導入 browse.sh 比價診斷計畫

tests/test_competitor_match_attempts_persistence.py

@@ -1,4 +1,5 @@
 from pathlib import Path
+import json
 import logging
 from datetime import datetime
 from types import SimpleNamespace
@@ -42,6 +43,7 @@ def test_competitor_feeder_persists_all_match_attempt_outcomes():
     source = (ROOT / "services/competitor_price_feeder.py").read_text(encoding="utf-8")
     migration = (ROOT / "migrations/023_competitor_match_attempts.sql").read_text(encoding="utf-8")
     diagnostics_migration = (ROOT / "migrations/041_competitor_match_diagnostics.sql").read_text(encoding="utf-8")
+    browse_migration = (ROOT / "migrations/042_add_browse_diagnostics_to_match_attempts.sql").read_text(encoding="utf-8")
 
     assert "attempts_written" in source
     assert "_ensure_competitor_match_attempts_table" in source
@@ -56,6 +58,9 @@ def test_competitor_feeder_persists_all_match_attempt_outcomes():
     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 "_prepare_browse_diagnostic" in source
+    assert "browse_diagnostic_json" in source
+    assert "PCHOME_FEEDER_BROWSE_SH_EXECUTE_ENABLED" in source
     assert 'attempt_status="protected_existing_match"' in source
     assert "_should_upsert_competitor_price" in source
     assert "_classify_low_score_attempt" in source
@@ -88,6 +93,8 @@ def test_competitor_feeder_persists_all_match_attempt_outcomes():
     assert "match_diagnostic_json" in diagnostics_migration
     assert "comparison_mode" in diagnostics_migration
     assert "diagnostic_codes" in diagnostics_migration
+    assert "browse_diagnostic_json" in browse_migration
+    assert "idx_comp_match_attempts_browse_diag_time" in browse_migration
     assert "competitor_product_url" in source
     assert "competitor_image_url" in source
     assert "competitor_stock" in source
@@ -95,6 +102,74 @@ def test_competitor_feeder_persists_all_match_attempt_outcomes():
     assert "idx_comp_match_attempts_sku_source_time" in migration
 
 
+def test_competitor_feeder_records_browse_sh_plan_for_no_result(monkeypatch):
+    from services.competitor_price_feeder import CompetitorPriceFeeder
+
+    class FakeCrawler:
+        def __init__(self, *_args, **_kwargs):
+            pass
+
+        def search_products(self, *_args, **_kwargs):
+            return True, "ok", []
+
+    monkeypatch.setattr("services.pchome_crawler.PChomeCrawler", FakeCrawler)
+    feeder = CompetitorPriceFeeder(engine=object())
+    attempts = []
+    monkeypatch.setattr(
+        feeder,
+        "_record_match_attempt",
+        lambda *args, **kwargs: attempts.append(kwargs),
+    )
+
+    result = feeder._run_sku_items([{
+        "sku": "BROWSE001",
+        "name": "MOMO 稀有專櫃組合 50ml",
+        "product_id": 901,
+        "momo_price": 1280,
+    }])
+
+    assert result.matched == 0
+    assert result.skipped_no_result == 1
+    browse_plan = attempts[0]["browse_diagnostic"]
+    assert browse_plan["tool"] == "browse.sh"
+    assert browse_plan["mode"] == "plan_only"
+    assert browse_plan["execute_enabled"] is False
+    assert browse_plan["reason"] == "no_result"
+    assert browse_plan["execution"]["status"] == "disabled"
+    assert browse_plan["suggested_commands"][0]["args"][0] == "get"
+    assert "ecshweb.pchome.com.tw/search" in browse_plan["urls"][0]
+
+
+def test_competitor_match_attempt_persists_browse_diagnostic_json():
+    from sqlalchemy import create_engine, text
+    from services.competitor_price_feeder import CompetitorPriceFeeder
+
+    engine = create_engine("sqlite:///:memory:")
+    feeder = CompetitorPriceFeeder(engine=engine)
+    feeder._record_match_attempt(
+        sku="BROWSE002",
+        momo_name="MOMO 取證測試商品",
+        search_terms=["取證 測試"],
+        attempt_status="no_result",
+        browse_diagnostic={
+            "tool": "browse.sh",
+            "mode": "plan_only",
+            "urls": ["https://ecshweb.pchome.com.tw/search/v3.3/?q=test"],
+        },
+    )
+
+    with engine.connect() as conn:
+        row = conn.execute(text("""
+            SELECT browse_diagnostic_json
+            FROM competitor_match_attempts
+            WHERE sku = 'BROWSE002'
+        """)).scalar_one()
+
+    payload = json.loads(row)
+    assert payload["tool"] == "browse.sh"
+    assert payload["mode"] == "plan_only"
+
+
 def test_match_diagnostics_payload_carries_professional_match_lanes():
     from services.competitor_price_feeder import _match_diagnostics_payload, _extend_match_tags
     from services.marketplace_product_matcher import score_marketplace_match
@@ -591,6 +666,166 @@ def test_competitor_feeder_marks_weak_identity_as_true_low_confidence(monkeypatc
     assert attempts[0]["attempt_status"] == "true_low_confidence"
 
