chore: 暂存本地修改

- Updated `.gitignore` to streamline ignored files and added logging for common sites.
- Expanded `config.py` with new configurations for Weixin and Redis, and improved database connection settings.
- Refined `README.md` to clarify project structure and usage instructions.
- Enhanced `requirements.txt` with additional dependencies for MongoDB and Redis support.
- Refactored multiple spider scripts to utilize a session-based approach for HTTP requests, improving error handling and proxy management.
- Updated `export_lawyers_excel.py` to include a default timestamp for data exports.

.gitignore

@@ -1,36 +1,6 @@
-# Python
-__pycache__/
-*.py[cod]
-*$py.class
-
-# Build / packaging
-build/
-dist/
-*.egg-info/
-.eggs/
-
-# Virtual environments
 .venv/
-venv/
-env/
-
-# Test / type caches
-.pytest_cache/
-.mypy_cache/
-.ruff_cache/
-
-# IDE
-.vscode/
-.idea/
-
-# OS
-.DS_Store
-Thumbs.db
-
-# Local runtime files
-*.log
-logs/
-data/
-
-# accidental local files
-=*
+__pycache__/
+*.pyc
+common_sites/*.log
+logs/*
+data/*

.playwright/cli.config.json

@@ -0,0 +1,5 @@
+{
+  "launchOptions": {
+    "chromiumSandbox": false
+  }
+}

README.md

@@ -1,62 +1,70 @@
-# lawyers
+# lawyers-common-sites
 
-`common_sites` 独立采集项目。
+从 `/www/wwwroot/lawyer` 中抽离出的 `common_sites` 独立项目。
 
 ## 目录
 
-- `common_sites/`：大律师、找法网、法律快车、律图、华律 5 个采集脚本
-- `request/proxy_config.py`：代理配置加载逻辑
-- `request/proxy_settings.json`：代理配置文件
-- `Db.py`：数据库连接与基础操作
-- `config.py`：数据库与请求头配置
+- `common_sites/`: 站点采集脚本
+- `request/`: 代理配置
+- `utils/`: 公共工具
+- `Db.py`: 数据库封装
+- `config.py`: 项目配置
 
-## 运行
+## 快速启动
 
 ```bash
 cd /www/wwwroot/lawyers
-python3 -m venv .venv
-.venv/bin/pip install -r requirements.txt
-./common_sites/start.sh
+python -m venv .venv
+source .venv/bin/activate
+pip install -r requirements.txt
+bash common_sites/start.sh
 ```
 
-## 启动参数
+## 拆分运行（直连/代理）
 
-`start.sh` 默认并行启动 5 个站点采集（大律师使用 `dls_fresh.py`）。
+本仓库支持用环境变量 `PROXY_ENABLED` 在一次运行内强制开/关代理：
 
-- 日志目录：`/www/wwwroot/lawyers/logs`
-- 大律师 JSON 输出：`/www/wwwroot/lawyers/data/dls_records.jsonl`
+- **直连**：`PROXY_ENABLED=0`（不使用代理 IP）
+- **代理**：`PROXY_ENABLED=1`（强制使用 `request/proxy_settings.json` 的代理配置）
+- **默认**：不设置（跟随 `request/proxy_settings.json` 的 `enabled` 字段）
 
-常用环境变量：
+对应提供两套入口脚本：
 
 ```bash
-# 顺序执行（默认 parallel）
-RUN_MODE=sequential ./common_sites/start.sh
+# 直连（默认包含：大律师/大律师PC/找法网/法律快车）
+bash common_sites/start_direct_twice_weekly.sh
 
-# 大律师限制采集范围
-DLS_CITY_FILTER=beijing DLS_MAX_CITIES=1 DLS_MAX_PAGES=1 ./common_sites/start.sh
-
-# 大律师直连（不走代理）/ 仅导出JSON不写库
-DLS_DIRECT=1 DLS_NO_DB=1 ./common_sites/start.sh
+# 代理（默认包含：华律/律图）
+bash common_sites/start_proxy_weekly.sh
 ```
 
-## 导出 Excel
+## cron 示例（每周两次直连 + 每周一次代理）
 
-新增导出脚本：`common_sites/export_lawyers_excel.py`
+> 下面仅给示例，你可以按机器负载调整时间；日志会输出到 `common_sites/*.log`。
 
 ```bash
-# 无参数：默认导出最近7天数据（含手机号/姓名/律所/省份/市区/站点名称）
-# 并默认解析 params 扩展信息（邮箱/地址/执业证号/执业年限/擅长领域等）
-./.venv/bin/python ./common_sites/export_lawyers_excel.py
+# 编辑定时任务
+crontab -e
 
-# 按 create_time 时间戳范围导出
-./.venv/bin/python ./common_sites/export_lawyers_excel.py \
-  --start-ts 1772380000 --end-ts 1772429999 \
-  --output ./data/lawyers_20260302.xlsx
+# 每周二、周五 02:10 直连跑一次
+10 2 * * 2,5 cd /www/wwwroot/lawyers && bash common_sites/start_direct_twice_weekly.sh
 
-# 只导出某站点，并带技术字段（url/域名/时间等）
-./.venv/bin/python ./common_sites/export_lawyers_excel.py \
-  --domain 大律师 --include-extra
-
-# 如果不需要解析 params 扩展信息
-./.venv/bin/python ./common_sites/export_lawyers_excel.py --no-parse-params
+# 每周日 03:20 走代理跑一次（你手动续费代理 IP）
+20 3 * * 0 cd /www/wwwroot/lawyers && bash common_sites/start_proxy_weekly.sh
 ```
+
+### 常用参数（可选）
+
+```bash
+# 限流（跨进程共享），直连可适当调高，代理建议保守
+export PROXY_MAX_REQUESTS_PER_SECOND=8
+
+# 代理连通性输出（部分脚本会打印测试信息）
+export PROXY_TEST=1
+```
+
+## 说明
+
+- 当前项目直接复用原项目数据库配置和代理配置。
+- 采集依赖原库中的 `lawyer`、`area_new`、`area`、`area2` 等表。
+- 日志默认输出到 `common_sites/*.log`。

