fix(growth): normalize nested marketplace packs

services/marketplace_product_matcher.py

@@ -858,6 +858,8 @@ PRODUCT_TYPES = {
 COUNT_UNITS = {"入", "組", "瓶", "支", "條", "盒", "包", "袋", "片", "顆", "粒", "錠", "枚", "件", "罐", "杯", "本", "刀把", "刀片", "刀頭", "蕊"}
 COUNT_UNIT_PATTERN = r"(?:刀把|刀片|刀頭|入|組|瓶|支|條|盒|包|袋|片|張|顆|粒|錠|枚|件|罐|杯|本|蕊)"
 PIECE_UNITS = {"包", "袋", "片", "張", "顆", "粒", "錠", "枚"}
+ATOMIC_PIECE_UNITS = {"片", "張", "顆", "粒", "錠", "枚"}
+PACK_MULTIPLIER_UNITS = {"包", "袋"}
 CONTAINER_UNITS = {"入", "組", "盒", "罐", "杯", "本", "瓶", "支", "條", "件"}
 COUNT_UNIT_FAMILIES = {
     "刀片": "blade",
@@ -1210,14 +1212,32 @@ def _extract_specs(
         counts.append((2, "入"))
 
     total_piece_count = None
-    explicit_total = re.search(r"共\s*(\d+)\s*([包袋片顆粒錠枚])", text)
+    explicit_total = re.search(r"共\s*(\d+)\s*([包袋片張顆粒錠枚])", text)
     if explicit_total:
         total_piece_count = int(explicit_total.group(1))
     else:
+        atomic_piece_counts = [
+            count for count, unit in counts if unit in ATOMIC_PIECE_UNITS
+        ]
         piece_counts = [count for count, unit in counts if unit in PIECE_UNITS]
-        container_counts = [count for count, unit in counts if unit in CONTAINER_UNITS]
-        if piece_counts and container_counts:
-            total_piece_count = max(piece_counts) * max(container_counts)
+        container_multiplier_by_unit = {
+            unit: max(count for count, candidate_unit in counts if candidate_unit == unit)
+            for unit in CONTAINER_UNITS
+            if any(candidate_unit == unit for _, candidate_unit in counts)
+        }
+        multiplier_by_unit = {
+            unit: max(count for count, candidate_unit in counts if candidate_unit == unit)
+            for unit in PACK_MULTIPLIER_UNITS | CONTAINER_UNITS
+            if any(candidate_unit == unit for _, candidate_unit in counts)
+        }
+        if atomic_piece_counts and multiplier_by_unit:
+            total_piece_count = max(atomic_piece_counts)
+            for multiplier in multiplier_by_unit.values():
+                total_piece_count *= multiplier
+        elif piece_counts and container_multiplier_by_unit:
+            total_piece_count = max(piece_counts)
+            for multiplier in container_multiplier_by_unit.values():
+                total_piece_count *= multiplier
         elif piece_counts:
             total_piece_count = max(piece_counts)
 
@@ -1782,6 +1802,17 @@ def _has_overlapping_base_spec(left: ProductIdentity, right: ProductIdentity) ->
             return False
         return _close_number(left_weights[0], right_weights[0])
 
+    if left.total_piece_count or right.total_piece_count:
+        if not left.total_piece_count or not right.total_piece_count:
+            return False
+        left_units = {
+            unit for _, unit in left.counts if unit in ATOMIC_PIECE_UNITS
+        }
+        right_units = {
+            unit for _, unit in right.counts if unit in ATOMIC_PIECE_UNITS
+        }
+        return bool(left_units & right_units)
+
     return False
 
 
@@ -1798,7 +1829,11 @@ def _single_unit_total(identity: ProductIdentity) -> tuple[Optional[str], Option
         multiplier = identity.total_piece_count or _pack_multiplier(identity)
         return "g", weights[0] * multiplier, "ok"
     if identity.total_piece_count:
-        return "入", float(identity.total_piece_count), "ok"
+        atomic_units = {
+            unit for _, unit in identity.counts if unit in ATOMIC_PIECE_UNITS
+        }
+        unit = next(iter(atomic_units)) if len(atomic_units) == 1 else "入"
+        return unit, float(identity.total_piece_count), "ok"
     return None, None, "missing_single_unit"