補 Gemini 出站守門測試與同款價差放行

services/marketplace_product_matcher.py

@@ -447,6 +447,16 @@ VARIANT_OPTION_COLOR_WORDS = {
     "月光銀影",
 }
 
+VARIANT_DESCRIPTOR_NOISE_KEYWORDS = {
+    "平輸航空版",
+    "多色任選",
+    "色號任選",
+    "任選色號",
+    "極細筆頭",
+    "筆頭",
+    "官方直營",
+}
+
 SEARCH_AMBIGUOUS_PRODUCT_TERMS = {
     "保護膜",
     "保護貼",
@@ -478,6 +488,8 @@ BRAND_ALIAS_OVERRIDES = {
     "febreze": ("febreze", "風倍清"),
     "jo malone": ("jo malone",),
     "prada": ("prada", "普拉達"),
+    "za": ("za",),
+    "xiaomi": ("小米有品", "小米", "xiaomi"),
 }
 
 PRODUCT_TYPES = {
@@ -1559,9 +1571,22 @@ def score_marketplace_match(
     try:
         if momo_price and competitor_price:
             ratio = float(competitor_price) / max(float(momo_price), 1.0)
+            allow_price_penalty_suppression = (
+                shared_anchor
+                and len(shared_anchor.replace(" ", "")) >= 7
+                and brand_score >= 0.95
+                and not hard_veto
+                and type_score >= 0.55
+                and spec_score >= 0.99
+                and token_score >= 0.68
+                and sequence_score >= 0.72
+            )
             if (ratio < 0.3 or ratio > 3.2) and token_score < 0.78:
-                price_penalty = 0.12
-                reasons.append("price_ratio_extreme")
+                if allow_price_penalty_suppression:
+                    reasons.append("price_penalty_suppressed_exact_identity")
+                else:
+                    price_penalty = 0.12
+                    reasons.append("price_ratio_extreme")
             elif (ratio < 0.48 or ratio > 2.2) and token_score < 0.68:
                 price_penalty = 0.06
                 reasons.append("price_ratio_wide")
@@ -1662,6 +1687,35 @@ def score_marketplace_match(
     ):
         score += 0.02
         reasons.append("shared_identity_anchor_core_line")
+    if (
+        shared_anchor
+        and len(shared_anchor.replace(" ", "")) >= 6
+        and brand_score >= 0.95
+        and not hard_veto
+        and price_penalty == 0
+        and type_score >= 0.55
+        and spec_score >= 0.45
+        and token_score >= 0.86
+        and sequence_score >= 0.75
+        and not variant_descriptor_conflict
+    ):
+        score += 0.07
+        reasons.append("shared_identity_anchor_exact_line")
+    if (
+        shared_anchor
+        and len(shared_anchor.replace(" ", "")) >= 5
+        and brand_score >= 0.95
+        and not hard_veto
+        and price_penalty == 0
+        and type_score >= 0.55
+        and spec_score >= 0.45
+        and token_score >= 0.74
+        and sequence_score >= 0.60
+        and _shared_variant_descriptors(left, right)
+        and not variant_descriptor_conflict
+    ):
+        score += 0.05
+        reasons.append("shared_variant_descriptor_alignment")
     if (
         shared_anchor
         and len(shared_anchor.replace(" ", "")) >= 6
@@ -1870,12 +1924,30 @@ def _variant_descriptors(identity: ProductIdentity) -> set[str]:
             continue
         if compact in SEARCH_NOISE_TOKENS or compact in SEARCH_BROAD_ANCHORS:
             continue
+        if any(keyword in compact for keyword in VARIANT_DESCRIPTOR_NOISE_KEYWORDS):
+            continue
         if re.fullmatch(r"[a-z0-9-]+", compact):
             continue
         descriptors.add(compact.removesuffix("款"))
     return {token for token in descriptors if token}
 
 
+def _shared_variant_descriptors(left: ProductIdentity, right: ProductIdentity) -> set[str]:
+    left_descriptors = _variant_descriptors(left)
+    right_descriptors = _variant_descriptors(right)
+    shared: set[str] = set()
+    for left_descriptor in left_descriptors:
+        for right_descriptor in right_descriptors:
+            if left_descriptor == right_descriptor:
+                shared.add(left_descriptor)
+                continue
+            if len(left_descriptor) >= 2 and left_descriptor in right_descriptor:
+                shared.add(left_descriptor)
+            elif len(right_descriptor) >= 2 and right_descriptor in left_descriptor:
+                shared.add(right_descriptor)
+    return shared
+
+
 def _is_variant_sensitive_identity(
     left: ProductIdentity,
     right: ProductIdentity,