common_sites/baidu_lvlin.py

@@ -0,0 +1,473 @@
+#!/usr/bin/env python3
+import argparse
+import json
+import os
+import sys
+import time
+from typing import Dict, List, Optional, Set, Tuple
+from urllib.parse import urlencode
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+project_root = os.path.dirname(current_dir)
+request_dir = os.path.join(project_root, "request")
+if request_dir not in sys.path:
+    sys.path.insert(0, request_dir)
+if project_root not in sys.path:
+    sys.path.append(project_root)
+
+import requests
+from requests.adapters import HTTPAdapter
+from urllib3.util.retry import Retry
+
+from Db import Db
+from request.proxy_config import get_proxies, report_proxy_status
+
+
+DOMAIN = "百度法行宝"
+BASE_URL = "https://lvlin.baidu.com"
+CITY_API = f"{BASE_URL}/pc/api/law/sync/city"
+LIST_API = f"{BASE_URL}/pc/api/law/api/lawyerlist"
+DETAIL_API = f"{BASE_URL}/pc/api/law/api/lawyerhome"
+DEFAULT_PAGE_SIZE = 16
+DEFAULT_MAX_PAGES = 30
+DEFAULT_STOP_ZERO_NEW_PAGES = 3
+DEFAULT_SLEEP_SECONDS = 0.1
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="采集百度法行宝律师信息并落库")
+    parser.add_argument("--province", default="", help="仅采集指定省份，例如：山东")
+    parser.add_argument("--city", default="", help="仅采集指定城市，例如：聊城 / 聊城市")
+    parser.add_argument(
+        "--areas",
+        default="",
+        help="指定案件类型，逗号分隔；不传时自动发现顶级类型并追加不限",
+    )
+    parser.add_argument(
+        "--limit-cities",
+        type=int,
+        default=0,
+        help="仅处理前 N 个城市，0 表示不限",
+    )
+    parser.add_argument(
+        "--page-size",
+        type=int,
+        default=DEFAULT_PAGE_SIZE,
+        help=f"每次列表请求条数，默认 {DEFAULT_PAGE_SIZE}",
+    )
+    parser.add_argument(
+        "--max-pages-per-query",
+        type=int,
+        default=DEFAULT_MAX_PAGES,
+        help=f"单城市单类型最大翻页数，默认 {DEFAULT_MAX_PAGES}",
+    )
+    parser.add_argument(
+        "--stop-zero-new-pages",
+        type=int,
+        default=DEFAULT_STOP_ZERO_NEW_PAGES,
+        help=f"连续多少页无新增就停止当前查询，默认 {DEFAULT_STOP_ZERO_NEW_PAGES}",
+    )
+    parser.add_argument(
+        "--sleep-seconds",
+        type=float,
+        default=DEFAULT_SLEEP_SECONDS,
+        help=f"请求间隔秒数，默认 {DEFAULT_SLEEP_SECONDS}",
+    )
+    return parser.parse_args()
+
+
+class BaiduLvlinSpider:
+    def __init__(self, db_connection: Db, args: argparse.Namespace):
+        self.db = db_connection
+        self.args = args
+        self.page_size = max(1, int(args.page_size or DEFAULT_PAGE_SIZE))
+        self.max_pages_per_query = max(1, int(args.max_pages_per_query or DEFAULT_MAX_PAGES))
+        self.stop_zero_new_pages = max(1, int(args.stop_zero_new_pages or DEFAULT_STOP_ZERO_NEW_PAGES))
+        self.sleep_seconds = max(0.0, float(args.sleep_seconds or 0.0))
+        self.proxy_enabled = False
+        self.session = self._build_session()
+        self.existing_urls = self._load_existing_urls()
+        self.cities = self._load_cities()
+        self.areas = self._load_areas()
+        self.inserted_count = 0
+
+    def _build_session(self) -> requests.Session:
+        report_proxy_status()
+        session = requests.Session()
+        session.trust_env = False
+        proxies = get_proxies()
+        if proxies:
+            session.proxies.update(proxies)
+            self.proxy_enabled = True
+        else:
+            session.proxies.clear()
+            self.proxy_enabled = False
+
+        retries = Retry(
+            total=3,
+            backoff_factor=1,
+            status_forcelist=(429, 500, 502, 503, 504),
+            allowed_methods=frozenset(["GET"]),
+            raise_on_status=False,
+        )
+        adapter = HTTPAdapter(max_retries=retries)
+        session.mount("https://", adapter)
+        session.mount("http://", adapter)
+        session.headers.update(
+            {
+                "User-Agent": (
+                    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+                    "AppleWebKit/537.36 (KHTML, like Gecko) "
+                    "Chrome/123.0.0.0 Safari/537.36"
+                ),
+                "Accept": "application/json, text/plain, */*",
+                "Referer": f"{BASE_URL}/pc/r?vn=law",
+                "Connection": "close",
+            }
+        )
+        return session
+
+    def _disable_proxy(self) -> None:
+        if not self.proxy_enabled:
+            return
+        self.session.proxies.clear()
+        self.proxy_enabled = False
+        print(f"[{DOMAIN}] 代理不可用，已切换直连")
+
+    def _sleep(self) -> None:
+        if self.sleep_seconds > 0:
+            time.sleep(self.sleep_seconds)
+
+    def _get_json(self, url: str, params: Optional[Dict[str, object]] = None, referer: str = "") -> Dict:
+        headers = {}
+        if referer:
+            headers["Referer"] = referer
+        try:
+            resp = self.session.get(url, params=params or {}, timeout=20, headers=headers)
+        except requests.exceptions.ProxyError:
+            self._disable_proxy()
+            resp = self.session.get(url, params=params or {}, timeout=20, headers=headers)
+        try:
+            resp.raise_for_status()
+            return resp.json()
+        finally:
+            resp.close()
+
+    def _load_existing_urls(self) -> Set[str]:
+        urls: Set[str] = set()
+        cursor = self.db.db.cursor()
+        try:
+            cursor.execute("SELECT url FROM lawyer WHERE domain=%s AND url IS NOT NULL", (DOMAIN,))
+            for row in cursor.fetchall():
+                url = (row[0] or "").strip()
+                if url:
+                    urls.add(url)
+        finally:
+            cursor.close()
+        print(f"[{DOMAIN}] 已存在 URL 数: {len(urls)}")
+        return urls
+
+    def _normalize_city_name(self, city_name: str) -> str:
+        text = str(city_name or "").strip()
+        if text.endswith("市"):
+            return text[:-1]
+        return text
+
+    def _city_matches(self, expected_city: str, actual_city: str) -> bool:
+        left = self._normalize_city_name(expected_city)
+        right = self._normalize_city_name(actual_city)
+        if not left or not right:
+            return False
+        return left == right
+
+    def _load_cities(self) -> List[Dict[str, str]]:
+        payload = self._get_json(CITY_API, params={"vn": "law"}, referer=f"{BASE_URL}/pc/r?vn=law")
+        all_city_list = payload.get("data", {}).get("AllCityList", []) or []
+        cities: List[Dict[str, str]] = []
+        province_filter = self.args.province.strip()
+        city_filter = self._normalize_city_name(self.args.city)
+
+        for block in all_city_list:
+            for item in block.get("cityList", []) or []:
+                city_name = str(item.get("name") or "").strip()
+                province = str(item.get("province") or "").strip()
+                city_code = str(item.get("code") or "").strip()
+                if not city_name or not province or not city_code:
+                    continue
+                if province_filter and province != province_filter:
+                    continue
+                if city_filter and self._normalize_city_name(city_name) != city_filter:
+                    continue
+                cities.append(
+                    {
+                        "province": province,
+                        "city": city_name,
+                        "city_code": city_code,
+                    }
+                )
+
+        cities.sort(key=lambda item: (item["province"], item["city"]))
+        if self.args.limit_cities and self.args.limit_cities > 0:
+            cities = cities[: self.args.limit_cities]
+        print(f"[{DOMAIN}] 本次待采城市数: {len(cities)}")
+        return cities
+
+    def _discover_top_level_areas(self) -> List[str]:
+        sample_city = self.cities[0]["city"] if self.cities else "北京"
+        payload = self._get_json(
+            LIST_API,
+            params={
+                "city_name": sample_city,
+                "page_num": 1,
+                "page_size": self.page_size,
+                "ts": int(time.time()),
+                "clientType": "pc",
+                "list_type": 1,
+            },
+            referer=f"{BASE_URL}/pc/r?vn=law",
+        )
+        filters = payload.get("data", {}).get("filters", []) or []
+        areas: List[str] = ["不限"]
+        seen = {"不限"}
+        for item in filters:
+            if item.get("key") != "type":
+                continue
+            for option in item.get("options", []) or []:
+                value = str(option.get("value") or "").strip()
+                if not value or value in seen:
+                    continue
+                seen.add(value)
+                areas.append(value)
+        return areas
+
+    def _load_areas(self) -> List[str]:
+        if self.args.areas.strip():
+            areas = [part.strip() for part in self.args.areas.split(",") if part.strip()]
+            unique: List[str] = []
+            seen = set()
+            for area in areas:
+                if area not in seen:
+                    seen.add(area)
+                    unique.append(area)
+            print(f"[{DOMAIN}] 使用指定案件类型: {unique}")
+            return unique
+
+        areas = self._discover_top_level_areas()
+        print(f"[{DOMAIN}] 自动发现案件类型: {areas}")
+        return areas
+
+    def _build_pc_detail_url(self, qc_no: str, rs_id: str) -> str:
+        return f"{BASE_URL}/pc/lawyer?vn=law&qc_no={qc_no}&rs_id={rs_id}"
+
+    def _build_list_page_url(self, city_name: str, area_name: str) -> str:
+        params = {"city": city_name, "vn": "law"}
+        if area_name and area_name != "不限":
+            params["expertiseArea"] = area_name
+        return f"{BASE_URL}/pc/r?{urlencode(params)}"
+
+    def _fetch_list(self, city_name: str, area_name: str, page_num: int) -> List[Dict]:
+        params: Dict[str, object] = {
+            "city_name": city_name,
+            "page_num": page_num,
+            "page_size": self.page_size,
+            "ts": int(time.time()),
+            "clientType": "pc",
+            "list_type": 1,
+        }
+        if area_name and area_name != "不限":
+            params["expertiseArea"] = area_name
+        payload = self._get_json(
+            LIST_API,
+            params=params,
+            referer=self._build_list_page_url(city_name, area_name),
+        )
+        return payload.get("data", {}).get("lawyer_list", []) or []
+
+    def _fetch_detail(self, qc_no: str, rs_id: str) -> Dict:
+        payload = self._get_json(
+            DETAIL_API,
+            params={"vn": "law", "qc_no": qc_no, "rs_id": rs_id},
+            referer=self._build_pc_detail_url(qc_no, rs_id),
+        )
+        return payload.get("data", {}).get("lawyer", {}) or {}
+
+    def _extract_phone(self, detail: Dict) -> Optional[str]:
+        for service in detail.get("lawyer_service", []) or []:
+            phone = str(service.get("phone_num") or "").strip()
+            if phone:
+                return phone
+        for service in detail.get("lawyer_service_new", []) or []:
+            phone = str(service.get("phone_num") or "").strip()
+            if phone:
+                return phone
+        return None
+
+    def _safe_json(self, payload: Dict) -> str:
+        return json.dumps(payload, ensure_ascii=False)
+
+    def _build_record(
+        self,
+        city_info: Dict[str, str],
+        area_name: str,
+        page_num: int,
+        list_item: Dict,
+        detail: Dict,
+    ) -> Dict[str, object]:
+        qc_no = str(list_item.get("qc_no") or detail.get("qc_no") or "").strip()
+        rs_id = str(list_item.get("rs_id") or detail.get("rs_id") or "").strip()
+        detail_url = self._build_pc_detail_url(qc_no, rs_id)
+        name = str(detail.get("lawyer_name") or list_item.get("lawyer_name") or "").strip()
+        law_firm = str(detail.get("practice_company") or list_item.get("practice_company") or "").strip()
