feat(ai): 補抓 PChome 待比對商品

CONSTITUTION.md

@@ -2,7 +2,7 @@
 
 > 本文件定義專案開發的核心準則與不可違反的規範
 > **建立日期**: 2026-01-12
-> **當前版本**: V10.47 (Auto-detect sales columns for product picks)
+> **當前版本**: V10.48 (Priority backfill for unmatched PChome products)
 > **最後更新**: 2026-05-01
 
 ---

app.py

@@ -95,8 +95,8 @@ except Exception as e:
     sys_log.error(f"無法檢測磁碟空間: {e}")
 
 # 🚩 系統版本定義 (備份與顯示用)
-# 🚩 2026-05-01 V10.47: Auto-detect sales columns for product picks
-SYSTEM_VERSION = "V10.47"
+# 🚩 2026-05-01 V10.48: Priority backfill for unmatched PChome products
+SYSTEM_VERSION = "V10.48"
 
 # ==========================================
 # 🔒 SQL Injection 防護函數

docs/AI_INTELLIGENCE_MODULE_SOT.md

@@ -35,6 +35,7 @@ SQL漏斗(~300筆)
 - 寫入策略使用 `strategy='product_pick'`，保留在既有 AI 決策表，不新增假頁面或暫存 JSON。
 - 後台入口：`POST /api/ai/product-picks/generate`，`/ai_intelligence` 可手動產生清單。
 - 配對來源仍以 PChome crawler 真實搜尋結果為準；無競品資料時不生成挑品。
+- 比對覆蓋率補強入口：`POST /api/ai/pchome-match/backfill`，優先補抓仍無有效 PChome 配對的高價 ACTIVE 商品。
 
 | 角色 | 模型 | 主機 | 成本 | 每日限額 |
 |------|------|------|------|---------|

routes/ai_routes.py

@@ -1647,6 +1647,46 @@ def api_generate_product_picks():
         return jsonify({'success': False, 'error': str(e)}), 500
 
 
+@ai_bp.route('/api/ai/pchome-match/backfill', methods=['POST'])
+@login_required
+def api_pchome_match_backfill():
+    """背景補抓尚未有有效 PChome 配對的高價商品，提高比價覆蓋率。"""
+    import threading
+
+    payload = request.get_json(silent=True) or {}
+    limit = max(5, min(int(payload.get('limit', 60)), 160))
+
+    def _run_backfill():
+        try:
+            from config import DATABASE_PATH
+            from sqlalchemy import create_engine
+            from services.competitor_price_feeder import CompetitorPriceFeeder
+
+            engine = create_engine(DATABASE_PATH)
+            result = CompetitorPriceFeeder(engine=engine).run_unmatched_priority(limit=limit)
+            logger.info(
+                "[PChomeBackfill] done total=%s matched=%s no=%s low=%s errors=%s history=%s duration=%ss",
+                result.total_skus,
+                result.matched,
+                result.skipped_no_result,
+                result.skipped_low_score,
+                result.errors,
+                result.history_written,
+                result.duration_sec,
+            )
+        except Exception as exc:
+            logger.error(f"[PChomeBackfill] 背景補抓失敗: {exc}")
+
+    thread = threading.Thread(target=_run_backfill, daemon=True)
+    thread.start()
+
+    return jsonify({
+        'success': True,
+        'message': f'已啟動 PChome 待比對補抓，優先處理 {limit} 筆高價未配對商品',
+        'limit': limit,
+    }), 202
+
+
 @ai_bp.route('/api/ai/icaim/trigger', methods=['POST'])
 @login_required
 def api_icaim_trigger():

services/competitor_price_feeder.py

@@ -358,6 +358,47 @@ class CompetitorPriceFeeder:
             rows = conn.execute(sql).fetchall()
         return [dict(r._mapping) for r in rows]
 
