- Cleaned CATEGORY_MAP: removed dead entries (Furniture, GF Bev,
Snack/bev, Snack/food), fixed duplicate 'Food' key, proper
distinct icons for all 6 real categories
- Added categoryInfo() helper: 🥤/🍿/🍔/🔧/🔌/📦 for known cats,
❓ (red) for unmapped categories, 📦 for null
- Updated all render sites (list, map pins, map list, detail view,
scan result, OCR result) to use categoryInfo()
- Added category icons to detail view and scan/OCR result headers
- Synced categories DB table: added Equipment/Appliances with icons,
removed unused Unknown
- Updated seed data in both server.py and admin_server.py to match
real asset categories
- Added category_health to /api/stats with unmapped category detection
- Added Category Health card to admin dashboard (shown only when
unmapped categories exist)
- Updated CSS backgrounds for all real categories + .cat-unmapped
- Auto-save photo_path to matched assets when photo uploaded via /api/ocr
- Auto-extract and save serial_number from OCR text to blank-serial matched assets
- Add _extract_serial_from_text helper for serial number pattern matching
- Create scripts/backfill_vision_photos.py for batch-processing unlinked photos
- Add docs/OCR_PIPELINE.md documenting the full OCR/vision pipeline
- Fix test DB schema: add connect_id, equipment_id, barcode, customer_name
- Fix OCR test assertions to match actual endpoint behavior
- normalize_identifier() strips dots/dashes/prefixes, keeps alphanumeric
- find_asset_by_normalized_id() searches serial_number, connect_id,
equipment_id, barcode with normalized comparison
- /api/ocr now returns matched_assets in addition to legacy machine_id
- New /api/match-text endpoint for client-side text matching
- scripts/match_label_photo.py CLI tool for OCR + DB matching
- Vision model fixed (mimo-v2-omni at opencode.ai, was using
truncated placeholder key)
- Add QR_CODE, DATA_MATRIX, PDF_417, AZTEC to barcode scanner hints
so serial number stickers with 2D barcodes can be scanned
- Update /api/assets/search to also match connect_id so full
Connect IDs (e.g. VM-05912-0000068454) work when scanned
- Bump SW cache v5→v6, footer v3.2→v3.3
- New script: scripts/populate_asset_company_place.py maps connect_id →
account name (company/customer_name) + city (place) + address via
extraction DB (msdyn_customerasset + account join)
- 5,743/7,488 assets now have company + customer_name populated
- 5,647/7,488 assets now have place (city) populated
- Fix: SELECT COALESCE(c.name, a.customer_name) so stored customer_name
isn't overridden to NULL when customer_id FK is missing
- Bump SW v4→v5, version v3.1→v3.2 to force cache refresh
- Add/Find tab: removed check-in card and add-asset form, now pure lookup
- OCR: stop camera after capture, added GPS capture at photo time
- OCR: increased timeout from 8s to 15s to reduce 500 errors
- Nav tab: added GPS prompt + asset search when navigating directly
- Route: include all 26 techs from bookableresource table (not just Shawn)
- Added import_seed.py — reads Machine List(8).xlsx, matches by Asset ID → machine_id
- 51 new columns on assets table (route, barcode, sales, contact info, etc.)
- seed_data table with full 62-field JSON per asset for detail enrichment
- Seed data card in frontend asset detail view
- GET /api/assets/{id} now returns result.seed and result.seed_imported_at
- GPS columns from Seed ignored per request
- 1,654 matched, 199 unmatched (Seed-only assets not in MSFS)
Refs #47
OCR endpoint (/api/ocr) now auto-saves EXIF GPS from the photo
to the asset's latitude/longitude fields when they are blank,
so the first photo scan at a machine permanently records its location.
GET /api/assets/{id} now includes an 'msfs' block enriched from
the merged Dynamics 365 Field Service data (merged-assets.json)
loaded at server start (1,834 matched assets). Fields shown:
equipment_id, connect_id, compass_asset_number, manufacturer,
model_text, dex_model, install_date, software_version, etc.
Frontend updates:
- New 'Field Service Data' card in asset detail view when MSFS
data is available (manufacturer, Connect ID, install date, etc.)
- '📍 GPS saved to asset' confirmation shown after OCR auto-save
- AppState.ocrGpsSaved flag to track the auto-save event
- Add dex_report_date field to AssetCreate Pydantic model
- Include dex_report_date in _build_asset_insert columns/values
- Add test_list_assets_filter_no_dex_days test (11/11 pass)
- Added is_disney INTEGER DEFAULT 0 column to assets table
- Backfilled 996 Disney assets based on D- customer prefix
- Server: is_disney in AssetCreate/AssetUpdate models, INSERT/UPDATE
- Server: disney_filter query param now uses is_disney column (no JOIN needed)
- import_cantaloupe.py: sets is_disney=1 when customer starts with D-
- import_machines.py: sets is_disney=1 when customer starts with D-
- disney_classify.py: sets is_disney=1 when classifying
- Frontend: new Location Type dropdown with Disney/Non-Disney toggle
(replaces the __disney__/__non_disney__ park dropdown pseudo-options)
Backend:
- disney_filter query param on /api/assets (values: 'disney' | 'non_disney')
- Detects Disney by checking disney_park IS NOT NULL OR customer name LIKE 'D-%'
- Catches ESPN, Wide World of Sports, all Disney resorts, etc.