+        city_name = str(list_item.get("city") or city_info.get("city") or "").strip()
+        avatar_url = str(detail.get("lawyer_avatar_big") or detail.get("lawyer_avatar") or list_item.get("lawyer_avatar_big") or list_item.get("lawyer_avatar") or "").strip()
+        phone = self._extract_phone(detail)
+
+        params = {
+            "source": {
+                "site": "baidu_lvlin",
+                "city_name": city_info.get("city"),
+                "city_code": city_info.get("city_code"),
+                "province": city_info.get("province"),
+                "expertise_area": area_name,
+                "page_num": page_num,
+                "list_url": self._build_list_page_url(city_info.get("city", ""), area_name),
+                "detail_url": detail_url,
+                "list_api": LIST_API,
+                "detail_api": DETAIL_API,
+            },
+            "list_item": list_item,
+            "detail": detail,
+        }
+
+        return {
+            "name": name or None,
+            "phone": phone or None,
+            "law_firm": law_firm or None,
+            "province": city_info.get("province") or None,
+            "city": city_name or city_info.get("city") or None,
+            "url": detail_url,
+            "avatar_url": avatar_url or None,
+            "domain": DOMAIN,
+            "create_time": int(time.time()),
+            "site_time": None,
+            "params": self._safe_json(params),
+        }
+
+    def _insert_record(self, record: Dict[str, object]) -> bool:
+        url = str(record.get("url") or "").strip()
+        if not url or url in self.existing_urls:
+            return False
+        self.db.insert_data("lawyer", record)
+        self.existing_urls.add(url)
+        self.inserted_count += 1
+        return True
+
+    def _iter_city_area(self, city_info: Dict[str, str], area_name: str) -> Tuple[int, int]:
+        inserted = 0
+        pages = 0
+        zero_new_pages = 0
+        city_name = city_info["city"]
+
+        for page_num in range(1, self.max_pages_per_query + 1):
+            pages = page_num
+            try:
+                items = self._fetch_list(city_name, area_name, page_num)
+            except Exception as exc:
+                print(f"[{DOMAIN}] 列表请求失败 {city_name}-{area_name}-p{page_num}: {exc}")
+                break
+
+            if not items:
+                print(f"[{DOMAIN}] {city_name}-{area_name} 第 {page_num} 页无数据，停止")
+                break
+
+            page_new = 0
+            for item in items:
+                qc_no = str(item.get("qc_no") or "").strip()
+                rs_id = str(item.get("rs_id") or "").strip()
+                actual_city = str(item.get("city") or "").strip()
+                if not qc_no or not rs_id:
+                    continue
+                if actual_city and not self._city_matches(city_name, actual_city):
+                    continue
+
+                detail_url = self._build_pc_detail_url(qc_no, rs_id)
+                if detail_url in self.existing_urls:
+                    continue
+
+                detail: Dict = {}
+                try:
+                    detail = self._fetch_detail(qc_no, rs_id)
+                except Exception as exc:
+                    print(f"[{DOMAIN}] 详情请求失败 {qc_no}-{rs_id}: {exc}")
+
+                record = self._build_record(city_info, area_name, page_num, item, detail)
+                try:
+                    if self._insert_record(record):
+                        page_new += 1
+                        inserted += 1
+                        print(
+                            f"[{DOMAIN}] -> 新增 {record.get('name') or qc_no} "
+                            f"| {city_name} | {area_name} | p{page_num}"
+                        )
+                except Exception as exc:
+                    print(f"[{DOMAIN}] 插入失败 {record.get('url')}: {exc}")
+                self._sleep()
+
+            print(
+                f"[{DOMAIN}] {city_name} | {area_name} | p{page_num} "
+                f"| 列表 {len(items)} | 新增 {page_new}"
+            )
+
+            if len(items) < self.page_size:
+                break
+            if page_new == 0:
+                zero_new_pages += 1
+                if zero_new_pages >= self.stop_zero_new_pages:
+                    print(
+                        f"[{DOMAIN}] {city_name}-{area_name} 连续 {zero_new_pages} 页无新增，停止"
+                    )
+                    break
+            else:
+                zero_new_pages = 0
+
+            self._sleep()
+
+        return inserted, pages
+
+    def run(self) -> None:
+        print(f"[{DOMAIN}] 启动采集")
+        if not self.cities:
+            print(f"[{DOMAIN}] 无可采城市")
+            return
+        if not self.areas:
+            print(f"[{DOMAIN}] 无可采案件类型")
+            return
+
+        total_queries = len(self.cities) * len(self.areas)
+        query_index = 0
+        for city_info in self.cities:
+            city_inserted = 0
+            for area_name in self.areas:
+                query_index += 1
+                print(
+                    f"[{DOMAIN}] 进度 {query_index}/{total_queries} | "
+                    f"{city_info['province']}-{city_info['city']} | {area_name}"
+                )
+                inserted, pages = self._iter_city_area(city_info, area_name)
+                city_inserted += inserted
+                print(
+                    f"[{DOMAIN}] 完成 {city_info['city']} | {area_name} "
+                    f"| 翻页 {pages} | 新增 {inserted}"
+                )
+            print(
+                f"[{DOMAIN}] 城市完成 {city_info['province']}-{city_info['city']} "
+                f"| 本城新增 {city_inserted} | 总新增 {self.inserted_count}"
+            )
+        print(f"[{DOMAIN}] 采集完成，总新增 {self.inserted_count}")
+
+
+if __name__ == "__main__":
+    cli_args = parse_args()
+    with Db() as db:
+        spider = BaiduLvlinSpider(db, cli_args)
+        spider.run()