+    def _fetch_unmatched_priority_skus(self, limit: int = 80) -> list:
+        """
+        取得目前沒有有效 PChome 配對的高價 ACTIVE 商品，供補強流程優先處理。
+        """
+        if self.engine is None:
+            raise RuntimeError("需要注入 SQLAlchemy engine")
+
+        from sqlalchemy import text
+        sql = text("""
+            WITH latest_momo AS (
+                SELECT
+                    p.id AS product_id,
+                    p.i_code AS sku,
+                    p.name,
+                    p.category,
+                    pr.price AS momo_price,
+                    ROW_NUMBER() OVER (PARTITION BY p.id ORDER BY pr.timestamp DESC) AS rn
+                FROM products p
+                JOIN price_records pr ON pr.product_id = p.id
+                WHERE p.status = 'ACTIVE'
+            )
+            SELECT
+                lm.product_id,
+                lm.sku,
+                lm.name,
+                lm.category,
+                lm.momo_price
+            FROM latest_momo lm
+            LEFT JOIN competitor_prices cp
+              ON cp.sku = lm.sku
+             AND cp.source = 'pchome'
+             AND (cp.expires_at IS NULL OR cp.expires_at > CURRENT_TIMESTAMP)
+            WHERE lm.rn = 1
+              AND cp.sku IS NULL
+            ORDER BY lm.momo_price DESC NULLS LAST, lm.sku
+            LIMIT :limit
+        """)
+        with self.engine.connect() as conn:
+            rows = conn.execute(sql, {"limit": max(1, min(int(limit), 300))}).fetchall()
+        return [dict(r._mapping) for r in rows]
+
     def _upsert_competitor_price(
         self,
         sku: str,
@@ -431,16 +472,7 @@ class CompetitorPriceFeeder:
                 "tags":            tags_json,
             })
 
-    def run(self, source: str = "pchome") -> FeederResult:
-        """
-        執行一輪競品價格抓取與寫入
-
-        Args:
-            source: 競品來源代碼（目前支援 'pchome'）
-
-        Returns:
-            FeederResult
-        """
+    def _run_sku_items(self, skus: list, source: str = "pchome", label: str = "PChome 競品價格") -> FeederResult:
         start = time.time()
 
         if source != "pchome":
@@ -450,14 +482,7 @@ class CompetitorPriceFeeder:
         from services.pchome_crawler import PChomeCrawler
         crawler = PChomeCrawler(timeout=30, delay=RATE_DELAY)
 
-        # Step 1: 取得監控清單
-        try:
-            skus = self._fetch_active_skus()
-        except Exception as e:
-            logger.error(f"[Feeder] 讀取商品清單失敗: {e}")
-            return FeederResult(0, 0, 0, 0, 1, time.time() - start)
-
-        logger.info(f"[Feeder] 開始抓取 {len(skus)} 支商品的 PChome 競品價格")
+        logger.info(f"[Feeder] 開始抓取 {len(skus)} 支商品的 {label}")
 
         matched = 0
         skipped_no = 0
@@ -529,6 +554,34 @@ class CompetitorPriceFeeder:
             history_written=history_written,
         )
 
+    def run(self, source: str = "pchome") -> FeederResult:
+        """
+        執行一輪競品價格抓取與寫入
+
+        Args:
+            source: 競品來源代碼（目前支援 'pchome'）
+
+        Returns:
+            FeederResult
+        """
+        try:
+            skus = self._fetch_active_skus()
+        except Exception as e:
+            logger.error(f"[Feeder] 讀取商品清單失敗: {e}")
+            return FeederResult(0, 0, 0, 0, 1, 0.0)
+
+        return self._run_sku_items(skus, source=source, label="PChome 競品價格")
+
+    def run_unmatched_priority(self, limit: int = 80, source: str = "pchome") -> FeederResult:
+        """優先補抓尚未有有效 PChome 配對的高價商品。"""
+        try:
+            skus = self._fetch_unmatched_priority_skus(limit=limit)
+        except Exception as e:
+            logger.error(f"[Feeder] 讀取待比對優先商品失敗: {e}")
+            return FeederResult(0, 0, 0, 0, 1, 0.0)
+
+        return self._run_sku_items(skus, source=source, label="待比對優先補抓")
+
 
