chore: commit remaining local changes

one_off_sites/zhongfali_single.py

@@ -191,7 +191,7 @@ def extract_records(url: str, state: Dict) -> List[Dict]:
             "source": {
                 "site": SITE_NAME,
                 "list_url": url,
-                "detail_url": url,
+                "detail_url": "",
                 "province": province,
                 "province_py": "",
                 "city": area or city,
@@ -199,6 +199,7 @@ def extract_records(url: str, state: Dict) -> List[Dict]:
                 "page": 1,
                 "group_id": group_id,
                 "module_id": module_id,
+                "detail_url_status": "unresolved_from_pool",
             },
             "list_snapshot": {
                 "name": "",
@@ -347,12 +348,68 @@ def write_records_to_db(db: Db, records: List[Dict]) -> Tuple[int, int]:
     return inserted, skipped
 
 
+def lookup_name_map_from_db(db: Db, phones: List[str]) -> Dict[str, str]:
+    deduped = sorted({phone for phone in phones if phone})
+    if not deduped:
+        return {}
+
+    name_map: Dict[str, str] = {}
+    cur = db.db.cursor()
+    try:
+        chunk_size = 500
+        for i in range(0, len(deduped), chunk_size):
+            chunk = deduped[i:i + chunk_size]
+            placeholders = ",".join(["%s"] * len(chunk))
+            sql = (
+                "SELECT phone, name, create_time FROM lawyer "
+                f"WHERE phone IN ({placeholders}) AND name<>'' "
+                "ORDER BY create_time DESC"
+            )
+            cur.execute(sql, chunk)
+            for phone, name, _ in cur.fetchall():
+                if phone not in name_map and name:
+                    name_map[phone] = str(name).strip()
+    finally:
+        cur.close()
+    return name_map
+
+
+def apply_name_backfill(records: List[Dict], name_map: Dict[str, str]) -> int:
+    updated = 0
+    if not name_map:
+        return updated
+
+    for record in records:
+        profile = record.get("profile", {}) or {}
+        list_snapshot = record.get("list_snapshot", {}) or {}
+        phone = normalize_phone(profile.get("phone", ""))
+        if not phone:
+            continue
+
+        backfill_name = name_map.get(phone, "")
+        if not backfill_name:
+            continue
+
+        current_name = str(profile.get("name") or "").strip()
+        if current_name:
+            continue
+
+        profile["name"] = backfill_name
+        list_snapshot["name"] = backfill_name
+        record["profile"] = profile
+        record["list_snapshot"] = list_snapshot
+        updated += 1
+
+    return updated
+
+
 def parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser(description="众法利单页律师电话采集")
     parser.add_argument("--url", default=DEFAULT_URL, help="详情页 URL")
     parser.add_argument("--output", default=DEFAULT_OUTPUT, help="输出 jsonl 文件路径")
     parser.add_argument("--direct", action="store_true", help="直连模式，不使用代理")
     parser.add_argument("--no-db", action="store_true", help="仅输出 JSON，不写入数据库")
+    parser.add_argument("--skip-name-backfill", action="store_true", help="跳过按手机号回填姓名")
     return parser.parse_args()
 
 
@@ -418,12 +475,24 @@ def main() -> None:
 
     db_new = 0
     db_skip = 0
+    name_backfill_count = 0
+    if not args.skip_name_backfill:
+        try:
+            with Db() as db:
+                name_map = lookup_name_map_from_db(
+                    db,
+                    [normalize_phone((record.get("profile", {}) or {}).get("phone", "")) for record in records],
+                )
+                name_backfill_count = apply_name_backfill(records, name_map)
+        except Exception as exc:
+            print(f"[name-backfill] 跳过，查询失败: {exc}")
+
     if not args.no_db:
         with Db() as db:
             db_new, db_skip = write_records_to_db(db, records)
 
     print(
-        f"[done] 采集{len(records)}条, JSON新增{json_new}条, "
+        f"[done] 采集{len(records)}条, 姓名回填{name_backfill_count}条, JSON新增{json_new}条, "
         f"DB新增{db_new}条, DB跳过{db_skip}条, 输出: {args.output}"
     )