common_sites/dls.py

@@ -1,14 +1,9 @@
 import json
 import os
-import random
-import re
 import sys
 import time
-from typing import Dict, List, Optional, Set, Tuple
-from urllib.parse import urljoin
-
-import urllib3
-from bs4 import BeautifulSoup
+import random
+from typing import Dict, Optional
 
 current_dir = os.path.dirname(os.path.abspath(__file__))
 project_root = os.path.dirname(current_dir)
@@ -18,144 +13,191 @@ if request_dir not in sys.path:
 if project_root not in sys.path:
     sys.path.append(project_root)
 
-from Db import Db
-from request.requests_client import (
-    RequestClientError,
-    RequestConnectTimeout,
-    RequestConnectionError,
-    RequestTimeout,
-    RequestsClient,
-)
-from utils.rate_limiter import wait_for_request
+import requests
+from requests.adapters import HTTPAdapter
+from urllib3.util.retry import Retry
+import urllib3
+from bs4 import BeautifulSoup
+from request.proxy_config import get_proxies, report_proxy_status
 
+# 禁用 SSL 警告
 urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
 
+from Db import Db
+from utils.rate_limiter import request_slot
+
 DOMAIN = "大律师"
-SITE_BASE = "https://m.maxlaw.cn"
-LIST_TEMPLATE = SITE_BASE + "/law/{pinyin}?page={page}"
-PHONE_PATTERN = re.compile(r"1[3-9]\d{9}")
-MAX_PAGES_PER_CITY = int(os.getenv("MAXLAW_MAX_PAGES", "0"))
-PROXY_TESTED = False
+LIST_TEMPLATE = "https://m.maxlaw.cn/law/{pinyin}?page={page}"
+_PROXY_TESTED = False
 
 
 class DlsSpider:
     def __init__(self, db_connection):
         self.db = db_connection
-        self.client = self._build_client()
+        self.session = self._build_session()
         self.areas = self._load_areas()
 
