cantaloupe_sync: fix column mapping heuristic, add comprehensive tests

- Fix greedy matching where 'name' keyword matched 'Location Name' header
- Use word-subset matching instead of substring matching
- Normalize keywords (replace _ and # with spaces) to match header normalization
- Reorder heuristics: multi-word specific patterns before single-word generics
- Add 10 unit tests for mapping, diff, and Excel parsing
- Add 23 integration tests for all sync API endpoints
- All 33 tests pass

cantaloupe_sync.py

@@ -98,37 +98,59 @@ ASSET_COLUMNS = [
 COLUMN_HEURISTICS = [
     # machine_id is the primary key for upserts
     (["asset id", "assetid", "machine id", "machineid", "machine_id", "equipment id", "unit id", "asset number"], "machine_id"),
-    (["serial", "serial number", "serialnumber", "serial_no", "s/n"], "serial_number"),
-    (["name", "asset name", "equipment name", "description"], "name"),
-    (["description", "notes", "detail", "comments"], "description"),
+    # Multi-word / specific heuristics first to avoid greedy single-word matches
+    (["location name", "location_name", "site name"], "_location_name"),
+    (["building name", "building_name", "bldg name"], "building_name"),
+    (["building number", "building_number", "building_no", "bldg number", "bldg #", "building #"], "building_number"),
+    (["trailer number", "trailer_number", "trailer_no", "trailer #", "mobile unit"], "trailer_number"),
+    (["walking directions", "walking_directions", "how to get there"], "walking_directions"),
+    (["map link", "map_link", "google maps", "map url", "location url"], "map_link"),
+    (["parking location", "parking_location", "parking spot"], "parking_location"),
+    (["street address", "location address", "site address"], "address"),
+    (["asset name", "equipment name"], "name"),
+    (["model number", "model_no", "model #"], "model"),
+    (["serial number", "serialnumber", "serial_no", "s/n"], "serial_number"),
+    (["room number", "room #", "suite"], "room"),
+    (["customer name", "client name", "account name", "company"], "_customer_name"),
+    (["photo path", "photo_path", "image path"], "photo_path"),
+    (["route number", "route #", "delivery route"], "_route"),
+    (["sales rep", "salesperson", "account manager"], "_sales_rep"),
+    (["qr code", "tag number", "tag #"], "_barcode"),
+    (["geofence radius", "geofence_radius", "geo radius"], "geofence_radius_meters"),
+    (["gps lat", "lat"], "latitude"),
+    (["gps lng", "gps long", "long", "lng", "lon"], "longitude"),
+    # Single-word/generic heuristics (last, after specific multi-word ones)
+    (["serial", "s/n"], "serial_number"),
     (["category", "type", "asset type", "equipment type", "class"], "category"),
     (["status", "state", "condition"], "status"),
     (["make", "manufacturer", "brand", "vendor"], "make"),
-    (["model", "model number", "model_no", "model #"], "model"),
-    (["address", "street", "street address", "location address", "site address"], "address"),
-    (["building name", "building_name", "bldg name", "site name"], "building_name"),
-    (["building number", "building_number", "building_no", "bldg number", "bldg #", "building #"], "building_number"),
+    (["address", "street"], "address"),
     (["floor", "level", "story"], "floor"),
-    (["room", "room number", "room #", "suite"], "room"),
-    (["trailer", "trailer number", "trailer_no", "trailer #", "mobile unit"], "trailer_number"),
-    (["walking directions", "walking_directions", "directions", "how to get there"], "walking_directions"),
-    (["map link", "map_link", "google maps", "map url", "location url"], "map_link"),
-    (["parking", "parking location", "parking_location", "parking spot"], "parking_location"),
-    (["latitude", "lat", "gps lat"], "latitude"),
-    (["longitude", "long", "lng", "lon", "gps lng", "gps long"], "longitude"),
-    (["geofence", "geo radius", "radius", "geofence radius", "geofence_radius"], "geofence_radius_meters"),
-    (["customer", "client", "account", "company", "customer name"], "_customer_name"),
-    (["location name", "location_name", "site", "site name", "location"], "_location_name"),
-    (["photo", "photo path", "photo_path", "image", "picture"], "photo_path"),
-    (["route", "route number", "route #", "delivery route"], "_route"),
-    (["sales", "sales rep", "salesperson", "account manager"], "_sales_rep"),
-    (["barcode", "qr code", "tag", "tag number", "tag #"], "_barcode"),
+    (["latitude", "gps"], "latitude"),
+    (["longitude", "geo"], "longitude"),
+    (["geofence", "radius"], "geofence_radius_meters"),
+    (["customer", "client", "account"], "_customer_name"),
+    (["location", "site"], "_location_name"),
+    (["name"], "name"),
+    (["description", "notes", "detail", "comments"], "description"),
+    (["model"], "model"),
+    (["building"], "building_name"),
+    (["room"], "room"),
+    (["trailer"], "trailer_number"),
+    (["walking", "directions"], "walking_directions"),
+    (["map"], "map_link"),
+    (["parking"], "parking_location"),
+    (["photo", "image", "picture"], "photo_path"),
+    (["route"], "_route"),
+    (["sales", "rep"], "_sales_rep"),
+    (["barcode", "tag", "qr"], "_barcode"),
 ]
 
 
 def _normalize(s: str) -> str:
-    """Lowercase, strip, collapse whitespace."""
-    return " ".join(str(s).lower().strip().split())
+    """Lowercase, strip, collapse whitespace. Replace underscores with spaces
+    so 'machine_id' matches 'machine id'."""
+    return " ".join(str(s).lower().strip().replace("_", " ").replace("#", " ").split())
 
 
 def map_columns(excel_headers: list[str]) -> dict[str, str]:
@@ -146,10 +168,17 @@ def map_columns(excel_headers: list[str]) -> dict[str, str]:
         if not norm:
             continue
         matched = False
+        norm_words = set(norm.split())
         for keywords, canonical in COLUMN_HEURISTICS:
             for kw in keywords:
-                # Match if the header contains the keyword or vice versa
-                if kw in norm or norm in kw:
+                kw_norm = _normalize(kw)
+                if not kw_norm:
+                    continue
+                kw_words = set(kw_norm.split())
+                # Match: all keyword words must be present in the header.
+                # This prevents short headers like "name" from matching
+                # multi-word keywords like "location name".
+                if kw_words.issubset(norm_words):
                     mapping[header] = canonical
                     discovered.append(f"{header} → {canonical}")
                     matched = True