直到 PASS
```
### 角色边界(重要)
- **Implementer 不得扩大范围**:只能改任务卡 `Files` 里列出的文件;超出范围的问题要在回报里以 `OUT-OF-SCOPE:` 标注,交给 Orchestrator 决定是否新开任务,而不是顺手改掉。
- **Reviewer 必须独立重跑校验闸门**,不能只信 implementer 的回报。
- **Reviewer 的返工清单必须可执行、带编号**`REWORK 1: ...`),不写主观感受。
- 一个任务**不通过校验闸门就不算完成**,Orchestrator 不得跳过。
---
## 2. 原子任务的定义
一个"原子任务"必须同时满足:
1. **单一关注点**:一个任务只解决一件事(一次 schema 变更、一个端点、一个模块迁移……)。
2. **PR 大小**:理想 diff < ~200 行(结构性 sweep 任务可放宽,但应在卡上标 `[structural]` 并优先派给较强 implementer)。
3. **边界处可绿**:任务完成时,整个仓库通过校验闸门(见下)。**一个任务"拥有"它所改代码对应的测试**——如果改动会让某些现有测试失败,修这些测试就属于这个任务的范围,不允许留红给下一个任务。
4. **可独立验收**:验收标准是客观、可机械检查的断言,不依赖人的主观判断。
5. **依赖显式**:通过 `Depends` 字段声明前置任务。没有声明依赖的任务,Orchestrator 可并行派发。
---
## 3. 任务卡格式
每个任务在里程碑文档中以如下结构出现。Implementer 和 Reviewer 都只需要任务卡 + 本约定文件,**不需要读其它任务**即可工作。
```markdown
### M{n}-T{nn} — <标题> [structural?]
- **Status**: `todo` | `in-progress` | `in-review` | `done`
- **Depends**: M{n}-T{nn}, …(或 `none`
- **Context**: 1–2 句,为什么要做这个、它在里程碑里的位置。
**Files**(精确到路径,标注动作)
- `create path/to/new_file.py`
- `modify path/to/existing.py`
- `delete path/to/old.py`
**Steps**(便宜模型可直接照做的有序步骤)
1. …
2. …
**Out of scope / 不要碰**
- …(明确列出容易被误改的相邻区域,约束便宜模型漂移)
**Acceptance criteria**(客观、可勾选;Reviewer 逐条核)
- [ ] …
- [ ] 校验闸门全绿(见 §4
**Reviewer checklist**(除验收标准外,强模型重点看的点)
- …
```
---
## 4. 校验闸门(每个任务结束都要全绿)
在仓库根目录、激活 `.venv` 后执行:
```bash
# 1) 单元 / 集成测试(CI 同款,权威闸门)
pytest
# 2) Lintpyproject 已配置 ruffline-length=100
ruff check .
# 3) 若本任务改动了任何 HTTP 路由 / schema:重导出 OpenAPI 并确认已提交
python scripts/export_openapi.py
git diff --exit-code openapi/ # 必须无未提交差异
```
- `pytest` 是**权威闸门**`.github/workflows/pytest.yml` 跑的就是它,任何任务都不得让它变红。
- 改了路由 / Pydantic schema 的任务,`openapi/openapi.json``openapi/openapi.yaml` 必须在同一任务里重新生成并提交(`openapi/` 纳入版本控制)。
- 前端(M2/M3)相关任务的前端侧闸门(lint / typecheck / build)在对应里程碑文档里单独定义。
---
## 5. 提交与集成约定
- 每个任务一个 commitmessage 前缀任务 ID,例如:`M1-T03: unify data layer onto single app DB engine`
- 一个里程碑在一个 feature 分支上推进(如 `feature/m1-db-consolidation`),按任务依赖顺序合并。
- 任务卡里的 `Status` 字段由 Orchestrator 维护,作为流水线的单一进度源。
- 涉及**不可逆 / 数据破坏**的步骤(删旧 DB 文件、删 Grafana volume 等)一律不进自动化任务,只在文档里标为人工步骤(见 M1 的"人工操作"小节)。
---
## 6. 数据安全红线(贯穿所有里程碑,不可违反)
1. **任何脚本 / migration 都不得删除或覆盖用户数据文件**(旧 `.db`、备份、volume)。删除只能是人工、事后、确认无误的独立步骤。
2. 涉及历史数据的迁移**先在备份副本上演练**,再对真实库执行。
3. 数据迁移脚本必须**幂等**且**搬完对账行数**,对不上立即中止并非零退出。
4. 破坏性 Reviewer 一票否决:只要任务里出现"删文件 / drop 有数据的表 / truncate"Reviewer 直接 REWORK,要求改为人工步骤。
+348
View File
@@ -0,0 +1,348 @@
# M1 — 单库化地基(DB Consolidation
> 阅读前提:先读 [`README.md`](./README.md)(协作模型、任务卡格式、校验闸门、数据安全红线)。本文档只展开 M1 的现状、目标与原子任务。
## 1. 目标
把 location、poo 两个独立 SQLite 库合并进 `app.db`,收敛成**单库 + 单 engine + 单 DeclarativeBase + 单 Alembic 链**,清理项目早期散落的数据层代码,并移除 Grafana。历史数据零丢失。
## 2. 现状(实现者可据此工作,不必重新通读全仓库)
**三套数据层(散落点)**
- `app/db.py``Base` + `engine`/`SessionLocal`/`get_db_session` —— 实际绑定 `settings.location_database_url`(即 location 库,命名有误导性)。
- `app/models/base.py`:仅 `from app.db import Base` 转出。
- `app/poo_db.py``PooBase` + `poo_engine`/`PooSessionLocal`/`get_poo_db_session` —— 绑定 `poo_database_url`
- `app/auth_db.py``AuthBase` + 带 `lru_cache``_get_auth_engine` / `get_auth_session_local` / `reset_auth_db_caches` / `get_auth_db_session` —— 绑定 `app_database_url`(真正的 app 库)。
**模型与归属**
- `app/models/auth.py``AuthUser``AuthSession``AuthBase` → app 库)
- `app/models/config.py``AppConfigEntry``AuthBase` → app 库)
- `app/models/public_ip.py``PublicIPState``PublicIPHistory``AuthBase` → app 库)
- `app/models/location.py``Location``Base` → location 库),表 `location`PK(`person`,`datetime`)`latitude`/`longitude` NOT NULL`altitude` nullable
- `app/models/poo.py``PooRecord``PooBase` → poo 库),表 `poo_records`PK(`timestamp`)`status`/`latitude`/`longitude` NOT NULL
- `app/models/__init__.py`:导出除 `PooRecord` 外的模型(`PooRecord` 单独存在)
**三条 Alembic 链**
- `alembic_app.ini` + `alembic_app/``env.py``AuthBase.metadata`),head = `20260429_05_public_ip_monitor`
- `alembic_location.ini` + `alembic_location/`head = `20260419_01_location_baseline`
- `alembic_poo.ini` + `alembic_poo/`head = `20260420_01_poo_baseline`
**adoption / 启动链路**
- `scripts/app_db_adopt.py`(常量 `APP_BASELINE_REVISION = "20260429_05_public_ip_monitor"`
- `scripts/location_db_adopt.py``scripts/poo_db_adopt.py`(含 legacy 校验:`EXPECTED_USER_VERSION`、表结构断言)
- `scripts/run_migrations.py`:依次调用三个 adopt 函数,返回 `{"app","location","poo"}`
- `app/main.py` lifespan`ensure_runtime_dirs`app/location/poo 三路径)、`ensure_auth_db_ready``ensure_location_db_ready``ensure_poo_db_ready`,再起 APScheduler 每 4h 检查 public IP
**依赖与路由**
- `app/dependencies.py``get_auth_db`app session)、`get_db`location session)、`get_poo_db`poo session)、`get_app_settings``get_current_auth_session``get_homeassistant_client``get_ticktick_client`
- `app/api/routes/location.py``POST /location/record`,依赖 `get_db`**无鉴权**
- `app/api/routes/poo.py``POST /poo/record``GET /poo/latest`,依赖 `get_poo_db`**无鉴权**
- `app/api/routes/homeassistant.py`:同时用 `get_db`location)和 `get_poo_db`
**config**
- `app/config.py``app_database_url` / `location_database_url` / `poo_database_url` 三字段 + computed `app_sqlite_path` / `location_sqlite_path` / `poo_sqlite_path`
- `app/services/config_page.py``build_runtime_settings` 用到 `reset_auth_db_caches`;配置页 sections 暴露 `location_database_url` / `poo_database_url`(约 263264 行)
**测试耦合点(M1 必然要改)**
- `tests/conftest.py``test_database_urls` 设三套环境变量;`ready_location_database` / `ready_poo_database` / `auth_database` / `location_client`monkeypatch `app_db.engine`/`SessionLocal`/ `poo_client`monkeypatch `poo_db.poo_engine`/`PooSessionLocal`
- `tests/test_location.py` / `tests/test_poo.py`:用上述 client + 各自 adopt 脚本的 adoption 测试
- `tests/test_deployment.py`:断言 `run_all_migrations()` 返回 `{app,location,poo}` 三库各自 revision;断言 entrypoint 不含 `*_db_adopt`
- `tests/test_homeassistant_inbound.py`monkeypatch `app.poo_db`
- `tests/test_config.py` / `tests/test_public_ip.py` / `tests/test_smtp.py`:硬编码三套 URL / 路径
- `reset_auth_db_caches``conftest``test_app``test_auth``test_deployment``test_ticktick` 引用
## 3. 目标架构(M1 完成态)
**单数据层 `app/db.py`**
```python
class Base(DeclarativeBase): ...
# 绑定 settings.app_database_url 的 cached engine;建连时启用 WALPRAGMA journal_mode=WAL
def get_engine() -> Engine: ...
def get_session_local() -> sessionmaker: ...
def reset_db_caches() -> None: ...
def get_db_session() -> Generator[Session, None, None]: ...
```
- 所有模型(auth / config / public_ip / location / poo)都继承这一个 `Base`
- 删除 `app/auth_db.py``app/poo_db.py``app/models/base.py`
- 依赖收敛为**单一** `get_db`app session);移除 `get_poo_db`、旧 `get_auth_db`
- 一条 Alembic 链(`alembic_app`),`location` / `poo_records` 成为其管理对象;删除 `alembic_location*` / `alembic_poo*`
- `config.py` 只保留 `app_database_url`;移除 location/poo 的 url 与 path。
- `docker-compose.yml` 去掉 grafana service;删除 `grafana/`
- 数据搬迁由 `scripts/migrate_legacy_data.py` 一次性完成(不进 Alembic 链)。
## 4. 任务依赖图
```
T01 (app 链建 location+poo 空表)
├─► T02 (数据搬迁脚本) # 逻辑上需要新表存在
└─► T03 [structural] (统一数据层/模型/依赖/路由)
└─► T04 (lifespan + run_migrations 收敛, 删 adopt 脚本)
└─► T05 (config 去 location/poo url + 配置页 + 测试硬编码)
T06 (删 Grafana) # 独立, 可并行
T07 (文档 + OpenAPI 重导出) # 收尾, 依赖 T03/T04/T05
```
`T01``T06` 无前置可先开;`T02` 依赖 `T01``T03` 依赖 `T01``T04`/`T05` 依赖 `T03``T07` 最后。
---
## 5. 原子任务
### M1-T01 — app 链新增 revision:建 `location` + `poo_records` 空表 `[schema]`
- **Status**: `todo`
- **Depends**: none
- **Context**: 让 app 库的 Alembic 链能建出这两张表,schema 与旧库**完全一致**。本任务只动 schema,不搬数据、不移模型。
**Files**
- `create alembic_app/versions/20260611_06_merge_location_poo_tables.py`
- `modify scripts/app_db_adopt.py`(更新 `APP_BASELINE_REVISION`
**Steps**
1. 新 revision`revision = "20260611_06_merge_location_poo_tables"``down_revision = "20260429_05_public_ip_monitor"`
2. `upgrade()``op.create_table` 手写建 `location``poo_records`,列/约束严格照抄现有 baseline(`location`: person TEXT, datetime TEXT, latitude REAL NOT NULL, longitude REAL NOT NULL, altitude REAL nullable, PK(person,datetime)`poo_records`: timestamp TEXT, status TEXT, latitude REAL NOT NULL, longitude REAL NOT NULL, PK(timestamp))。
3. `downgrade()``op.drop_table("poo_records")` + `op.drop_table("location")`
4. 把 `scripts/app_db_adopt.py``APP_BASELINE_REVISION` 更新为新 head。
**Out of scope / 不要碰**
- 不要把 `Location` / `PooRecord` 模型改到 app Base(那是 T03)。
- 不要触碰 `alembic_location*` / `alembic_poo*`T03/T04 删)。
- 不要在本 revision 里写任何数据拷贝。
**Acceptance criteria**
- [ ] 在一个全新临时 app 库上 `command.upgrade(alembic_app head)` 后,`sqlite_master``location``poo_records`、且与旧 baseline 表结构一致(`PRAGMA table_info` 对齐)。
- [ ] `downgrade -1` 能干净回滚这两张表。
- [ ] `APP_BASELINE_REVISION == "20260611_06_merge_location_poo_tables"`
- [ ] 校验闸门全绿(`pytest``test_deployment` 对 app head 的断言仍通过,因为它用的是常量)。
**Reviewer checklist**
- 表结构与旧 baseline **逐列逐约束**一致(类型 TEXT/REAL、nullable、PK 顺序)。
- `down_revision` 正确指向旧 head,链上只有一个 head。
---
### M1-T02 — 数据搬迁脚本 `scripts/migrate_legacy_data.py`
- **Status**: `todo`
- **Depends**: M1-T01
- **Context**: 把旧 `locationRecorder.db` / `pooRecorder.db` 的行幂等拷进 app 库的新表,搬完对账。**不进 Alembic 链**,人工运行一次。
**Files**
- `create scripts/migrate_legacy_data.py`
- `create tests/test_migrate_legacy_data.py`
**Steps**
1. 入口 `migrate_legacy_data(app_url, location_url, poo_url, *, dry_run=False) -> dict`CLI 默认从 env 读三个 url(即便 location/poo url 已从 `Settings` 移除,本脚本可直接读环境变量或接受 `--location-db`/`--poo-db` 参数,保持自包含)。
2. 对每个旧库:若文件不存在 → 该表 `skipped`(**不报错**,保证 CI / 全新部署可安全 no-op)。
3. 拷贝用 SQLite `ATTACH DATABASE '<old>' AS legacy` + `INSERT OR IGNORE INTO main.<table> SELECT <显式列> FROM legacy.<table>`(显式列名,禁用 `SELECT *`)。`INSERT OR IGNORE` 保证幂等(PK 冲突跳过)。
4. 搬完对账:对每张表比对 `源行数``目标行数中来自源的部分`;目标行数 < 源行数则 `raise` 并以非零码退出。
5. `dry_run` 模式只读统计、不写入。
6. 打印每表结果:`{location: {source, copied, skipped, final}, poo_records: {...}}`
**Out of scope / 不要碰**
- **绝不** `os.remove` / 覆盖任何旧文件(数据安全红线)。
- 不修改 Alembic 链,不在 app 启动链路里调用本脚本。
- 不改 `config.py`
**Acceptance criteria**
- [ ] 单测:给定含 N 行的临时旧库 + 已 upgrade 的临时 app 库,运行后 app 库对应表有 N 行;**再运行一次**仍是 N 行(幂等)。
- [ ] 单测:旧库文件不存在时该表返回 `skipped`,不抛异常,app 库该表保持为空。
- [ ] 单测:构造"目标缺行"场景,断言对账失败抛错且退出码非零。
- [ ] 脚本中不出现任何文件删除/覆盖调用(`grep -nE "os\.remove|unlink|shutil|truncate|DROP TABLE" scripts/migrate_legacy_data.py` 为空)。
- [ ] 校验闸门全绿。
**Reviewer checklist**
- 幂等机制确实是 PK 冲突安全(`INSERT OR IGNORE` 或等价 upsert),不是靠"先清空目标"。
- 对账逻辑会在丢行时**真的中止**(非零退出),不是只打印 warning。
- 列名显式,与两表 schema 完全对应。
---
### M1-T03 — 统一数据层、模型、依赖、路由到单库 `[structural]`
- **Status**: `todo`
- **Depends**: M1-T01
- **Context**: M1 的核心 sweep。把三套 engine/Base/session 收敛成 `app/db.py` 一套(绑 app 库、开 WAL),所有模型挂到同一个 `Base`,依赖收敛为单一 `get_db`,所有路由改用它。**本任务必须原子落地**——删除旧模块会同时打断所有 importer,无法分多次保持绿色。Orchestrator 可按下方 Steps 的自然分段派给较强 implementer。
**Files**
- `modify app/db.py`(改写为统一数据层:`Base` + 绑 `app_database_url` 的 cached engine + WAL + `get_session_local` + `reset_db_caches` + `get_db_session`
- `delete app/auth_db.py`
- `delete app/poo_db.py`
- `delete app/models/base.py`
- `modify app/models/location.py``from app.db import Base`
- `modify app/models/poo.py`(改继承统一 `Base`import 改 `app.db`
- `modify app/models/auth.py``app/models/config.py``app/models/public_ip.py``AuthBase` → 统一 `Base`
- `modify app/models/__init__.py`(补导出 `PooRecord`,保证 `from app import models` 注册所有表到同一 metadata
- `modify app/dependencies.py`(单一 `get_db`;删 `get_poo_db``get_app_settings`/`get_current_auth_session` 改用 `get_db`
- `modify app/api/routes/auth.py``pages.py``public_ip.py``ticktick.py``get_auth_db``get_db`
- `modify app/api/routes/location.py``poo.py``homeassistant.py`location/poo session 改用 `get_db`;删 `get_poo_db` 引用)
- `modify app/services/config_page.py``reset_auth_db_caches``reset_db_caches`
- `modify app/main.py``import app.auth_db as auth_db` → 统一层;`get_auth_session_local``get_session_local`
- `modify tests/conftest.py``tests/test_app.py``tests/test_auth.py``tests/test_ticktick.py``tests/test_homeassistant_inbound.py``tests/test_location.py``tests/test_poo.py`import sweep + 把 location/poo client 改成写 app 库的统一 session;移除对 `app.poo_db`/`app.db`(location) monkeypatch 的依赖)
**Steps**
1. 改写 `app/db.py``Base(DeclarativeBase)`;沿用 `auth_db.py` 的 cached-engine + reset 模式但绑 `app_database_url`;为 sqlite 连接注册 `PRAGMA journal_mode=WAL`(用 `event.listens_for(engine, "connect")` 或建连后执行)。导出 `get_engine`/`get_session_local`/`reset_db_caches`/`get_db_session`
2. 模型 sweep:所有 `from app.auth_db import AuthBase` / `from app.poo_db import PooBase` / `from app.db import Base` 统一成 `from app.db import Base`;类继承统一 `Base``app/models/__init__.py` 增加 `from app.models.poo import PooRecord` 并补进 `__all__`
3. 删 `app/auth_db.py``app/poo_db.py``app/models/base.py`
4. 依赖 sweep`app/dependencies.py` 留单一 `get_db`yield 统一 session),删 `get_poo_db``get_app_settings``get_current_auth_session``Depends(get_auth_db)``Depends(get_db)`
5. 路由 sweep:所有 `Depends(get_auth_db)``Depends(get_poo_db)``Depends(get_db)` 统一为 `Depends(get_db)`(变量名 `auth_db_session`/`poo_db`/`db` 可保留,不强制改)。
6. `app/services/config_page.py``reset_auth_db_caches``reset_db_caches`
7. `app/main.py`:把 `_run_scheduled_public_ip_check` / `ensure_auth_db_ready` 里的 `auth_db.get_auth_session_local()` 换成统一 `get_session_local()`。(lifespan 里 location/poo 的 ready 检查留到 T04 删。)
8. 测试 sweep`reset_auth_db_caches``reset_db_caches`6 个文件);conftest 的 `location_client`/`poo_client` 改成"写入统一 app session 即可"的形式(不再 monkeypatch 已删除的 `app.poo_db`/location `app.db`);`test_homeassistant_inbound` 同理。
**Out of scope / 不要碰**
- 不删 `scripts/location_db_adopt.py` / `scripts/poo_db_adopt.py`,不改 lifespan 的 location/poo ready 调用(那是 T04,避免与本任务交叉冲突)。
- 不动 `config.py` 的字段(T05)。
- 不改业务逻辑(service 内部算法、HA 集成行为保持不变)。
**Acceptance criteria**
- [ ] `grep -rnE "auth_db|poo_db|PooBase|AuthBase|get_auth_db|get_poo_db|reset_auth_db_caches|app\.models\.base" app | grep -v __pycache__` 结果为空。
- [ ] `app/db.py` 的 engine 绑定 `app_database_url`sqlite 下 `PRAGMA journal_mode` 实测为 `wal`
- [ ] 所有模型 `Base.metadata.tables` 同时包含 auth/config/public_ip/location/poo_records 五类表。
- [ ] `pytest` 全绿(含 location/poo/homeassistant_inbound 测试在单库下通过)。
- [ ] `ruff check .` 无新增告警。
**Reviewer checklist**
- WAL 真的生效(实际连接 `PRAGMA journal_mode` 返回 `wal`),不是只写了注释。
- location/poo 的读写在单库下行为不变(端点仍返回 200、行落库)。
- 没有遗留指向已删模块的死 import;没有把业务逻辑顺手改了。
- `get_db` 现在产出的是 app 库 session(不是旧 location 库)。
---
### M1-T04 — 收敛启动链路:lifespan + run_migrations,删除 location/poo adopt 脚本
- **Status**: `todo`
- **Depends**: M1-T03
- **Context**: 单库后只需保证 app 库就绪;location/poo 的 adoption 链路整条退役。
**Files**
- `modify app/main.py``ensure_runtime_dirs` 只建 app 路径;删 `ensure_location_db_ready`/`ensure_poo_db_ready` 及其调用与 import
- `modify scripts/run_migrations.py`(只 `adopt_or_initialize_app_db`,返回 `{"app": ...}`
- `delete scripts/location_db_adopt.py`
- `delete scripts/poo_db_adopt.py`
- `delete alembic_location.ini``alembic_location/`(含 env.py、versions
- `delete alembic_poo.ini``alembic_poo/`
- `modify tests/test_deployment.py``run_all_migrations` 期望值改为单 `{"app": ...}`;删/改 legacy location/poo 迁移断言;保留"app DB 不存在则 fail-closed"用例)
- `modify tests/test_location.py``tests/test_poo.py`(删除针对已删 adopt 脚本的 adoption 测试;保留端点行为测试)
- `modify tests/conftest.py`(删 `_make_alembic_config`/`_make_poo_alembic_config`/`ready_location_database`/`ready_poo_database` 等已无意义的 fixture
**Steps**
1. `app/main.py`:移除 `from scripts.location_db_adopt ...` / `poo_db_adopt` import;删两个 `ensure_*_db_ready` 函数及 lifespan 中调用;`ensure_runtime_dirs` 只处理 `settings.app_sqlite_path`
2. `scripts/run_migrations.py``run_all_migrations` 只返回 app 一项。
3. 删除两套 adopt 脚本与两套 alembic 环境/ini。
4. 测试:把 `test_migration_runner_*` 改成单库口径;删掉引用已删脚本常量(`LOCATION_BASELINE_REVISION` 等)的用例。
**Out of scope / 不要碰**
- 不动 `scripts/app_db_adopt.py` 的核心逻辑(仅 T01 已更新其常量)。
- 不动数据搬迁脚本(T02)。
**Acceptance criteria**
- [ ] `grep -rnE "location_db_adopt|poo_db_adopt|alembic_location|alembic_poo" app scripts tests | grep -v __pycache__` 为空。
- [ ] 仓库不再有 `alembic_location*` / `alembic_poo*` 文件。
- [ ] `python -m scripts.run_migrations` 在全新临时 app 库上成功初始化(含 location/poo_records 表)。
- [ ] 校验闸门全绿。
**Reviewer checklist**
- lifespan 仍对 app 库 fail-closed(缺库时明确报错),未弱化启动安全。
- 没有残留对已删 alembic 环境的引用(包括 `.ini` 路径字符串)。
---
### M1-T05 — config 去除 location/poo URL 与路径,清理配置页与测试硬编码
- **Status**: `todo`
- **Depends**: M1-T03
- **Context**: 配置层只剩 `app_database_url`,运行时不再有 location/poo 库概念。
**Files**
- `modify app/config.py`(删 `location_database_url`/`poo_database_url` 字段与 `location_sqlite_path`/`poo_sqlite_path` computed 属性)
- `modify app/services/config_page.py`(配置页 sections 移除 `location_database_url`/`poo_database_url` 展示项)
- `modify .env.example`(移除两行 legacy DB URL;保留 `APP_DATABASE_URL`
- `modify tests/test_config.py`(删对两个 URL/路径的断言)
- `modify tests/test_public_ip.py``tests/test_smtp.py`(构造 `Settings` 时去掉 location/poo url 入参)
- `modify tests/conftest.py``test_database_urls` 不再 set `LOCATION_DATABASE_URL`/`POO_DATABASE_URL`
**Out of scope / 不要碰**
- 不动 `migrate_legacy_data.py`(它自带读旧库路径的能力,与 `Settings` 解耦)。
- 不改其它配置项(SMTP / TickTick / HA 等)。
**Acceptance criteria**
- [ ] `grep -rnE "location_database_url|poo_database_url|location_sqlite_path|poo_sqlite_path" app tests | grep -v __pycache__` 为空。
- [ ] 配置页渲染不再出现 location/poo DB URL 字段。
- [ ] 校验闸门全绿。
**Reviewer checklist**
- 没有别的代码还假设 `Settings` 上存在这两个属性(运行期不会 AttributeError)。
---
### M1-T06 — 移除 Grafana
- **Status**: `todo`
- **Depends**: none(可与 T01 并行)
- **Context**: 可视化将由 M2 的 React 承担;Grafana 直接删除,不再 re-point。
**Files**
- `modify docker-compose.yml`(删 `grafana` service 及其 `depends_on`/挂载;删顶层 `volumes.homeautomation_grafana_storage`
- `delete grafana/``provisioning/``dashboards/` 全部)
- `modify tests/test_deployment.py`(若有针对 grafana service 的断言则同步移除)
- `modify README.md`(删"Grafana Provisioning"整节——也可并入 T07,二选一,避免重复改同段)
**Out of scope / 不要碰**
- **不在脚本里删除** named volume `homeautomation_grafana_storage` 的实际数据卷——这是人工 ops 步骤(见 §6),compose 里移除声明即可。
- 不动 app/migration service。
**Acceptance criteria**
- [ ] `docker-compose.yml` 不再含 `grafana``homeautomation_grafana_storage`
- [ ] 仓库不再有 `grafana/` 目录。
- [ ] `docker compose config` 能成功解析(语法有效)。
- [ ] 校验闸门全绿。
**Reviewer checklist**
- 没有遗留对 `./grafana/...` 挂载路径的引用。
- 没有顺手删 `./data` 卷或改动 app service 端口/卷。
---
### M1-T07 — 文档与 OpenAPI 收尾
- **Status**: `todo`
- **Depends**: M1-T03, M1-T04, M1-T05
- **Context**: 让文档反映单库现实,并把"前后端不分离 / 三库不合并 / Grafana"约束在 architecture 文档中正式退役。
**Files**
- `modify README.md`(三库 → 单库;删 location/poo DB 初始化与 adopt 说明;更新"运行测试"段落使其与实际测试一致)
- `modify docs/architecture-overview.md`(退役"三库不合并"location/poo Alembic 链合并说明)
- `modify docs/roadmap.md`(勾掉 M1 范围项)
- `run python scripts/export_openapi.py` 并提交 `openapi/` 变更(location/poo 路由依赖在 T03 改过,schema 可能变化)
**Acceptance criteria**
- [ ] README / architecture 不再描述 location/poo 独立库与 adopt 脚本。
- [ ] `python scripts/export_openapi.py``git diff --exit-code openapi/` 无未提交差异。
- [ ] 校验闸门全绿。
**Reviewer checklist**
- 文档无残留的旧命令(`location_db_adopt.py` 等)。
- OpenAPI 已重导出且入库。
---
## 6. 人工操作 runbook(生产切换,不进自动化任务)
按数据安全红线,下列步骤由人执行,**不**写进 implementer 任务:
1. **备份**:停服前复制 `data/app.db``data/locationRecorder.db``data/pooRecorder.db` 到带时间戳的归档目录。
2. **演练**:把上述备份恢复到 scratch 目录,先在副本上跑完整流程(升级 + `migrate_legacy_data.py --dry-run` 再实跑),核对行数。
3. **部署新镜像**:新镜像的 migration job 会把 app 库升级到新 head,建出空的 `location` / `poo_records`
4. **搬数据**:在生产机运行 `python scripts/migrate_legacy_data.py`(指向归档前的旧库),核对对账输出。
5. **验证**app 起来后确认 location/poo 端点与历史查询正常、行数与旧库一致。
6. **(事后,确认无误再做)撤旧库**:归档旧 `.db` 文件、删除 `homeautomation_grafana_storage` 卷。**这一步人工、可回退地保留归档,永不在脚本中自动执行。**
## 7. 里程碑完成定义(Definition of Done
- 运行期只存在 `app.db` 一个库、一个 engine、一个 `Base`、一条 Alembic 链。
- `grep` 不到任何 `auth_db` / `poo_db` / location 独立库 / adopt 脚本 / grafana 的残留引用。
- 旧库历史数据已通过 `migrate_legacy_data.py` 搬入且对账通过。
- `pytest``ruff check .``export_openapi` 全绿且 `openapi/` 已入库。
- README / architecture / roadmap 反映单库现实。
+250
View File
@@ -0,0 +1,250 @@