-    def _build_client(self) -> RequestsClient:
-        client = RequestsClient(
-            headers={
-                "User-Agent": (
-                    "Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) "
-                    "AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 "
-                    "Mobile/15E148 Safari/604.1"
-                ),
-                "Host": "m.maxlaw.cn",
-                "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
-                "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
-                "Connection": "close",
-            },
-            retry_total=3,
-            retry_backoff_factor=1,
-            retry_status_forcelist=(429, 500, 502, 503, 504),
-            retry_allowed_methods=("GET", "POST"),
+    def _build_session(self) -> requests.Session:
+        """构建带重试机制的 session"""
+        report_proxy_status()
+        s = requests.Session()
+        s.trust_env = False
+        proxies = get_proxies()
+        if proxies:
+            s.proxies.update(proxies)
+        else:
+            s.proxies.clear()
+        self._proxy_test(s, proxies)
+        # 配置重试策略
+        retries = Retry(
+            total=3,  # 总共重试3次
+            backoff_factor=1,  # 重试间隔：1s, 2s, 4s
+            status_forcelist=(429, 500, 502, 503, 504),  # 对这些状态码进行重试
+            allowed_methods=frozenset(["GET", "POST"]),
+            raise_on_status=False  # 不立即抛出异常，让代码处理
         )
-        self._proxy_test(client, client.proxies or None)
-        return client
+        adapter = HTTPAdapter(max_retries=retries)
+        s.mount("https://", adapter)
+        s.mount("http://", adapter)
+        s.headers.update({
+            "User-Agent": "Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 Mobile/15E148 Safari/604.1",
+            "Host": "m.maxlaw.cn",
+            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
+            "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
+            "Connection": "close",
+        })
+        return s
 
-    def _refresh_client(self) -> None:
-        self.client.refresh()
-        self._proxy_test(self.client, self.client.proxies or None)
+    def _refresh_session(self) -> None:
+        try:
+            self.session.close()
+        except Exception:
+            pass
+        self.session = self._build_session()
 
-    def _proxy_test(self, client: RequestsClient, proxies: Optional[Dict[str, str]]) -> None:
-        global PROXY_TESTED
-        if PROXY_TESTED or not os.getenv("PROXY_TEST"):
+    def _proxy_test(self, session: requests.Session, proxies: Optional[Dict[str, str]]) -> None:
+        global _PROXY_TESTED
+        if _PROXY_TESTED or not os.getenv("PROXY_TEST"):
             return
-        PROXY_TESTED = True
+        _PROXY_TESTED = True
         if not proxies:
             print("[proxy] test skipped: no proxy configured")
             return
         test_url = os.getenv("PROXY_TEST_URL", "https://dev.kdlapi.com/testproxy")
         timeout = float(os.getenv("PROXY_TEST_TIMEOUT", "10"))
         try:
-            resp = client.get_text(test_url, timeout=timeout, headers={"Connection": "close"})
+            with request_slot():
+                resp = session.get(
+                    test_url,
+                    timeout=timeout,
+                    headers={"Connection": "close"},
+                )
             print(f"[proxy] test {resp.status_code}: {resp.text.strip()[:200]}")
         except Exception as exc:
             print(f"[proxy] test failed: {exc}")
 
-    def _load_areas(self) -> List[Dict[str, str]]:
-        tables = ("area_new", "area2", "area")
-        last_error = None
-        for table in tables:
-            try:
-                rows = self.db.select_data(table, "province, city, pinyin", "domain='maxlaw'") or []
-            except Exception as exc:
-                last_error = exc
-                continue
-            if rows:
-                missing_pinyin = sum(1 for row in rows if not (row.get("pinyin") or "").strip())
-                print(f"[大律师] 地区来源表: {table}, 城市数: {len(rows)}, 缺少pinyin: {missing_pinyin}")
-                return rows
-        if last_error:
-            print(f"[大律师] 加载地区失败: {last_error}")
-        print("[大律师] 无地区数据（已尝试 area_new/area2/area）")
-        return []
+    def _load_areas(self):
+        try:
+            return self.db.select_data(
+                "area_new",
+                "province, city, pinyin",
+                "domain='maxlaw'"
+            ) or []
+        except Exception as exc:
+            print(f"加载地区失败: {exc}")
+            return []
 