- Adds LEFT JOIN to customers table when disney_filter is active
Frontend:
- '🏰 Disney only' and '🏢 Non-Disney' options in Disney Park dropdown
- Active filter tag shows which is selected
- Works together with other dropdown filters
Backend:
- GET /api/customers — lists all customers for filter dropdown
- GET /api/locations?customer_id=X — cascading locations per customer
- GET /api/users — lists users for assigned_to filter
Frontend:
- Collapsible 'Filters' panel with Customer→Location cascade, Status, Assigned To
- Active filter tags showing what's currently filtering, removable individually
- 'Clear all filters' button to reset all at once
- Result count display ('X assets found')
- Better search placeholder showing searchable fields
- Filter count badge on the toggle button
- Added disney_park TEXT DEFAULT NULL to assets table in _create_v2_tables
- Created service_entrances table with id, location_id, name, latitude,
longitude, notes, created_at, updated_at; location_id FK to locations
with ON DELETE CASCADE; name/lat/lon are NOT NULL
- Added v4 migration in init_db: ALTER TABLE for disney_park on existing
DBs, schema-aware recreation of service_entrances if name column lacks
NOT NULL constraint
- 44/45 tests pass (1 pre-existing 404 failure on stats endpoint)
The test suite login as 'admin'/'changeme' but only 'shawn' was seeded.
Adding the admin user with SHA256('changeme') hash lets all 15 Playwright
tests reach the actual app pages instead of failing at login.
- Added .dng to PHOTO_ALLOWED_EXTS
- Bumped PHOTO_MAX_SIZE from 10MB to 20MB
- DNG files (~13MB from Pixel 9a) now accepted by all upload paths:
/api/upload/photo, /api/ocr, connect_labels
- Server validates extension+size only; DNG EXIF preserved naturally
by write_bytes(); _re_embed_exif() correctly skips non-JPEG;
_extract_gps_from_bytes() opens DNG as TIFF via PIL (already works)
Verified: .dng upload via /api/upload/photo and /api/ocr both succeed,
file saved byte-identical.
- Removed duplicate clearPickedPhoto() at old line ~1772
(kept newer version at line ~3857 with pickedPhotoGps clearing)
- Removed duplicate extractGpsFromPicked() at old line ~1831
(kept newer async version at line ~3766 that uses exifr)
- Both ocrPickedPhoto() and createAssetFromPicked() were already
singular — they were refactored in a prior edit
Verified: no duplicate function names, JS syntax OK, no regressions
- Client: exifr.parse() reads full EXIF before upload, sent as exif_data
- Server: piexif re-embeds EXIF into saved JPEG after upload
- Auto-extract GPS from picked gallery photos on selection
- Removed Google Photos button (needs OAuth, not feasible)
- _extract_gps_from_bytes helper reads GPSInfo IFD via PIL get_ifd()
- Converts DMS to decimal degrees, returns {lat, lng} or None
- /api/upload/photo: reads bytes first, extracts GPS before saving
- /api/ocr: extracts GPS from raw bytes before OCR processing
- exif_gps returned alongside path/machine_id in both endpoints
- Photos without EXIF GPS omit the exif_gps key entirely
- Raw file bytes preserved (no re-compression, no EXIF stripping)
- Removed extension whitelist — no more 'unsupported format' errors
- Falls back to .jpg temp file for unknown/no extension
- PIL detects actual image format from file bytes (handles DNG as TIFF)
- Android browsers often strip DNG extensions when picking from gallery
- Tested: works with .dng, .jpg, no extension, and .bogus
- Server: accept .dng in OCR endpoint (PIL opens as TIFF)
- Server: bump OCR size limit to 20MB (DNG files can be large)
- Client: gallery upload tries server OCR first (with 6s timeout),
then falls back to client-side Tesseract.js — matches 📸 capture flow
- DNG photos now work with 'Pick from Gallery'
- Improved geolocation error messages (maps error codes to friendly text: Location blocked, GPS unavailable, GPS timed out)
- Added location data to list_assets API (location_latitude, longitude, address, building_name, building_number, name) via LEFT JOIN
- Map pins now use location coordinates as direct fallback when asset lacks own lat/lng
- GeocodeAndPin now falls back to location address fields when asset address is empty
- All 96 map-related tests pass (66 API + 30 frontend)
- Fix Active Technicians: query users WHERE role='technician' instead of counting from visit data
- Interactive bar charts: hover tooltips show label+count, click drills down to filtered asset list
- Quick action icons: coin for Assets, Sites, Check-in replaces CSV export buttons
- No visit data: suggestion card with 'Add a Check-in' button
- No manufacturer data: hide section entirely when empty
- 30-second TTL cache on both backend (/api/stats) and frontend (loadDashboard)
- Add dashboard cache module with time-based invalidation
F1: Scan existing asset → auto check-in (handleBarcode now autoCheckins)
F2: Auto-generate address from GPS reverse geocode (Nominatim, all 3 modes)
F3: Larger camera viewport (3:4 aspect ratio, full-width)
F4: Building # already syncs to Address/Trailer field (existing)
F5: Parking location GPS button already exists (fillGpsParking)
F6: Navigation tool with route line on map (existing navigateToAsset)
Backend: Added GET /api/geocode endpoint for reverse geocoding
Previously assets never had coordinates populated - only checkins did.
Now when a checkin is created with GPS data, the asset's latitude/longitude
fields are auto-updated. Also backfilled existing assets from their latest
checkin history. This fixes empty maps — loadAssetPins() filters by
a.latitude != null.
- OCR: Extract only last 5 digits from Connect ID (XXXXX-XXXXXX → last 5)
- Barcode: Switch to decodeFromVideoDevice with 1D-only format hints
(Code 128/39/EAN/UPC/ITF/Codabar), TRY_HARDER, 50ms scan interval
- Auto check-in: New assets auto-checkin with GPS on creation via
all three entry modes (barcode, OCR, manual)
- Tests: Updated e2e tests for login wait and API flow
shawn