# M2 — 前端 v2React SPA
> 阅读前提:先读 [`README.md`](./README.md)。M2 依赖 M1 完成(单库 + 干净的数据层 + API 建立在合并后的 schema 上)。
## 1. 目标
**React SPA** 取代现有 Jinja 页面,由 FastAPI **同源**托管(同一容器、同一 origin)。一步合并 roadmap 的"前端重写"与"前端做厚":配置界面 + 数据可视化(热力图 / 地图,接管 Grafana)+ 记录的按需展示与小幅增删改。
> **元目标(agentic 实验)**:这是用 agent 写 React 的试水,全程尽量不读代码。因此本里程碑**强约束 OpenAPI → 类型化 TS client 作为契约护栏**:后端 API 先稳,前端永远对着强类型契约写,便宜模型不易跑偏,reviewer 也有客观依据。
## 2. 现状(M1 完成后)
- 页面仍是服务端 Jinja`app/api/routes/pages.py``GET/POST /config``/``/admin``POST /config/smtp/test`+ `app/templates/``base/config/home/login.html``styles.css`)。
- 鉴权:`get_current_auth_session`(读 `auth_session_cookie_name` cookie),server-side session + 每会话 `csrf_token` 内嵌在表单。
- `app/main.py``app.mount("/static", StaticFiles(...))`
- 配置读写逻辑在 `app/services/config_page.py``build_config_sections` / `save_config_updates` / `build_runtime_settings`)。
- 业务数据:单库中的 `location``poo_records``public_ip_state``public_ip_history`
## 3. 目标架构
### 3.1 后端:JSON API + SPA 托管
- 所有数据交互走 **JSON API**,统一前缀 `/api`(SPA 是客户端渲染,必须有 API——这与"同源/同容器"无关)。
- FastAPI 既挂 `/api/*`,又挂 SPA 静态产物,并对非 `/api`、非静态资源的路径**回退到 `index.html`**(支持前端路由 deep-link)。
- Jinja 页面在 SPA 达到功能对齐后移除。
### 3.2 鉴权:复用 session cookie + SPA 版 CSRF
- 继续用现有 **HttpOnly session cookie**(同源自动携带),M2 **不引入 token**token 属 M3)。
- CSRF(已定·简化版):依赖 `SameSite=Lax` 的 session cookie——跨站发起的写请求(POST/PUT/PATCH/DELETE**不会自动携带 cookie**,经典 CSRF 主路已被堵;再要求所有写请求带一个**自定义 header**(跨站无 CORS 预检发不出,且本应用不对外站开放 CORS)作为纵深防御。**不做 per-session token 比对**(个人自用场景足够)。`GET /api/session` 仍保留,用途是返回当前登录用户、引导 SPA(不再以下发/校验 `csrf_token` 为目的)。
- 浏览器面向的所有新端点一律 session 保护;**裸 ingestion 端点(设备调用的 `POST /location/record``POST /poo/record`)维持现状到 M3**。
### 3.3 前端工程
- `frontend/`**Vite + React + TypeScript**。
- 组件库:**Mantine**(已定;批电池齐、TS 优先、视觉中性,最贴近此前 Vue 侧 Naive UI 的用法)。
- API client:由后端 `openapi/openapi.json` **自动生成** TS 类型与请求函数(如 `openapi-typescript` + 轻量 fetch 封装)。**生成物入库** + `npm run codegen` + CI 校验"生成物与 openapi 同步"(已定)。fetch 封装统一带 cookie、写请求注入自定义 CSRF header、401 跳登录。
- 可视化:**Leaflet**(已定)—— `react-leaflet` + `leaflet.heat`(热力图,**头号功能**+ `leaflet.markercluster`(点多时聚合)+ OSM 栅格瓦片(零 key)。**地图封在一个自包含组件后面**(如 `<RecordsMap points mode onSelect>`,全应用只此处 import leaflet),数据获取/时间窗 state 在外面;这样将来若要换 **MapLibre GL** 是被隔离的局部重写,不波及其它。
- 状态/数据请求:轻量即可(**TanStack Query**,已定),不引入重型框架。
### 3.4 构建与部署
- 多阶段 `Dockerfile`node 阶段 `npm ci && npm run build` → 把 `frontend/dist` 拷进 python 镜像的静态目录;运行镜像不带 node。
- compose 仍是单 app 容器(同源)。
## 4. API 契约(M2 要落地的端点)
> 全部 `/api` 前缀、session 保护、JSON 进出。具体 schema 在各任务里用 Pydantic 定义,并经 `export_openapi.py` 固化。
| 分组 | 端点 | 用途 |
| --- | --- | --- |
| 会话 | `GET /api/session` | 返回当前用户 + csrf_token;未登录 401 |
| 会话 | `POST /api/auth/login` | 账号密码登录,下发 session cookie |
| 会话 | `POST /api/auth/logout` | 注销 |
| 会话 | `POST /api/auth/password` | 改密(沿用现有强制改密语义)|
| 配置 | `GET /api/config` | 返回配置 sectionssecret 不回显)|
| 配置 | `PUT /api/config` | 保存配置(留空保留旧 secret 语义不变)|
| 配置 | `POST /api/config/smtp/test` | 触发测试发信 |
| 数据 | `GET /api/locations` | location 记录查询(时间范围/分页,供地图/热力图)|
| 数据 | `GET /api/poo` | poo 记录列表(分页)|
| 数据 | `GET /api/public-ip` | 当前状态 + 变化历史 |
| CRUD | `PATCH /api/locations/{person}/{datetime}` | 修正单条 location |
| CRUD | `DELETE /api/locations/{person}/{datetime}` | 删除单条 location |
| CRUD | `PATCH /api/poo/{timestamp}` | 修正单条 poo |
| CRUD | `DELETE /api/poo/{timestamp}` | 删除单条 poo |
> 记录 CRUD 依赖现有 PK 作行标识(location PK=`person+datetime`poo PK=`timestamp`)。路径参数需对 `datetime`/`timestamp` 做 URL 编码处理。
## 5. 已锁定决策(讨论后拍板)
> 以下为与项目所有者讨论后**已定**的选择。**线框图本里程碑不画**——按本节 + 各任务卡描述,由实现侧自行合理排版(含移动端布局)。
**技术选型**
1. **组件库 = Mantine**。批电池齐、TS 优先、视觉中性、文档好,最贴近此前 Naive UI 的用法,利于 agent 产出一致 UI。
2. **地图库 = Leaflet**`react-leaflet` + `leaflet.heat` + `leaflet.markercluster`,OSM 栅格、零 key)。**封在自包含组件后**,预留将来迁 MapLibre 的接缝(见 §3.3)。
3. **OpenAPI client = 生成物入库** + `npm run codegen` + CI 校验"与 openapi 同步"。
4. **CSRF = 简化版**`SameSite=Lax` cookie + 写请求带自定义 header**不做 per-session token**(见 §3.2)。
5. **前端栈**Vite + React + TS + TanStack Query + Mantine。
6. **Jinja**SPA 功能对齐后**全量移除** `templates/``pages.py`
**信息架构 / UX**
7. **首页主视图 = 地图(热力图为主)+ 时间范围选择器**。可视化优先级:**热力图(最重要)> 时间选择器(必须)> 散点点位/列表(辅助)**。
8. **列表 = 辅助页面,分页**(默认页大小 ~100、有上限;前端换页取数,不拉全量)。
9. **记录编辑/删除****location 靠点地图上的点**触发(不做 75k 行大列表);**poo 靠列表 + 地图点位**。
10. **配置入口**:config 作为普通页之一,由界面上一个**齿轮图标**进入。`/admin``/` 现状只是重定向到 `/config`SPA **不需要单独 admin 页**`/` 首页直接给地图主视图(概览 dashboard 列为**可选/后续**,非 M2 核心)。
11. **响应式 = 要**(手机浏览器可用、合理移动端布局)。**PWA** 列为近期 backlog(见 `docs/future-ideas.md`),M2 设计即按移动端友好铺路。
**范围边界**
12. **CRUD = 改非主键字段 + 删单行**;主键(location=`person+datetime`、poo=`timestamp`)**不可改**;**不提供 UI 新建**(记录由设备 ingestion 产生)。
13. **裸 ingestion 端点**`POST /location/record``POST /poo/record`)**维持现状到 M3**,本里程碑不加保护、不改动。
14. **trip / 轨迹连线**为**可选 / 后续**(5 分钟一点 + 手机记录较糙,先不做核心)。
> 项目定位:个人自用、家庭特化、不开源——设计可按单用户场景简化,不为通用性过度抽象。
## 6. 任务依赖图
```
后端 API(可与前端 scaffold 并行)
M2-T01 config API
M2-T02 session/auth API ─┐
M2-T03 data read API ├─► 都产出 OpenAPI 契约
M2-T04 record CRUD API │
M2-T05 smtp/action API ─┘
│ (openapi 稳定后)
M2-T06 前端 scaffold + codegen ──► M2-T07 auth UI
├─► M2-T08 config UI
├─► M2-T09 可视化 UI
└─► M2-T10 records 管理 UI
M2-T11 FastAPI 托管 SPA + 移除 Jinja(依赖 T07T10 达到对齐)
M2-T12 多阶段 Dockerfile + CI/compose
M2-T13 文档 + OpenAPI 收尾
```
---
## 7. 原子任务(任务卡)
> 后端任务沿用 M1 的校验闸门(`pytest` / `ruff` / `export_openapi`)。前端任务的闸门见 §8。
### M2-T01 — config JSON API
- **Status**: `done` · **Depends**: noneM1 完成后)
- **Context**: 把 `config_page` 的读写能力暴露成 JSON,复用现有 service,不重写业务逻辑。
- **Files**: `create app/api/routes/api/config.py``create app/schemas/config.py``modify app/main.py`(注册路由);`create tests/test_api_config.py`
- **Steps**: 用 `build_config_sections`/`save_config_updates` 包出 `GET/PUT /api/config`session 保护;secret 不回显、留空保留旧值语义照搬。
- **Acceptance**:
- [ ] 未登录访问 `GET /api/config` 返回 401。
- [ ] 登录后 `GET` 返回 sectionssecret 字段被遮罩。
- [ ] `PUT` 留空 secret 时保留旧值;非法值返回 4xx 且不写库。
- [ ] 校验闸门全绿(含 `openapi/` 重导出入库)。
- **Reviewer**: 复用了 service 而非复制逻辑;CSRF 校验存在;secret 不泄漏到响应或 OpenAPI 示例。
### M2-T02 — session / auth JSON API
- **Status**: `done` · **Depends**: none
- **Context**: 给 SPA 提供登录/注销/会话探测 + CSRF 下发。
- **Files**: `create app/api/routes/api/session.py``app/schemas/session.py``modify app/main.py``create tests/test_api_session.py`
- **Steps**: `GET /api/session`401 或 user+csrf)、`POST /api/auth/login``POST /api/auth/logout``POST /api/auth/password`,复用 `app/services/auth.py`
- **Acceptance**:
- [ ] 正确账号密码登录后置下 HttpOnly session cookie`GET /api/session` 返回 user + csrf_token。
- [ ] 错误凭据 401,不下发 cookie。
- [ ] 写端点缺 `X-CSRF-Token` 或不匹配 → 403。
- [ ] 强制改密语义与现有一致。
- [ ] 校验闸门全绿。
- **Reviewer**: cookie 仍 HttpOnly、`Secure` 跟随 `app_env``SameSite=Lax`;密码仍 Argon2,不明文。
### M2-T03 — 数据读取 APIlocations / poo / public-ip
- **Status**: `done` · **Depends**: none
- **Files**: `create app/api/routes/api/data.py``app/schemas/data.py``modify app/main.py``create tests/test_api_data.py`
- **Steps**: `GET /api/locations`(时间范围 + 分页)、`GET /api/poo`(分页)、`GET /api/public-ip`state + history);session 保护;查询参数有上限防全表导出。
- **Acceptance**:
- [ ] 分页/时间范围参数生效且有上限;越权未登录 401。
- [ ] 返回 schema 经 OpenAPI 固化。
- [ ] 校验闸门全绿。
- **Reviewer**: 查询走索引/PK,无 N+1;时间过滤边界正确。
### M2-T04 — 记录 CRUD API(修正 / 删除)
- **Status**: `done` · **Depends**: M2-T03
- **Files**: `modify app/api/routes/api/data.py``app/services/location.py``app/services/poo.py``create tests/test_api_record_crud.py`
- **Steps**: `PATCH`/`DELETE` locationPK person+datetime)与 pooPK timestamp);session + CSRF 保护;PK 路径参数 URL 解码;删除是**硬删单行**(不是清表)。
- **Acceptance**:
- [ ] PATCH 改单行字段、DELETE 删单行,行数变化精确为 1。
- [ ] 不存在的 PK → 404。
- [ ] 缺 CSRF → 403。
- [ ] 没有任何"批量删/清表"路径。
- [ ] 校验闸门全绿。
- **Reviewer**: 删除限定单 PK;编辑校验输入;ingestion 裸端点未被顺手加保护或改动。
### M2-T05 — SMTP 测试 / 动作类 JSON API
- **Status**: `done` · **Depends**: M2-T01
- **Files**: `modify app/api/routes/api/config.py``modify tests/test_api_config.py`
- **Steps**: `POST /api/config/smtp/test` 复用 `send_smtp_test_email`,返回结构化结果(success / config-error / failed)。
- **Acceptance**:
- [ ] 三种结果都有明确 JSON 状态码/字段;session + CSRF 保护。
- [ ] 校验闸门全绿。
### M2-T06 — 前端 scaffold + OpenAPI codegen `[structural]`
- **Status**: `done` · **Depends**: M2-T01..T05OpenAPI 已稳定)
- **Context**: 建 `frontend/` 工程与类型化 client 流水线,这是后续所有前端任务的地基。
- **Files**: `create frontend/`Vite+React+TS 脚手架、`package.json``tsconfig.json`、eslint、vitest、`.gitignore`)、`frontend/src/api/`codegen 产物 + fetch 封装,自动注入 `X-CSRF-Token`)、`frontend/README.md``npm run codegen` 脚本
- **Steps**: 初始化 Vite React-TS;接 `openapi/openapi.json` 生成类型;写一个最小 App 壳 + 受保护路由骨架;fetch 封装统一带 cookie、写请求注入 CSRF header、401 跳登录。
- **Acceptance**:
- [ ] `npm ci && npm run build` 成功产出 `frontend/dist`
- [ ] `npm run lint``npm run typecheck``npm run test` 全绿(哪怕只有 1 个 smoke 测试)。
- [ ] `npm run codegen` 生成物与当前 `openapi/openapi.json` 一致(CI 可校验)。
- **Reviewer**: client 全部基于生成类型;CSRF/cookie/401 处理在统一封装层;无手写、与契约不符的请求类型。
### M2-T07 — 鉴权 UI(登录 / 会话引导 / 改密)
- **Status**: `done` · **Depends**: M2-T06
- **Acceptance**: 登录成功进受保护区;未登录访问受保护路由跳登录;强制改密流程可走完;`build/lint/typecheck/test` 全绿。
### M2-T08 — 配置 UI(取代 Jinja config 页)
- **Status**: `done` · **Depends**: M2-T06
- **Acceptance**: 能读/存所有现有配置 section;secret 不回显、留空保留;SMTP 测试按钮反映三态;前端闸门全绿。
### M2-T09 — 数据可视化 UI(热力图为主的地图)
- **Status**: `done` · **Depends**: M2-T06(数据来自 T03
- **Context**: 接管 Grafana 原职责,且**首页主视图就是这张地图**。优先级:**① 热力图(最重要)② 时间范围选择器(必须)③ 散点点位(辅助,主要服务编辑/删除)**。location:去过哪的密度;poo:狗最爱在哪拉。
- **Acceptance**: 首页渲染热力图(location / poo);**时间范围选择器生效、只取窗口内数据**(不拉全量);散点层可切换、点选某点可进入编辑/删除(接 T10/T04);location 点多时聚合;响应式(手机浏览器可用);前端闸门全绿。
### M2-T10 — 记录管理 UI(按需展示 + 增删改)
- **Status**: `done` · **Depends**: M2-T06CRUD 来自 T04
- **Acceptance**: 列表分页展示 poo/location;可编辑、可删除单条并即时刷新;删除有二次确认;前端闸门全绿。
### M2-T11 — FastAPI 托管 SPA + 移除 Jinja
- **Status**: `done` · **Depends**: M2-T07, T08, T09, T10
- **Files**: `modify app/main.py`(挂载 SPA 静态目录 + 非 `/api` 路径回退 `index.html`);`delete app/templates/``app/api/routes/pages.py`(功能对齐后);`modify tests`(移除 Jinja 页面测试,新增 SPA fallback 测试)
- **Acceptance**:
- [ ] `/config` 等路径返回 SPA`index.html`),`/api/*` 不被 fallback 吞掉,`/static`/资源正常。
- [ ] 旧 Jinja 模板与 pages 路由移除后 `pytest` 全绿。
- [ ] 校验闸门全绿(含 OpenAPI 重导出)。
- **Reviewer**: fallback 不拦截 `/api``/docs``/openapi.json`、静态资源;未登录访问 API 仍 401(不是被 SPA 壳吞掉)。
### M2-T12 — 多阶段 Dockerfile + CI/compose
- **Status**: `done` · **Depends**: M2-T11
- **Files**: `modify Dockerfile`node build 阶段 → 拷 `dist` 进 python 镜像);`modify .github/workflows/*`(加前端 build/lint/typecheck);`modify tests/test_deployment.py`(镜像断言更新)
- **Acceptance**:
- [ ] 镜像构建成功且运行镜像不含 node 运行时。
- [ ] CI 跑前端闸门 + 后端 `pytest`
- [ ] 校验闸门全绿。
### M2-T13 — 文档 + OpenAPI 收尾
- **Status**: `done` · **Depends**: M2-T12
- **Acceptance**: README 增"前端 v2"段(开发/构建说明);architecture 退役"不前后端分离"约束;roadmap 勾选 M2`openapi/` 已同步入库。
---
## 8. 前端校验闸门(前端任务每次结束都要全绿)
`frontend/` 下:
```bash
npm ci
npm run codegen # 生成类型化 client;产物须与 openapi/openapi.json 同步
npm run lint
npm run typecheck
npm run test
npm run build # 必须产出 dist
```
- 后端若同任务改了路由/schema,仍需根目录 `python scripts/export_openapi.py` 并提交 `openapi/`
- "codegen 产物与 OpenAPI 同步"应在 CI 校验(生成后 `git diff --exit-code`)。
## 9. 里程碑完成定义(DoD
- 访问应用得到 React SPA;配置、可视化、记录增删改都在 SPA 内完成。
- 所有浏览器交互走 `/api` JSON 端点,session + CSRF 保护;ingestion 裸端点维持现状(留给 M3)。
- Jinja `templates/` 与 pages 路由移除;FastAPI 同源托管 SPA。
- 多阶段镜像构建通过;CI 含前端闸门。
- 后端 `pytest`/`ruff`/`export_openapi` + 前端 `build/lint/typecheck/test` 全绿。
+109
View File
@@ -0,0 +1,109 @@
# M3 — Token 鉴权与移动端(远期试水)
> 阅读前提:先读 [`README.md`](./README.md)。M3 依赖 M2(已有 `/api` JSON 契约与 session 鉴权)。
>
> **定位**:远期、低投入、探索性。React Native 部分主要是"没做过、试试水"。范围**可能收缩**——其中**token 鉴权 + ingestion 端点收口**是有持久价值的安全改进,应优先;RN app 是加分项。Orchestrator 可只取前半。
## 1. 目标
1. 引入 **bearer token 鉴权**,让非浏览器客户端(移动端、设备脚本)能安全访问。
2. 把 M2 暂时维持裸奔的 **ingestion 端点**`POST /location/record``POST /poo/record`)收口到 token 鉴权下。
3. 做一个 **React Native** 移动端,用类 OAuth 流程拿 token 后消费现有 `/api`
## 2. 现状(M2 完成后)
- `/api/*` 走 session cookie + `X-CSRF-Token`
- `app/services/auth.py` 有 server-side session`auth_sessions` 表,token_hash 存储)。
- `POST /location/record``POST /poo/record` 仍**无鉴权**(设备/脚本裸调用)。
## 3. 目标架构
### 3.1 Token 模型
- 新表 `auth_tokens``id``user_id``token_hash`(仅存哈希)、`label`(设备名)、`created_at``expires_at`(可空=长期)、`revoked_at`
- bearer 校验:`Authorization: Bearer <token>` → 哈希比对 `auth_tokens` → 命中且未撤销未过期则认定身份。
### 3.2 类 OAuth 签发流程(无第三方的 Authorization Code 简化版)
1. 移动端在**内置浏览器**打开 `/authorize?...`
2. 用户账号密码登录(走现有 session),页面展示"授权此设备"。
3. 批准后服务端生成**一次性 authorization code**,重定向到 app 深链 `homeautomation://callback?code=...`
4. app 用 code 调 `POST /api/auth/token` 换取 bearer token 并存本地。
> 简化兜底:批准页直接展示一次性 token 由 app 捕获。优先实现重定向 + code 交换的正规版。
### 3.3 统一鉴权依赖
- `/api` 的数据/CRUD 端点接受**session cookie 或 bearer**两者之一(同一套端点同时服务 Web 与移动端)。
- ingestion 端点(location/poo record)改为**要求 bearer**。
### 3.4 React Native
- **Expo + TypeScript**,复用 M2 的 OpenAPI 类型化 client(共享契约)。
- 内置浏览器走 §3.2 流程拿 token;之后所有请求带 `Authorization: Bearer`
## 4. 迁移注意(重要)
- ingestion 端点一旦要求 bearer,**现有调用方(HA/设备脚本)必须先配置 token**,否则记录会中断。
- 上线顺序:先签发 token 能力(T01–T02)→ 给现有设备配 token → 再对 ingestion 端点强制 bearer(T03),避免断流。可设一个过渡开关或灰度。
## 5. 任务依赖图
```
M3-T01 token 模型 + 迁移
└─► M3-T02 签发流程(authorize + code 交换)
└─► M3-T03 统一鉴权依赖 + ingestion 端点收口(含过渡开关)
├─► M3-T04 Web 端 token 管理 UI(列出/撤销设备)
└─► M3-T05 React Native app(试水)
└─► M3-T06 文档收尾
```
## 6. 原子任务(任务卡)
### M3-T01 — token 数据模型 + Alembic 迁移
- **Status**: `todo` · **Depends**: noneM2 完成后)
- **Files**: `create app/models/token.py``alembic_app/versions/<new>_auth_tokens.py``modify app/models/__init__.py``create tests/test_token_model.py`
- **Acceptance**:
- [ ] 迁移在全新库 upgrade 后建出 `auth_tokens` 表;downgrade 可回滚。
- [ ] token 仅以哈希存储(与 `auth_sessions` 同等强度),明文不入库。
- [ ] 校验闸门全绿。
- **Reviewer**: 哈希算法/长度与现有 session token 一致;`expires_at` 可空语义明确。
### M3-T02 — 签发流程:authorize 页 + code 交换端点
- **Status**: `todo` · **Depends**: M3-T01
- **Files**: `create app/api/routes/api/token.py``app/schemas/token.py`;前端 `/authorize` 页(M2 SPA 内);`create tests/test_api_token.py`
- **Acceptance**:
- [ ] 登录用户在 `/authorize` 批准后得到一次性 code`POST /api/auth/token` 用 code 换取 bearercode 一次性且短时效。
- [ ] 未登录访问 `/authorize` 跳登录;无效/过期 code 换取失败。
- [ ] 返回的 bearer 仅此一次明文出现,库中只存哈希。
- [ ] 校验闸门全绿。
- **Reviewer**: code 一次性、绑定用户、短 TTL;深链 redirect 白名单校验,防开放重定向。
### M3-T03 — 统一鉴权依赖 + ingestion 端点收口
- **Status**: `todo` · **Depends**: M3-T02
- **Files**: `modify app/dependencies.py`(新增"cookie 或 bearer"统一身份依赖);`modify app/api/routes/location.py``poo.py`(要求 bearer,带过渡开关);`modify tests`
- **Acceptance**:
- [ ] `/api` 数据/CRUD 端点用合法 bearer 可访问(等价于 session)。
- [ ] ingestion 端点:带合法 bearer 通过,缺/错 token 在强制模式下 401;过渡开关可临时放行(默认关)。
- [ ] 撤销的 token 立即失效。
- [ ] 校验闸门全绿。
- **Reviewer**: 过渡开关默认安全(强制);bearer 与 session 两路鉴权不产生绕过;ingestion 行为变更有测试覆盖。
### M3-T04 — Web 端 token 管理 UI
- **Status**: `todo` · **Depends**: M3-T03
- **Acceptance**: 在 SPA 内可列出已签发设备 token(label/创建时间/最近使用)、可撤销;撤销后该 token 立即失效;前端闸门全绿。
### M3-T05 — React Native app(试水)
- **Status**: `todo` · **Depends**: M3-T03 · `[experimental]`
- **Files**: `create mobile/`Expo + TS,复用 OpenAPI 类型化 client
- **Acceptance**:
- [ ] 内置浏览器走签发流程拿到 token 并安全存储(Keychain/Keystore)。
- [ ] 至少跑通:登录拿 token → 拉取一类数据展示 → 记一条 ingestion。
- [ ] `npm run lint`/`typecheck`/`build`(或 Expo 等价) 全绿。
- **Reviewer**: token 存安全存储而非明文;client 基于共享 OpenAPI 类型。
### M3-T06 — 文档收尾
- **Status**: `todo` · **Depends**: M3-T05
- **Acceptance**: README/architecture 增 token 鉴权与移动端说明;roadmap 勾选 M3;`openapi/` 同步。
## 7. 里程碑完成定义(DoD
- 存在 bearer token 鉴权与签发流程;token 仅哈希存储、可撤销。
- ingestion 端点已收口到 bearer(过渡完成后强制)。
- `/api` 同时支持 session 与 bearer。
- (加分)React Native app 能拿 token 并消费 `/api`
- 后端 + 前端 + 移动端各自校验闸门全绿,`openapi/` 入库。
> 提醒:本里程碑探索性强,T05 可作为独立试水随时叫停,不影响 T01–T04 带来的安全收口价值。
+309
View File
@@ -0,0 +1,309 @@
# M4 — 登录加固(Login Hardening
> 阅读前提:先读 [`README.md`](./README.md)(协作模型、任务卡格式、校验闸门、数据安全红线)。
> **排期:本里程碑排在 [M5](./m5-iot-energy.md) 之前。** 系统现为公网 + 仅密码 + 无限重试,是现成风险;M4 与 M5 零文件重叠,先堵这个洞再做 IoT。
## 1. 目标
给暴露在公网的单 admin 登录做纵深防御:
1. **防爆破 / 指数退避**:失败登录按指数增长延迟,拖垮暴力枚举;成功即清零。
2. **CLI 逃生通道**:纯命令行重置密码 / 解锁 / 关停 TOTP,**不依赖任何已存恢复凭据**——拿到服务器 CLI 权限的人本来就无所不能,所以这是可接受、且必须存在的最终逃生口(消除"锁死就彻底进不去"的死角)。
3. **TOTP 二次验证(可选配)**admin 可**自选启用** RFC 6238 TOTP;启用后登录需密码 + 6 位动态码;提供一次性恢复码;CLI 能在恢复码全丢时关掉 TOTP。**非强制**,不启用则维持纯密码登录。
> 三段顺序:A(防爆破,紧急、可独立先发版)→ B(TOTP,可选配)→ 收尾。Phase A 自成可用增量,Phase B 失败不影响 A。
## 2. 现状(实现者可据此工作,不必通读全仓库)
**单 admin + Argon2**
- `app/models/auth.py``AuthUser``id / username unique / password_hash / is_active / force_password_change / created_at`)、`AuthSession``token_hash / csrf_token / expires_at / revoked_at`)。单库 `Base`
- `app/services/auth.py`
- `authenticate_user(session, *, username, password) -> AuthUser | None`:查用户→`verify_password`Argon2);失败只 `logger.info`**无任何节流**。
- `create_session(session, *, user, settings) -> (AuthSession, raw_token)``secrets.token_urlsafe`SHA256 存 `token_hash`TTL `auth_session_ttl_hours`(默认 12h)。
- `change_password(...)``get_authenticated_session(...)``initialize_auth_schema(session, settings)`**仅当无任何用户时**用 `AUTH_BOOTSTRAP_USERNAME/PASSWORD` 建初始 admin)。
- `app/api/routes/api/session.py`
- `POST /api/auth/login`body `{username, password}`):`authenticate_user` → None 则 401;否则 `create_session` + `set_cookie`(HttpOnly/SameSite=Lax/`auth_cookie_secure`) → 返回 `{user, csrf_token}`
- `GET /api/session`401 或 user+csrf)、`POST /api/auth/logout`CSRF)、`POST /api/auth/password`CSRF + 校验当前密码)。
**配置**
- `app/config.py``auth_bootstrap_username/password``auth_session_cookie_name``auth_session_ttl_hours``auth_cookie_secure_override`computed `auth_cookie_secure`)。
- 配置系统:扁平 KV`CONFIG_FIELDS` 注册表 + `Settings` + `_settings_payload`),新增标量配置项前端自动渲染(见 M5 文档 §2)。
**无现成 CLI**
- `scripts/``run_migrations.py` / `app_db_adopt.py` / `migrate_legacy_data.py`(argparse) / `export_openapi.py`。**没有**任何 admin/密码/auth 的 CLI。
- `pyproject.toml`**无 `console_scripts`**。CLI 入口沿用 `python -m scripts.<mod>` 风格(与 `python -m scripts.run_migrations` 一致)。
**前端登录(单步)**
- `frontend/src/pages/LoginPage.tsx``POST /api/auth/login {username,password}` → 401 显示通用错误;成功存 csrf、跳转。
- `frontend/src/auth/SessionProvider.tsx``GET /api/session` 引导。
- 无"锁定/稍后再试"提示、无二步流程。
**测试**
- `tests/test_api_session.py`(凭据/cookie/登出/改密/强制改密);`tests/conftest.py` 用 env 设 bootstrap 凭据。
## 3. 目标架构
### 3.1 防爆破 / 指数退避
- **状态表 `auth_login_throttle`**:按 key 记失败窗口(`key / scope / failures / first_failed_at / last_failed_at / next_allowed_at`),成功即删该 key 的行。
- **双键**:同时按 **client IP****username** 计;本次请求需等待 = 两者退避的**较大值**。
- 按 IP:堵单点暴力(attacker IP 越敲越慢,合法用户换 IP 不受影响)。
- 按 username:单 admin 的全局兜底(即便分布式换 IP 也有一层)。
- **DoS 权衡**:单 admin 下纯按 username 硬锁会被攻击者故意失败把你锁死,所以退避是**延迟(429 + Retry-After)而非永久锁定**,且 CLI 是最终逃生口。
- **client IP 来源**:默认用 socket IP`AUTH_TRUST_FORWARDED_FOR=true` 时取 `X-Forwarded-For` 最左(部署在反代后才开)。**这点必须显式配置**,否则反代后所有请求同一 IP,按 IP 退避失效。
- **指数公式(默认,常量可后调)**:前 `N_free=3` 次不延迟;之后 `wait = min(cap, base * 2^(failures - N_free))``base=1s``cap=900s`
- **接入点**`POST /api/auth/login` 开头先查退避 → 在窗口内直接 429(带 `Retry-After`),**不验密码**(省 Argon2、防枚举);验密码失败 → 记一次失败;成功 → 清该 IP + username 的退避行。
- **全局开关** `AUTH_LOGIN_THROTTLE_ENABLED`CONFIG_FIELDS,默认 true)。
### 3.2 CLI 逃生通道
`scripts/admin_cli.py``python -m scripts.admin_cli <cmd>`,argparse 子命令;直接连本地 DB,无网络,复用 `get_session_local()` + 模型 + Argon2 hasher):
| 命令 | 作用 |
| --- | --- |
| `reset-password <username> [--password ...]` | 重置密码(不给则交互式 prompt,不回显);逃生:忘密码 |
| `unlock [--all | --ip <ip> | --username <u>]` | 清 `auth_login_throttle` 行;逃生:被退避锁住 |
| `disable-totp <username>` | 关 TOTP(清 secret + 恢复码);逃生:**恢复码全丢也能进**(Phase B 才有意义)|
| `reissue-totp <username>` | 重新发放 TOTP secret(可选)|
| `list-admin` | 列用户/状态(排障)|
- **数据安全**:CLI 只动 auth 状态行(人工执行的 admin 动作),**绝不**触碰用户数据表(location/poo/energy 等),不 drop 有数据的表。