 
+def test_competitor_feeder_downgrades_variant_selection_gap_from_recoverable(monkeypatch):
+    from services.competitor_price_feeder import CompetitorPriceFeeder
+    from services.pchome_crawler import PChomeProduct
+
+    products = [
+        PChomeProduct(
+            product_id="DDAB01-08",
+            name="PERIPERA 雙頭旋轉極細眉筆 08深杏色 0.05g",
+            price=180,
+            original_price=220,
+            discount=18,
+            image_url="",
+            product_url="https://24h.pchome.com.tw/prod/DDAB01-08",
+            stock=20,
+            store="24h",
+            rating=4.7,
+            review_count=8,
+            is_on_sale=True,
+            crawled_at=datetime.now(),
+        ),
+        PChomeProduct(
+            product_id="DDAB01-09",
+            name="PERIPERA 雙頭旋轉極細眉筆 09灰褐棕 0.05g",
+            price=180,
+            original_price=220,
+            discount=18,
+            image_url="",
+            product_url="https://24h.pchome.com.tw/prod/DDAB01-09",
+            stock=20,
+            store="24h",
+            rating=4.7,
+            review_count=8,
+            is_on_sale=True,
+            crawled_at=datetime.now(),
+        ),
+        PChomeProduct(
+            product_id="DDAB01-11",
+            name="PERIPERA 雙頭旋轉極細眉筆 11摩卡灰褐 0.05g",
+            price=180,
+            original_price=220,
+            discount=18,
+            image_url="",
+            product_url="https://24h.pchome.com.tw/prod/DDAB01-11",
+            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", products
+
+    def fake_score(_momo_name, competitor_name, **_kwargs):
+        return SimpleNamespace(
+            score=0.734 if "09灰褐棕" in competitor_name else 0.733,
+            brand_score=1.0,
+            token_score=0.74,
+            spec_score=0.55,
+            sequence_score=0.66,
+            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 = []
+    monkeypatch.setattr(
+        feeder,
+        "_record_match_attempt",
+        lambda *args, **kwargs: attempts.append(kwargs),
+    )
+
+    result = feeder._run_sku_items([{
+        "sku": "P001",
+        "name": "【peripera官方直營】雙頭旋轉極細眉筆_多色任選(1.5mm極細筆頭)",
+        "product_id": 11,
+        "momo_price": 180,
+    }])
+
+    assert result.matched == 0
+    assert result.skipped_low_score == 1
+    assert attempts[0]["attempt_status"] == "true_low_confidence"
+
+
+def test_competitor_feeder_treats_choose_one_offer_as_missing_variant_signal(monkeypatch):
+    from services.competitor_price_feeder import CompetitorPriceFeeder
+    from services.pchome_crawler import PChomeProduct
+
+    product = PChomeProduct(
+        product_id="DDAB01-YSL",
+        name="【YSL聖羅蘭】恆久完美透膚煙染腮紅 6g ( #12/ #57/ #93)",
+        price=1650,
+        original_price=1780,
+        discount=7,
+        image_url="",
+        product_url="https://24h.pchome.com.tw/prod/DDAB01-YSL",
+        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=0.735,
+            brand_score=1.0,
+            token_score=0.74,
+            spec_score=0.55,
+            sequence_score=0.66,
+            type_score=1.0,
+            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 = []
+    monkeypatch.setattr(
+        feeder,
+        "_record_match_attempt",
+        lambda *args, **kwargs: attempts.append(kwargs),
+    )
+
+    result = feeder._run_sku_items([{
+        "sku": "Y001",
+        "name": "【YSL】官方直營 恆久完美透膚煙染腮紅(腮紅/任選1款/新品上市)",
+        "product_id": 12,
+        "momo_price": 1650,
+    }])
+
+    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
 

tests/test_marketplace_product_matcher.py

@@ -503,6 +503,49 @@ def test_marketplace_matcher_promotes_recipe_box_marketing_line_drift():
     assert "shared_identity_anchor_recipe_box_line" in diagnostics.reasons
 
 
+def test_marketplace_matcher_promotes_st_deodorizer_with_brand_alias_and_line_anchor():
+    from services.marketplace_product_matcher import score_marketplace_match
+
+    diagnostics = score_marketplace_match(
+        "【日本雞仔牌ST】室內消臭力智能光感應3段定時無線自動除臭芳香噴霧機(內贈芳香劑39ml 衛浴精油擴香瓶棒組)",
+        "日本ST雞仔牌-室內消臭力智能光感應3段定時無線自動除臭芳香噴霧機1入(含芳香劑39ml)",
+        momo_price=699,
+        competitor_price=699,
+    )
+
+    assert diagnostics.score >= 0.76
+    assert "shared_identity_anchor_exact_line" in diagnostics.reasons or "shared_identity_anchor_packaging_variant" in diagnostics.reasons
+
+
+def test_marketplace_matcher_promotes_nivea_dry_lotion_with_long_shared_anchor():
+    from services.marketplace_product_matcher import score_marketplace_match
+
+    diagnostics = score_marketplace_match(
+        "【NIVEA 妮維雅】男士無印乾爽止汗爽身乳液(無印止汗滾珠/德國妮維雅)",
+        "【NIVEA 妮維雅】止汗爽身乳液 無印乾爽50ml",
+        momo_price=129,
+        competitor_price=129,
+    )
+
+    assert diagnostics.score >= 0.76
+    assert "shared_identity_anchor_nivea_dry_lotion" in diagnostics.reasons
+
+
+def test_marketplace_matcher_rejects_refill_core_vs_case_only_pack():
+    from services.marketplace_product_matcher import score_marketplace_match
+
+    diagnostics = score_marketplace_match(
+        "【KATE 凱婷】3D造型眉彩餅補充芯(眉彩刷、眉餅盒分開販售)",
+        "【KATE 凱婷】眉彩餅盒一入款(搭配3D造型眉彩餅補充芯)",
+        momo_price=280,
+        competitor_price=280,
+    )
+
+    assert diagnostics.score < 0.76
+    assert diagnostics.hard_veto is True
+    assert "accessory_case_conflict" in diagnostics.reasons or "refill_pack_conflict" in diagnostics.reasons
+
+
 def test_marketplace_matcher_suppresses_wide_price_penalty_for_exact_lip_product():
     from services.marketplace_product_matcher import score_marketplace_match