[V10.355] rescue near-threshold identity cohorts

2026-05-21 09:21:46 +08:00
parent b272abccca
commit 88101cf5d0
7 changed files with 433 additions and 29 deletions
--- a/tests/test_competitor_match_attempts_persistence.py
+++ b/tests/test_competitor_match_attempts_persistence.py
@@ -18,14 +18,16 @@ def test_competitor_feeder_persists_all_match_attempt_outcomes():
    assert "INSERT INTO competitor_match_attempts" in source
    assert "CAST(:search_terms AS jsonb)" in source
    assert 'attempt_status="matched"' in source
-    assert '"low_score"' in source
+    assert '"recoverable_low_score"' in source
+    assert '"true_low_confidence"' in source
    assert '"identity_veto"' in source
    assert 'attempt_status="no_result"' in source
    assert 'attempt_status="no_match"' in source
    assert 'attempt_status="error"' in source
    assert "_search_pchome_candidates(crawler, momo_name, search_terms, momo_price=momo_price)" in source
-    assert 'attempt_status="needs_review"' in source
+    assert 'attempt_status="protected_existing_match"' in source
    assert "_should_upsert_competitor_price" in source
+    assert "_classify_low_score_attempt" in source
    assert "replace_legacy_unverified" in source
    assert "identity_v2" in source
    assert "_fetch_expired_identity_skus" in source
@@ -35,8 +37,7 @@ def test_competitor_feeder_persists_all_match_attempt_outcomes():
    retryable_source = source.split("def _fetch_retryable_candidate_skus", 1)[1].split(
        "def _fetch_expired_identity_skus", 1
    )[0]
-    assert "la.attempt_status = 'low_score'" in retryable_source
-    assert "refresh_low_score')" not in retryable_source
+    assert "la.attempt_status IN ('low_score', 'refresh_low_score', 'recoverable_low_score')" in retryable_source
    latest_attempt_source = retryable_source.split("latest_attempt AS", 1)[1].split(
        "SELECT\n                lm.product_id", 1
    )[0]
@@ -144,7 +145,7 @@ def test_reject_review_expires_current_formal_price():
                 best_competitor_price, best_match_score, error_message, attempted_at)
            VALUES
                ('A005', 'pchome', 1, '舒特膚 AD 乳液 200ml', 980,
-                 '[]', 1, 'needs_review',
+                 '[]', 1, 'protected_existing_match',
                 'DDAB01-REJECT', '舒特膚 AD 乳液 200ml', 899, 0.84,
                 'score=0.84', '2026-05-20 09:10:00')
        """))
@@ -408,6 +409,134 @@ def test_competitor_feeder_splits_hard_veto_from_low_score(monkeypatch):
    assert attempts[0]["diagnostics"].hard_veto is True