### 3.3 TOTP 二次验证(可选配)
- **库**`pyotp`(Python 侧);二维码在**前端**用 JS 库(`qrcode.react`)从 `otpauth://` URI 渲染——后端不引图像依赖。
- **存储**
- `AuthUser` 增列 `totp_secret`(nullable)、`totp_enabled`(bool, default false)。secret 落库明文(与现有 secret 处理一致,靠文件权限保护;文档注明)。
- `auth_recovery_code` 表(`user_id / code_hash / used_at`):恢复码**哈希**存(复用 Argon2 hasher),一次性。
- **启用流程(opt-in**
1. `POST /api/auth/totp/setup`(已登录)→ 生成 pending secret + `otpauth://` URI + 一组恢复码(**仅此一次明文返回**),尚未启用。
2. `POST /api/auth/totp/enable`(带一个当前 6 位码确认)→ `totp_enabled=true`,持久化恢复码哈希。
3. `POST /api/auth/totp/disable`(带密码或当前码)→ 关闭、清 secret/恢复码。
- **登录二因子(单端点、无独立 challenge token,保持无状态)**
- `POST /api/auth/login` body 增可选 `totp_code`
- 密码通过且 `totp_enabled`:无 `totp_code` → 返回 401 + `{totp_required: true}`**不发 session**);有 `totp_code` → 校验 TOTP **或**恢复码(命中恢复码则消费它)→ 通过才发 session。
- 未启用 TOTP:维持现状单步。
- 退避(§3.1)对两步都生效。
- **issuer 标签**`AUTH_TOTP_ISSUER`(默认 `app_name`),显示在 Authenticator app 里。
### 3.4 前端
- **登录页两步**`LoginPage` 在收到 `401 + totp_required` 时切到第二屏(6 位码输入,也接受恢复码),再次 `POST /api/auth/login``totp_code`。被退避(429)显示"稍后再试"(用 `Retry-After`)。
- **设置页 TOTP 区**:未启用→「启用」走 setup(展示二维码 + 恢复码一次性、要求输入码确认);已启用→「停用」。恢复码只在生成时展示一次,提示妥善保存。
## 4. API 契约(M4 要落地的端点)
> 全部 `/api`、写端点 session + CSRF 保护、JSON 进出;schema 经 `export_openapi.py` 固化。
| 分组 | 端点 | 用途 |
| --- | --- | --- |
| 登录 | `POST /api/auth/login` | 加退避(429+Retry-After);body 增可选 `totp_code`;启用 TOTP 且缺码→401 `{totp_required:true}` |
| TOTP | `POST /api/auth/totp/setup` | 生成 pending secret + otpauth URI + 恢复码(一次性)|
| TOTP | `POST /api/auth/totp/enable` | 带当前码确认启用 |
| TOTP | `POST /api/auth/totp/disable` | 关闭 TOTP |
| TOTP | `GET /api/auth/totp` | 返回当前 TOTP 状态(enabled 与否)|
> CLI 不是 HTTP 端点。throttle 开关与 issuer 走现有 `GET/PUT /api/config`(仅加 CONFIG_FIELDS)。
## 5. 已锁定决策(讨论后拍板)
1. **M4 独立里程碑、排在 M5 之前**
2. **范围**:防爆破 + 指数退避 + CLI 重置(密码/解锁/关 TOTP)+ TOTP(**可选配,非强制**)。
3. **退避双键(IP + username)、延迟非永久锁定、成功清零、CLI 兜底**;反代后需开 `AUTH_TRUST_FORWARDED_FOR`
4. **CLI 逃生通道不依赖任何已存恢复凭据**;只动 auth 行,不碰用户数据。
5. **TOTP 单端点二因子**login 加可选 `totp_code`,无独立 challenge token);恢复码哈希存、一次性;secret 明文存(与现有一致)。
6. **二维码前端渲染**`qrcode.react`),后端只给 `otpauth://` URI + secretPython 仅加 `pyotp`
7. **CLI 入口 = `python -m scripts.admin_cli`**(无 console_scripts)。
> 项目定位:个人自用、单 admin——可按单用户简化,不为多租户/找回邮件等过度设计。
## 6. 任务依赖图
```
Phase A(防爆破,紧急,自成可发版增量)
M4-T01 [schema] auth_login_throttle 表 + 模型
└─► M4-T02 退避 service + 接入 login429/Retry-After/成功清零)
└─► M4-T03 CLI 骨架 + reset-password + unlock
Phase BTOTP,可选配)
M4-T04 [schema] AuthUser TOTP 字段 + auth_recovery_code 表
├─► M4-T05 TOTP service + setup/enable/disable/status API
├─► M4-T06 登录二因子(login 加 totp_code;依赖 T02 已改过 login
└─► M4-T07 CLI disable-totp / reissue-totp
M4-T08 前端:两步登录 + 设置页 TOTP(依赖 T05+T06;引入 qrcode.react
收尾
M4-T09 文档 + OpenAPI + roadmap(依赖全部)
```
`T01` 可先开。Phase A(T01–T03)跑完即可发一版只含防爆破 + CLI 的安全增量。
---
## 7. 原子任务(任务卡)
> 后端沿用校验闸门(`pytest`/`ruff`/改路由或 schema 则 `export_openapi` 重导出入库)。前端闸门见 §8。新增依赖(`pyotp``qrcode.react`)须同步 `requirements.in/.txt``frontend/package.json` 重新锁定。
### M4-T01 — `auth_login_throttle` 表 + 模型 `[schema]`
- **Status**: `done` · **Depends**: none
- **Context**: 存按 IP / username 的失败退避状态。仅建 schema。
- **Files**: `create app/models/auth_throttle.py`(或并入 `app/models/auth.py`);`create alembic_app/versions/<date>_NN_auth_login_throttle.py``modify alembic_app/env.py``scripts/app_db_adopt.py`baseline 常量);`create tests/test_auth_throttle_model.py`
- **Steps**: 模型 `LoginThrottle``id / key str / scope str('ip'|'user') / failures int / first_failed_at / last_failed_at / next_allowed_at``(scope,key)` unique 索引);新 revision 建表 + 索引;更新 baseline 常量。
- **Out of scope / 不要碰**: 不写退避逻辑(T02);不动登录端点。
- **Acceptance criteria**:
- [ ] 临时库 upgrade 到 head 含该表与唯一索引;`downgrade -1` 干净。
- [ ] `APP_BASELINE_REVISION` 更新;`env.py` 已 import。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: `(scope,key)` 唯一;链上单 head;env.py 注册模型。
### M4-T02 — 退避 service + 接入登录
- **Status**: `done` · **Depends**: M4-T01
- **Context**: 指数退避核心 + 接 `POST /api/auth/login`
- **Files**: `create app/services/login_throttle.py``modify app/api/routes/api/session.py``app/services/config_page.py`(+`CONFIG_FIELDS` `AUTH_LOGIN_THROTTLE_ENABLED`)、`app/config.py`(+`auth_login_throttle_enabled``auth_trust_forwarded_for`)`modify tests/test_api_session.py``create tests/test_login_throttle.py`
- **Steps**:
1. `check_and_get_wait(session, *, ip, username) -> int`(秒;>0 表示仍在窗口);`register_failure(...)`(更新双键 failures/next_allowed_at,按 §3.1 公式);`clear(session, *, ip, username)`(成功清零)。
2. 登录端点:取 client IP(按 `auth_trust_forwarded_for` 决定是否信 XFF 最左);**先**查退避,>0 → 429 + `Retry-After`,不验密码;验密码失败 → `register_failure` 后 401;成功 → `clear` 再发 session。
3. 开关关闭时整段 no-op。
- **Out of scope / 不要碰**: 不做 TOTPPhase B);不改 logout/password 端点。
- **Acceptance criteria**:
- [ ] 单测:连续失败后 `wait` 按指数增长且封顶;窗口内请求 429 带 `Retry-After` 且不验密码。
- [ ] 单测:成功登录清零;下次无延迟。
- [ ] 401 仍为通用文案(不泄露用户是否存在);现有登录测试仍绿。
- [ ] 开关 false 时无节流。
- [ ] 校验闸门全绿(OpenAPI 若变重导出)。
- **Reviewer checklist**: 退避在验密码**之前**生效;双键较大值;XFF 仅在配置开启时信任;无把密码写进日志。
### M4-T03 — CLI 骨架 + reset-password + unlock
- **Status**: `done` · **Depends**: M4-T01
- **Context**: 逃生通道第一批命令。
- **Files**: `create scripts/admin_cli.py``create tests/test_admin_cli.py`
- **Steps**: argparse 子命令;`reset-password <username> [--password|prompt(getpass)]` 用 Argon2 hasher 重置 `password_hash``unlock [--all|--ip|--username]``auth_login_throttle` 行;`list-admin`;复用 `get_session_local()` + 模型;无网络。
- **Out of scope / 不要碰**: 不动 HTTP 端点;不碰用户数据表;TOTP 命令在 T07。
- **Acceptance criteria**:
- [ ] 单测:`reset-password` 后新密码可 `verify_password` 通过、旧密码失败。
- [ ] 单测:`unlock` 清掉指定/全部退避行。
- [ ] 不存在的 user 友好报错、非零退出。
- [ ] `grep` 确认 CLI 不含对用户数据表的删除/drop。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 密码可交互输入且不回显;只动 auth 行;幂等可重跑。
### M4-T04 — AuthUser TOTP 字段 + `auth_recovery_code``[schema]`
- **Status**: `done` · **Depends**: none(与 Phase A 并行的 schema,但 T06 依赖 T02
- **Files**: `modify app/models/auth.py``AuthUser.totp_secret`/`totp_enabled`);`create` 模型 `RecoveryCode``create alembic_app/versions/<date>_NN_totp.py``modify alembic_app/env.py``scripts/app_db_adopt.py``create tests/test_totp_models.py`
- **Steps**: `totp_secret` nullable、`totp_enabled` bool default false`auth_recovery_code(user_id FK, code_hash, used_at nullable)`revision 建列 + 表;更新 baseline。
- **Out of scope / 不要碰**: 不写 TOTP 逻辑(T05)。
- **Acceptance criteria**:
- [ ] upgrade/downgrade 干净;默认 `totp_enabled=false``totp_secret=null`
- [ ] baseline 常量更新;env.py import。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 现有用户迁移后默认未启用 TOTP(不破坏现有登录)。
### M4-T05 — TOTP service + setup/enable/disable/status API
- **Status**: `done` · **Depends**: M4-T04
- **Files**: `create app/services/totp.py``app/schemas/totp.py``modify app/api/routes/api/session.py`(或新 `auth_totp.py` 路由)、`app/config.py`(+`auth_totp_issuer`)`modify requirements.in/.txt`(+`pyotp`)`create tests/test_api_totp.py`
- **Steps**: pyotp 生成 secret + `otpauth://` URIissuer=`auth_totp_issuer`);恢复码生成(N=10、`xxxx-xxxx`Argon2 哈希存、明文仅 setup 返回一次);`setup`pending,不启用)/`enable`(带当前码确认)/`disable`(带密码或当前码)/`GET status`;全部 session+CSRF。
- **Out of scope / 不要碰**: 不改 login 校验(T06)。
- **Acceptance criteria**:
- [ ] `setup` 返回 secret+URI+恢复码(一次性);未确认前 `totp_enabled` 仍 false。
- [ ] `enable` 用错码失败、对码成功;恢复码以哈希持久化。
- [ ] `disable` 关闭并清 secret/恢复码。
- [ ] schema 经 OpenAPI 固化;secret/恢复码不回显于 `GET status`
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 恢复码只存哈希、明文仅一次;secret 不进 OpenAPI 示例/日志;`requirements.txt` 锁定 `pyotp`
### M4-T06 — 登录二因子(login 加 `totp_code`
- **Status**: `done` · **Depends**: M4-T05, M4-T02
- **Files**: `modify app/api/routes/api/session.py``app/schemas/session.py``LoginRequest.totp_code` 可选)、`app/services/totp.py`(校验 + 消费恢复码);`modify tests/test_api_session.py``tests/test_api_totp.py`
- **Steps**: 密码通过后:若 `totp_enabled` 且无 `totp_code` → 401 `{totp_required:true}`(不发 session、退避照算);有 `totp_code` → 校验 TOTP 或恢复码(命中恢复码置 `used_at`)→ 通过发 session;未启用 → 现状。
- **Out of scope / 不要碰**: 不改 setup/enableT05);不动退避公式(T02)。
- **Acceptance criteria**:
- [ ] 未启用 TOTP 的用户登录与现状完全一致。
- [ ] 启用后:缺码→401 totp_required;错码→401;对码→发 session;恢复码可用且一次性。
- [ ] 退避对二步同样生效。
- [ ] OpenAPI 重导出入库;校验闸门全绿。
- **Reviewer checklist**: `totp_required` 时确未发 cookie;恢复码消费后不可复用;通用错误不泄露细节。
### M4-T07 — CLI disable-totp / reissue-totp
- **Status**: `done` · **Depends**: M4-T04(逻辑上配合 T03 的 CLI 骨架)
- **Files**: `modify scripts/admin_cli.py``modify tests/test_admin_cli.py`
- **Steps**: `disable-totp <username>``totp_enabled=false` + 清 secret + 删恢复码(**不需任何恢复码即可执行**);`reissue-totp <username>`:生成新 secret(可选打印新 URI)。
- **Out of scope / 不要碰**: 不碰用户数据表。
- **Acceptance criteria**:
- [ ] 单测:启用 TOTP 的用户经 `disable-totp` 后可纯密码登录(结合 T06 行为)。
- [ ] 不依赖任何恢复码即可关停。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 逃生不依赖已存凭据;只动 auth 行。
### M4-T08 — 前端:两步登录 + 设置页 TOTP
- **Status**: `done` · **Depends**: M4-T05, M4-T06
- **Files**: `modify frontend/src/pages/LoginPage.tsx``create frontend/src/pages/.../TotpSettings.tsx`(或并入 ConfigPage)、`frontend/src/auth/totp.ts``modify frontend/package.json`(+`qrcode.react`lock 同步)`create` 对应测试
- **Steps**: 登录页:401+`totp_required`→第二屏 6 位码(也接受恢复码);429→"稍后再试"(读 `Retry-After`)。设置页:未启用→启用流程(二维码由 `otpauth` URI 渲染 + 恢复码一次性展示 + 输码确认);已启用→停用。
- **Out of scope / 不要碰**: 不改后端;不碰防爆破逻辑。
- **Acceptance criteria**:
- [ ] 未启用 TOTP 的登录体验不变;启用后两步可走通;恢复码可登录。
- [ ] 二维码可被 Authenticator 扫描;恢复码仅展示一次。
- [ ] 429 有"稍后再试"提示。
- [ ] 前端闸门全绿。
- **Reviewer checklist**: 全走类型化 clientsecret/恢复码不落 localStorage/日志;空/错/加载态有处理。
### M4-T09 — 文档 + OpenAPI + roadmap 收尾
- **Status**: `done` · **Depends**: 全部
- **Files**: `modify README.md``docs/auth.md`(防爆破 + CLI + TOTP 说明、CLI 用法)、`docs/architecture-overview.md``modify docs/roadmap.md`(把"TOTP 二次验证"与新"登录防爆破"从「下一阶段」毕业、加 M4 行、注明 M4 先于 M5);`modify docs/design/README.md`(已含 m4);`run python scripts/export_openapi.py` 提交 `openapi/`
- **Acceptance criteria**:
- [ ] 文档含 CLI 逃生用法与 TOTP 启停说明;`git diff --exit-code openapi/` 无差异。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: roadmap 反映 M4 已落地且先于 M5;无残留旧描述。
---
## 8. 前端校验闸门(前端任务每次结束都要全绿)
`frontend/` 下:`npm ci && npm run lint && npm run typecheck && npm run test && npm run build`。新增依赖 `qrcode.react` 须提交 `package.json` + `package-lock.json`。后端同任务改路由/schema 仍需根目录 `export_openapi.py` 并提交 `openapi/`
## 9. 构建上下文 / 依赖完整性
- 新增 Python 依赖 `pyotp`、前端 `qrcode.react`:同步重新锁定 `requirements.txt` / `package-lock.json`,否则镜像缺包。
- 新文件均在既有 `app/``scripts/``frontend/` 的 COPY 范围内;`scripts/admin_cli.py` 须在镜像里可 `python -m scripts.admin_cli` 调用(确认 `scripts/` 已进镜像)。
- `tests/test_deployment.py::test_dockerfile_copy_sources_exist` 仍应通过。
## 10. 人工验收 walkthrough(实现完成后)
> 重点:连续输错 → 系统自动退避锁住 → CLI 解锁后恢复登录。可用 `docker compose` 起服务,CLI 在容器内或容器外跑均可。
**准备**:起服务(任选其一)
- Docker`docker compose up -d`(应用容器名见 `docker-compose.yml`,下文记为 `<app>`)。
- 或本地:`source .venv/bin/activate && uvicorn app.main:app --port 8000`
**1) 触发退避锁定** —— 用**错误密码**连续登录同一账号:
```bash
for i in $(seq 1 8); do \
code=$(curl -s -o /dev/null -w "%{http_code}" -X POST http://localhost:8000/api/auth/login \
-H 'Content-Type: application/json' -d '{"username":"admin","password":"wrong"}'); \
echo "尝试 $i -> HTTP $code"; done
```
- 预期:前几次 `401`,超过免费次数后变 `429`(被锁/退避)。单独跑一次带 `-i` 可看到 `Retry-After` 头逐次增大(指数退避)。
- 在退避窗口内,**即使输正确密码**也应被 `429` 挡住(验证锁定确实生效)。
**2) CLI 解锁**
- 容器内:`docker compose exec <app> python -m scripts.admin_cli unlock --all`
- 或容器外:`python -m scripts.admin_cli unlock --all`
- 预期:打印清除的退避条目数。
**3) 验证恢复** —— 立即用**正确密码**登录:
```bash
curl -i -X POST http://localhost:8000/api/auth/login \
-H 'Content-Type: application/json' -d '{"username":"admin","password":"<正确密码>"}'
```
- 预期:`200``Set-Cookie` 下发 session → 解锁成功。
**4)(可选)其它逃生命令**
- 重置密码:`python -m scripts.admin_cli reset-password admin`(交互输入新密码、不回显)。
- 关停 TOTP(若已启用):`python -m scripts.admin_cli disable-totp admin`,随后纯密码即可登录。
## 11. 里程碑完成定义(DoD
- 登录失败按指数退避(429 + Retry-After),成功清零;可经 `AUTH_LOGIN_THROTTLE_ENABLED` 开关。
- `python -m scripts.admin_cli` 能在**无任何 Web 访问、无恢复码**的情况下重置密码 / 解锁 / 关停 TOTP。
- TOTP 可由 admin 自选启用:启用后两步登录、恢复码可用且一次性;不启用维持纯密码。
- 后端 `pytest`/`ruff`/`export_openapi` + 前端 `lint/typecheck/test/build` 全绿且 `openapi/` 入库。
- README/auth/architecture/roadmap 反映 M4 现实与"先于 M5"的排期。
+533
View File
@@ -0,0 +1,533 @@
# M5 — IoT 集成与能耗采集(Modbus 设备采集 + MQTT/HA Discovery + 前端侧边栏)
> 阅读前提:先读 [`README.md`](./README.md)(协作模型、任务卡格式、校验闸门、数据安全红线)。本里程碑建立在 M1 单库 + M2 React SPA 之上。
> 配套参考:电表协议见 [`../references/SDM120-Modbus-Protocol.md`](../references/SDM120-Modbus-Protocol.md)。
## 1. 目标
给后端接入家庭 IoT 生态,并新增一条通用的 Modbus 设备采集链路(首个落地的领域是能耗):
1. **Modbus 设备采集(通用管道)**:通过 Modbus-TCP 网关周期读取挂在网关后面的 Modbus slave 设备(首个 profile 为 SDM120 单相电表),按设备的 **YAML profile** 解码为工程量,存入单库的**通用读数表**。后台静默轮询,支持多设备、多 profile。
2. **MQTT + Home Assistant Discovery**:后端作为 MQTT 发布方,按"**可勾选暴露**"的方式把数据以 HA Discovery 自动注册成 device/entity(不止 sensor)。
3. **前端侧边栏 + Energy 视图**:把现有顶栏改成侧边导航,承载首个领域视图 **Energy**(设备管理 + 最新读数 + 走势图)。
> **命名分层(本里程碑的核心决策)**:**存储 / 采集 / API 是通用的 `modbus_*`**(不锁死在"电表"上,以后接别的 Modbus 设备无需改表);**面向用户的呈现是领域特定的**——本里程碑只落第一个领域视图 **Energy**,消费通用 device 数据、用电表视角展示。
> 三段有依赖关系,按 §6 的 `Depends` 顺序推进:A(侧边栏,独立)→ B(Modbus 采集后端 + Energy 前端)→ CMQTT/Discovery,消费 B 的数据)。
## 2. 现状(实现者可据此工作,不必通读全仓库)
**单库数据层**M1 完成态)
- `app/db.py``class Base(DeclarativeBase)`,绑 `settings.app_database_url` 的 cached engineWAL 已开),`get_engine` / `get_session_local` / `reset_db_caches` / `get_db_session`
- 模型都继承同一 `Base``app/models/{auth,config,public_ip,location,poo}.py`
- 单 Alembic 链 `alembic_app/`head = `20260611_06_merge_location_poo_tables`(见 `alembic_app/versions/`);`alembic_app/env.py` 逐个 import 所有模型。
- 迁移命名惯例:`YYYYMMDD_NN_<desc>.py``revision` / `down_revision` 串链。
- `scripts/app_db_adopt.py` 常量 `APP_BASELINE_REVISION` 指向当前 head`scripts/run_migrations.py` 负责把 app 库升到 head。
**配置系统(扁平 KV,自动渲染)**
- `app/config.py``class Settings(BaseSettings)`,每个配置项一个带类型的字段 + 默认值。
- `app/services/config_page.py``CONFIG_FIELDS: tuple[ConfigField, ...]` 是**注册表**`section / env_name / setting_attr / label / secret / input_type`);`build_config_sections`(读,secret 回空串)、`save_config_updates`(写,空 secret 保留旧值)、`build_runtime_settings`DB override 合并进 Settings)、`_settings_payload`(把 Settings 摊平成 dict,新字段要在此补一行)。
- `app/api/routes/api/config.py``GET/PUT /api/config`session + CSRF 保护;非法值 422 且不写库。
- 前端 `frontend/src/pages/ConfigPage.tsx`**通用渲染**——按 section 分组、按 `input_type`/`secret` 渲染输入框。**新增标量配置项零前端改动**(追加 `CONFIG_FIELDS` + `Settings` 字段 + `_settings_payload` 一行即可)。
- ⚠️ 扁平 KV **装不下"设备列表"和"逐实体勾选"**——这两者走专用表 + 专用 API + 自定义 UI(见 §3.2 / §3.4)。
**后台调度(APScheduler**
- `app/main.py` lifespan`BackgroundScheduler(timezone="UTC")``scheduler.add_job(_run_scheduled_public_ip_check, IntervalTrigger(hours=4), id=..., max_instances=1, coalesce=True)``scheduler.start()``yield``scheduler.shutdown(wait=False)`
- 周期任务惯例:一个**同步** wrapper 自己开/关 session`session = get_session_local()(); try: service(session, ...) finally: session.close()`service 内部不抛崩溃。
**Home Assistant 现状(REST,不碰 MQTT**
- `app/integrations/homeassistant.py``HomeAssistantClient.publish_sensor()``POST /api/states/{entity}`)、`trigger_webhook()`
- 入站 webhook `app/api/routes/homeassistant.py``POST /homeassistant/publish`envelope `target/action/content`)。
- 新 MQTT Discovery 与此**并行、不冲突**,是第二条独立通道。
**前端(M2**
- React + react-router v6 + Mantine + TanStack Query + `openapi-fetch` 生成的类型化 client`frontend/src/api/client.ts` + `schema.d.ts`)。
- `frontend/src/App.tsx``AppLayout`(当前是**顶栏**),包住所有受保护页;路由 `/`(HomePage 地图)、`/config``/records``/login``/change-password` 不带 layout。
- 数据请求惯例:`useQuery`/`useMutation` + `apiClient.GET/POST/...`(见 `frontend/src/records/hooks.ts`)。
- **无图表库**(只有 Leaflet 地图);走势图需新引入 **Recharts**
## 3. 目标架构
### 3.0 两层数据模型(本里程碑的地基决策)
把"设备是什么、怎么读"与"读到了什么"彻底分成两层,**协议知识与部署信息分离**:
```
┌─────────────────────────────────────────────────────────────┐
│ 协议层(固定,随代码走) 部署层(可配置,落 DB) │
│ ── YAML profile ────────── ── modbus_device 行 ────── │
│ • 读哪些寄存器(FC/地址/块) • friendly_name(可改) │
│ • 每个量:key / 类型 / 解码 • host / port(网关) │
│ • 每个量:unit / device_class • unit_id=电表 Meter ID
│ / ha_component 设备上可设,故落 DB) │
│ • 唯一真相源,喂给"读/存/HA注册" • profile 名(选哪个 YAML
│ • uuid / poll_interval / enabled│
└─────────────────────────────────────────────────────────────┘
▲ 多个 device 可共享同一 profile │
└─────────────────────────────────────────┘
│ 采集
── modbus_reading 行(通用遥测)──
device_id · recorded_at · payload(JSON blob)
```
- **profile = 仓库内只读 YAML**(如 `app/integrations/modbus/profiles/sdm120.yaml`),随镜像打包、纳入版本控制,**只描述固定协议知识**:读哪些寄存器、怎么解码、每个量的 `key`/`unit`/`device_class`/`ha_component`。**绝不**放 friendly_name / unit_id 这类每设备各异的可配置项。
- **可配置项落 `modbus_device` 行**,由前端 UI 设置:friendly_name、host、port、**unit_idModbus 从机地址,设备面板上可改)**、选用的 profile 名、poll_interval、enabled。
- **多设备共享一个 profile**:例如两块 SDM120 都 `profile="sdm120"`,一块 friendly_name="SDM120 空调"、unit_id=1,另一块 friendly_name="SDM120 服务器"、unit_id=2——共用同一份解码规则,部署参数各自独立。
- **读数存通用 JSON `payload`**,不再用固定列。profile 是解释器:知道 payload 里有哪些 key、各自单位与 device_class。
- 聚合不必搬到后端硬算:**SQLite 用 `json_extract` 在 SQL 端做 `AVG`/`MAX`/`GROUP BY`**(走势图都是带时间窗的查询,被 `(device_id, recorded_at)` 索引圈住,只扫窗内行)。
- 真有某个量需要高频聚合 → 后续给该量加 **generated column + 表达式索引**(非破坏性,要哪个补哪个),无需一开始把表锁成固定列。
### 3.1 Modbus 采集
- **传输:仅 Modbus-TCP**(用户的 Waveshare RTU↔TCP 网关,RJ45 以太网;服务器无串口)。用 **pymodbus**`ModbusTcpClient`,由它处理封帧 / CRC / 超时重试 / float 解码。
- 角色厘清:**我们的后端 = Modbus client/master****网关 = TCP server**(在 "Modbus TCP" 模式下终结 TCP 再转 RTU);**电表 = 网关后面的 RTU slave**,由 `unit_id` 寻址。`modbus_device` 一行同时装下网关地址(host:port)与 slave 地址(unit_id)。
- 网关若开"Modbus TCP"协议转换 → pymodbus 默认 framer 直连。
- 网关若是透传(RTU-over-TCP,裸 RTU 帧含 CRC)→ pymodbus 用 RTU framer over TCP。
- **二者实现时对一次即可确定**(连上读 Voltage 寄存器验证),不影响表结构与上层。
- **协议知识在 YAML profile,部署信息在 DB**(见 §3.0):
- `app/integrations/modbus/driver.py`:薄封装 pymodbus 的连接 + 块读 + 大端 float32 解码(word/byte 都大端,高寄存器在前)。
- `app/integrations/modbus/profiles.py`YAML profile 的**加载 + 校验 + 解码 + 实体枚举**——纯"数据 + 函数"**不做 OOP 抽象基类/继承**。`load_profile(name) -> ModbusProfile`(pydantic 模型,启动期校验)、`decode(profile, registers) -> dict[key -> value]``enumerate_entities(device, profile) -> list[ExposableEntity]`(供 §3.3)。
- `app/integrations/modbus/profiles/*.yaml`:每个设备型号一个 YAML。首个 = `sdm120.yaml`(寄存器地址见参考文档 §4)。
- profile 输出统一的 `dict[key -> value]`,由 service 整个塞进 `modbus_reading.payload`JSON);profile 没列出的量根本不读。
- **只读**:不写电表配置寄存器(改 Meter ID/波特率有通信中断风险)。
- **轮询**:每个设备一个 `poll_interval_s`(默认 5s),APScheduler 一个 job 扫所有 `enabled` 设备,逐设备块读→解码→落库。9600 总线上多设备串行,5s 余量充足。
- 全局开关 `MODBUS_POLLING_ENABLED`CONFIG_FIELDS)可一键停采。
- 采集成功/失败写入设备的 `last_poll_at` / `last_poll_ok`(供 §3.3 的 online binary_sensor)。
- **两级周期 / 降采样 / 保留是后续杠杆**(见 §10),本里程碑用单周期读全。
**`sdm120.yaml` 示例(只含固定协议知识)**
```yaml
name: sdm120
description: Eastron SDM120 single-phase energy meter
function_code: 4 # input registers (FC04)
word_order: big # 高寄存器在前
byte_order: big
blocks: # 块读,减少 Modbus 事务
- { start: 0x0000, count: 0x0060 } # 30001..30095 连续段
- { start: 0x0156, count: 0x0004 } # total active/reactive energy
metrics:
- { key: voltage, address: 0x0000, type: float32, unit: V, device_class: voltage, ha_component: sensor }
- { key: current, address: 0x0006, type: float32, unit: A, device_class: current, ha_component: sensor }
- { key: active_power, address: 0x000C, type: float32, unit: W, device_class: power, ha_component: sensor }
- { key: power_factor, address: 0x001E, type: float32, unit: "", device_class: power_factor, ha_component: sensor }
- { key: frequency, address: 0x0046, type: float32, unit: Hz, device_class: frequency, ha_component: sensor }
- { key: import_energy, address: 0x0048, type: float32, unit: kWh, device_class: energy, state_class: total_increasing, ha_component: sensor }