-    def _get(
-        self,
-        url: str,
-        *,
-        headers: Optional[Dict[str, str]] = None,
-        max_retries: int = 3,
-        timeout: Tuple[int, int] = (10, 30),
-    ) -> Optional[str]:
-        wait_for_request()
+    def _get(self, url: str, max_retries: int = 3, headers: Optional[Dict[str, str]] = None) -> Optional[str]:
+        """发送 GET 请求，带重试机制"""
         for attempt in range(max_retries):
             try:
-                resp = self.client.get_text(url, timeout=timeout, verify=False, headers=headers)
-                if resp.status_code == 403:
+                # 使用更长的超时时间，分别设置连接和读取超时
+                with request_slot():
+                    resp = self.session.get(
+                        url,
+                        timeout=(10, 30),  # (connect_timeout, read_timeout)
+                        verify=False,
+                        headers=headers,
+                    )
+                status_code = resp.status_code
+                content = resp.text
+                resp.close()
+                if status_code == 403:
                     if attempt < max_retries - 1:
-                        wait_time = (2 ** attempt) + random.uniform(0.3, 1.0)
-                        print(f"请求403，{wait_time:.2f}s后重试 ({attempt + 1}/{max_retries}): {url}")
-                        self._refresh_client()
+                        wait_time = 2 ** attempt + random.uniform(0.3, 1.0)
+                        print(f"403被拦截，{wait_time}秒后重试 ({attempt + 1}/{max_retries}): {url}")
+                        self._refresh_session()
                         time.sleep(wait_time)
                         continue
                     print(f"请求失败 {url}: 403 Forbidden")
                     return None
-                if resp.status_code >= 400:
-                    raise RequestClientError(f"{resp.status_code} Error: {url}")
-                return resp.text
-            except RequestConnectTimeout as exc:
+                if status_code >= 400:
+                    raise requests.exceptions.HTTPError(f"{status_code} Error: {url}")
+                return content
+            except requests.exceptions.ConnectTimeout as exc:
+                if attempt < max_retries - 1:
+                    wait_time = 2 ** attempt  # 指数退避：2s, 4s, 8s
+                    print(f"连接超时，{wait_time}秒后重试 ({attempt + 1}/{max_retries}): {url}")
+                    time.sleep(wait_time)
+                else:
+                    print(f"连接超时，已达到最大重试次数 {url}: {exc}")
+                    return None
+            except requests.exceptions.Timeout as exc:
                 if attempt < max_retries - 1:
                     wait_time = 2 ** attempt
-                    print(f"连接超时，{wait_time}s后重试 ({attempt + 1}/{max_retries}): {url}")
+                    print(f"请求超时，{wait_time}秒后重试 ({attempt + 1}/{max_retries}): {url}")
                     time.sleep(wait_time)
-                    continue
-                print(f"连接超时，已达到最大重试次数 {url}: {exc}")
-                return None
-            except RequestTimeout as exc:
+                else:
+                    print(f"请求超时，已达到最大重试次数 {url}: {exc}")
+                    return None
+            except requests.exceptions.ConnectionError as exc:
                 if attempt < max_retries - 1:
                     wait_time = 2 ** attempt
-                    print(f"请求超时，{wait_time}s后重试 ({attempt + 1}/{max_retries}): {url}")
+                    print(f"连接错误，{wait_time}秒后重试 ({attempt + 1}/{max_retries}): {url}")
                     time.sleep(wait_time)
-                    continue
-                print(f"请求超时，已达到最大重试次数 {url}: {exc}")
-                return None
-            except RequestConnectionError as exc:
-                if attempt < max_retries - 1:
-                    wait_time = 2 ** attempt
-                    print(f"连接错误，{wait_time}s后重试 ({attempt + 1}/{max_retries}): {url}")
-                    time.sleep(wait_time)
-                    continue
-                print(f"连接错误，已达到最大重试次数 {url}: {exc}")
-                return None
-            except RequestClientError as exc:
+                else:
+                    print(f"连接错误，已达到最大重试次数 {url}: {exc}")
+                    return None
+            except requests.exceptions.RequestException as exc:
                 print(f"请求失败 {url}: {exc}")
                 return None
+        
         return None
 
+    def _parse_list(self, html: str, province: str, city: str, list_url: str) -> int:
+        soup = BeautifulSoup(html, "html.parser")
+        cards = soup.find_all("div", class_="lstx")
+        if not cards:
+            return 0
+
+        inserted = 0
+        for card in cards:
+            link = card.find("a")
+            if not link or not link.get("href"):
+                continue
+            detail = self._parse_detail(link['href'], province, city, list_url)
+            if not detail:
+                continue
+            phone = detail.get("phone")
+            if not phone:
+                continue
+            condition = f"phone='{phone}' and domain='{DOMAIN}'"
+            if self.db.is_data_exist("lawyer", condition):
+                print(f"  -- 已存在: {detail['name']} ({phone})")
+                time.sleep(0.3)
+                continue
+            try:
+                self.db.insert_data("lawyer", detail)
+                inserted += 1
+                print(f"  -> 新增: {detail['name']} ({phone})")
+            except Exception as exc:
+                print(f"  插入失败: {exc}")
+            time.sleep(1)
+            time.sleep(0.3)
+        # 列表页结束后再缓一缓，降低风控
+        time.sleep(0.6)
+        return inserted
+
     def _detail_headers(self, referer: str) -> Dict[str, str]:
         return {
             "Referer": referer,
@@ -166,215 +208,72 @@ class DlsSpider:
             "Upgrade-Insecure-Requests": "1",
         }
 