+def test_competitor_feeder_marks_near_threshold_same_line_as_recoverable(monkeypatch):
+    from services.competitor_price_feeder import CompetitorPriceFeeder
+    from services.pchome_crawler import PChomeProduct
+
+    product = PChomeProduct(
+        product_id="DDAB01-RECOVERABLE",
+        name="Recipe Box 韓兔 兒童防曬氣墊粉餅",
+        price=699,
+        original_price=799,
+        discount=12,
+        image_url="",
+        product_url="https://24h.pchome.com.tw/prod/DDAB01-RECOVERABLE",
+        stock=20,
+        store="24h",
+        rating=4.7,
+        review_count=8,
+        is_on_sale=True,
+        crawled_at=datetime.now(),
+    )
+
+    class FakeCrawler:
+        def __init__(self, *_args, **_kwargs):
+            pass
+
+        def search_products(self, *_args, **_kwargs):
+            return True, "ok", [product]
+
+    def fake_score(*_args, **_kwargs):
+        return SimpleNamespace(
+            score=0.754,
+            brand_score=1.0,
+            token_score=0.59,
+            spec_score=0.55,
+            sequence_score=0.53,
+            type_score=1.0,
+            price_penalty=0.0,
+            hard_veto=False,
+            reasons=("strong_product_line_match",),
+            comparison_mode="exact_identity",
+            tags=["identity_v2", "comparison_exact_identity", "brand_match"],
+        )
+
+    monkeypatch.setattr("services.pchome_crawler.PChomeCrawler", FakeCrawler)
+    monkeypatch.setattr("services.marketplace_product_matcher.score_marketplace_match", fake_score)
+    feeder = CompetitorPriceFeeder(engine=object())
+    attempts = []
+    monkeypatch.setattr(
+        feeder,
+        "_record_match_attempt",
+        lambda *args, **kwargs: attempts.append(kwargs),
+    )
+
+    result = feeder._run_sku_items([{
+        "sku": "RB001",
+        "name": "【Recipebox】Recipe Box兒童防曬氣墊粉餅(兒童化妝品/無毒防曬粉餅/天然彩妝)",
+        "product_id": 8,
+        "momo_price": 699,
+    }])
+
+    assert result.matched == 0
+    assert result.skipped_low_score == 1
+    assert attempts[0]["attempt_status"] == "recoverable_low_score"
+
+
+def test_competitor_feeder_marks_weak_identity_as_true_low_confidence(monkeypatch):
+    from services.competitor_price_feeder import CompetitorPriceFeeder
+    from services.pchome_crawler import PChomeProduct
+
+    product = PChomeProduct(
+        product_id="DDAB01-WEAK",
+        name="韓系彩妝 十色眼影盤",
+        price=499,
+        original_price=699,
+        discount=28,
+        image_url="",
+        product_url="https://24h.pchome.com.tw/prod/DDAB01-WEAK",
+        stock=20,
+        store="24h",
+        rating=4.2,
+        review_count=8,
+        is_on_sale=True,
+        crawled_at=datetime.now(),
+    )
+
+    class FakeCrawler:
+        def __init__(self, *_args, **_kwargs):
+            pass
+
+        def search_products(self, *_args, **_kwargs):
+            return True, "ok", [product]
+
+    def fake_score(*_args, **_kwargs):
+        return SimpleNamespace(
+            score=0.733,
+            brand_score=0.95,
+            token_score=0.51,
+            spec_score=0.45,
+            sequence_score=0.44,
+            type_score=0.55,
+            price_penalty=0.0,
+            hard_veto=False,
+            reasons=(),
+            comparison_mode="exact_identity",
+            tags=["identity_v2", "comparison_exact_identity"],
+        )
+
+    monkeypatch.setattr("services.pchome_crawler.PChomeCrawler", FakeCrawler)
+    monkeypatch.setattr("services.marketplace_product_matcher.score_marketplace_match", fake_score)
+    feeder = CompetitorPriceFeeder(engine=object())
+    attempts = []
+    monkeypatch.setattr(
+        feeder,
+        "_record_match_attempt",
+        lambda *args, **kwargs: attempts.append(kwargs),
+    )
+
+    result = feeder._run_sku_items([{
+        "sku": "RM001",
+        "name": "【rom&nd】勝過眼皮十色眼影盤",
+        "product_id": 9,
+        "momo_price": 499,
+    }])
+
+    assert result.matched == 0
+    assert result.skipped_low_score == 1
+    assert attempts[0]["attempt_status"] == "true_low_confidence"
+
+
 def test_should_upsert_allows_same_identity_candidate_to_replace_lower_score():
    from sqlalchemy import create_engine, text

@@ -455,6 +584,83 @@ def test_should_upsert_allows_same_identity_candidate_to_replace_lower_score():
    assert reason.startswith("replace_same_identity_better_score=0.788->0.811")


+def test_competitor_feeder_marks_existing_stronger_match_as_protected(monkeypatch):
+    from services.competitor_price_feeder import CompetitorPriceFeeder
+    from services.pchome_crawler import PChomeProduct
+
+    product = PChomeProduct(
+        product_id="DDAB01-NEW",
+        name="PONY EFFECT 絕對持久定妝噴霧",
+        price=599,
+        original_price=699,
+        discount=14,
+        image_url="",
+        product_url="https://24h.pchome.com.tw/prod/DDAB01-NEW",
+        stock=20,
+        store="24h",
+        rating=4.7,
+        review_count=8,
+        is_on_sale=True,
+        crawled_at=datetime.now(),
+    )
+
+    class FakeCrawler:
+        def __init__(self, *_args, **_kwargs):
+            pass
+
+        def search_products(self, *_args, **_kwargs):
+            return True, "ok", [product]
+
+    def fake_score(*_args, **_kwargs):
+        return SimpleNamespace(
+            score=0.781,
+            brand_score=1.0,
+            token_score=0.79,
+            spec_score=0.55,
+            sequence_score=0.68,
+            type_score=0.55,
+            price_penalty=0.0,
+            hard_veto=False,
+            reasons=("shared_identity_anchor_packaging_variant",),
+            comparison_mode="exact_identity",
+            tags=["identity_v2", "comparison_exact_identity", "brand_match"],
+        )
+
+    monkeypatch.setattr("services.pchome_crawler.PChomeCrawler", FakeCrawler)
+    monkeypatch.setattr("services.marketplace_product_matcher.score_marketplace_match", fake_score)
+    feeder = CompetitorPriceFeeder(engine=object())
+    attempts = []
+    writes = []
+    monkeypatch.setattr(
+        feeder,
+        "_record_match_attempt",
+        lambda *args, **kwargs: attempts.append(kwargs),
+    )
+    monkeypatch.setattr(
+        feeder,
+        "_should_upsert_competitor_price",
+        lambda *_args, **_kwargs: (False, "existing_match_conflict;existing_score=0.948;incoming_score=0.781"),
+    )
+    monkeypatch.setattr(
+        feeder,
+        "_upsert_competitor_price",
+        lambda *args, **kwargs: writes.append((args, kwargs)),
+    )
+
+    result = feeder._run_sku_items([{
+        "sku": "14133077",
+        "name": "【PONY EFFECT】絕對持久定妝噴霧",
+        "product_id": 10,
+        "momo_price": 599,
+    }])
+
+    assert result.matched == 0
+    assert result.skipped_low_score == 1
+    assert writes == []
+    assert attempts[0]["attempt_status"] == "protected_existing_match"
+    assert "existing_match_conflict" in attempts[0]["error_message"]
+
+
 def test_search_candidates_does_not_stop_on_merely_acceptable_match(monkeypatch):
    from services.competitor_price_feeder import _search_pchome_candidates
    from services.pchome_crawler import PChomeProduct
--- a/tests/test_marketplace_product_matcher.py
+++ b/tests/test_marketplace_product_matcher.py
@@ -573,6 +573,22 @@ def test_marketplace_matcher_does_not_promote_different_option_without_spec():