- { key: export_energy, address: 0x004A, type: float32, unit: kWh, device_class: energy, state_class: total_increasing, ha_component: sensor }
- { key: total_energy, address: 0x0156, type: float32, unit: kWh, device_class: energy, state_class: total_increasing, ha_component: sensor }
# 冷门量(demands / maxima / 无功电能…)可后续按需补进 metrics;未列出的就不读。
```
> 注意 YAML 里**没有** unit_id / friendly_name——那些是 `modbus_device` 行各自带的。
### 3.2 数据模型(新增两张通用表,单库 app 链)
**`modbus_device`**(设备定义 = 部署/可配置项,CRUD 管理)
| 列 | 类型 | 说明 |
| --- | --- | --- |
| id | int PK | 内部代理主键(FK join 用)|
| uuid | str unique | uuid4 生成的**稳定内部身份**;也作 API 路径键与 HA `unique_id` 的锚 |
| friendly_name | str | 显示名(可改;改名重发 discovery,HA 显示名跟着变)|
| transport | str | 现仅 `'tcp'` |
| host | str | 网关 IP |
| port | int | 网关端口(默认 502|
| unit_id | int | Modbus 从机地址 = 电表 Meter ID(设备面板可改,默认 1)|
| profile | str | 用哪个 YAML profile(首个 `'sdm120'`|
| poll_interval_s | int | 采样周期(默认 5|
| enabled | bool | 是否轮询 |
| last_poll_at | datetime null | 最近一次轮询时刻(online 判定)|
| last_poll_ok | bool null | 最近一次轮询成败(online 判定)|
| created_at / updated_at | datetime | |
**`modbus_reading`**(通用遥测表,一行 = 一个设备的一次采样)
| 列 | 类型 | 说明 |
| --- | --- | --- |
| id | int PK | |
| device_id | int FK→modbus_device.id | **ON DELETE RESTRICT**(见 §5 删除语义)|
| recorded_at | datetime (UTC) | 采样时刻,**真实列、带索引**(所有查询按它走时间窗)|
| payload | JSON | profile 解码出的全部工程量 `{key: value}`;无固定列 |
- 索引:`(device_id, recorded_at)`
- **SDM120 单相 payload 示例**`{"voltage": 230.2, "current": 1.3, "active_power": 295.0, "power_factor": 0.98, "frequency": 50.0, "import_energy": 123.4, "export_energy": 0.0, "total_energy": 123.4}`——key 由 `sdm120.yaml``metrics[].key` 决定。
- **接入新设备型号无需改表**:换 profilepayload 里的 key 集合随之变;表结构不动。
- **三相电表**以后用三相 profilepayload 里多几个相位 key(如 `voltage_l1/l2/l3`),仍是同一张表。
### 3.3 MQTT + HA Discovery(通用 expose 框架,元数据由 profile 派生)
HA MQTT Discovery 模型 = **device → entities**:往 `<prefix>/<component>/<node>/<object>/config` 发 retained 消息定义一个 entityconfig 内 `device.identifiers` 相同的 entity 归到同一个 HA device 卡片下;之后往 `state_topic` 推值。
- **可暴露实体目录由 provider 动态产出**
- `app/integrations/expose.py`:定义 `ExposableEntity``key`(稳定)、`component`sensor/binary_sensor/switch…)、`device`(归属,决定 HA device 分组)、`device_class``unit`、取值来源)+ 一个 provider 注册表。
- **Modbus/energy provider**:每个 `enabled` 设备 = 一个 HA **device**`identifiers` 用设备 `uuid`),其各工程量 = 一组 sensor entity——**`device_class`/`unit`/`component` 直接取自该设备 profile 的 `metrics[]`**(不再单独维护一份映射),外加一个 `binary_sensor`「online」(取 `last_poll_ok`)——这就是"**不止 sensor**"的体现。
- 其它 provider(如 public-ip、poo)可后续挂入,本里程碑只接 Modbus/energy provider。
- **HA 实体身份锚定(Z2M 模型)**discovery config 的 **`unique_id` 用设备 `uuid` + 量的 `key` 派生**(稳定,不随改名变);可见的 **`name` 用 friendly_name**。改 friendly_name → 重发 discovery → HA 显示名跟着变、但 `unique_id` 不变故历史不丢。topic 的 object_id 也用 uuid 派生(稳定、丑无所谓)。
- **`exposed_entities` 表**:只存"逐 key 的开关"`key` unique + `enabled` + `updated_at`)。目录本身由 provider 计算,表只记被勾选的状态(默认未勾 = 不暴露)。
- **MQTT 客户端**`app/integrations/mqtt.py`**paho-mqtt**`loop_start()` 后台线程;在 lifespan 起/停;支持配置变更后**重连 + 重发 discovery**。
- **发布时机**
- discovery configretained):连接成功时、目录/勾选变更时全量发;取消勾选时发空 payload 清除该 entity。
- state:采集在每次轮询后推最新值;另有一个周期 job 兜底重发所有 enabled 实体的 state + availability(在线)topic。
- **配置**(走现有扁平 CONFIG_FIELDS):`MQTT_ENABLED``MQTT_BROKER_HOST/PORT/USERNAME/PASSWORD(secret)``MQTT_TLS_ENABLED``HA_DISCOVERY_ENABLED``HA_DISCOVERY_PREFIX`(默认 `homeassistant`)。
### 3.4 前端
- **侧边栏**:把 `AppLayout` 从顶栏重构为侧边导航(Mantine `AppShell` 或 flex sidebar),导航项:Home / Records / Energy / Config + 主题切换 + 注销;当前路由高亮;移动端可折叠。仅改 `App.tsx`+ 可抽 `AppSidebar`/`NavItem` 组件),各页面主体不动。
- **Energy 视图**(新页 `/energy`,首个领域视图,消费通用 `/api/modbus` 数据):
- 设备管理:列表 + 新建/编辑/删除(删除有二次确认;后端对有读数的设备拒删,引导改用"禁用")。新建/编辑表单里设 friendly_name、host、port、unit_id、profile(下拉选 `sdm120` 等)、poll_interval、enabled。
- 最新读数卡片(每设备当前各工程量;字段标签/单位取自 profile 的 metrics 元数据,见 `/metrics` 端点)。
- 走势图:用 **Recharts** 画时间序列(电压/电流/功率/电能),时间范围选择,取数走 readings API(窗口 + 上限),从 `payload` 里按 key 取序列。
- **Expose 设置**:设置页内一块「Home Assistant Expose」——列出可暴露实体目录、逐项勾选、显示 MQTT/Discovery 连接状态、一个"重新发布 discovery"按钮。
## 4. API 契约(M5 要落地的端点)
> 全部 `/api` 前缀、session + CSRF(写)保护、JSON 进出。schema 经 `export_openapi.py` 固化入库。路径键用设备 `uuid`(稳定、非自增)。
| 分组 | 端点 | 用途 |
| --- | --- | --- |
| Modbus | `GET /api/modbus/devices` | 列出设备 |
| Modbus | `POST /api/modbus/devices` | 新建设备 |
| Modbus | `GET /api/modbus/devices/{uuid}` | 单个设备 |
| Modbus | `PATCH /api/modbus/devices/{uuid}` | 修改设备(含 enable/disable|
| Modbus | `DELETE /api/modbus/devices/{uuid}` | 删除设备;**有读数时 409**,引导改 disable |
| Modbus | `GET /api/modbus/devices/{uuid}/metrics` | 该设备 profile 的量目录(key/label/unit/device_class),供前端渲染卡片与图表标签 |
| Modbus | `GET /api/modbus/devices/{uuid}/latest` | 该设备最新一条读数(payload)|
| Modbus | `GET /api/modbus/devices/{uuid}/readings` | 时间范围读数(`start/end/limit`limit 有上限),返回 `recorded_at + payload`,供走势图 |
| Modbus | `POST /api/modbus/devices/{uuid}/test` | 即时试读一次(验证网关连通/地址),返回解码 payload,不落库 |
| Modbus | `GET /api/modbus/profiles` | 列出可用 profile 名 + 描述(前端建设备时的下拉选项)|
| Expose | `GET /api/expose` | 返回可暴露实体目录 + 勾选状态 + MQTT/Discovery 状态 |
| Expose | `PUT /api/expose` | 设置逐 key 勾选(map key→bool|
| Expose | `POST /api/expose/republish` | 手动重发 discovery |
| 配置 | `POST /api/config/mqtt/test` | 试连 broker **并发布一条测试消息**MQTT Explorer 可见),仿 SMTP 测试三态 |
> MQTT broker / discovery 的**标量配置**复用现有 `GET/PUT /api/config`(只新增 CONFIG_FIELDS,不新增端点)。
## 5. 已锁定决策(讨论后拍板)
1. **里程碑编排**:一个 M5 文档分三段,`Depends` 串顺序(A 侧边栏 → B Modbus/Energy → C MQTT)。
2. **两层数据模型,协议与部署分离**`modbus_device`(部署/可配置项:friendly_name、host、port、unit_id、profile 名、poll、enabled+ `modbus_reading`(通用遥测:device_id、recorded_at、`payload` JSON)。取代原"宽表 + 固定列"。
3. **读数 = JSON `payload`,无固定列**;聚合走 SQLite `json_extract`DB 端做 AVG/MAX/GROUP BY),热点量后补 generated column + 表达式索引。两级周期/降采样为后续杠杆。
4. **协议知识 = 仓库内只读 YAML profile**(声明式"数据 + 函数",无 OOP 继承,pydantic 启动期校验);YAML 携带寄存器 + 每个量的 key/unit/device_class/ha_component,是**唯一真相源**,同时驱动"读 / 存 / HA 注册"。多设备可共享一个 profile。首个 profile `sdm120`
5. **Modbus 仅 TCP**pymodbusframer 配网关模式,**只读**采集。设备是网关后面的 slave,`unit_id` 寻址(设备面板可改,故落 DB)。
6. **命名分层(方案 C**:存储/采集/API 一律通用 `modbus_*``/api/modbus/devices`;前端首个领域视图叫 **Energy**(消费通用 device 数据)。
7. **UUID = 内部生成(uuid4)的稳定身份**:既做内部索引/ API 路径键,也做 HA discovery `unique_id` 的锚;friendly_name 可改,改名重发 discovery、HA 显示名跟着变(Z2M 模型,历史不丢)。
8. **MQTT = 通用 expose 框架**:provider 动态产出可暴露实体目录,**实体元数据(device_class/unit/component)从 profile 派生**`exposed_entities` 只存逐 key 开关;支持 sensor/binary_sensor/switch 等多 component;设备自动注册(每设备一 device、各量为 entity + 一个 online binary_sensor)。
9. **MQTT 库 = paho-mqtt**,lifespan 长连接,配置变更后重连 + 重发 discovery。
10. **MQTT broker/discovery 标量配置走现有扁平 CONFIG_FIELDS**(自动渲染);设备清单与 expose 勾选走专用表 + 自定义 UI。
11. **图表库 = Recharts**(封在自包含组件后,仿 M2 对 Leaflet 的隔离)。
12. **删除设备安全**FK `ON DELETE RESTRICT`,有读数拒删(避免一键删表丢历史);"停用"用 `enabled=false`
13. **Config 页用 Accordion 分区**(进页见大类、逐类展开),**不在 config 页内再放第二个 side nav**——避免与主侧栏(T01)的"双抽屉"冲突;纯前端、独立任务 M5-T01B。
14. **CLI 手工测试工具为"受控手工链路验证"而设**(设备接市电、非随时在线,不进自动化):`scripts/modbus_cli` 提供 `read`profile 解码)与 `probe`(手工指定请求内容、看原始回复)两个子命令,**一律只读**(仅 FC03/04),不暴露写寄存器。
15. **MQTT/HA 发布链路的手工验证走 UI + 外部工具,不另做 CLI**Config 页「发送测试消息」(`mqtt/test` 发一条到测试 topic)→ 在 **MQTT Explorer** 查看;Expose 勾选实体 + 开 `HA_DISCOVERY_ENABLED` + 「重新发布 discovery」(`/api/expose/republish`)→ 到 **Home Assistant** 查看。不通则迭代配置再重发。
> 项目定位:个人自用、家庭特化、不开源——可按单用户场景简化,不过度抽象。
## 6. 任务依赖图
```
Phase A(独立,可最先做,纯前端)
M5-T01 [structural] 侧边栏布局重构
M5-T01B Config 页分区折叠(Accordion) ← 与 T01 互不依赖,可并行/先后任意
Phase BModbus 采集 + Energy 前端)
M5-T02 [schema] modbus_device + modbus_reading 表 + 模型
├─► M5-T03 Modbus 驱动 + YAML profile 框架(pymodbus + sdm120.yaml,纯模块)
│ └─► M5-T04 采集 service + APScheduler 轮询(接 lifespan,落 payload
└─► M5-T05 Modbus JSON APIdevice CRUD + readings + metrics + test
└─► M5-T06 前端:设备管理 UI(依赖 T01 侧栏 + T05 API
└─► M5-T07 前端:读数展示 + Recharts 走势图(依赖 T05;引入 recharts
Phase CMQTT / Discovery,依赖 B 的设备数据与 provider 接口)
M5-T08 MQTT/Discovery 配置项(CONFIG_FIELDS
M5-T09 [schema] exposed_entities 表 + ExposableEntity/provider 框架(modbus provider 从 profile 派生)
├─► M5-T10 MQTT 客户端(paholifespan 连接 + 重连 + config/mqtt/test
│ └─► M5-T11 Discovery 发布 + state 发布(连 T04 轮询推 state
└─► M5-T12 前端:Expose 勾选 UI + /api/expose API
收尾
M5-T13 文档 + OpenAPI + roadmap 收尾(依赖全部)
```
`T01``T01B``T02``T08` 无前置可先开。
---
## 7. 原子任务(任务卡)
> 后端任务沿用校验闸门(`pytest` / `ruff` / 改路由或 schema 则 `export_openapi` 重导出入库)。前端任务闸门见 §8。新增依赖(`pymodbus``PyYAML``paho-mqtt``recharts`)须在对应任务里同步 `requirements.in/.txt``frontend/package.json` 并重新锁定。
### M5-T01 — 侧边栏布局重构 `[structural]`
- **Status**: `done` · **Depends**: none
- **Context**: 把 `AppLayout` 从顶栏改为侧边导航,给后续 Energy 等视图腾入口。纯前端,不碰各页主体。
- **Files**: `modify frontend/src/App.tsx``create frontend/src/components/AppSidebar.tsx``frontend/src/components/NavItem.tsx`(可选);`modify` 受影响的 `frontend/src/pages/*.test.tsx`(导航断言)
- **Steps**:
1. 用 Mantine `AppShell`(或 flex sidebar)重构 `AppLayout`:左侧竖直导航(Home/Records/Config + 主题切换 + 注销),`<Outlet/>` 在右。
2. 当前路由高亮(`useLocation` 比对 `pathname`);移动端可折叠(burger)。
3. 导航项图标沿用 `react-feather`;样式走 Mantine(暗色模式自动适配)。
4. 不在本任务加 Energy 项(页面还不存在,T06 加),保持导航无死链。
- **Out of scope / 不要碰**: 不改各页面主体;不动鉴权(SessionProvider/ProtectedRoute);不引入图表库。
- **Acceptance criteria**:
- [ ] 受保护页都在侧边栏布局内;`/login``/change-password` 不带布局(与现状一致)。
- [ ] 当前路由在侧栏高亮;移动端宽度下可折叠/展开。
- [ ] 前端闸门全绿(`lint`/`typecheck`/`test`/`build`)。
- **Reviewer checklist**: 布局只在 `App.tsx`/新组件内变动,未误改页面或鉴权;无死链导航项。
### M5-T01B — Config 页分区折叠(Accordion
- **Status**: `done` · **Depends**: none
- **Context**: config 内容会越来越多(M5 还要加 MQTT / HA Discovery / Modbus 配置 + Expose 面板)。把 ConfigPage 从一长串 section 改为 Mantine `Accordion`:进页只见各大类标题,逐类展开编辑。**纯前端、单列内容**——它不是"第二个 app 级侧栏",与主侧栏(T01)无冲突,故 `Depends: none`、可独立先做。
- **Files**: `modify frontend/src/pages/ConfigPage.tsx``modify frontend/src/pages/ConfigPage.test.tsx`(折叠/展开断言)
- **Steps**:
1. 用 Mantine `Accordion` 包住现有"按 section 分组"的渲染:每个 config section = 一个 `Accordion.Item`(标题 = section 名,面板 = 该 section 的字段表单)。
2. 默认折叠(或首个展开);**保留现有保存逻辑与 config API 不变**M2 的整页/按 section 保存语义照旧,本任务不碰后端)。
3. 预留:将来 T12 的「Home Assistant Expose」自定义面板也作为一个 `Accordion.Item` 接入,保持一致。
4. 移动端单列堆叠即可,**不引入第二个侧栏/抽屉**。
- **Out of scope / 不要碰**: 不改后端 config API/字段(`CONFIG_FIELDS` / `_settings_payload`);不动主侧栏(T01);不改保存语义;不在 config 页内放任何 app 级 side nav。
- **Acceptance criteria**:
- [ ] ConfigPage 以 accordion 呈现,每大类可独立展开/折叠;字段渲染与保存行为与现状一致。
- [ ] 与主侧栏无视觉/交互冲突(单列内容;移动端不出现双抽屉)。
- [ ] 前端闸门全绿(`lint`/`typecheck`/`test`/`build`)。
- **Reviewer checklist**: 仅改 ConfigPage+其测试),未碰 config 后端或主 layout;保存逻辑无回归;页面内**无第二个 app 级 sidebar**accordion 是内容、非 chrome)。
### M5-T02 — `modbus_device` + `modbus_reading` 表与模型 `[schema]`
- **Status**: `done` · **Depends**: none
- **Context**: 单库 app 链新增两张**通用**表,建出 §3.2 结构(设备 = 部署层,读数 = JSON payload 通用遥测层)。本任务只建 schema + 模型,不写采集/接口。
- **Files**: `create app/models/modbus.py``ModbusDevice``ModbusReading`,继承 `app.db.Base`);`create alembic_app/versions/<date>_07_modbus_tables.py``modify alembic_app/env.py`import 新模型);`modify scripts/app_db_adopt.py``APP_BASELINE_REVISION` → 新 head);`create tests/test_modbus_models.py`
- **Steps**:
1. 模型按 §3.2 列定义(`Mapped[...]` 2.0 风格):`ModbusDevice``uuid`(unique)、`friendly_name``transport``host``port``unit_id``profile``poll_interval_s``enabled``last_poll_at`/`last_poll_ok`(nullable)、时间戳;`uuid``default` 生成 uuid4 字符串。`ModbusReading``device_id` FK→`modbus_device.id`**`ondelete="RESTRICT"`**)、`recorded_at``payload`JSON 列,非 null)。
2. 新 revision`down_revision` = 当前 head`upgrade()``op.create_table` 建两表 + 索引 `(device_id, recorded_at)``downgrade()` 反向 drop。
3. 更新 `APP_BASELINE_REVISION`
- **Out of scope / 不要碰**: 不写 pymodbus/采集(T03/T04);不加路由(T05);不动其它模型。
- **Acceptance criteria**:
- [ ] 全新临时 app 库 upgrade 到 head 后含 `modbus_device``modbus_reading` 及索引;`downgrade -1` 干净回滚。
- [ ] `Base.metadata.tables` 含两新表;FK 为 RESTRICT`uuid` 唯一且自动生成。
- [ ] `APP_BASELINE_REVISION` == 新 head。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 列/约束与 §3.2 一致;`payload` 为 JSON 列;`recorded_at` 为真实索引列(非塞进 payload);链上单 head`env.py` 已 import 新模型(否则 autogenerate/建表漏表)。
### M5-T03 — Modbus 驱动 + YAML profile 框架(`sdm120`
- **Status**: `done` · **Depends**: M5-T02
- **Context**: 薄封装 pymodbus 的连接/块读/大端 float 解码,加 YAML profile 加载/校验/解码框架与 SDM120 profile。纯模块,mock client 单测。
- **Files**: `create app/integrations/modbus/__init__.py``app/integrations/modbus/driver.py``app/integrations/modbus/profiles.py``app/integrations/modbus/profiles/sdm120.yaml``scripts/modbus_cli.py``modify requirements.in`/`requirements.txt`(加 `pymodbus``PyYAML`,重新锁定);`create tests/test_modbus_driver.py``tests/test_modbus_profiles.py``tests/test_modbus_cli.py`
- **Steps**:
1. `driver.py``read_blocks(host, port, unit_id, blocks) -> dict[int,int]`(按 profile 的 blocks 块读 input registersFC04),用 `ModbusTcpClient`;超时/连接失败抛明确异常;大端 float32 解码 helper`registers_to_float`,高寄存器在前)。framer 选择留可配置/可探测。
2. `profiles.py`:定义 pydantic `ModbusProfile` / `MetricSpec``key`/`address`/`type`/`unit`/`device_class`/`state_class`?/`ha_component`);`load_profile(name) -> ModbusProfile`(读 `profiles/<name>.yaml`,校验失败抛错);`decode(profile, registers) -> dict[key -> value]`(按各 metric 的 address+type 从寄存器对解码);`list_profiles() -> list[(name, description)]`。**不做抽象基类/继承**,全是数据 + 模块函数。
3. `profiles/sdm120.yaml`:按 §3.1 示例写全核心量(参考文档 §4 地址)。
4. `scripts/modbus_cli.py`:纯命令行、**不依赖 DB / 不需先配设备**,供**受控手工测试**(设备接市电、非随时在线,不进自动化)。两个**只读**子命令:
- `read --host H --port P --unit U --profile sdm120`:按 profile 读一次、解码后**把各工程量打印成可读结果**(表格/JSON)——验证整套解码链路。
- `probe --host H --port P --unit U --fc 4 --address 0x0000 --count 2 [--decode float32]`:**手工指定要发送的请求内容**(功能码 FC03/04 + 起始地址 + 数量),打印**原始寄存器(hex)+ 可选大端 float 解码**——first-contact 验证网关连通、framer 模式与 unit 地址,可单读一个寄存器,不依赖 profile。
连接失败给清晰报错 + 非零退出。**只读**:CLI 仅暴露读功能码(FC03/04),**不提供任何写寄存器子命令**(数据红线 + 防改坏电表通信参数)。
- **Out of scope / 不要碰**: 不连真实硬件(单测用 mock/fake 返回已知寄存器字节);不写调度(T04);不写电表配置寄存器;不在 profile 里放 unit_id/friendly_name。
- **Acceptance criteria**:
- [ ] 单测:给定 `0x4366,0x3334` 解码为 `230.2`(参考文档实例);字序/字节序正确。
- [ ] 单测:`load_profile("sdm120")` 校验通过;`decode(profile, ...)` 把已知寄存器映射到正确 key 与值。
- [ ] 单测:profile YAML 缺字段/类型错时 `load_profile` 抛可识别校验错。
- [ ] 连接失败/超时抛可识别异常,不静默返回错值。
- [ ] `python -m scripts.modbus_cli read ...` 能(对 mock/真实网关)打印解码后的各工程量;连接失败非零退出。
- [ ] `python -m scripts.modbus_cli probe --fc 4 --address 0x0000 --count 2 ...` 能打印原始寄存器与可选解码值;CLI **无任何写寄存器子命令**
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 解码确为大端、高寄存器在前;地址与参考文档一致;**CLI 与 driver 都无任何写寄存器路径(仅 FC03/04)**;profile 纯协议知识、无部署项;`requirements.txt` 已同步锁定 `pymodbus``PyYAML`
### M5-T04 — 采集 service + APScheduler 轮询
- **Status**: `done` · **Depends**: M5-T03
- **Context**: 周期扫所有 enabled 设备,调用 driver 读+解码,落 `modbus_reading.payload`。仿 public-ip 的同步 job 模式。
- **Files**: `create app/services/modbus_poll.py``modify app/main.py`lifespan 注册 job);`create tests/test_modbus_poll.py`
- **Steps**:
1. `modbus_poll.py``poll_device(session, device) -> ModbusReading | None``load_profile` → driver 读 → `decode` → 把 dict 存进 `payload` 插一行;更新 `last_poll_at`/`last_poll_ok`);`poll_all_enabled_devices(session)` 遍历 enabled 设备;service 内吞异常并日志,不让 job 崩。
2. `main.py`:加同步 wrapper `_run_scheduled_modbus_poll`(自管 session),`add_job(IntervalTrigger(seconds=...), id="modbus-poll", max_instances=1, coalesce=True)`。周期取**最小 per-device interval 或一个基础 tick**(实现可用单一基础 tick + 各设备按自身 interval 取模决定本 tick 是否读,保持 job 简单);受全局 `MODBUS_POLLING_ENABLED` 控制。
3. 失败的设备记 `last_poll_ok=false`(供 T11 暴露),不影响其它设备。
- **Out of scope / 不要碰**: 不发 MQTTT11);不加 HTTP 路由(T05);不引入两级周期(后续杠杆)。
- **Acceptance criteria**:
- [ ] 单测:mock driver 返回已知 dict`poll_all_enabled_devices``modbus_reading` 精确 +N 行、`payload` 内容正确。
- [ ] 单测:某设备读失败时其它设备仍正常落库,job 不抛;失败设备 `last_poll_ok=false`
- [ ] `MODBUS_POLLING_ENABLED=false` 时不轮询。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: session 在 wrapper 内开关、try/finally 关闭;job `max_instances=1` 防叠加;无 N+1/每行单独 connect 的明显低效;异常不外泄崩 job;payload 为解码后的 dict(非裸寄存器)。
### M5-T05 — Modbus JSON APIdevice CRUD + readings + metrics + test
- **Status**: `done` · **Depends**: M5-T02
- **Context**: 给前端提供设备 CRUD、最新读数、时间范围读数、量目录、即时试读。
- **Files**: `create app/api/routes/api/modbus.py``app/schemas/modbus.py``modify app/main.py`(注册路由);`create tests/test_api_modbus.py`
- **Steps**:
1. devices`GET`(list)/`POST`/`GET{uuid}`/`PATCH{uuid}`/`DELETE{uuid}`session+CSRF`POST` 校验 profile 名存在;`DELETE` 有读数 → 409。
2. readings`GET {uuid}/latest`(最新一行 payload)、`GET {uuid}/readings``start/end/limit`limit 有上限防全表导出,按 `recorded_at` 升序,返回 `recorded_at + payload`)。
3. `GET {uuid}/metrics`:返回该设备 profile 的量目录(key/label/unit/device_class),供前端渲染。
4. `GET /api/modbus/profiles``list_profiles()` 的名+描述。
5. `POST {uuid}/test`:用 driver 即时读一次返回解码 payload(或错误),**不落库**。
- **Out of scope / 不要碰**: 不在此处发 MQTT;不写采集逻辑(复用 T03/T04 的 driver/service)。
- **Acceptance criteria**:
- [ ] CRUD 行为正确:创建/改/删行数精确;删有读数的设备返回 409;未登录 401、缺 CSRF 403;建设备引用不存在 profile → 422。
- [ ] readings 时间范围 + limit 上限生效;latest 返回最新一条 payloadmetrics 返回 profile 量目录。
- [ ] schema 经 OpenAPI 固化入库。
- [ ] 校验闸门全绿(含 `openapi/` 重导出)。
- **Reviewer checklist**: 删除受 RESTRICT 保护、无批量删/清表路径;查询走 `(device_id, recorded_at)` 索引;test 端点确不落库;路径键用 `uuid`
### M5-T06 — 前端:设备管理 UIEnergy 视图)
- **Status**: `done` · **Depends**: M5-T01, M5-T05
- **Context**: 在侧栏加 Energy 入口与 `/energy` 路由;设备增删改 + 试读。
- **Files**: `create frontend/src/pages/EnergyPage.tsx``frontend/src/energy/DeviceForm.tsx``frontend/src/energy/hooks.ts``modify frontend/src/App.tsx`(路由)、`frontend/src/components/AppSidebar.tsx`Energy 项);`create` 对应 `*.test.tsx`
- **Steps**: `useQuery`/`useMutation``/api/modbus/devices` API;列表 + 新建/编辑表单(friendly_name/host/port/unit_id/profile 下拉/poll/enabled+ 删除二次确认(删失败 409 提示改用禁用);"试读"按钮调 `POST {uuid}/test` 显示结果。
- **Out of scope / 不要碰**: 走势图在 T07;不碰其它页面。
- **Acceptance criteria**:
- [ ] 能增/改/删设备并即时刷新;删除有二次确认;409 有友好提示;profile 走 `/api/modbus/profiles` 下拉。
- [ ] 侧栏出现 Energy 入口、`/energy` 可达。
- [ ] 前端闸门全绿。
- **Reviewer checklist**: 全部走生成的类型化 client;删除走确认;无与契约不符的手写请求。
### M5-T07 — 前端:读数展示 + Recharts 走势图
- **Status**: `done` · **Depends**: M5-T05(数据), M5-T06(页面壳)
- **Context**: 在 Energy 页展示每设备最新读数 + 时间序列走势。
- **Files**: `modify frontend/src/pages/EnergyPage.tsx``create frontend/src/energy/EnergyCharts.tsx`(封装 Recharts);`modify frontend/package.json`(加 `recharts``package-lock.json` 同步);`create` 对应测试
- **Steps**: 最新读数卡片(接 `latest`,字段标签/单位取自 `/metrics`);时间范围选择 + 折线图(电压/电流/功率/电能,从 `payload` 按 key 取序列),接 `readings`(窗口 + limit);图表封在 `EnergyCharts` 内(仿 Leaflet 隔离,便于将来换库)。
- **Out of scope / 不要碰**: 不做服务端降采样(后续);不改后端。
- **Acceptance criteria**:
- [ ] 最新读数与走势图渲染正确;时间范围只取窗口数据(不拉全量);标签/单位来自 metrics。
- [ ] Recharts 封装自包含、仅此处 import。
- [ ] 前端闸门全绿(`build` 通过,注意 chunk 体积提示)。
- **Reviewer checklist**: 图表组件隔离;查询有窗口/上限;空数据/加载/错误态有处理;从 payload 取 key 的逻辑容忍缺 key。
### M5-T08 — MQTT / Discovery 配置项(CONFIG_FIELDS
- **Status**: `done` · **Depends**: none
- **Context**: 把 MQTT broker 与 discovery 的标量配置接入扁平配置系统(前端自动渲染)。
- **Files**: `modify app/config.py`(新增 Settings 字段)、`app/services/config_page.py`(追加 CONFIG_FIELDS + `_settings_payload`);`modify .env.example``modify tests/test_api_config.py`
- **Steps**: 加字段 `mqtt_enabled``mqtt_broker_host/port/username/password`(secret)、`mqtt_tls_enabled``ha_discovery_enabled``ha_discovery_prefix`(默认 `homeassistant`)、`modbus_polling_enabled`CONFIG_FIELDS 归入「MQTT」「Home Assistant Discovery」「Modbus」section`_settings_payload` 补齐对应行。
- **Out of scope / 不要碰**: 不建 MQTT 客户端(T10);不动 expose 表(T09)。
- **Acceptance criteria**:
- [ ] 新配置项在 `GET /api/config` 出现且分 sectionpassword 为 secret(回空、留空保留);port 为 number。
- [ ] 非法值(端口非数字)422 不写库。
- [ ] 校验闸门全绿(OpenAPI 若变化则重导出)。
- **Reviewer checklist**: secret 不回显/不入 OpenAPI 示例;`_settings_payload` 未漏字段(否则运行期 override 丢失)。