 # ─────────────────────────────────────────────
 # CLI 測試（不依賴 DB，直接測試爬蟲 + 比對邏輯）

templates/ai_intelligence.html

@@ -23,6 +23,9 @@
             <button class="btn btn-outline-primary btn-sm" id="btnPickList" onclick="generatePickList()">
                 <i class="fas fa-wand-magic-sparkles me-1"></i>產生挑品清單
             </button>
+            <button class="btn btn-outline-warning btn-sm" id="btnBackfill" onclick="backfillPchomeMatches()">
+                <i class="fas fa-magnifying-glass-chart me-1"></i>補抓待比對
+            </button>
             <button class="btn btn-outline-secondary btn-sm" onclick="loadDashboard()">
                 <i class="fas fa-redo me-1"></i>重新整理
             </button>
@@ -390,6 +393,41 @@ async function generatePickList() {
     }
 }
 
+// ── 補抓 PChome 待比對商品 ─────────────────────────
+async function backfillPchomeMatches() {
+    const btn = document.getElementById('btnBackfill');
+    const toast = document.getElementById('triggerToast');
+    const msg = document.getElementById('triggerToastMsg');
+
+    btn.disabled = true;
+    btn.innerHTML = '<span class="spinner-border spinner-border-sm me-1"></span>補抓中...';
+
+    try {
+        const res = await fetch('/api/ai/pchome-match/backfill', {
+            method: 'POST',
+            headers: { 'Content-Type': 'application/json' },
+            body: JSON.stringify({ limit: 60 })
+        });
+        const data = await res.json();
+
+        msg.innerHTML = data.success
+            ? `<i class="fas fa-check-circle me-1"></i>${data.message}`
+            : `<i class="fas fa-times-circle me-1"></i>${data.error || '補抓啟動失敗'}`;
+        toast.className = 'toast align-items-center text-white border-0 ' +
+                          (data.success ? 'bg-success' : 'bg-danger');
+        new bootstrap.Toast(toast, { delay: 6000 }).show();
+
+        if (data.success) setTimeout(loadDashboard, 90000);
+    } catch (e) {
+        msg.innerHTML = '<i class="fas fa-times-circle me-1"></i>補抓失敗：' + e.message;
+        toast.className = 'toast align-items-center text-white border-0 bg-danger';
+        new bootstrap.Toast(toast, { delay: 4000 }).show();
+    } finally {
+        btn.disabled = false;
+        btn.innerHTML = '<i class="fas fa-magnifying-glass-chart me-1"></i>補抓待比對';
+    }
+}
+
 // ── 手動觸發分析 ────────────────────────────────────
 async function triggerAnalysis() {
     const btn   = document.getElementById('btnTrigger');

tests/test_frontend_v2_assets.py

@@ -139,6 +139,8 @@ def test_ai_product_pick_agent_uses_real_competitor_data_and_dashboard_action():
     assert "_build_search_keywords" in feeder_source
     assert "_search_pchome_candidates" in feeder_source
     assert "crawler.search_products(keyword, limit=SEARCH_LIMIT)" in feeder_source
+    assert "_fetch_unmatched_priority_skus" in feeder_source
+    assert "run_unmatched_priority" in feeder_source
 
     assert "generate_product_pick_list" in agent_source
     assert "competitor_prices" in agent_source
@@ -155,12 +157,17 @@ def test_ai_product_pick_agent_uses_real_competitor_data_and_dashboard_action():
 
     assert "@ai_bp.route('/api/ai/product-picks/generate', methods=['POST'])" in route_source
     assert "generate_product_pick_list(engine" in route_source
+    assert "@ai_bp.route('/api/ai/pchome-match/backfill', methods=['POST'])" in route_source
+    assert "run_unmatched_priority(limit=limit)" in route_source
     assert "match_rate" in route_source
     assert "product_pick_count" in route_source
 
     assert "產生挑品清單" in template
+    assert "補抓待比對" in template
     assert "generatePickList" in template
+    assert "backfillPchomeMatches" in template
     assert "/api/ai/product-picks/generate" in template
+    assert "/api/ai/pchome-match/backfill" in template
     assert "'product_pick':['bg-success'" in template
     assert "kpiMatchRate" in template