    assert diagnostics.score < 0.76
    assert "strong_exact_spec_match" not in diagnostics.reasons
+    assert "variant_descriptor_conflict" in diagnostics.reasons
+
+
+def test_marketplace_matcher_promotes_variant_safe_exact_option():
+    from services.marketplace_product_matcher import score_marketplace_match
+
+    diagnostics = score_marketplace_match(
+        "【DASHING DIVA】MAGICPRESS時尚潮流美甲片_極光之藍",
+        "Dashing Diva/F 時尚潮流美甲片-極光之藍 MDF5F001AG",
+        momo_price=331,
+        competitor_price=420,
+    )
+
+    assert diagnostics.score >= 0.76
+    assert diagnostics.hard_veto is False
+    assert "shared_identity_anchor_variant_safe" in diagnostics.reasons


 def test_marketplace_matcher_promotes_shared_identity_anchor_near_threshold():
@@ -649,6 +665,36 @@ def test_marketplace_matcher_promotes_shared_anchor_without_spec_conflict():
    assert "shared_identity_anchor_no_spec" in diagnostics.reasons


+def test_marketplace_matcher_promotes_recipe_box_near_threshold_with_variant_safe_anchor():
+    from services.marketplace_product_matcher import score_marketplace_match
+
+    diagnostics = score_marketplace_match(
+        "【Recipebox】Recipe Box兒童防曬氣墊粉餅(兒童化妝品/無毒防曬粉餅/天然彩妝)",
+        "Recipe Box 韓兔 兒童防曬氣墊粉餅",
+        momo_price=699,
+        competitor_price=699,
+    )
+
+    assert diagnostics.score >= 0.76
+    assert diagnostics.hard_veto is False
+    assert "shared_identity_anchor_variant_safe" in diagnostics.reasons
+
+
+def test_marketplace_matcher_promotes_romand_palette_exact_line():
+    from services.marketplace_product_matcher import score_marketplace_match
+
+    diagnostics = score_marketplace_match(
+        "【rom&nd】勝過眼皮十色眼影盤",
+        "rom&nd X ZO&FRIENDS 勝過眼皮十色眼影盤 8g/7g",
+        momo_price=499,
+        competitor_price=499,
+    )
+
+    assert diagnostics.score >= 0.76
+    assert diagnostics.hard_veto is False
+    assert "shared_identity_anchor_variant_safe" in diagnostics.reasons
+
+
 def test_marketplace_matcher_promotes_shared_model_token_for_exact_model():
    from services.marketplace_product_matcher import score_marketplace_match

@@ -735,6 +781,16 @@ def test_marketplace_search_terms_keep_professional_product_phrase():
    assert not any("卸除防曬" in term or "外出清潔" in term for term in mustela_terms)


+def test_marketplace_search_terms_keep_variant_descriptor_for_sensitive_lines():
+    from services.marketplace_product_matcher import build_search_terms
+
+    dashing_terms = build_search_terms("【DASHING DIVA】MAGICPRESS時尚潮流美甲片_極光之藍", max_terms=5)
+    romand_terms = build_search_terms("【rom&nd】勝過眼皮十色眼影盤", max_terms=5)
+
+    assert dashing_terms[0] == "dashing diva 時尚潮流美甲片 極光之藍"
+    assert romand_terms[0] == "romand 勝過眼皮十色眼影盤"
+
+
 def test_marketplace_search_terms_prefer_specific_line_over_generic_usage_words():
    from services.marketplace_product_matcher import build_search_terms