### M5-T09 — `exposed_entities` 表 + ExposableEntity/provider 框架 `[schema]`
- **Status**: `done` · **Depends**: M5-T02
- **Context**: 建"可暴露实体目录"的抽象与开关存储;modbus provider 把设备映射成 device/entities**实体元数据从 profile 派生**。
- **Files**: `create app/integrations/expose.py``ExposableEntity`、provider 协议、注册表、modbus provider);`create app/models/expose.py``ExposedEntityToggle``key` unique + `enabled` + `updated_at`);`create alembic_app/versions/<date>_08_exposed_entities.py``modify alembic_app/env.py``scripts/app_db_adopt.py``create tests/test_expose_catalog.py`
- **Steps**:
1. `ExposableEntity``key/component/device/device_class/unit/value_getter`+ provider 接口 `enumerate(session) -> list[ExposableEntity]`
2. modbus provider:每个 enabled 设备 → 一个 device`identifiers` 用设备 `uuid`),其各量 → sensor entity**device_class/unit/component 取自 profile 的 `metrics[]`**),加一个 `binary_sensor` online(取 `last_poll_ok`)。
3. `build_catalog(session)` 合并所有 provider 的实体 + 各自 `enabled`(来自 toggle 表,缺省 false)。
4. migration 建 toggle 表;更新 baseline 常量。
- **Out of scope / 不要碰**: 不发 MQTTT11);不加 HTTPT12)。
- **Acceptance criteria**:
- [ ] 单测:建若干设备后 `build_catalog` 产出每设备对应 entity(含 online binary_sensor+ 正确 device 分组、device_class、unit(与 profile 一致)。
- [ ] toggle 表 migration 可升/降;缺省 enabled=false。
- [ ] 至少含一个非 sensor componentonline binary_sensor)。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: `key` 稳定(用设备 `uuid` + 量 key,不用自增 id,避免重建漂移);component 支持多类型;目录由 provider 计算、元数据源自 profile 而非写死。
### M5-T10 — MQTT 客户端(paholifespan 连接 + 重连)
- **Status**: `done` · **Depends**: M5-T08
- **Context**: 长连接 MQTT 客户端,配置变更可重连;含连接测试端点。
- **Files**: `create app/integrations/mqtt.py``modify app/main.py`lifespan 起/停)、`app/api/routes/api/config.py``POST /api/config/mqtt/test`);`modify requirements.in`/`requirements.txt`(加 `paho-mqtt` 重新锁定);`create tests/test_mqtt_client.py`
- **Steps**:
1. `MqttManager``is_configured()``connect()/disconnect()/reconnect(settings)``publish(topic, payload, retain)`paho `loop_start()` 后台线程;未配置/未启用则 no-op。
2. lifespan:启用则 connectshutdown disconnect。
3. 配置保存后若 MQTT 设置变化 → 触发 manager 重连(在 config 保存路径加 hook 或保存后比对)。
4. `POST /api/config/mqtt/test`:用提交/现存配置试连**并发布一条测试消息**到一个测试 topic(如 `<discovery_prefix>/home-automation/test`),返回三态(success/config-error/failed)。仿 SMTP 测试"发一封测试邮件"的语义——用户随后在 **MQTT Explorer** 里就能看到这条消息,确认 broker 发布链路通(这是 MQTT 端的手工验证手段,**不另做 CLI**)。
- **Out of scope / 不要碰**: 不构建 discovery/state 消息(T11)。
- **Acceptance criteria**:
- [ ] 单测(fake broker/paho mock):configured 时 connect 调用正确;未配置 no-oppublish 透传 topic/payload/retain。
- [ ] `POST /api/config/mqtt/test` 试连**并发布一条测试消息**(可在 MQTT Explorer 看到);三态有明确返回;session+CSRF 保护。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 断网/连接失败不崩主进程;线程在 shutdown 正确停止;`requirements.txt` 同步锁定 `paho-mqtt`;密码不进日志。
### M5-T11 — Discovery 发布 + state 发布
- **Status**: `done` · **Depends**: M5-T09, M5-T10
- **Context**: 把 enabled 实体发成 HA discovery configretained)并周期推 state;采集轮询后推最新值。
- **Files**: `create app/services/ha_discovery.py``modify app/services/modbus_poll.py`(轮询后推 state)、`app/main.py`state 周期 job + 连接后/勾选变更后发 discovery)、`app/api/routes/api/...``/api/expose/republish` 在 T12 接,本任务提供 service);`create tests/test_ha_discovery.py`
- **Steps**:
1. `build_discovery_payload(entity)` → HA 规范 config`<prefix>/<component>/<node>/<object>/config`,含 `device` 块、`state_topic``device_class``unit_of_measurement``availability`**`unique_id` 用设备 uuid + 量 key 派生,`name` 用 friendly_name**)。
2. `publish_discovery(session)`:对 enabled 实体发 retained config;对取消勾选的发空 payload 清除。
3. `publish_states(session)`:取各实体当前值发 state;采集在 `poll_device` 成功后顺带推该设备实体 state + online。
4. lifespan:连接成功 / 目录或勾选变更后 `publish_discovery`;周期 job 兜底 `publish_states` + availability。
- **Out of scope / 不要碰**: 不做前端(T12);不改采集解码逻辑。
- **Acceptance criteria**:
- [ ] 单测:discovery payload 符合 HA 结构(device 分组正确、topic/device_class/unit 正确、`unique_id` 取自 uuid);取消勾选发空 payload。
- [ ] 单测:采集轮询成功后推对应 state topic;失败推 online=false。
- [ ] discovery 用 retained。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 仅发 enabled 实体;entity `unique_id` 稳定(源自 uuid,不随改名变);MQTT 未启用时整链 no-op;不阻塞轮询。
### M5-T12 — 前端:Expose 勾选 UI + `/api/expose`
- **Status**: `done` · **Depends**: M5-T09, M5-T11
- **Context**: 后端 expose 读写端点 + 设置页勾选界面。
- **Files**: `create app/api/routes/api/expose.py``app/schemas/expose.py``modify app/main.py``create tests/test_api_expose.py`;前端 `create frontend/src/pages/.../ExposeSettings.tsx`(或并入 ConfigPage)、`modify` 路由/设置入口;`create` 前端测试
- **Steps**: `GET /api/expose`(目录 + 勾选 + MQTT/Discovery 状态)、`PUT /api/expose`key→bool)、`POST /api/expose/republish`(调 T11 service);前端列出目录、按 device 分组、逐项开关、显示连接状态、"重新发布"按钮。
- **Out of scope / 不要碰**: 不改采集/发布逻辑(T11)。
- **Acceptance criteria**:
- [ ] `GET/PUT /api/expose` 正确读写勾选;session+CSRFOpenAPI 固化。
- [ ] 勾选变更后(或点重新发布)触发 discovery 重发。
- [ ] 前端能勾选并显示状态;前后端闸门全绿。
- **Reviewer checklist**: PUT 只改 toggle 不误碰其它配置;republish 真触发 T11;类型化 client。
### M5-T13 — 文档 + OpenAPI + roadmap 收尾
- **Status**: `done` · **Depends**: 全部
- **Files**: `modify README.md`Modbus/Energy/MQTT 段、新依赖)、`docs/roadmap.md`M5 行 + 把"MQTT/IoT"从"下一阶段"毕业、新增 Modbus 采集方向)、`docs/architecture-overview.md`(新增 MQTT 通道与 Modbus 采集);`modify docs/design/README.md`(列入 m5);`run python scripts/export_openapi.py` 并提交 `openapi/`
- **Acceptance criteria**:
- [ ] 文档反映新链路;`git diff --exit-code openapi/` 无未提交差异。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 无残留旧描述(含旧 `energy_meters`/`/api/energy` 字样);OpenAPI 已入库。
---
## 8. 前端校验闸门(前端任务每次结束都要全绿)
`frontend/` 下:
```bash
npm ci
npm run lint
npm run typecheck
npm run test
npm run build # 必须产出 dist;留意 chunk 体积告警
```
- 后端若同任务改了路由/schema,仍需根目录 `python scripts/export_openapi.py` 并提交 `openapi/`
- 新增前端依赖(recharts)须提交 `package.json` + `package-lock.json`
## 9. 构建上下文完整性(M1 教训)
- 本里程碑**不删/移文件**,但新增了 Python 依赖(`pymodbus``PyYAML``paho-mqtt`)与前端依赖(`recharts`):必须同步 `requirements.in/.txt` 的重新锁定与 `package-lock.json`,否则镜像构建会缺包。
- 新增源文件都在 `app/``scripts/``frontend/` 既有 COPY 范围内,无需改 `Dockerfile``COPY`**但 `app/integrations/modbus/profiles/*.yaml` 是非 .py 资源**——确认 `COPY app ...` 把整个目录(含 YAML)带进镜像,且运行期能按相对路径定位 YAML(建议用 `importlib.resources` 或基于 `__file__` 的路径,别用 CWD 相对路径)。`scripts/modbus_cli.py` 须在镜像里可 `python -m scripts.modbus_cli` 调用;`tests/test_deployment.py::test_dockerfile_copy_sources_exist` 仍应通过。
- 发版前置走查(见 CLAUDE.md):真起 app 跑一次轮询、真连一次 broker、前端 Energy 视图人工瞄一眼渲染,再打 tag。
## 10. 后续杠杆(本里程碑不做,文档留痕)
- **热点量的 generated column + 索引**JSON `payload` 默认无法对单个量建索引;某个量若需高频聚合,给 `modbus_reading` 加一列 `GENERATED ALWAYS AS (json_extract(payload,'$.<key>'))` 并建索引——非破坏性、要哪个补哪个。
- **两级采样周期**:瞬时量(V/I/P/PF/Hz)快、累计电能慢——9600 总线吃紧或要把功率压到 5s 以下时再开(device 表加 `slow_interval_s`,读数表是通用 payload、无需改结构)。
- **保留 / 降采样**:5s 采样长期行数大(≈630 万行/年/设备),加定期降采样或保留窗口任务(独立于本里程碑),`GROUP BY` + `AVG(json_extract(...))` 在 SQL 端做。
- **更多 device profile**:三相电表(如 SDM630)新增一个 YAML profilepayload 里多几个相位 key),不改表、不改采集主链。
- **更多 expose provider**public-ip / poo 等挂入 expose 框架。
- **写 Modbus 寄存器**:当前只读;如需经 MQTT/UI 控制设备(switch 类),再单独评估安全边界。
## 11. 人工验收 walkthrough(实现完成后)
> 重点:用命令行**手工**读到设备数据并展示结果。这些是**受控手工测试**——设备接市电、并非随时在线,故**不纳入自动化测试**(自动化只用 mock);CLI 工具(`modbus_cli read/probe`)就是为这种"我想测的时候手动测一次"而设。可用 `docker compose` 起环境,命令在容器内或容器外跑均可。
**前提**:网关(Waveshare RTU↔TCP)已上电接入网络,电表 Meter ID 已知(默认 1)。
**1) 命令行直接试读(不依赖 DB,最快验证)**
- 容器内:`docker compose exec <app> python -m scripts.modbus_cli read --host <网关IP> --port 502 --unit 1 --profile sdm120`
- 或容器外:`source .venv/bin/activate && python -m scripts.modbus_cli read --host <网关IP> --port 502 --unit 1 --profile sdm120`
- 预期:打印解码后的工程量(电压/电流/有功功率/功率因数/频率/导入导出电能等),数值合理;连不上则清晰报错。
- **first-contact / 原始验证**`python -m scripts.modbus_cli probe --host <网关IP> --port 502 --unit 1 --fc 4 --address 0x0000 --count 2 --decode float32` —— 手工指定请求内容、看原始回复(电压寄存器应解出约 230V),用于确认网关 framer 模式与 unit 地址;建 profile / 设备前就能跑。
**2)(可选)经 API 试读已配置的设备**
- 先在前端 Energy 页或 `POST /api/modbus/devices` 建一个设备;
- `POST /api/modbus/devices/{uuid}/test` 即时试读,返回解码 payload(不落库)。
**3)(可选)验证后台轮询落库**
- 确认 `MODBUS_POLLING_ENABLED=true` 且设备 `enabled`;等一个采样周期;
- 看前端 Energy 视图的最新读数/走势图,或查 `GET /api/modbus/devices/{uuid}/readings` 有新行。
**4)(可选)手工验证 MQTT / HA Discovery 发布链路**(受控手工步骤,不进自动化)
- **broker 发布链路**:配好 MQTT broker 后,在 Config 页点「发送测试消息」(`POST /api/config/mqtt/test`)——它试连并发一条测试消息;打开 **MQTT Explorer** 确认能收到,即链路通。
- **HA Discovery**:在 Expose 设置勾选若干实体、开 `HA_DISCOVERY_ENABLED`,点「重新发布 discovery」(`POST /api/expose/republish`);到 **Home Assistant** 确认对应 device/entity 出现、值正确。
- **改名验证**:改某设备 friendly_name 后重发,确认 HA 显示名跟着变、历史不丢(`unique_id` 稳定)。
- 不通则按需调整配置/勾选再重发即可。
## 12. 里程碑完成定义(DoD
- 后端能按 per-device 周期静默轮询 Modbus-TCP 设备、按 YAML profile 解码落 `modbus_reading.payload`,支持多设备 CRUD。
- MQTT 启用时,勾选的实体以 HA Discovery 注册成 device/entities(含非 sensor),state 周期发布;配置变更可重连重发;改 friendly_name 重发后 HA 显示名跟着变、`unique_id` 不变。
- 前端侧边栏可切换功能;Energy 视图能管理设备、看最新读数与走势图;设置页可勾选 expose。
- 后端 `pytest`/`ruff`/`export_openapi` + 前端 `lint/typecheck/test/build` 全绿且 `openapi/` 已入库。
- README / architecture / roadmap / design 索引反映 M5 现实。
+483
View File
@@ -0,0 +1,483 @@
# M6 — 通用电价层 + DSMR 实时数据 → 实时买卖电费计算(含 Tibber 动态 / 固定合同两种 profile)
> 阅读前提:先读 [`README.md`](./README.md)(协作模型、任务卡格式、校验闸门、数据安全红线)。本里程碑建立在 **M5MQTT + HA Discovery + expose 框架 + Energy 视图 + Recharts** 之上,复用其大部分基建,并把 M5 的"两层模型"YAML profile 定结构 + DB 行存部署值)从设备采集复制到**电价合同**上。
> **状态**:架构与原子任务已拆到可开工。少数价格常数(卖价残差、能源税年值、双费率寄存器映射)等真实 Tibber token / 账单确认,已落成 §13 的可调值 + TODO,不阻塞实现。
## 1. 目标
把"电价合同"与"DSMR 实时电表数据"接起来,按 **15 分钟**周期算出**实际买电支出 / 卖电收入**,自己落库留底(不可变、可审计),并通过 MQTT + HA Discovery 反哺 Home Assistant 的 Energy 仪表盘:
1. **通用电价层(两层模型)**:仓库内**只读 YAML profile 定义合同结构**`manual` 固定/可变费率、`tibber` 动态电价各一个结构);DB 的 **`EnergyContract`(+ 版本)存可改的数值**UI 填);**price strategy** 按 `kind` 出价。一次激活一个合同。
2. **DSMR 实时数据接入**:订阅 DSMR Reader 的 `dsmr/json`(每秒一条完整 telegram),**整帧存为 JSON blob、按 10 秒降采样**落库(电、气、各相全存,未来加供暖不改表)。
3. **实时买卖电费计算(两层)**:每 15 分钟用**电表累计寄存器差值** × 当时合同的买/卖价算计量电费(不可变、快照价);日/月/年**汇总**再加固定费、减能源税抵扣。**不做净计量**(saldering 2027 取消)。
4. **反哺 Home Assistant**:复用 M5 expose/Discovery,把当前买/卖价、累计买电支出、累计卖电收入发成 HA 实体(累计用 `total_increasing`)。
5. **前端**:Energy 视图下新增合同管理(按 profile 结构生成表单)+ 价格曲线 + 费用走势/明细 + Tibber 测试。
> **命名分层(延续 M5**:存储/采集/计算层中性命名(`dsmr_*` / `tibber_price` / `energy_cost_period` / `energy_contract`);面向用户并入既有 **Energy** 视图。
## 2. 现状(实现者可据此工作,不必通读全仓库)
**单库 + AlembicM5 完成态)**
- `app/db.py``class Base(DeclarativeBase)`cached engine(已开 `journal_mode=WAL` + `foreign_keys=ON`)。
- 单 Alembic 链 `alembic_app/`**当前 head = `20260622_10_exposed_entities`**`scripts/app_db_adopt.py``APP_BASELINE_REVISION` 同。迁移命名 `YYYYMMDD_NN_<desc>.py`,串 `down_revision``alembic_app/env.py` 逐个 import 模型。
- `modbus_reading` 已确立**通用 JSON payload 遥测**范式(`recorded_at` 真实索引列 + `payload` JSON)——`dsmr_reading` 直接沿用。
- M5 两层模型(`app/integrations/modbus/profiles/*.yaml` 定结构 + `modbus_device` 存部署值 + pydantic 启动期校验 + `decode`)是本里程碑电价层的范本。
**配置系统(扁平 KV,自动渲染)**
- `app/config.py` `Settings(BaseSettings)``app/services/config_page.py``CONFIG_FIELDS` 注册表 + `build_runtime_settings(session, bootstrap)`DB override 合并,**消费方都用它**+ `_settings_payload`(新字段补一行)。前端 `ConfigPage.tsx` 按 section/`input_type`/`secret` 通用渲染,bool 渲染成开关。
- ⚠️ 扁平 KV **装不下"合同清单 + 逐字段数值 + 版本"**——这些走专用表 + 专用 API + 自定义 UI(仿 M5 `modbus_device`)。
**MQTT(M5,关键:当前只发不收)**
- `app/integrations/mqtt.py` `MqttManager``is_configured/is_connected/connect/disconnect/reconnect/publish`paho `loop_start()` 后台线程,VERSION2 回调。**无 `subscribe`/`on_message`**——M6-T06 在此扩订阅端。
- `app/integrations/expose.py``ExposableEntity`dataclass`value_getter: Callable[[Session],Any]`)、`EntityProvider = Protocol(session)->list[ExposableEntity]``register_provider` + `build_catalog`。已有 `_modbus_provider`。**M6 加 `_energy_cost_provider` 复用整条 Discovery 链**。
- `app/services/ha_discovery.py``build_discovery_payload`/`publish_discovery`/`publish_states`/`publish_device_state``app/models/expose.py` `ExposedEntityToggle`(默认未勾不暴露)。
**后台调度(APSchedulerUTC**
- `app/main.py` lifespan`BackgroundScheduler(timezone="UTC")``add_job(..., max_instances=1, coalesce=True)`;同步 wrapper 自管 session、吞异常不崩 job。已有 public-ip / modbus-poll / discovery-state 等 job——**新增 job/订阅不得破坏它们**。
**依赖**`httpx`Tibber GraphQL)、`paho-mqtt``pyyaml``apscheduler` 均已在 `requirements.in/.txt`。**M6 不新增 Python 依赖**。前端已有 Recharts / TanStack Query / `openapi-fetch` 类型化 client / Mantine。
## 3. 目标架构
### 3.0 数据流总览
```
EnergyContract(active, 版本带生效日期) DSMR Reader
│ kind=manual → 常数双费率 │ MQTT dsmr/json (1s)
│ kind=tibber → API 15min ▼
│ 订阅 + 10s 降采样 → dsmr_reading(整帧 JSON blob)
│ (kind=tibber 时) │
▼ │ 累计寄存器=真值
Tibber GraphQL → tibber_price(15min, 不可变) │
│ ▼
└──────────────► price strategy ◄────── 计费引擎(每 15min 边界)
(manual/tibber 出价) │ 寄存器差 × 价,快照
energy_cost_period(每15min 计量电费, 不可变)
┌────────────────────────┼───────────────────────────┐
▼ ▼ ▼
日/月/年汇总(+固定费 −抵扣) expose provider + HA Discovery /api/energy/* + 前端
```
三条核心原则:
1. **算钱用寄存器差值,不用功率积分**:每 15 分钟能量 = 累计寄存器@末 @初,电表真值、与采样频率无关。瞬时功率只用于显示(且 HA 里已有,后端不专门用它算钱)。
2. **结构固定(YAML)、数值可改(UI/DB)、历史不可变(快照 + 版本)**profile YAML 定合同长什么样;UI 填数值、改价存新版本;算过的周期快照当时的价,永不被回改。
3. **两层费用**:每 15 分钟只算**计量电费**;按天/年的**固定费、能源税抵扣**在**汇总层**加,不污染每周期引擎。三相 / 双费率 / 燃气都靠"通用 blob + profile 结构"容纳,不改表。
### 3.1 两层电价模型(本里程碑地基,仿 M5)
```
┌──────────────────────────────────────────────────────────────┐
│ 知识层(固定,随代码走) 部署层(可改,落 DB,UI 填) │
│ ── pricing profile YAML ── ── EnergyContract 行 ── │
│ • kind=manual / tibber • nameUI 自由填) │
│ • 有哪些字段、单位、计费语义 • kind(UI 选) │
│ • 哪些分档 / 是否动态源 • active(一次一个) │
│ • UI 不可编辑 • currency │
│ ── EnergyContractVersion ── │
│ • effective_from / to │
│ • values(JSON,符合 profile) │
│ • 改价=加新版本,旧版本保留 │
└──────────────────────────────────────────────────────────────┘
│ kind 决定 strategy
price strategy(代码): manual=常数双费率出价; tibber=读 tibber_price 出价
```
- **profile YAML(仓库内只读,定结构)**`app/integrations/pricing/profiles/manual.yaml``tibber.yaml`,pydantic 启动期校验。声明该 kind **有哪些字段、单位、计费语义、哪些分档**——**不放具体数值**。
- **`EnergyContract` 行(部署,UI 填值)**`name`(自由填,不 hardcode "fixed")、`kind`UI 选)、`active``currency`
- **`EnergyContractVersion`(版本/时段,可审计)**`effective_from`/`effective_to`(null=开口) + `values`(JSON,符合 profile 结构)。**改价 = 加一个新版本行,旧版本保留**;"2026 上半年一个价、下半年另一个价" = 同合同两个版本。引擎算某周期时挑**覆盖该时刻的版本**。
- **price strategy(代码,按 kind**`manual` 用版本里的双费率常数出价;`tibber``tibber_price` + 版本里的 energy_tax/sell_adjust 出价。计费引擎对两者一视同仁。
**`manual.yaml` 结构(块状、可读;值在 UI 填)**
```yaml
kind: manual
label: 固定 / 可变费率(NL,双费率)
energy:
dual_tariff: true # 用 delivered_1/2、returned_1/2 分高低谷
buy:
normal: { unit: EUR/kWh } # 高价档(_2
dal: { unit: EUR/kWh } # 低价档(_1
sell:
normal: { unit: EUR/kWh } # 回送高价
dal: { unit: EUR/kWh } # 回送低价(现两档同价,仍分开填)
energy_tax: { unit: EUR/kWh } # 能源税,加到买价上(含 VAT)
ode: { unit: EUR/kWh, default: 0 } # 现并入能源税
standing: # 固定费:UI 按月填,内部摊到每天
network_fee: { unit: EUR/month }
management_fee: { unit: EUR/month }
credits:
heffingskorting: { unit: EUR/year } # 能源税抵扣,年度,汇总时扣
```
**`tibber.yaml` 结构**
```yaml
kind: tibber
label: Tibber 动态电价(15 分钟)
energy:
source: tibber_api # buy = total; sell = total energy_tax sell_adjust
energy_tax: { unit: EUR/kWh }
sell_adjust: { unit: EUR/kWh, default: 0 }
standing:
management_fee: { unit: EUR/month, default: 5.99 }
network_fee: { unit: EUR/month }
credits:
heffingskorting: { unit: EUR/year }
```
> 所有金额含 VAT(用户口径)。**Tibber token** 不在合同里、走 secret config(凭据归 config)。
### 3.2 DSMR 接入
- **数据源**DSMR Reader 的 **Telegram JSON**(单 topic `dsmr/json`,一条 = 一帧完整 telegram,已是解析后的表寄存器值)。**每秒一条**。
- **整帧存 JSON blob**(沿用 `modbus_reading` payload 哲学):**不做字段 allowlist**,整帧存下来(电、气 `extra_device_*`、各相全要)。理由:现有燃气、未来供暖、三相——blob 全装,**不改表**;读取时按 key 取。
- **实测 JSON 的坑(已确认)**:数值是 **JSON 字符串**`"20915.154"`)→ 计费读取时转 **Decimal**;缺测相位是 **`null`** → 视为缺测;`id`telegram 自增)做幂等去重。
- **降采样**:仅当 `timestamp` 秒数落在 `dsmr_sample_interval_s`(默认 10)整数倍时落盘,其余丢弃(量从 60 行/分降到 ~6 行/分;秒=00 在网格上,每个 15 分钟边界都抓得到)。
- **订阅**:扩 `MqttManager` 订阅端(`on_connect` 里 subscribe、`topic→handler``on_message` 分发,handler 在网络线程内只做"解析+降采样+一次短事务"、吞异常不崩连接)。`dsmr_ingest_enabled`(默认 falseopt-in)。
- **进口/出口总量** = `delivered_1+delivered_2` / `returned_1+returned_2`**双费率分档** = `_1`(dal/低) vs `_2`(normal/高)(NL 惯例,§13 确认别接反)。
- **三相 / 燃气 / 供暖**:blob 天然容纳,单相→三相只是多几个 key,不改表(详 §10)。
### 3.3 Tibber 价格接入(仅 `kind=tibber` 时)
- **客户端**`app/integrations/tibber/client.py`httpx POST `https://api.tibber.com/v1-beta/gql``Bearer <token>`
- **15 分钟查询**introspection + demo 实证):
```graphql
{ viewer { homes { id currentSubscription {
priceInfoRange(resolution: QUARTER_HOURLY, first: 96) {
nodes { startsAt total energy tax currency level }
} } } } }
```
- `Price` 字段:`total`=energy+tax**全包价**demo 已证 `total==energy+tax`)、`energy`(纯 spot ex-VAT)、`tax``startsAt`(含时区偏移)、`currency``level`
- 枚举 **`PriceInfoRangeResolution`** 值 `DAILY/HOURLY/QUARTER_HOURLY`(不是 `PriceResolution`);老的 `priceInfo.range` 已废弃。
- **已验证(demo key**:返回 96 节点、间隔 15 分钟、`total==energy+tax`;老数据无真 15min 价时 API 把小时价**重复成 4 个刻钟**,当前 NL 数据则是真 15 分钟。**解析不假设固定间隔/节点数**,按 `startsAt` 逐点存。
- **抓取调度**:启动 + 每日 job 拉「今天+明天」,按 `starts_at` **upsert** `tibber_price`(幂等)。次日价约 13:00 CET 发布 → 每日 14:00 CET12:00 UTC)抓明天,缺失每小时重试。**仅当 active 合同 kind=tibber 且 token 存在时跑**。
- **连接测试** `POST /api/energy/tibber/test`:拉一次 current price,三态(success 带当前价 / config-error / failed),仿 M5 `mqtt/test`
### 3.4 计费引擎(两层)
**第一层 — 每 15 分钟计量电费(`energy_cost_period`,不可变、快照价)**
- 触发:APScheduler 1 分钟 tick,找"已闭合未算"的 15 分钟周期算(也支持按区间手动重算)。
- 单周期 `[t0,t1)``t1` 已过):
1. 取各寄存器在 `t0`/`t1` 的值(`recorded_at ≤ 边界` 的最后一行,Decimal),算 **per-register 差**`Δd1,Δd2,Δr1,Δr2`
2. 取 active 合同**在 t0 生效的版本** + 其 strategy
- `manual``import_cost = Δd1×(buy_dal) + Δd2×(buy_normal)``buy_x = energy_buy_x + energy_tax + ode`);`export_revenue = Δr1×sell_dal + Δr2×sell_normal`
- `tibber`:取覆盖 t0 的 `tibber_price``starts_at ≤ t0` 最近一条);`buy = total``sell = total energy_tax sell_adjust``import_cost = (Δd1+Δd2)×buy``export_revenue = (Δr1+Δr2)×sell`
3. `net_cost = import_cost export_revenue`**upsert** `energy_cost_period`**快照**当时用的价 + `contract_version_id`
- 缺价/缺数据:跳过或标 `degraded`,留待重算。**不做净计量**(进出口分开累加)。
**第二层 — 汇总(日/月/年,读时计算,不存表)**
- `Σ 周期 net_cost`(区间内)** 固定费**`network_fee + management_fee` 按月→按天 × 天数)**− 能源税抵扣**(`heffingskorting` 按年→按天 × 天数)= **实际应付**
- 固定费/抵扣取 active 合同版本的值;UI 按月/按年填,引擎内部摊到每天。
### 3.5 数据模型(新增 5 张表,单库 app 链)
> 单一 P1 智能电表,`dsmr_reading` 不设 device 表;将来第二块表加 `source` 列即可。
| 表 | 关键列 | 说明 |
| --- | --- | --- |
| **`dsmr_reading`** | `recorded_at`(idx)、`source_id`(unique=telegram id)、`payload`(JSON) | 整帧 telegram blob10s 降采样 |
| **`energy_contract`** | `name``kind`(manual/tibber)、`active``currency`、时间戳 | 合同头;一次一个 active |
| **`energy_contract_version`** | `contract_id`(FK)、`effective_from``effective_to`(null)、`values`(JSON)、`created_at` | 版本/时段;改价加新版本、旧版本保留(审计)|
| **`tibber_price`** | `starts_at`(unique)、`resolution``energy``tax``total``level``currency``fetched_at` | 15min 价缓存,**不可变**;仅 tibber kind |
| **`energy_cost_period`** | `period_start`(unique)、`d1/d2/r1/r2_kwh`(周期差)、`import_cost``export_revenue``net_cost``currency``pricing`(JSON 快照)、`contract_version_id`(FK)、`degraded``computed_at` | 每15min 计量电费,**不可变**(快照价+来源版本)|
- `energy_cost_period` 存 per-register 差 + 快照价 → 完全可审计、source-agnostic、可重算覆盖。
- FK`energy_contract_version.contract_id → energy_contract.id`ON DELETE RESTRICT);`energy_cost_period.contract_version_id → energy_contract_version.id`RESTRICT,保审计链)。
### 3.6 不可变 / 审计(贯穿)
1. **算账快照**`energy_cost_period` 存死当时实际用的价 + `contract_version_id`;改合同 / 改价**不回改**已算周期——"以前的价就是以前的价"。重算是**显式 opt-in**`POST /recompute`),不自动覆盖历史。
2. **合同版本只增不改**:改价 = 新 `energy_contract_version`(带生效日期),旧版本保留,全程审计链。
3. **Tibber 价不可变**:过去某 15 分钟的价不会变。
4. 三者叠加 → manual / tibber 都满足审计与不可变。
### 3.7 MQTT / HA expose(复用 M5
`app/integrations/expose.py::_energy_cost_provider``register_provider`),归一个 HA device`identifiers="energy-cost"`):
| key | component | 取值 | 用途 |
| --- | --- | --- | --- |
| `buy_price_now` | sensor | 当前周期/当前档买价(€/kWhmonetary | HA 当前单价 |
| `sell_price_now` | sensor | 当前卖价(€/kWh | |
| `import_cost_total` | sensor | 累计 import_cost`total_increasing`monetarycurrency | HA Energy "总成本实体" |
| `export_revenue_total` | sensor | 累计 export_revenue`total_increasing` | HA Energy 回送补偿 |
- `value_getter(session)``energy_cost_period`(累计/当前)取值。沿用 M5 勾选(默认未勾)、`publish_discovery``publish_states`(计费 job 后顺带推 + 周期兜底)。
### 3.8 前端(并入 Energy 视图)
- **合同管理**:列表 + 新建/编辑(**表单按选中 profile 结构渲染字段**:manual 双费率/能源税/固定费/抵扣;tibber token 提示走 Config+ 激活(一次一个 active)+ 改价=新版本(带生效日期)+ 版本历史只读展示。
- **价格曲线**:今天+明天 15min 买/卖价(tibber)或固定档位(manual)折线(复用 `EnergyCharts`),取 `/api/energy/prices`
- **费用走势 + 明细 + 汇总**:当日/月 import_cost/export_revenue/net 走势 + 每 15min 明细 + 汇总卡片(含固定费/抵扣),取 `/api/energy/costs` + `/costs/summary`
- **Tibber 测试**Config 页一块「Tibber」,三态(仿 MQTT 测试)。
- 全走类型化 client;空/加载/错误态齐全;缺 key 容错。
## 4. API 契约(M6 要落地的端点)
> 全部 `/api` 前缀、session + CSRF(写)保护、JSON 进出;schema 经 `scripts/export_openapi.py` 固化入库。
| 分组 | 端点 | 用途 |
| --- | --- | --- |
| 合同 | `GET /api/energy/contracts` | 列出合同 + active 标记 |
| 合同 | `POST /api/energy/contracts` | 新建合同(kind + 首版本值,按 profile 校验)|
| 合同 | `GET /api/energy/contracts/{id}` | 单个合同 + 版本历史 |
| 合同 | `PATCH /api/energy/contracts/{id}` | 改名 / 激活(一次一个 active)|
| 合同 | `POST /api/energy/contracts/{id}/versions` | 加新版本(改价,带生效日期)|
| 合同 | `GET /api/energy/profiles` | 列出 pricing profile 结构(前端按它渲染表单)|
| 价格 | `GET /api/energy/prices?start&end` | 区间价格点(曲线)|
| 费用 | `GET /api/energy/costs?start&end` | 区间 `energy_cost_period`(走势/明细)|
| 费用 | `GET /api/energy/costs/summary?start&end` | 区间汇总(计量电费 + 固定费 − 抵扣)|
| 费用 | `POST /api/energy/costs/recompute?start&end` | 幂等重算(补数据/改价后,显式)|
| DSMR | `GET /api/energy/dsmr/latest` | 最新 `dsmr_reading` |
| 测试 | `POST /api/energy/tibber/test` | 试连 Tibber + 拉当前价,三态 |
> DSMR/Tibber 的**标量配置 + token** 复用现有 `GET/PUT /api/config`(只加 CONFIG_FIELDS)。**价格数值全在合同里**,不进 flat config。
## 5. 已锁定决策(讨论后拍板)
1. **两层电价模型**profile YAML 定结构(仓库、固定、UI 不可编辑)+ `EnergyContract`(+版本) 存数值(UI 填、版本化)+ strategy 按 kind 出价。仿 M5。
2. **kind 不叫 "fixed"**`manual`(人工填、可双费率、可带时段)/ `tibber`API 动态);合同 `name` UI 自由填。
3. **买价**tibber = API `total`(全包,已证 total=energy+tax);manual = `energy_buy_档 + energy_tax`。**卖价**tibber = `total energy_tax sell_adjust`manual = `sell_档`(回送价,无能源税)。均含 VAT。
4. **双费率**manual 用 `delivered_1/2``returned_1/2` 分 dal/normal 计价(`_1`=dal/低、`_2`=normal/高);tibber 求和、15min 价不分档。
5. **两层费用**:每 15min `energy_cost_period` 只算计量电费(不可变、快照价);日/月/年汇总再加固定费(按月→天)− heffingskorting(按年→天)。
6. **回送阶梯罚金(terugleverkosten)不做**:按自然年累计、用户住不到年底算不准——不算、不记、不加功能(留痕见 §10)。
7. **DSMR 整帧存 JSON blob**(不做字段 allowlist),10s 降采样;数值转 Decimal、null 容错、`id` 幂等。扩 `MqttManager` 订阅端。
8. **5 张表**`dsmr_reading``energy_contract``energy_contract_version``tibber_price``energy_cost_period`。单电表不设 device 表。
9. **不可变 / 审计**:算账快照价 + 合同版本只增不改 + tibber 价天然不可变;重算显式 opt-in。
10. **Tibber 用 httpx + `priceInfoRange(QUARTER_HOURLY)`**,仅 active=tibber 时抓;token 走 secret config。
11. **复用 M5 expose/Discovery**:加 `_energy_cost_provider`;累计成本/收入用 `total_increasing` 喂 HA Energy。
12. **前端并入 Energy 视图**:合同管理表单按 profile 结构渲染;Recharts 画价/费。
13. **opt-in 开关**`dsmr_ingest_enabled` 默认 falseTibber 抓取/计费由 active 合同存在性驱动。
14. **周期边界用 UTC 刻钟**(NL 偏移整小时,本地/UTC 重合)。**不新增依赖**。
> 项目定位:个人自用、家庭特化、不开源——按单用户场景简化,不过度抽象。
## 6. 任务依赖图
```
Phase A(地基)
M6-T01 [schema] 5 张表 + 模型 + 迁移
M6-T02 flat 配置(dsmr_* + tibber token/home_id ← 与 T01 并行
M6-T03 pricing profile 框架(manual.yaml/tibber.yaml + pydantic + strategy 注册表) ← 纯模块,可早做
Phase B(合同 + 数据接入)
M6-T04 EnergyContract CRUD + 版本 + 按 profile 校验(API (Depends T01,T03)
M6-T05 Tibber 客户端 + 抓价 service + 调度 + tibber/test (Depends T01,T02,T03)
M6-T06 MqttManager 扩订阅 + DSMR ingestblob/10s (Depends T01,T02)
Phase C(计算 + 反哺)
M6-T07 计费引擎(每15min 计量 + 汇总)+ 周期 job + 重算 (Depends T01,T03,T04,T05,T06)
└─► M6-T08 energy_cost expose provider + state 发布 (Depends T07)
Phase DAPI + 前端)
M6-T09 Energy 数据 APIprices/costs/summary/dsmr-latest/recompute (Depends T05,T07)
└─► M6-T10 前端:合同管理 + 价格/费用视图 + Tibber 测试 (Depends T04,T09)
收尾
M6-T11 文档 + OpenAPI + roadmapDepends 全部)
```
`T01``T02``T03` 无前置可先开。
---
## 7. 原子任务(任务卡)
> 后端任务沿用校验闸门(`pytest` / `ruff` / 改路由或 schema 则 `export_openapi` 重导出入库);前端闸门见 §8。**不新增依赖**。纯新增、不删/移文件;改 `app/main.py` lifespan 时不得破坏既有 job/连接。
### M6-T01 — 5 张表与模型 `[schema]`
- **Status**: `todo` · **Depends**: none
- **Context**: 单库 app 链新增 5 张表(§3.5)。只建 schema + 模型,不写采集/计算/接口。
- **Files**: `create app/models/energy.py``DsmrReading`/`EnergyContract`/`EnergyContractVersion`/`TibberPrice`/`EnergyCostPeriod`);`create alembic_app/versions/20260623_11_energy_tables.py``modify alembic_app/env.py``scripts/app_db_adopt.py``APP_BASELINE_REVISION`→新 head);`create tests/test_energy_models.py`
- **Steps**:
1. 按 §3.5 定义模型(`Mapped[...]`):`DsmrReading``recorded_at` idx、`source_id` unique nullable、`payload` JSON);`EnergyContract``name`/`kind`/`active`/`currency`/时间戳);`EnergyContractVersion``contract_id` FK RESTRICT、`effective_from`/`effective_to` nullable、`values` JSON、`created_at`);`TibberPrice``starts_at` unique、`resolution``energy/tax/total` float、`level` nullable、`currency``fetched_at`);`EnergyCostPeriod``period_start` unique、`d1/d2/r1/r2_kwh` float、`import_cost/export_revenue/net_cost` float、`currency``pricing` JSON、`contract_version_id` FK RESTRICT nullable、`degraded` bool、`computed_at`)。
2. 新 revision`down_revision="20260622_10_exposed_entities"``upgrade()` 建 5 表 + 索引/唯一/FK`downgrade()` 反向 drop。更新 `APP_BASELINE_REVISION`
- **Out of scope**: 不写客户端/采集/计费/接口;不动其它模型。
- **Acceptance criteria**:
- [ ] 全新临时库 upgrade 到 head 含 5 表及约束;`downgrade -1` 干净回滚;链上单 head。
- [ ] FK 为 RESTRICT;唯一/索引正确;`payload`/`values`/`pricing` 为 JSON 列。
- [ ] `APP_BASELINE_REVISION` == 新 head。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 列/约束/FK 与 §3.5 一致;`recorded_at`/`period_start`/`starts_at` 真实索引或唯一列;`env.py` 已 import 新模型;迁移 `down_revision` 指当前真实 head;无破坏性操作。
### M6-T02 — flat 配置(DSMR + Tibber 凭据)
- **Status**: `todo` · **Depends**: none
- **Context**: DSMR 订阅设置 + Tibber 凭据接入扁平配置(前端自动渲染)。**价格数值不在这里**(在合同里)。
- **Files**: `modify app/config.py``app/services/config_page.py`CONFIG_FIELDS + `_settings_payload`);`modify .env.example``tests/test_api_config.py`
- **Steps**: `Settings``dsmr_ingest_enabled: bool=False``dsmr_mqtt_topic: str="dsmr/json"``dsmr_sample_interval_s: int=10``tibber_api_token: str=""``tibber_home_id: str=""`CONFIG_FIELDS 归「DSMR」「Tibber」section`tibber_api_token` 为 secretbool=checkbox、数值=number`_settings_payload` 补齐;`.env.example` 加注释样例(默认 off)。
- **Out of scope**: 不建客户端/采集;不动既有字段;不放任何价格常数。
- **Acceptance criteria**:
- [ ] 新项在 `GET /api/config` 分 sectiontoken 为 secret(回空/留空保留);非法值 422 不写库。
- [ ] 校验闸门全绿(OpenAPI 变化则重导出)。
- **Reviewer checklist**: secret 不回显/不进 OpenAPI 示例;`_settings_payload` 无遗漏;默认 off。
### M6-T03 — pricing profile 框架 + strategy 注册表
- **Status**: `todo` · **Depends**: none
- **Context**: 仓库内 `manual.yaml`/`tibber.yaml` 定结构(pydantic 校验),加 price strategy 注册表(manual/tibber 出价)。纯模块,单测。
- **Files**: `create app/integrations/pricing/__init__.py``pricing/profiles.py`pydantic 模型 + `load_profile`/`list_profiles`/`validate_values`)、`pricing/profiles/manual.yaml``pricing/profiles/tibber.yaml``pricing/strategies.py``register_strategy`/`get_strategy``manual`/`tibber` 实现);`create tests/test_pricing_profiles.py``tests/test_pricing_strategies.py`
- **Steps**:
1. `profiles.py`pydantic 模型描述 §3.1 结构;`load_profile(kind)`(读 YAML,校验失败抛错);`validate_values(kind, values)`(合同数值是否符合结构)。
2. `profiles/*.yaml`:按 §3.1 写 manual / tibber 结构。
3. `strategies.py`:策略接口 `price_period(deltas, t0, values, session) -> {import_cost, export_revenue, net, pricing_snapshot}`Decimal):
- `manual`:双费率公式(§3.4),`buy_x = energy_buy_x + energy_tax + ode``sell_x` 直接用。
- `tibber`:取覆盖 t0 的 `tibber_price``buy=total``sell=totalenergy_taxsell_adjust`,进出口求和。
- **Out of scope**: 不连真实 APItibber 策略只读 `tibber_price` 表,由 T05 填);不写采集/调度/接口。
- **Acceptance criteria**:
- [ ] 单测:`load_profile("manual"/"tibber")` 校验通过;缺字段/类型错抛可识别错;`validate_values` 正确判合规。
- [ ] 单测:manual 策略双费率出价正确(`import = Δd1×buy_dal + Δd2×buy_normal` 等);tibber 策略 `buy=total``sell=totalenergy_taxsell_adjust`,负 total 出口得负值。
- [ ] Decimal 计算无 float 误差。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: profile 纯结构、无具体数值;策略用 Decimal;进出口分开(不相抵);tibber 价匹配用 `starts_at ≤ t0` 最近一条;无 OOP 过度抽象(数据+函数,仿 M5 profiles.py)。
### M6-T04 — EnergyContract CRUD + 版本 + 校验(API
- **Status**: `todo` · **Depends**: M6-T01, M6-T03
- **Context**: 合同增删改、版本(改价加新版本、生效日期)、激活(一次一个)、按 profile 校验数值。
- **Files**: `create app/api/routes/api/energy_contracts.py``app/schemas/energy_contract.py``app/services/contracts.py``modify app/main.py`(注册路由);`create tests/test_api_energy_contracts.py`
- **Steps**:
1. `contracts.py` service:建合同(含首版本,`validate_values` 校验)、加版本(带 `effective_from`,自动闭合上一版本 `effective_to`)、激活(把其它 active 置 false)、`active_contract_version_at(session, ts)`(取 active 合同在 ts 生效的版本)。
2. 路由:`GET/POST/GET{id}/PATCH{id}``POST {id}/versions``GET /profiles`(返回结构供前端渲染表单);session+CSRF;数值不合规 422。
- **Out of scope**: 不算费用(T07);不发 MQTT;不写采集。
- **Acceptance criteria**:
- [ ] CRUD + 版本 + 激活行为正确:新建按 profile 校验(不合规 422);加版本自动闭合上一段;激活互斥(同时只一个 active);`GET /profiles` 返回结构。
- [ ] 未登录 401、缺 CSRF 403schema 经 OpenAPI 固化入库。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 版本只增不改(不覆盖旧版本);激活互斥;删除受 FK RESTRICT(有版本/有费用记录不可裸删);数值校验走 T03 `validate_values`;路径键用合同 id。
### M6-T05 — Tibber 客户端 + 抓价 service + 调度 + tibber/test
- **Status**: `todo` · **Depends**: M6-T01, M6-T02, M6-T03
- **Context**: httpx 拉 15min 价 upsert `tibber_price`;启动+每日抓(仅 active=tibber);连接测试。
- **Files**: `create app/integrations/tibber/__init__.py``tibber/client.py``app/services/tibber_prices.py``modify app/main.py`(抓价 job);`create tests/test_tibber_client.py``tests/test_tibber_prices.py`
- **Steps**:
1. `client.py``fetch_price_range(token, home_id|None) -> list[PricePoint]`(§3.3 query,httpx 超时/认证失败抛错,不假设固定节点数);`fetch_current_price(...)`(供 test)。
2. `tibber_prices.py``refresh_prices(session, settings)`(拉今天+明天 upsert,幂等);仅当 active 合同 kind=tibber 且 token 存在时执行。
3. `main.py`:同步 wrapper + `IntervalTrigger``max_instances=1, coalesce=True`,吞异常不崩。
- **Out of scope**: 不算费用(T07);不发 MQTT;test 路由在 T09(本任务备 service)。
- **Acceptance criteria**:
- [ ] 单测(httpx mock):解析返回为 PricePoint;认证失败/超时抛可识别异常。
- [ ] 单测:`refresh_prices` 幂等(按 `starts_at` upsert,不重复插);非 tibber active 时 no-op。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: token 不进日志;httpx 超时;upsert 幂等;时间存 UTC;无对 `tibber_price` 的破坏性批删。
### M6-T06 — MqttManager 扩订阅 + DSMR ingest
- **Status**: `todo` · **Depends**: M6-T01, M6-T02
- **Context**: 给只发不收的 `MqttManager` 扩订阅;DSMR ingest 整帧 blob + 10s 降采样落库。
- **Files**: `modify app/integrations/mqtt.py`(订阅注册 + on_message 分发);`create app/services/dsmr_ingest.py``modify app/main.py`(启用时注册订阅);`create tests/test_mqtt_subscribe.py``tests/test_dsmr_ingest.py`
- **Steps**:
1. `mqtt.py``subscribe(topic, handler)``on_connect` 里对已注册 topic subscribe`on_message` 按 topic 分发,handler 异常吞掉不崩连接);保持既有 publish 不回归。
2. `dsmr_ingest.py``handle_message(payload_bytes, settings)`parse JSON → 取 `timestamp`(UTC) → 降采样(`second % dsmr_sample_interval_s != 0` 丢弃)→ **整帧存 payload**(值原样/Decimal 友好)→ 短 session 写 `dsmr_reading``source_id`=`id`,存在则跳过)。
3. `main.py``dsmr_ingest_enabled` 时把 handler 注册到 `dsmr_mqtt_topic`
- **Out of scope**: 不算费用(T07);不改既有 publish/discovery 语义;不做字段 allowlist(整帧存)。
- **Acceptance criteria**:
- [ ] 单测:`subscribe` 注册后 `on_message` 分发到 handler;既有 publish 无回归。
- [ ] 单测:喂样本 JSON(字符串数值、null 相位),秒=00 整帧落盘、秒≠10 整数倍丢弃;同 `id` 不重插。
- [ ] `dsmr_ingest_enabled=false` 不订阅/不落盘。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: on_message 网络线程内只做短事务、吞异常不崩连接;整帧存(含 gas/各相);null 容错;`source_id` 幂等;10s 降采样正确。
### M6-T07 — 计费引擎 + 周期 job + 汇总 + 重算
- **Status**: `todo` · **Depends**: M6-T01, M6-T03, M6-T04, M6-T05, M6-T06
- **Context**: 每 15min 用寄存器差 × active 合同 strategy 算计量电费(快照、不可变);汇总加固定费 − 抵扣;周期 job + 重算。
- **Files**: `create app/services/energy_cost.py``modify app/main.py`(周期 job);`create tests/test_energy_cost.py`
- **Steps**:
1. `energy_cost.py``register_at(session, boundary)`(取 ≤boundary 最后一行的 d1/d2/r1/r2Decimal);`compute_period(session, t0)`(取 active 版本 + strategyT03/T04)→ 算 → upsert + 快照 `pricing` + `contract_version_id`,缺价/缺数据标 `degraded` 或跳过);`compute_closed_periods(session)``recompute_range(session, start, end)`(幂等覆盖);`summarize(session, start, end)`(Σnet + 固定费按天 heffingskorting 按天,取 active 版本值)。
2. `main.py`1 分钟 tick job`max_instances=1, coalesce=True`),有 active 合同 + DSMR 数据时算;吞异常不崩。
- **Out of scope**: 不发 MQTTT08);不加 HTTPT09,备 service);不改采集/价格抓取。
- **Acceptance criteria**:
- [ ] 单测:构造 `dsmr_reading`+合同版本(manual 双费率 / tibber + `tibber_price`),`compute_period` 算出正确 cost/revenue/net,快照价 + `contract_version_id` 落库;upsert 幂等。
- [ ] 单测:跨价格版本时按 t0 选对版本;缺价跳过、缺读数标 `degraded`
- [ ] 单测:`summarize` = Σnet + 固定费(月→天×天数)− heffingskorting(年→天×天数)。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: Decimal 算钱;寄存器差为"末−初";进出口分开;周期边界 UTC 刻钟;快照不可变(重算才覆盖,且显式);版本选择按 `effective_from≤t0<effective_to`job 不崩。
### M6-T08 — energy_cost expose provider + state 发布
- **Status**: `todo` · **Depends**: M6-T07
- **Context**: 复用 M5 expose/Discovery,发当前买/卖价 + 累计成本/收入。
- **Files**: `modify app/integrations/expose.py``_energy_cost_provider` + 注册);`modify app/services/energy_cost.py``app/main.py`(算完顺带 `publish_states`);`create tests/test_energy_expose.py`
- **Steps**: provider 产出 §3.7 实体(`buy_price_now`/`sell_price_now`/`import_cost_total`/`export_revenue_total`),累计项 `total_increasing`+`monetary`+currency`value_getter``energy_cost_period` 取;计费 job 后 `publish_states`
- **Out of scope**: 不改 Discovery 机制本身;不改采集/计费。
- **Acceptance criteria**:
- [ ] 单测:`build_catalog` 含 energy_cost 实体;累计项 `total_increasing`;勾选默认未勾。
- [ ] 单测:`value_getter` 取到正确当前价/累计值;MQTT 未启用整链 no-op。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: `key` 稳定(固定字符串);复用既有 provider 协议、未改其它 provider;累计 `total_increasing` 语义正确;不阻塞计费 job。
### M6-T09 — Energy 数据 API
- **Status**: `todo` · **Depends**: M6-T05, M6-T07
- **Context**: 价格/费用/汇总/最新 DSMR/重算/Tibber 测试端点(合同 CRUD 在 T04)。
- **Files**: `create app/api/routes/api/energy.py``app/schemas/energy.py``modify app/main.py`(注册);`create tests/test_api_energy.py`
- **Steps**: `GET /prices``GET /costs``GET /costs/summary``GET /dsmr/latest``POST /costs/recompute`(调 T07)、`POST /tibber/test`(调 T05 client,三态);session+CSRF;查询走索引 + 窗口/上限。
- **Out of scope**: 不写采集/计费/发布逻辑(复用 service)。
- **Acceptance criteria**:
- [ ] 各端点行为正确:窗口+limit 上限生效;recompute 幂等;tibber/test 三态;401/403 正确;schema 入库(`git diff --exit-code openapi/` 无差异)。
- [ ] 校验闸门全绿。
- **Reviewer checklist**: 查询有时间窗+上限;recompute 复用 T07、无破坏删;test 不泄 token。
### M6-T10 — 前端:合同管理 + 价格/费用视图 + Tibber 测试
- **Status**: `todo` · **Depends**: M6-T04, M6-T09
- **Files**: `modify frontend/src/pages/EnergyPage.tsx``create frontend/src/energy/ContractManager.tsx``ContractForm.tsx``TibberPrices.tsx``CostView.tsx`、扩 `hooks.ts``modify` Config 页 Tibber 测试入口;`create` 对应 `*.test.tsx`
- **Steps**: 合同列表/新建/编辑/激活/加版本(**表单字段按 `/profiles` 结构渲染**,manual 双费率/税/固定费/抵扣)+ 版本历史只读;价格曲线 + 费用走势/明细 + 汇总卡片;Tibber 测试三态;全走类型化 client;空/加载/错误态 + 缺 key 容错。
- **Out of scope**: 不做服务端降采样;不改后端。
- **Acceptance criteria**:
- [ ] 合同 CRUD/激活/版本可用,表单按 profile 渲染;价格/费用/汇总渲染正确;区间只取窗口;币种/单位来自 API。
- [ ] Tibber 测试三态可用。
- [ ] 前端闸门全绿(`lint`/`typecheck`/`test`/`build`)。
- **Reviewer checklist**: 全走类型化 client;图表组件隔离复用;表单按结构渲染、不 hardcode 字段;窗口/上限;空/错/加载/缺 key 容错。
### M6-T11 — 文档 + OpenAPI + roadmap 收尾
- **Status**: `todo` · **Depends**: 全部
- **Files**: `modify README.md``docs/roadmap.md`M6 行 + 毕业)、`docs/architecture-overview.md``docs/design/README.md`(列入 m6);`run python scripts/export_openapi.py` 并提交 `openapi/`
- **Acceptance criteria**:
- [ ] 文档反映新链路;`git diff --exit-code openapi/` 无未提交差异;校验闸门全绿。
- **Reviewer checklist**: 无残留过时描述;OpenAPI 入库;§13 已敲定项同步回文档。
---
## 8. 前端校验闸门
`frontend/` 下:`npm ci && npm run lint && npm run typecheck && npm run test && npm run build`(留意 chunk 体积告警)。后端同任务改路由/schema 仍需根目录 `export_openapi.py` 并提交 `openapi/`。**不新增前端依赖**(Recharts 已在)。
## 9. 构建上下文完整性(M1 教训)
- **纯新增、不删/移文件****不新增依赖**httpx/paho/pyyaml 已在)。新增源文件都在 `app/``scripts/``frontend/` 既有 `COPY` 范围内,无需改 `Dockerfile`**`app/integrations/pricing/profiles/*.yaml` 是非 .py 资源**——确认随 `COPY app` 进镜像、运行期按 `__file__`/`importlib.resources` 定位(别用 CWD 相对路径)。`tests/test_deployment.py::test_dockerfile_copy_sources_exist` 仍应通过。
- 改 `app/main.py` lifespan 注册新 job/订阅时**不得破坏**既有 public-ip / modbus-poll / discovery job 与 MQTT 连接。
- 发版前置走查(CLAUDE.md):真起 app → 订到 `dsmr/json`、(合同生效后)拉一次 Tibber 价、看费用落库、前端合同/价格/费用视图人工瞄一眼,再打 tag。
## 10. 后续杠杆(本里程碑不做,文档留痕)
- **回送阶梯罚金(terugleverkosten**:按自然年累计、阶梯式;本期不做。将来做就把阶梯表存进合同结构,在**年度汇总**层按 YTD 累计回送量套阶梯(不进每周期引擎)。
- **燃气 / 区域供暖计费**:数据已在 blob 里(gas `extra_device_*`);将来在合同结构加"商品价"+ `energy_cost_period``commodity` 维度即可,不改主链。
- **能源税分档**:年用电跨 10000 kWh 档时税率变;现按单档配置。
- **降采样 / 保留**`dsmr_reading` 10s 长期行数大,加定期降采样/保留窗口任务(SQL 端 `AVG(json_extract(...))`)。
- **价格 level 配色 / 便宜时段提示**:用 Tibber `level` 驱动前端配色与用电建议。
- **HA 让 HA 算 vs 后端算**:当前后端算累计;也可只发买/卖价 €/kWh 让 HA Energy 自乘(实体都已具备)。
## 11. 人工验收 walkthrough(实现完成后)
> 受控手工验证(依赖家里 DSMR Reader + brokerTibber 部分待合同生效),不进自动化。
1. **合同(先用 manual,立即可用)**Energy 页新建一个 `manual` 合同,按表单填双费率/能源税/固定费/抵扣,激活。
2. **DSMR 落库**:开 `dsmr_ingest_enabled`、确认 topicMQTT Explorer 看 `dsmr/json` 有消息;等若干 10 秒 → `GET /api/energy/dsmr/latest` 有最新整帧;`dsmr_reading` 约每 10 秒一行(整帧 blob)。
3. **费用计算**:跨过一个整刻钟边界 → `GET /api/energy/costs` 出现该周期 import/export/cost/revenue/netmanual 双费率分档对得上手算);`GET /costs/summary` 含固定费/抵扣;改价加新版本后旧周期不变、`POST /recompute` 才覆盖。
4. **Tibber(合同生效后)**:建 `tibber` 合同、Config 填 token、点「Tibber 测试」三态成功;激活后抓价 → `GET /prices` 有 15min 价、费用切到动态。
5. **反哺 HA**Expose 勾选 energy_cost 实体、开 Discovery、重发 → HA 出现当前买/卖价 + 累计成本/收入;HA Energy 挂累计实体。
## 12. 里程碑完成定义(DoD
- 后端能:订 `dsmr/json` 按 10s 整帧落库;按 active 合同(manual 双费率 / tibber 动态)每 15min 用寄存器差 × 价算计量电费(不可变、快照);日/月/年汇总加固定费 − 抵扣;不做净计量;改价走版本、历史不回改。
- 复用 M5 Discovery 把当前买/卖价 + 累计成本/收入(`total_increasing`)发成 HA 实体。
- 前端 Energy 视图能管理合同(表单按 profile 渲染、激活、版本)、看价格曲线与费用走势/明细/汇总、测 Tibber。
- 后端 `pytest`/`ruff`/`export_openapi` + 前端 `lint/typecheck/test/build` 全绿且 `openapi/` 入库。
- README / architecture / roadmap / design 索引反映 M6。
## 13. 待确认 / TODO(拿到真实 token + 账单后钉死,均已落成配置/默认值,不阻塞实现)
1. **买价**:✅ tibber = API `total`demo 已证 total=energy+tax);manual = energy_buy_档 + energy_tax。无待办。
2. **卖价残差(tibber**`sell = total energy_tax sell_adjust``sell_adjust` 默认 0(买卖费相等抵消)。真实账单确认后若有残差再调。
3. **双费率寄存器映射**`_1`=dal/低、`_2`=normal/高(NL 惯例)——接价前用真实数据确认别接反(差价小但要对)。
4. **能源税年值**manual/tibber 的 `energy_tax` 默认 ~0.11082026 第一档含 VAT),按当年实际值核。
5. **Tibber 15min + 币种**:✅ 查询/分辨率已 demo 证实;仍需合同生效后用**真实 token** 确认 NL 返回真 15 分钟价 + 币种 EUR。
6. **manual 现行数值**(你的固定合同):normal 0.133 / dal 0.127 / 回送两档(现同价)/ 管理费 ≈€0.329/天(≈€9.87/月)/ 电网费 + heffingskorting 待填。
</content>
+20
View File
@@ -0,0 +1,20 @@
# Future Ideas / Backlog(暂无 Milestone
记录尚未排期的想法。等某条成形、值得集中推进时,再升级为 `docs/roadmap.md` 里的 milestone 并展开成 `docs/design/` 任务卡。**这里只是备忘,不是承诺。**
> 项目定位:**个人自用、针对自家场景特化,不开源**。因此设计可按单用户 / 自家需求简化,不必为通用性、多租户、对外发布做过度抽象。
## 数据与存储
- 增加更多数据类型 / 来源(持续扩展)。
- 针对**需要长期保存**的数据,考虑更合适的存储方案(当前全 SQLite;长期 / 大量数据可能需要更强的数据库)。
- 把 **Home Assistant 接收到的数据**纳入本系统做持久化 / 展示。
## 集成
- **MQTT**:让后端作为一个 MQTT client,双向收发数据。
## 前端 / 移动端
- **PWA**(**近期、可能并入 M2 或单独小里程碑**):在 React NativeM3)之前,用 PWA 把 web SPA 包装成"准手机 App"——可安装到桌面、响应式、离线壳。
- 影响当下设计:**M2 的 UI 从一开始就按移动端布局考虑**(响应式 + 合理的参数显示),为之后加 PWA 铺路。
## 备注
- 以上为临时记录(讨论 M2 范围时随手想到),后续可增删、重排优先级。
Binary file not shown.
+134
View File
@@ -0,0 +1,134 @@
# SDM120 Modbus 协议(从官方 PDF 提取)
> 来源:`docs/references/SDM120-MODBUS_Protocol.pdf`
> Eastron SDM120 Modbus Smart Meter Modbus Protocol Implementation **V2.4**
> 本文件是 PDF 的可读化提取,供本项目的 Modbus 采集驱动设计参考。**以官方 PDF 为准**,本文件如有出入以 PDF 为准。
## 0. 本项目的接入方式(重要)
SDM120 物理层是 **Modbus RTURS-485 串口)**。本项目通过一个 **Modbus-TCP 网关**接入:
- 后端用 **Modbus TCP**`IP:port`)连到网关,网关在串口侧转成 RTU 与电表通信。
- TCP 帧用 MBAP header、**无 CRC**CRC/Error Check 由网关在 RTU 侧处理)。本文档里 RTU 帧的 `Error Check (Lo/Hi)` 字段在 TCP 模式下不需要我们关心。
- **Slave Address / Unit ID = 电表的 Meter ID**(默认 `1`,范围 1–247),在 TCP 请求里作为 unit id 传入。
- 若以后直连串口(RTU),才需要管波特率 / 校验位 / CRC(见 §5 holding 寄存器)。
## 1. 协议帧格式
MODBUS 定义 master 查询 / slave 响应的格式。Eastron 电表用 16-bit 寄存器在主从间传值,**但实际数据是 32-bit IEEE-754 浮点**,因此每个测量参数占**两个相邻的 16-bit 寄存器**。
### Querymaster → slaveRTU 帧)
| 字段 | 说明 |
| --- | --- |
| Slave Address | 8-bit,目标从机地址 12470=广播,Eastron 不支持广播) |
| Function Code | 8-bit,功能码(Eastron 支持 **03 / 04 / 08 / 16(=0x10)** |
| Start Address (Hi/Lo) | 16-bit 起始寄存器地址;**寄存器成对使用、从 0 开始,所以起始地址必须是偶数** |
| Number of Points (Hi/Lo) | 16-bit 请求的寄存器数量;**也必须是偶数**(成对读浮点) |
| Error Check (Lo/Hi) | 16-bit CRCRTU 模式;TCP 网关模式无此字段) |
### Responseslave → master
| 字段 | 说明 |
| --- | --- |
| Slave Address | 响应从机地址 |
| Function Code | 与查询相同的功能码(表示识别并已响应) |
| Byte Count | 8-bit,本次返回的数据字节数 |
| Data (寄存器对) | 每个寄存器 Hi byte / Lo byte,按"高寄存器在前"排列 |
| Error Check (Lo/Hi) | 16-bit CRCRTU 模式) |
### Exception Response(异常响应)
- 异常响应的 Function Code = 查询功能码 **OR 0x80**(最高位置 1)。
- 数据是单字节 Error Code(异常码)。
- PDF 正文写了"见后文 Table Of Exception Codes",但提取的 8 页里**没有附上该表**。标准 Modbus 异常码(供参考,非本 PDF 内容):`01` 非法功能、`02` 非法数据地址、`03` 非法数据值、`04` 从机设备故障。
## 2. 功能码
| 功能码 | 作用 | 寄存器区 |
| --- | --- | --- |
| **04** | Read Input Registers(读输入寄存器,3X)—— **所有测量值都在这里** | 30001+ |
| **03** | Read Holding Registers(读保持寄存器,4X)—— 配置项 | 40001+ |
| **16 / 0x10** | Write Holding Registers(写保持寄存器,4X)—— 改配置 | 40001+ |
| 08 | Diagnostics(诊断) | — |
> ⚠️ **测量值用 FC 04(输入寄存器),不是 FC 03。** 这是最常见的踩坑点。
## 3. 浮点数据编码
- 每个参数 = **32-bit IEEE-754 float**,占两个相邻 16-bit 寄存器。
- **字序(word order= 大端:高寄存器在前。**
- **字节序(byte order)= 大端:寄存器内高字节在前。**
- 即整体就是标准大端 float`>f`),4 字节顺序 = `[Reg1 Hi][Reg1 Lo][Reg2 Hi][Reg2 Lo]`
**实例(来自 PDF**
| 含义 | 原始 4 字节 (hex) | 解码值 |
| --- | --- | --- |
| Volts 1 | `43 66 33 34` | `230.2` V |
| Demand Time | `3F 80 00 00` | `1.0` |
| Network Node | `42 70 00 00` | `60.0` |
> Python 解码:`struct.unpack('>f', bytes([0x43,0x66,0x33,0x34]))[0]``230.2`
> pymodbus 用 `BinaryPayloadDecoder.fromRegisters(regs, byteorder=Endian.BIG, wordorder=Endian.BIG)`
## 4. 输入寄存器表(FC 04 读测量值)
全部为 `Float`,长度 4 字节,每项占 2 个寄存器。"Hex 起始"是 Modbus 协议起始地址(即 Start Address Hi/Lo)。
| 寄存器 | 参数 | 单位 | Hex 起始 |
| --- | --- | --- | --- |
| 30001 | Voltage(电压) | Volts | `0000` |
| 30007 | Current(电流) | Amps | `0006` |
| 30013 | Active power(有功功率) | Watts | `000C` |
| 30019 | Apparent power(视在功率) | VA | `0012` |
| 30025 | Reactive power(无功功率) | VAr | `0018` |
| 30031 | Power factor(功率因数) | — | `001E` |
| 30071 | Frequency(频率) | Hz | `0046` |
| 30073 | Import active energy(导入有功电能) | kWh | `0048` |
| 30075 | Export active energy(导出有功电能) | kWh | `004A` |
| 30077 | Import reactive energy(导入无功电能) | kvarh | `004C` |
| 30079 | Export reactive energy(导出无功电能) | kvarh | `004E` |
| 30085 | Total system power demand | W | `0054` |
| 30087 | Maximum total system power demand | W | `0056` |
| 30089 | Import system power demand | W | `0058` |
| 30091 | Maximum import system power demand | W | `005A` |
| 30093 | Export system power demand | W | `005C` |
| 30095 | Maximum export system power demand | W | `005E` |
| 30259 | Current demand | Amps | `0102` |
| 30265 | Maximum current demand | Amps | `0108` |
| 30343 | Total active energy(总有功电能) | kWh | `0156` |
| 30345 | Total reactive energy(总无功电能) | kvarh | `0158` |
**读取分块建议**`0x00000x005E`(30001–30095)地址连续,可一次块读;`0x0102/0x0108``0x0156/0x0158` 各为独立小块。整表用 2–3 次块读即可覆盖,减少 Modbus 事务数。
### 常用核心子集(日常监控够用)
电压 `0000`、电流 `0006`、有功功率 `000C`、功率因数 `001E`、频率 `0046`、导入有功电能 `0048`、导出有功电能 `004A`、总有功电能 `0156`
## 5. 保持寄存器表(FC 03 读 / FC 16 写配置)
| 寄存器 | 参数 | Hex 起始 | 格式 | 说明 |
| --- | --- | --- | --- | --- |
| 40013 | Relay Pulse Width | `000C` | Float | 继电器脉宽 60/100/200 ms,默认 100ms |
| 40019 | Network Parity Stop | `0012` | Float | 0=1停止位无校验(默认),1=1停止位偶校验,2=1停止位奇校验,3=2停止位无校验;**改后需重启生效** |
| 40021 | Meter ID | `0014` | Float | 从机地址 1247,默认 1 |
| 40029 | Baud rate | `001C` | Float | 0=2400(默认)1=48002=96005=1200 |
| 40087 | Pulse 1 output mode | `0056` | Float | 0001 导入有功,0002 导入+导出有功,0004 导出有功(默认),0005 导入无功,0006 导入+导出无功,0008 导出无功 |
| 463745 | Time of scroll display | `F900` | HEX 2字节 | 滚动显示时间 0–30s,默认 0(不滚动) |
| 463761 | Pulse 1 output | `F910` | HEX 2字节 | 0000:0.001kWh/imp(默认)0001:0.010002:0.10003:1 kWh/imp |
| 463777 | Measurement mode | `F920` | HEX 2字节 | 1:total=import2:total=import+export(默认)3:total=import-export |
| 464513 | Serial number | `FC00` | uint32 4字节 | 序列号,**只读** |
| 464515 | Meter code | `FC02` | Hex 2字节 | 设备码=`0020`**只读** |
| 464516 | Software version | `FC03` | Hex 2字节 | 软件版本,**只读** |
> ⚠️ 写保持寄存器(改 Meter ID / 波特率 / 校验位等)会改变电表通信参数,配错可能导致**通信中断**。本项目默认**只读采集**,不建议在自动化链路里写电表配置。
## 6. 给本项目采集驱动的要点小结
1. 走 **Modbus TCP 网关**`ModbusTcpClient(host, port)``slave=<Meter ID>`
2. 测量值用 **FC 04 / 输入寄存器**,按 §4 地址读。
3. 解码 **大端 float32**word & byte 都大端,高寄存器在前)。
4. 起始地址与数量**都用偶数**(成对读)。
5. 默认**只读**,不写电表配置寄存器。
6. 不同型号电表 → 不同"寄存器 profile"。本表是 `sdm120` 这一个 profile 的定义。
@@ -0,0 +1,249 @@
# Tibber API + 荷兰电价 + DSMR Reader 参考
> 本文件汇总 M6(动态/固定电价 → 实时买卖电费计算)所需的全部外部事实,供 `docs/design/m6-tibber-dynamic-energy.md` 与实现者直接 refer。
> 来源:Tibber GraphQL introspection(实证)、Tibber Explorer demo 数据、Tibber NL support 文档、用户实际合同与 DSMR Reader 配置(2026-06 整理)。
> 凡标「✅ 实证」的为代码/数据验证过的事实;金额类以用户真实账单 / 合同生效后真实 token 为最终准(见末尾「待真实数据核对」)。
---
## 1. Tibber GraphQL API
- **Endpoint**`POST https://api.tibber.com/v1-beta/gql`
- **认证**HTTP header `Authorization: Bearer <token>`token 在 [developer.tibber.com](https://developer.tibber.com) 登录后生成(每个 Tibber 账户一个)。
- **在线 Explorer**[developer.tibber.com/explorer](https://developer.tibber.com/explorer),自带一个公共 **demo token**
- ⚠️ **demo token 现在只能做 schema introspection 与历史 demo 数据查询**;对 `viewer` 的真实数据查询返回 `UNAUTHENTICATED`。要拿自己家的实时价,必须用**自己账户的 token**。
### 1.1 `Price` 类型(✅ introspection 实证)
`Price` 对象恰好 6 个字段:
| 字段 | 类型 | 含义 |
| --- | --- | --- |
| `total` | Float | **全包价 = energy + tax**(即 App 里的 All-in Price,含 spot+能源税+VAT 等)|
| `energy` | Float | 纯能源/现货分量(ex-VAT 市场价)|
| `tax` | Float | 税费分量(能源税 + VAT 合并)|
| `startsAt` | String | 该价格段起点(ISO8601,**含时区偏移**,如 `+02:00`|
| `currency` | String! | 币种(唯一 non-null 字段;NL 为 EURdemo 为 SEK|
| `level` | PriceLevel | 价格档枚举(`VERY_CHEAP/CHEAP/NORMAL/EXPENSIVE/VERY_EXPENSIVE`|
**✅ 实证**`total == energy + tax` 精确成立(见 §1.4 样本)。
### 1.2 15 分钟价格:`Subscription.priceInfoRange`(✅ introspection 实证)
- **字段**`currentSubscription.priceInfoRange`
- **参数**`resolution: PriceInfoRangeResolution!`(必填)、`first: Int``last: Int``before: String``after: String`(游标分页)
- **resolution 枚举 `PriceInfoRangeResolution`**:值 `DAILY` / `HOURLY` / `QUARTER_HOURLY`
- ⚠️ 注意别用错枚举:另有 `PriceResolution`(只有 HOURLY/DAILY**无 QUARTER_HOURLY**)和 `PriceInfoResolution`——`priceInfoRange` 用的是 **`PriceInfoRangeResolution`**。
- **返回**`SubscriptionPriceConnection { nodes: [Price]!, edges: [...], pageInfo }`
- **废弃说明**:老的 `priceInfo.range``deprecationReason: "use Subscription.priceInfoRange instead"`**不支持 QUARTER_HOURLY**,必须改用 `Subscription.priceInfoRange`
- **✅ 实证(demo key2018 历史日)**`QUARTER_HOURLY` 返回 **96 个节点、间隔精确 15 分钟**;无真 15 分钟价的历史日,API 把**小时价重复成 4 个相同刻钟**(00/15/30/45 同值,到下一整点才变);NL 当前数据则是真 15 分钟。
- **解析注意**:**不要假设固定 96 个节点 / 固定 15 分钟间隔**,按 `startsAt` 逐点存最稳。
**15 分钟查询(拉今天 96 个刻钟)**:
```graphql
{ viewer { homes { id currentSubscription {
priceInfoRange(resolution: QUARTER_HOURLY, first: 96) {
nodes { startsAt total energy tax currency level }
} } } } }
```
### 1.3 小时价 / 当前价 / 探测:`priceInfo`
`currentSubscription.priceInfo``current`(当前价)、`today[]``tomorrow[]`(次日价约 13:00 CET 发布)。探测账户能力 + 当前价:
```graphql
{ viewer { homes {
id appNickname
features { realTimeConsumptionEnabled }
currentSubscription { priceInfo {
current { total energy tax startsAt level currency }
today { total energy tax startsAt level }
tomorrow { total energy tax startsAt level }
} } } } }
```
### 1.4 现成 curl + 样本响应
```bash
# 15 分钟价(换成自己账户 tokendemo token 拿不到 viewer 真实数据)
curl -s -X POST https://api.tibber.com/v1-beta/gql \
-H "Authorization: Bearer <YOUR_TOKEN>" -H "Content-Type: application/json" \
-d '{"query":"{ viewer { homes { id currentSubscription { priceInfoRange(resolution: QUARTER_HOURLY, first: 96) { nodes { startsAt total energy tax currency } } } } } }"}'
```
**样本:`priceInfo`demo,瑞典家庭,SEK**
```json
{"current": {"total": 0.8239, "energy": 0.5663, "tax": 0.2576,
"startsAt": "2026-06-23T14:00:00.000+02:00", "level": "NORMAL", "currency": "SEK"},
"today": [{"total": 1.276, "energy": 0.928, "tax": 0.348,
"startsAt": "2026-06-23T00:00:00.000+02:00", "level": "VERY_EXPENSIVE"}, … 24 项],
"tomorrow":[… 24 项]}
```
**样本:`priceInfoRange` QUARTER_HOURLYdemo2018-11-0296 节点)**
```json
{"nodes": [
{"startsAt":"2018-11-02T00:00:00.000+01:00","total":0.63, "energy":0.435,"tax":0.195,"currency":"SEK"},
{"startsAt":"2018-11-02T00:15:00.000+01:00","total":0.63, "energy":0.435,"tax":0.195,"currency":"SEK"},
{"startsAt":"2018-11-02T00:30:00.000+01:00","total":0.63, "energy":0.435,"tax":0.195,"currency":"SEK"},
{"startsAt":"2018-11-02T00:45:00.000+01:00","total":0.63, "energy":0.435,"tax":0.195,"currency":"SEK"},
{"startsAt":"2018-11-02T01:00:00.000+01:00","total":0.6141,"energy":0.4223,"tax":0.1918,"currency":"SEK"},
… 共 96 个,末节点 23:45 total 0.6342]}
```
(注意 00:00–00:45 四个刻钟同值 = 小时价被重复;`total==energy+tax`。)
---
## 2. 荷兰电价构成(Tibber NL + 用户合同)
> 所有金额**含 VATincl. btw**,除非另注。VAT/BTW 标准税率 **21%**
| 分量 | 荷兰语 | 单位 | 值 / 说明 |
| --- | --- | --- | --- |
| 现货/市场价 | marktprijs / dynamische kwartierprijs | €/kWh | 每 15 分钟跟随交易所;= API `energy` |
| 买侧采购费 | **inkoopvergoeding** | €/kWh | **€0.0248**(覆盖 onbalans + 绿证;见下「相等」事实)|
| 卖侧上网费 | **verkoopvergoeding** | €/kWh | **€0.0248**2026-01-01 起)|
| 能源税 | energiebelasting | €/kWh(含 VAT| 2026 第一档(010000 kWh)≈ **€0.1108**2025 ≈ €0.1228;高档更低 |
| 增值税 | BTW | % | **21%** |
| 固定月费 | vast bedrag / leveringskosten | €/月/合同 | Tibber **€5.99/月**(电、气各一次)|
| 电网维护费 | netbeheerkosten / systeembeheerkosten | €/天或/月 | 电网公司定、供应商代收;按地区,2026 约 +3.38% |
| 能源税抵扣 | heffingskorting / vermindering energiebelasting | €/年/连接 | 政府年度减免,从总费用扣 |
| ODE | Opslag Duurzame Energie | €/kWh | **现为 0**(已并入 energiebelasting|
**✅ 关键事实(Tibber NL 文档原文)**
> "De verkoopvergoeding van 2,48 cent is gelijk aan de inkoopvergoeding die je bij je afgenomen stroom betaalt."
> (卖侧 verkoopvergoeding 2.48 分 = 买侧 inkoopvergoeding。)
**买卖服务费相等(均 €0.0248/kWh)**,在买卖里一进一出**相互抵消**。
---
## 3. 净计量(saldering)、回送(teruglevering)、负电价、2027
### 3.1 回送价(净计量期内,文档原文)
> "Op het moment dat je teruglevert geven we je per kWh de beursprijs die op dat moment geldt …, inclusief energiebelasting en inkoopvergoeding plus de btw minus de verkoopvergoeding."
即净计量期内回送价 = `beursprijs + energiebelasting + inkoopvergoeding + btw verkoopvergoeding`。因 inkoopvergoeding = verkoopvergoeding 抵消 → **= 全额零售价**spot+能源税+VAT),正是 saldering "回送 1 度 = 用 1 度"的本质。
### 3.2 年末盈余 / 取消净计量后(文档原文,Scenario 2)
> "Voor de overproductie van 500 kWh heb je recht op de beursprijs en de inkoopvergoeding, maar heb je geen recht op de energiebelasting. … ontvang je nog een factuur van ons voor de te veel uitgekeerde belastingen …"
即**超额回送(或 2027 取消净计量后)**:盈余按 `beursprijs + inkoopvergoeding verkoopvergoeding`(**无能源税**)计价 → 因两费抵消 → **= 纯 spot**。
### 3.3 净计量机制
- 法定**按自然年**结算;可抵扣到「用电量」为止(回送抵到用电为止)。
- Tibber 不用预付,**净计量周期在第 12 张账单后结束**。
- **2027 起荷兰取消净计量(saldering stopt in 2027**——用户据此设计:**不再考虑净计量,直接按实时买卖算**。
### 3.4 负电价
- 负价时**用电**:理论上你拿钱(但仍计能源税,净计量期内税会退回)。
- 负价时**回送**:你要为回送的电付那个负价(= 倒贴)。
- 因 Tibber 给的是 all-in `total`(已含能源税),**只有 spot 负到比能源税还多,total 才转负**;中等负价时 total 仍正(照付)。
- **结论:负电价不需要特殊处理**——`buy=total``sell=total−能源税` 符号自洽。
---
## 4. 本项目采用的买/卖价公式(M6)
> spot 取 API `energy``total = energy + tax`(全包)。**买价直接用 `total`**,卖价从 `total` 扣掉卖电不交的能源税。
- **Tibber 动态合同**post-2027 口径):
- 买价 `buy = price.total`
- 卖价 `sell = price.total energy_tax_per_kwh sell_adjust``sell_adjust` 默认 0;含 VAT 归己;买卖费抵消已隐含在 total 里)
- **固定合同(manual,双费率)**
- 买价 `buy_档 = energy_buy_档 + energy_tax`(档 ∈ {normal, dal}
- 卖价 `sell_档 = sell_档`(回送价,**无能源税**
- **固定费/抵扣不进每度**`network_fee``management_fee`(月→天)、`heffingskorting`(年→天)在**日/月/年汇总**层加减。
- 详见 `docs/design/m6-tibber-dynamic-energy.md` §3.4。
---
## 5. 用户当前固定合同(manual 实例,2026-06
> 双费率(NL`_1`=dal/低谷、`_2`=normal/高峰)。以下为当前值,会随合同阶段/年度变(系统按"加新版本+生效日期"留底)。
| 项 | 当前值 | 备注 |
| --- | --- | --- |
| 能源价 normal(高) | €0.133/kWh | 买侧,含 VAT,不含能源税 |
| 能源价 dal(低) | €0.127/kWh | 买侧 |
| 回送价 normal | (现与 dal 同)| 卖侧,分档但现同价;系统仍分开填 |
| 回送价 dal | — | 卖侧 |
| 能源税 energiebelasting | 待填(≈€0.1108| 含 VAT,加到买价 |
| ODE | 0 | 已并入能源税 |
| 电网维护费 | 待填 | 按天(合同按天收)|
| 供应商管理费 | ≈€0.329/天(≈€9.87/月)| 对应 Tibber 的 €5.99/月 |
| heffingskorting | 待填 | 年度减免,汇总时扣 |
| 回送阶梯罚金 terugleverkosten | **不做** | 按自然年累计、阶梯式;用户住不到年底算不准,M6 不实现 |
---
## 6. DSMR Reader → MQTT
### 6.1 取数方式
- 用 **Telegram JSON**(单 topic **`dsmr/json`**,一条 = 一帧完整 telegram 的 JSON),**不用** split-topic(每字段一 topic、要按 id 拼)、**不用** raw(未解析 OBIS)。
- DSMR Reader **每秒一条**。本项目按 **10 秒降采样**(仅秒数整 10 落盘)、**整帧存 JSON blob**(不做字段 allowlist)。
### 6.2 telegram JSON 字段(DSMR Reader 配置的 JSON mapping
```
id, timestamp,
electricity_delivered_1, electricity_delivered_2, # 进口累计 kWh_1=dal 低谷, _2=normal 高峰)
electricity_returned_1, electricity_returned_2, # 出口/回送累计 kWh
electricity_currently_delivered, electricity_currently_returned, # 瞬时进/出口功率 kW
phase_currently_delivered_l1/l2/l3, phase_currently_returned_l1/l2/l3, # 各相瞬时功率 kW
phase_voltage_l1/l2/l3, # 各相电压 V
phase_power_current_l1/l2/l3, # 各相电流 A
extra_device_timestamp, extra_device_delivered # 燃气表(m³,每 5 分钟更新)
```
### 6.3 实测样本(单相,`dsmr/json`
```json
{
"id": 200086230,
"timestamp": "2026-06-23T12:16:48Z",
"electricity_delivered_1": "20915.154",
"electricity_returned_1": "2979.905",
"electricity_delivered_2": "15212.090",
"electricity_returned_2": "6786.406",
"electricity_currently_delivered": "0.000",
"electricity_currently_returned": "2.704",
"phase_currently_delivered_l1": "0.000",
"phase_currently_delivered_l2": null,
"phase_currently_delivered_l3": null,
"extra_device_timestamp": "2026-06-23T12:15:00Z",
"extra_device_delivered": "6208.234",
"phase_currently_returned_l1": "2.704",
"phase_currently_returned_l2": null,
"phase_currently_returned_l3": null,
"phase_voltage_l1": "237.0",
"phase_voltage_l2": null,
"phase_voltage_l3": null
}
```
### 6.4 解析要点(实测确认)
- **数值都是 JSON 字符串**`"20915.154"``"0.000"`)→ 计费读取转 **Decimal**(累计寄存器算钱要精度)。
- **缺测相位 = `null`**(不是缺 key)→ 视为缺测;单相只有 `*_l1`,三相后 `_l2/_l3` 由 null 变数字。
- **进口总量** = `electricity_delivered_1 + _2`**出口总量** = `electricity_returned_1 + _2`(双费率档对动态合同无意义,求和;对固定合同分档计价)。
- **`timestamp` 为 UTCZ**`id` 自增,做幂等去重。
- 历史样本里电压 key 曾被错配成带前缀 `dsmr/reading/phase_voltage_l1`DSMR Reader 的 JSON mapping 配置笔误,用户已改对)——解析仍按"键名容错"。
- **整表寄存器与相数无关**`delivered/returned_1/2` 是整表总量,三相只多了各相瞬时通道 → **计费逻辑相数无关**
- 燃气 `extra_device_delivered`(m³)一并存入 blob;燃气计费 M6 不做(数据先留,未来按 commodity 扩展)。
---
## 7. 来源 URL
- Tibber GraphQL reference / explorer`https://developer.tibber.com/docs/reference#rootsubscription``https://developer.tibber.com/explorer`
- Tibber NL 费用构成:`https://support.tibber.com/nl/articles/5605892-de-kosten-bij-tibber`
- Tibber NL 净计量与回送:`https://support.tibber.com/nl/articles/4669873-salderen-en-terugleveren-bij-tibber`
- DSMR ReaderHA 集成):`https://www.home-assistant.io/integrations/dsmr_reader/`
---
## 8. 待真实数据核对(合同生效后用真实 token / 账单)
1. **真实 token 复核**:跑 §1.4 的 15 分钟 curl,确认 NL 返回**真** 15 分钟价(非重复小时价)+ 币种 EUR。
2. **卖价残差**:确认 `total` 里 purchase fee 是否被卖侧 sales fee 完全抵掉、回送 VAT 口径 → 调 `sell_adjust`(默认 0)。
3. **双费率寄存器映射**:确认 `_1`=dal/`_2`=normal 没接反(差价小但要对)。
4. **能源税年值**:按当年实际值与年用电档位核 `energy_tax`
5. **固定合同数值**:回送两档价、电网费、heffingskorting 待用户从账单填。
</content>
+220
View File
@@ -0,0 +1,220 @@
# Roadmap
本文档记录 `home-automation``v1.0.3` 之后的下一阶段规划。这一阶段不是小修补,而是几次较大的结构性改动:单库化、前端重写、以及远期的移动端试水。
> 每个里程碑的**可执行原子任务**展开在 [`docs/design/`](./design/README.md)M1 [`m1-db-consolidation.md`](./design/m1-db-consolidation.md)、M2 [`m2-frontend-v2.md`](./design/m2-frontend-v2.md)、M3 [`m3-token-mobile.md`](./design/m3-token-mobile.md)、M4 [`m4-login-hardening.md`](./design/m4-login-hardening.md)、M5 [`m5-iot-energy.md`](./design/m5-iot-energy.md)。这些文档为 Orchestrator→Implementer→Reviewer 的多模型流水线设计。
## 当前基线(v1.0.3
- FastAPI + 服务端 Jinja 模板页面(目前只有 `/login``/config`
- 三个独立 SQLite 库:
- App DB`sqlite:///./data/app.db`
- Location DB`sqlite:///./data/locationRecorder.db`
- Poo DB`sqlite:///./data/pooRecorder.db`
- 三条独立 Alembic 链:`alembic_app/``alembic_location/``alembic_poo/`
- 单 admin 鉴权(Argon2 + server-side session cookie
- Public IPv4 monitor、SMTP 通知、Location / Poo recorder、Home Assistant in/out、TickTick OAuth
- 数据可视化目前由 Grafana provisioning 承担(仅 location / poo dashboard
- 已有 OpenAPI 导出脚本:`scripts/export_openapi.py`
## 本阶段正式退役的架构约束
`docs/architecture-overview.md` 里有几条当时刻意写死的约束,这一阶段明确退役:
- **“不引入前后端分离”** → 退役。本阶段改为 React SPA(仍由 FastAPI 同源托管,但渲染移到客户端)。
- **“三个独立 DB 不合并”** → 退役。本阶段把 location / poo 合并进 `app.db`
- **Grafana 作为可视化方案** → 退役。可视化由 React 前端自己承担(热力图、地图等)。
保持不变的约束:
- 继续使用 **SQLite**,本阶段不上 Postgres。
- 不引入 Notion。
## 里程碑总览
| 里程碑 | 主题 | 一句话 |
| --- | --- | --- |
| **M1** ✅ | 单库化地基 | 把三库合并成单一 `app.db`,清理散落数据层,删掉 Grafana |
| **M2** ✅ | 前端 v2 | React SPA 取代 Jinja,承载 config + 可视化 + 记录增删改 |
| **M4** ✅ | 登录加固 | 防爆破/指数退避 + CLI 逃生通道 + 可选 TOTP 二次验证(**先于 M5** |
| **M5** ✅ | IoT / 能耗采集 | 通用 Modbus 采集(YAML profile + JSON readings+ MQTT/HA Discovery + 前端侧边栏 + Energy 视图 |
| **M3** | 开放与移动端(远期试水) | token 鉴权 + React Native 移动端 |
排序原则:**先清地基,再在干净结构上盖楼。** M2 的新 API 和 React 必须建立在合并后的单库之上;M4 是公网安全加固,在 M5 IoT 集成之前先堵住裸密码这个洞;M5 在安全基座就绪后再做 IoT 接入。
---
## M1 — 单库化地基(✅ 已完成)
### 目标
把 location / poo 两个独立库合并进 `app.db`,借机清理项目早期散落各处的数据访问代码,并移除 Grafana。
### 范围
- **Alembic 收敛为单链(app 链)**location / poo 的表此后纳入 app 链管理;`alembic_location/``alembic_poo/` 退出活跃使用(保留在 git 历史)。
- **新建表(schema only**:在 app 链上加一条 upgrade revision,把原来两个旧库里的表**原样**建到 `app.db` 中。Alembic **不需要知道任何旧数据**——它只负责把 app DB 往上升一个版本、建出这两张新表。
- **数据搬迁交给独立脚本**`scripts/migrate_legacy_data.py`(见下方“迁移策略”),手动跑一次。
- **配置层收敛**:去掉 `LOCATION_DATABASE_URL` / `POO_DATABASE_URL`,统一到 `APP_DATABASE_URL`
- **开启 SQLite WAL**:单文件 + Web + APScheduler 并发写入,开 WAL 更稳。
- **删除 Grafana**:移除 compose 中的 grafana service、`grafana/provisioning/``grafana/dashboards/`。直接删除,不再 re-point datasource。
- **更新文档**README、architecture-overview 同步反映单库现实。
### 注意
- **可视化空窗可接受**M1 删掉 Grafana 后、到 M2 React 可视化落地之前会有一段没有可视化面板的时间。已确认可以接受。
- **历史数据是第一优先级,绝不能丢**(见“数据安全原则”)。
---
## 迁移策略(M1 核心)
职责拆分得很清楚:**Alembic 管 schema,脚本管数据。**
### Alembic revision(只建结构)
- 一条 app 链上的 upgrade revision,建出与旧库**完全相同**的表结构。
- 确定性、与环境无关:在生产机、CI、全新部署上都一样地建空表,不依赖任何旧文件是否存在。
- 本步**只原样挪表,不顺手改 schema**。任何表结构清理留到之后一条单独的 migration 去做——不可替代的历史数据,一次只承担一种风险。
### 数据搬迁脚本(`scripts/migrate_legacy_data.py`
- 把旧 `locationRecorder.db` / `pooRecorder.db` 里的行,拷进 `app.db` 的新表(SQLite `ATTACH DATABASE` 或单独连接均可)。
- **幂等**:重复运行不会重复插入。
- **搬完对账**:逐表核对源 / 目标行数,对不上就报错中止。
- 只在生产机上**手动跑一次**,不进 Alembic 永久链路(避免把一次性历史搬迁焊死进每次全新建库都要跑的链路里)。
### 旧库的“撤掉”
- “撤掉旧库” = ① 配置不再指向它们 + ② 文件**归档保留**。
- **绝不**在任何脚本 / migration 里 `os.remove` 旧文件——那不可逆,且踩数据安全红线。
- 真正的删除是**人工、最后、确认无误之后**单独的一步。
---
## 数据安全原则
历史数据(location / poo 记录)是这个项目里最不可替代的东西,迁移期间一律按以下原则:
1. **迁移前先归档**旧 `.db` 文件一份。
2. **先在副本上演练**:把每日备份恢复到一个 scratch 目录,在副本上跑完整迁移、核对行数无误,再对真实库动手。
3. **脚本幂等 + 行数对账**,对不上立即中止。
4. **旧文件只读归档、绝不自动删除**,删除是事后人工动作。
---
## M2 — 前端 v2React SPA)✅ 已完成
### 目标
用 React SPA 取代现有 Jinja 页面,由 FastAPI 同源托管(同一容器、同一 origin)。这一步合并了“前端重写为 React”和“前端做厚”两件原本分开的事——它们本质是同一坨活。
> 备注:React 是一次 agentic programming 试水。之前只手写过 Vue、没手写过 React,这一轮想全程靠 agent、尽量不读代码地把它做出来。OpenAPI 导出 → 生成类型化 TS client 作为 agent 的契约护栏,正好服务这个目标。
### 范围
- **React SPA**,FastAPI 挂载打包后的静态产物(同源,省掉 CORS)。
- **Config 界面**:取代现有 Jinja config 页。
- **数据可视化**:热力图、地图等,接管原先 Grafana 干的事。
- **按需展示 DB 数据**(例如 poo 记录)。
- **记录的小幅增删改**:用于修正不准确的记录。
### 后端配套
- **补一套 JSON API**:SPA 是客户端渲染,需要后端提供 config 读写、数据查询、记录 CRUD 等 JSON 端点。(同源不等于不需要 API——API 是“客户端怎么拿数据”,与文件托管在哪无关。)
- **鉴权**:浏览器面向的新端点(含记录 CRUD)复用现有 session cookie 保护。
- **类型化 client**:用 `scripts/export_openapi.py` 的输出生成 TS client。
### 鉴权边界(与 M3 衔接)
- 现在那个”裸 API 记小狗日志”的 ingestion 端点(设备 / 脚本调用,非浏览器)**维持现状到 M3**。
- M2 新增的、浏览器调用的 CRUD 端点,用 session 保护即可,本步不引入 token。
> **M2 已完成**M2-T01 至 M2-T13 全部 done)。Jinja 模板已移除,React SPA 同源托管,多阶段 Docker 构建通过,所有校验闸门绿。
---
## M4 — 登录加固(✅ 已完成,排在 M5 之前)
### 目标
给暴露在公网的单 admin 登录做纵深防御,先于 M5 IoT 集成关闭暴力枚举和单因子风险。
### 范围
- **防爆破 / 指数退避**:失败登录按双键(IP + username)指数增长延迟,成功即清零。退避是延迟(429 + Retry-After),不是永久封号。全局开关 `AUTH_LOGIN_THROTTLE_ENABLED`CONFIG_FIELDS);反代后需开 `AUTH_TRUST_FORWARDED_FOR``.env` 部署级)。
- **CLI 逃生通道**`python -m scripts.admin_cli` 直连本地 DB,无需 HTTP 服务、无需任何已存凭据;支持重置密码、解锁退避、关停/重发 TOTP、查看用户。
- **可选 TOTP 二次验证**:admin 可自选启用;启用后两步登录(密码 + 6 位动态码或一次性恢复码);不启用维持纯密码。CLI `disable-totp` 是连恢复码都丢了时的最终逃生口。
> 详细设计与任务卡:[`docs/design/m4-login-hardening.md`](./design/m4-login-hardening.md)
---
## M5 — IoT 集成与能耗采集(✅ 已完成)
### 目标
给后端接入家庭 IoT 生态,建立通用 Modbus 设备采集链路(首个领域:能耗),接入 MQTT + Home Assistant Discovery,并重构前端为侧边导航并新增 Energy 视图。
### 范围
- **两层数据模型(协议与部署分离)**YAML profile(随代码走,声明协议知识:寄存器/解码/key/unit/ha_component+ `modbus_device`(部署层 DB 行:friendly_name/host/port/unit_id/profile/poll_interval/enabled+ `modbus_reading`(通用遥测:device_id + recorded_at + JSON payload)。多设备可共享同一 profile。
- **通用 Modbus-TCP 采集**pymodbusAPScheduler 后台 job 轮询所有 enabled 设备;per-device `last_poll_at`/`last_poll_ok`;全局开关 `MODBUS_POLLING_ENABLED`CONFIG_FIELDS)。首个 profileSDM120 单相电表。
- **手工 CLI 试读**`scripts/modbus_cli``read`/`probe` 两个只读子命令),供受控手工验证,不进自动化。
- **MQTT + HA Discovery**paho-mqtt 长连接;通用 expose 框架(provider 动态产出可暴露实体目录,元数据从 YAML profile 派生);`exposed_entity_toggle` 表存逐 key 开关(默认不暴露);每设备 = 一个 HA device,各量 = sensor entity + online binary_sensor`unique_id` 锚定于设备 `uuid`(稳定,不随改名变);配置变更可重连重发 discovery。
- **前端侧边栏**`AppShell` 侧边导航(Home / Records / Energy / Config + 主题切换 + 注销),移动端可折叠,当前路由高亮。
- **Energy 视图**`/energy`):设备 CRUD;最新读数卡片(标签/单位取自 profile metrics 端点);Recharts 走势图(时间范围 + limit)。
- **Config 页 Accordion**:各大 section 折叠/展开;「Home Assistant Expose」面板勾选可暴露实体 + 连接状态 + 重新发布按钮。
- **API**`/api/modbus/*` + `/api/expose` + `/api/config/mqtt/test`):完整 CRUD + readings + metrics + test + expose 勾选 + 重发 discovery。
- `pytest`/`ruff`/`export_openapi` + 前端 `lint/typecheck/test/build` 全绿,`openapi/` 已入库。
### 命名决策(已锁定)
存储/采集/API 全部使用通用 `modbus_*``/api/modbus/devices`),**不锁死"电表"**;面向用户的领域呈现叫 **Energy**。接入新 Modbus 设备型号只需新增 YAML profile,无需改表/改 API。
> 详细设计与任务卡:[`docs/design/m5-iot-energy.md`](./design/m5-iot-energy.md)
---
## M3 — 开放与移动端(远期试水)
### 目标
引入 token 鉴权并做一个 React Native 移动端。**明确是很远期、低投入的试水**——先把 React 前端做出来,之后才会碰移动端,且主要是想试试没做过的 React / React Native。
### 范围
- **OAuth-lite token 签发**:移动端在内置浏览器里用账号密码登录,走一遍类 OAuth 流程,服务端签发一个 bearer token 给 app 存起来使用。(本质是没有第三方的 Authorization Code 简化版。)
- **React Native 移动端**:试水性质。
- **给 ingestion 端点上 token**:把 M2 暂时维持裸奔的设备端点收口到 token 鉴权下。
### 为什么放最后
- 移动端是这一阶段最远期、最不确定的部分。
- token 主要是移动端的前置条件;Web 端 React 用现有 session cookie 即可,不需要为它提前引入 token。
## 下一阶段:已确定要做(尚未拆解为任务卡)
> 这些是 M5 之后**已经定下来要做**的方向——区别于下面的 Future Ideas(仅备忘、未必做)。这里只记到 roadmap 粒度:确定**做什么、为什么**;具体排期、依赖与原子任务,等动手时再展开成 `docs/design/` 的任务卡。**先后顺序未定**,具体排期等动手时再定。
### 1. 前端优化
**动机**M2 的 React SPA 先把功能跑通,性能 / 体验层面的打磨还没做。这一项**确定要做,但具体优化什么还没定**。
**范围(待定)**:方向先留空,想清楚再细化。可能的候选(仅占位、非承诺):打包体积与代码分割(M2/M5 构建已提示存在 > 500 kB 的单 chunk)、首屏加载、热力图 / 地图的渲染性能、移动端适配、可访问性等。等确定具体目标后再拆任务卡。
### 2. 设置页生成 Long-lived Token(供 API 调用)
**动机**:浏览器端走 session cookie 即可,但**脚本 / 设备 / 外部程序调用 API** 需要一种长期有效、可随身携带的凭据。在设置页加一组功能,由 admin **手动签发 long-lived token**,之后用它来调 API。
**范围(粗略,待细化)**
- 设置页新增「API Token」区:生成 / 命名 / 吊销 long-lived token;明文只在**生成时展示一次**,此后只存哈希。
- 后端支持用该 token 鉴权访问 API(与现有 session cookie 并存,互不影响)。
- 与 [M3](#m3--开放与移动端远期试水) 的 token 主题相关,但**这条是 Web 设置页手动签发的 PAT 风格**,不依赖移动端 OAuth 流程;两者实现时可复用同一套 token 存储 / 校验。
## Future Ideas(暂不排期,想到先记下)
> 这里收集**还没排进里程碑、也还没决定要不要做**的想法。不是承诺、也没有先后顺序;想做时再从这里捞出来——先升进上面的「下一阶段」,再细化成 `docs/design/` 的任务卡。
_(暂无条目。)_
+7
View File
@@ -0,0 +1,7 @@
node_modules/
dist/
dist-ssr/
*.local
.env
.env.*
!.env.example
+209
View File
@@ -0,0 +1,209 @@
# Home Automation — Frontend
React SPA for the home-automation backend. Built with Vite + React 18 + TypeScript.
Scaffolded in M2-T06; feature pages filled in by T07T10.
## Stack
| Layer | Library | Version |
|---|---|---|
| Build | Vite | 6.x |
| UI framework | React | 18.x |
| Language | TypeScript | 5.x |
| Component library | Mantine | 7.x |
| Data fetching | TanStack Query | 5.x |
| Routing | react-router-dom | 6.x |
| API client codegen | openapi-typescript | 7.x |
| API client runtime | openapi-fetch | 0.17.x |
| Testing | Vitest + @testing-library/react | 4.x / 14.x |
## npm Scripts
| Command | What it does |
|---|---|
| `npm run dev` | Start Vite dev server (with backend proxy — see below) |
| `npm run build` | `tsc -b && vite build` — type-check then build to `dist/` |
| `npm run preview` | Serve the built `dist/` locally |
| `npm run lint` | ESLint (flat config, React + TypeScript rules) |
| `npm run typecheck` | `tsc --noEmit` — type-check without emitting files |
| `npm run test` | Vitest (run once, no watch) |
| `npm run codegen` | Regenerate `src/api/schema.d.ts` from `../openapi/openapi.json` |
All frontend gates must pass before any task is considered done:
```bash
npm run codegen
npm run lint
npm run typecheck
npm run test
npm run build # must produce dist/
```
## Directory Structure
```
frontend/
├── index.html Vite entry HTML
├── vite.config.ts Vite + Vitest config; dev proxy
├── tsconfig.json References tsconfig.app.json + tsconfig.node.json
├── tsconfig.app.json App source TS config (strict, react-jsx)
├── tsconfig.node.json Vite config TS config
├── eslint.config.js Flat ESLint config (React + TypeScript rules)
├── package.json Dependencies + npm scripts
├── package-lock.json Lockfile (committed; CI uses npm ci)
└── src/
├── main.tsx Entry point; mounts <App> into #root
├── App.tsx Provider stack + route tree (MantineProvider → QueryClient → Router → SessionProvider)
├── vite-env.d.ts /// <reference types="vite/client" /> for CSS imports
├── test-setup.ts Vitest global setup (@testing-library/jest-dom)
├── api/
│ ├── schema.d.ts AUTO-GENERATED from openapi/openapi.json (committed)
│ ├── client.ts openapi-fetch client + CSRF/cookie/401 middleware
│ └── csrf.ts Module-level CSRF token holder (setCsrfToken / getCsrfToken)
├── auth/
│ ├── SessionProvider.tsx TanStack Query against GET /api/session; exposes useSession()
│ └── ProtectedRoute.tsx Redirects to /login when unauthenticated
└── pages/
├── LoginPage.tsx Placeholder → T07 builds the real form
├── HomePage.tsx Placeholder → T09 builds the map/heatmap view
└── ConfigPage.tsx Placeholder → T08 builds the config editor
```
## Dev Proxy (local development)
`npm run dev` starts Vite on port 5173. The Vite config proxies API/auth paths
to the FastAPI backend running on port 8000:
| Proxied path | Backend URL |
|---|---|
| `/api/*` | `http://localhost:8000` |
| `/login` | `http://localhost:8000` |
| `/logout` | `http://localhost:8000` |
| `/static/*` | `http://localhost:8000` |
| `/docs` | `http://localhost:8000` |
| `/openapi.json` | `http://localhost:8000` |
To develop locally:
1. Start the backend: `uvicorn app.main:app --reload --host 0.0.0.0 --port 8000`
2. Start the frontend: `cd frontend && npm run dev`
3. Open `http://localhost:5173` — the app proxies all API calls to the backend.
Since the dev server proxies the session cookie path, auth flows work exactly as
they would in the deployed (same-origin) setup.
## Adding a New Page + Typed Query
This is the pattern every task T07T10 follows to wire up a real page:
### 1. Run codegen (if the OpenAPI contract changed)
```bash
npm run codegen
```
The generated `src/api/schema.d.ts` is committed to the repo. CI enforces that
the file is in sync with `openapi/openapi.json` via:
```bash
npm run codegen && git diff --exit-code frontend/src/api/schema.d.ts
```
### 2. Import the typed client
```typescript
// src/pages/SomePage.tsx
import apiClient from '../api/client'
```
### 3. Write a typed TanStack Query
```typescript
import { useQuery } from '@tanstack/react-query'
import apiClient from '../api/client'
function usePooRecords(limit = 100) {
return useQuery({
queryKey: ['poo', { limit }],
queryFn: async () => {
const res = await apiClient.GET('/api/poo', { params: { query: { limit } } })
// res.data is typed as PooResponse | undefined
// On non-2xx the middleware throws ApiError; TanStack Query catches it.
return res.data
},
})
}
```
The `params.query` and `params.path` objects are fully typed from `schema.d.ts`.
TypeScript will error if you pass unknown query params or mistype a path param.
### 4. Write a typed mutation (write request)
```typescript
import { useMutation, useQueryClient } from '@tanstack/react-query'
import apiClient from '../api/client'
function useDeletePoo() {
const qc = useQueryClient()
return useMutation({
mutationFn: (timestamp: string) =>
apiClient.DELETE('/api/poo/{timestamp}', {
params: { path: { timestamp } },
}),
onSuccess: () => qc.invalidateQueries({ queryKey: ['poo'] }),
})
}
```
The middleware (`src/api/client.ts`) automatically injects the `X-CSRF-Token` header
on all non-GET/HEAD requests (sourced from `getCsrfToken()`). You do not need to
handle CSRF manually in page code.
### 5. Add the route in App.tsx
```typescript
// App.tsx
import { SomePage } from './pages/SomePage'
// Inside <Routes>:
<Route path="/some-path" element={<SomePage />} />
// or, if protected:
<Route
element={
<ProtectedRoute>
<AppLayout />
</ProtectedRoute>
}
>
<Route path="/some-path" element={<SomePage />} />
</Route>
```
## OpenAPI codegen + CI sync rule
`src/api/schema.d.ts` is committed to the repository (not gitignored).
**Rule**: whenever `openapi/openapi.json` changes (any backend task that modifies
a route or schema), CI must run:
```bash
cd frontend && npm run codegen
git diff --exit-code frontend/src/api/schema.d.ts
```
If the file has changed but the new version was not committed, CI fails.
To update manually after a backend change:
```bash
cd frontend
npm run codegen
git add src/api/schema.d.ts
git commit -m "M2-Txx: update generated OpenAPI types"
```
## Production Build
The production build (`npm run build`) writes static files to `frontend/dist/`.
In the deployed setup (M2-T11 onwards), FastAPI serves `dist/` as a static
directory and falls back to `dist/index.html` for all non-`/api` paths,
enabling client-side routing with deep links.
The multi-stage Dockerfile (M2-T12) builds the frontend in a Node container and
copies only `dist/` into the Python image — the production image does not
contain Node or npm.
+29
View File
@@ -0,0 +1,29 @@
import js from '@eslint/js'
import globals from 'globals'
import reactHooks from 'eslint-plugin-react-hooks'
import reactPlugin from 'eslint-plugin-react'
import tseslint from 'typescript-eslint'
export default tseslint.config(
{ ignores: ['dist', 'src/api/schema.d.ts'] },
{
extends: [js.configs.recommended, ...tseslint.configs.recommended],
files: ['**/*.{ts,tsx}'],
languageOptions: {
ecmaVersion: 2020,
globals: globals.browser,
},
plugins: {
'react-hooks': reactHooks,
react: reactPlugin,
},
rules: {
...reactHooks.configs.recommended.rules,
'react/react-in-jsx-scope': 'off',
'@typescript-eslint/no-unused-vars': [
'error',
{ argsIgnorePattern: '^_', varsIgnorePattern: '^_' },
],
},
},
)
+13
View File
@@ -0,0 +1,13 @@
<!doctype html>
<html lang="en">
<head>
<meta charset="UTF-8" />
<link rel="icon" type="image/svg+xml" href="/vite.svg" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Home Automation</title>
</head>
<body>
<div id="root"></div>
<script type="module" src="/src/main.tsx"></script>
</body>
</html>
+7682
View File
File diff suppressed because it is too large Load Diff
+51
View File
@@ -0,0 +1,51 @@
{
"name": "home-automation-frontend",
"private": true,
"version": "0.0.0",
"type": "module",
"scripts": {
"dev": "vite",
"build": "tsc -b && vite build",
"preview": "vite preview",
"lint": "eslint .",
"typecheck": "tsc --noEmit",
"test": "vitest run",
"codegen": "openapi-typescript ../openapi/openapi.json -o ./src/api/schema.d.ts"
},
"dependencies": {
"@mantine/core": "^7.17.8",
"@mantine/hooks": "^7.17.8",
"@tanstack/react-query": "^5.101.0",
"@types/leaflet": "^1.9.21",
"@types/leaflet.markercluster": "^1.5.6",
"leaflet": "^1.9.4",
"leaflet.heat": "^0.2.0",
"leaflet.markercluster": "^1.5.3",
"openapi-fetch": "^0.17.0",
"qrcode.react": "^4.2.0",
"react": "^18.3.1",
"react-dom": "^18.3.1",
"react-feather": "^2.0.10",
"react-leaflet": "^4.2.1",
"react-router-dom": "^6.30.4",
"recharts": "^3.8.1"
},
"devDependencies": {
"@eslint/js": "^9.39.4",
"@testing-library/jest-dom": "^6.9.1",
"@testing-library/react": "^14.3.1",
"@testing-library/user-event": "^14.6.1",
"@types/react": "^18.3.31",
"@types/react-dom": "^18.3.7",
"@vitejs/plugin-react": "^4.7.0",
"eslint": "^9.39.4",
"eslint-plugin-react": "^7.37.5",
"eslint-plugin-react-hooks": "^7.1.1",
"jsdom": "^29.1.1",
"openapi-typescript": "^7.13.0",
"typescript": "^5.9.3",
"typescript-eslint": "^8.61.0",
"vite": "^6.4.3",
"vitest": "^4.1.8"
}
}
+138
View File
@@ -0,0 +1,138 @@
/**
* App top-level provider stack and route tree.
*
* Provider order (outermost first):
* MantineProvider QueryClientProvider BrowserRouter SessionProvider routes
*
* Route tree:
* /login LoginPage (public)
* /change-password ProtectedRoute ChangePasswordPage (forced password change gate)
* / ProtectedRoute AppLayout HomePage
* /config ProtectedRoute AppLayout ConfigPage
* /records ProtectedRoute AppLayout RecordsPage
* /energy ProtectedRoute AppLayout EnergyPage
*
* AppLayout renders a sidebar (AppSidebar) on the left; page content via <Outlet/> on the right.
* /login and /change-password have no sidebar layout.
*/
import { useState } from 'react'
import { QueryClient, QueryClientProvider } from '@tanstack/react-query'
import { BrowserRouter, Routes, Route, Outlet } from 'react-router-dom'
import { MantineProvider, AppShell, Burger } from '@mantine/core'
// Mantine requires its CSS to be imported once.
import '@mantine/core/styles.css'
import { SessionProvider } from './auth/SessionProvider'
import { ProtectedRoute } from './auth/ProtectedRoute'
import { LoginPage } from './pages/LoginPage'
import { HomePage } from './pages/HomePage'
import { ConfigPage } from './pages/ConfigPage'
import { RecordsPage } from './pages/RecordsPage'
import { EnergyPage } from './pages/EnergyPage'
import { ChangePasswordPage } from './pages/ChangePasswordPage'
import { AppSidebar } from './components/AppSidebar'
// ---------------------------------------------------------------------------
// TanStack Query client (singleton, created outside render to avoid re-creation)
// ---------------------------------------------------------------------------
const queryClient = new QueryClient({
defaultOptions: {
queries: {
// Don't retry on 4xx — we handle 401 in the middleware
retry: (failureCount, error) => {
if (error instanceof Error && 'status' in error) {
const status = (error as unknown as { status: number }).status
if (status >= 400 && status < 500) return false
}
return failureCount < 2
},
},
},
})
// ---------------------------------------------------------------------------
// App shell layout (used by all protected pages)
// ---------------------------------------------------------------------------
function AppLayout() {
// Mobile navbar open/close state — controlled here so the Burger and
// AppShell share the same state.
const [navbarOpen, setNavbarOpen] = useState(false)
return (
<AppShell
navbar={{
width: 220,
breakpoint: 'sm',
collapsed: { mobile: !navbarOpen },
}}
header={{ height: { base: 48, sm: 0 } }}
padding="md"
>
{/* Burger shown on mobile only (hidden on sm+) */}
<AppShell.Header hiddenFrom="sm" p="xs">
<Burger
opened={navbarOpen}
onClick={() => setNavbarOpen((o) => !o)}
size="sm"
aria-label="Toggle navigation"
data-testid="nav-burger"
/>
</AppShell.Header>
<AppSidebar onNavClick={() => setNavbarOpen(false)} />
<AppShell.Main>
<Outlet />
</AppShell.Main>
</AppShell>
)
}
// ---------------------------------------------------------------------------
// Root app
// ---------------------------------------------------------------------------
export default function App() {
return (
<MantineProvider defaultColorScheme="auto">
<QueryClientProvider client={queryClient}>
<BrowserRouter>
<SessionProvider>
<Routes>
{/* Public routes */}
<Route path="/login" element={<LoginPage />} />
{/* Forced password change — protected (must be logged in) but outside AppLayout */}
<Route
path="/change-password"
element={
<ProtectedRoute>
<ChangePasswordPage />
</ProtectedRoute>
}
/>
{/* Protected routes — all nested under AppLayout (sidebar) */}
<Route
element={
<ProtectedRoute>
<AppLayout />
</ProtectedRoute>
}
>
<Route index element={<HomePage />} />
<Route path="/config" element={<ConfigPage />} />
<Route path="/records" element={<RecordsPage />} />
<Route path="/energy" element={<EnergyPage />} />
</Route>
</Routes>
</SessionProvider>
</BrowserRouter>
</QueryClientProvider>
</MantineProvider>
)
}
+62
View File
@@ -0,0 +1,62 @@
/**
* csrfMiddleware 401-handling regression tests.
*
* Bug: clicking Logout (or landing on /login) flooded GET /api/session with 401s
* and the page hung instead of returning to the login screen.
*
* Root cause: the middleware redirected on EVERY 401, including the session
* probe's own 401. The redirect invalidated the ['session'] query, which
* refetched GET /api/session, which 401'd, which redirected again an infinite
* refetch loop. These tests pin the fix: the session probe and the login
* endpoint own their 401s (no redirect); any other endpoint's 401 still
* redirects (session expired mid-use).
*
* We call onResponse() directly (rather than going through apiClient.GET) so the
* test exercises the exact 401 branch without the singleton's relative baseUrl,
* which has no absolute origin to resolve against under jsdom.
*/
import { describe, it, expect, vi, beforeEach } from 'vitest'
import type { Middleware } from 'openapi-fetch'
import { csrfMiddleware, registerLoginRedirect } from './client'
type OnResponse = NonNullable<Middleware['onResponse']>
type OnResponseParams = Parameters<OnResponse>[0]
/** Build the minimal onResponse params for the given schema path + response. */
function params(schemaPath: string, response: Response): OnResponseParams {
return { schemaPath, response, request: new Request('http://test.local' + schemaPath) } as OnResponseParams
}
function response401(): Response {
return new Response(JSON.stringify({ detail: 'unauthorized' }), {
status: 401,
headers: { 'Content-Type': 'application/json' },
})
}
const onResponse = csrfMiddleware.onResponse as OnResponse
describe('csrfMiddleware 401 redirect (session-flood regression)', () => {
const redirect = vi.fn()
beforeEach(() => {
redirect.mockReset()
registerLoginRedirect(redirect)
})
it('does NOT redirect when GET /api/session returns 401 (probe owns its 401)', async () => {
await onResponse(params('/api/session', response401()))
expect(redirect).not.toHaveBeenCalled()
})
it('does NOT redirect when POST /api/auth/login returns 401 (bad credentials)', async () => {
await onResponse(params('/api/auth/login', response401()))
expect(redirect).not.toHaveBeenCalled()
})
it('redirects when a normal endpoint returns 401 (session expired mid-use)', async () => {
await onResponse(params('/api/locations', response401()))
expect(redirect).toHaveBeenCalledTimes(1)
})
})
+125
View File
@@ -0,0 +1,125 @@
/**
* Typed API client built on openapi-fetch + generated schema.d.ts.
*
* Middleware contract (orchestrator-decisions.md §11):
* 1. Always send cookies (credentials: "include"; same-origin auto-sends but explicit is clear).
* 2. Non-GET/HEAD requests inject X-CSRF-Token from the csrf holder.
* Exception: POST /api/auth/login skips injection (unauthenticated endpoint).
* 3. 401 responses clear session state + navigate to /login.
* 4. Other non-2xx responses throw an ApiError carrying the parsed JSON body,
* so callers (e.g. SMTP test) can inspect body.result.
*/
import createClient, { type Middleware } from 'openapi-fetch'
import type { paths } from './schema.d.ts'
import { getCsrfToken } from './csrf'
// ---------------------------------------------------------------------------
// Error type
// ---------------------------------------------------------------------------
/** Error thrown for non-2xx, non-401 responses. Carries the parsed JSON body. */
export class ApiError extends Error {
constructor(
public readonly status: number,
// eslint-disable-next-line @typescript-eslint/no-explicit-any
public readonly body: any,
) {
super(`API error ${status}`)
this.name = 'ApiError'
}
}
// ---------------------------------------------------------------------------
// Internal navigation helper (avoids React-router import at module level)
// ---------------------------------------------------------------------------
let _navigateToLogin: (() => void) | null = null
/**
* Register a callback that the middleware calls on 401.
* SessionProvider calls this during its setup.
*/
export function registerLoginRedirect(fn: () => void): void {
_navigateToLogin = fn
}
// ---------------------------------------------------------------------------
// CSRF middleware
// ---------------------------------------------------------------------------
const WRITE_METHODS = new Set(['POST', 'PUT', 'PATCH', 'DELETE'])
const LOGIN_PATH = '/api/auth/login'
/**
* Endpoints where a 401 is an EXPECTED, locally-handled outcome and must NOT
* trigger the global login redirect:
* - GET /api/session the session probe; 401 means "not logged in", handled
* by SessionProvider's queryFn (returns null unauthenticated state).
* - POST /api/auth/login bad-credentials check; 401 handled by LoginPage.
*
* Redirecting on these would invalidate the session query, which refetches
* /api/session, which 401s, which redirects again an infinite loop that
* floods GET /api/session after logout and on the login page.
*/
const SESSION_PATH = '/api/session'
const NO_REDIRECT_ON_401 = new Set<string>([SESSION_PATH, LOGIN_PATH])
export const csrfMiddleware: Middleware = {
async onRequest({ request }) {
// Always include cookies (same-origin; explicit for clarity)
// Note: credentials is set at client level; this is belt-and-suspenders doc.
const method = request.method.toUpperCase()
const url = new URL(request.url)
if (WRITE_METHODS.has(method) && url.pathname !== LOGIN_PATH) {
const token = getCsrfToken()
if (token) {
request.headers.set('X-CSRF-Token', token)
}
}
return request
},
async onResponse({ schemaPath, response }) {
if (response.status === 401) {
// The session probe and the login endpoint own their 401s (see
// NO_REDIRECT_ON_401). For any OTHER endpoint, a 401 means the session
// expired mid-use → redirect to /login. Crucially, NOT redirecting on the
// session probe breaks the refetch→401→redirect→refetch flood loop.
if (!NO_REDIRECT_ON_401.has(schemaPath) && _navigateToLogin) {
_navigateToLogin()
}
// Return the original response so callers can handle 401 if needed.
return response
}
if (!response.ok) {
// Parse body and throw; caller can catch ApiError and read .body
let body: unknown
try {
body = await response.clone().json()
} catch {
body = null
}
throw new ApiError(response.status, body)
}
return response
},
}
// ---------------------------------------------------------------------------
// Client instance
// ---------------------------------------------------------------------------
const apiClient = createClient<paths>({
baseUrl: '/',
credentials: 'include',
})
apiClient.use(csrfMiddleware)
export default apiClient

Some files were not shown because too many files have changed in this diff Show More