-    def _extract_detail_urls(self, html: str) -> List[str]:
-        soup = BeautifulSoup(html, "html.parser")
-        urls: List[str] = []
-        seen: Set[str] = set()
-
-        # 主选择器：当前站点列表卡片
-        for a_tag in soup.select("div.lstx a[href]"):
-            href = (a_tag.get("href") or "").strip()
-            if not href:
-                continue
-            url = urljoin(SITE_BASE, href)
-            if url in seen:
-                continue
-            seen.add(url)
-            urls.append(url)
-
-        # 回退选择器：页面结构轻微变化时尽量保活
-        if not urls:
-            for a_tag in soup.select("a[href]"):
-                href = (a_tag.get("href") or "").strip()
-                if "/lawyer/" not in href:
-                    continue
-                url = urljoin(SITE_BASE, href)
-                if url in seen:
-                    continue
-                seen.add(url)
-                urls.append(url)
-        return urls
-
-    def _extract_name(self, soup: BeautifulSoup) -> str:
-        for selector in ("h2.lawyerName", "h1.lawyerName", "h1", "h2"):
-            tag = soup.select_one(selector)
-            if tag:
-                name = tag.get_text(strip=True)
-                if name:
-                    return name
-        title = soup.title.get_text(strip=True) if soup.title else ""
-        match = re.search(r"(\S+律师)", title)
-        return match.group(1) if match else ""
-
-    def _extract_law_firm(self, soup: BeautifulSoup) -> str:
-        for selector in ("p.law-firm", "div.law-firm", "p[class*=firm]"):
-            tag = soup.select_one(selector)
-            if tag:
-                text = tag.get_text(strip=True)
-                if text:
-                    return text
-        page_text = soup.get_text(" ", strip=True)
-        match = re.search(r"(执业机构|律所)\s*[:：]?\s*([^\s，。,；;]{2,40})", page_text)
-        if match:
-            return match.group(2).strip()
-        return ""
-
-    def _normalize_phone(self, text: str) -> str:
-        compact = re.sub(r"\D", "", text or "")
-        match = PHONE_PATTERN.search(compact)
-        return match.group(0) if match else ""
-
-    def _extract_phone(self, soup: BeautifulSoup) -> str:
-        contact = soup.select_one("ul.contact-content")
-        if contact:
-            phone = self._normalize_phone(contact.get_text(" ", strip=True))
-            if phone:
-                return phone
-        for selector in ("a[href^='tel:']", "span.phone", "p.phone"):
-            tag = soup.select_one(selector)
-            if tag:
-                phone = self._normalize_phone(tag.get_text(" ", strip=True))
-                if phone:
-                    return phone
-        return self._normalize_phone(soup.get_text(" ", strip=True))
-
-    def _parse_detail(self, detail_url: str, province: str, city: str, list_url: str) -> Optional[Dict[str, str]]:
-        print(f"  详情: {detail_url}")
-        html = self._get(detail_url, headers=self._detail_headers(list_url))
+    def _parse_detail(self, path: str, province: str, city: str, list_url: str) -> Optional[Dict[str, str]]:
+        url = f"https://m.maxlaw.cn{path}"
+        print(f"  详情: {url}")
+        html = self._get(url, headers=self._detail_headers(list_url))
         if not html:
             return None
 
         soup = BeautifulSoup(html, "html.parser")
-        name = self._extract_name(soup)
-        phone = self._extract_phone(soup)
+        name_tag = soup.find("h2", class_="lawyerName")
+        law_firm_tag = soup.find("p", class_="law-firm")
+        contact_list = soup.find("ul", class_="contact-content")
+
+        name = name_tag.get_text(strip=True) if name_tag else ""
+        law_firm = law_firm_tag.get_text(strip=True) if law_firm_tag else ""
+        phone = ""
+
+        if contact_list:
+            items = contact_list.find_all("li")
+            if len(items) > 2:
+                phone_tag = items[2].find("p")
+                if phone_tag:
+                    phone = phone_tag.get_text(strip=True)
+                    phone = phone.split("咨询请说明来自大律师网")[0].strip()
+
+        phone = phone.replace('-', '').strip()
         if not name or not phone:
             print("    信息不完整，跳过")
             return None
 
-        safe_city = city or province
+        safe_city = city if city else province
         return {
             "name": name,
-            "law_firm": self._extract_law_firm(soup),
+            "law_firm": law_firm,
             "province": province,
             "city": safe_city,
             "phone": phone,
-            "url": detail_url,
+            "url": url,
             "domain": DOMAIN,
             "create_time": int(time.time()),
-            "params": json.dumps({"province": province, "city": safe_city}, ensure_ascii=False),
+            "params": json.dumps({"province": province, "city": safe_city}, ensure_ascii=False)
         }
 
-    def _existing_phones(self, phones: List[str]) -> Set[str]:
-        if not phones:
-            return set()
-        existing: Set[str] = set()
-        cur = self.db.db.cursor()
-        try:
-            chunk_size = 500
-            for idx in range(0, len(phones), chunk_size):
-                chunk = phones[idx:idx + chunk_size]
-                placeholders = ",".join(["%s"] * len(chunk))
-                sql = f"SELECT phone FROM lawyer WHERE domain=%s AND phone IN ({placeholders})"
-                cur.execute(sql, [DOMAIN, *chunk])
-                for row in cur.fetchall():
-                    existing.add(row[0])
-        finally:
-            cur.close()
-        return existing
-
-    def _save_lawyers(self, lawyers: List[Dict[str, str]]) -> Tuple[int, int]:
-        if not lawyers:
-            return 0, 0
-        phones = [row["phone"] for row in lawyers if row.get("phone")]
-        existing = self._existing_phones(phones)
-        inserted = 0
-        skipped = 0
-
-        for row in lawyers:
-            phone = row.get("phone", "")
-            if not phone:
-                skipped += 1
-                continue
-            if phone in existing:
-                skipped += 1
-                print(f"  -- 已存在: {row.get('name', '')} ({phone})")
-                continue
-            try:
-                self.db.insert_data("lawyer", row)
-                existing.add(phone)
-                inserted += 1
-                print(f"  -> 新增: {row.get('name', '')} ({phone})")
-            except Exception as exc:
-                skipped += 1
-                print(f"  插入失败 {row.get('url', '')}: {exc}")
-        return inserted, skipped
-
-    def _crawl_city(self, area: Dict[str, str]) -> Tuple[int, int]:
-        pinyin = (area.get("pinyin") or "").strip()
-        province = area.get("province", "")
-        city = area.get("city", "")
-        if not pinyin:
-            return 0, 0
-
-        total_inserted = 0
-        total_parsed = 0
-        page = 1
-        prev_fingerprint = ""
-
-        while True:
-            if MAX_PAGES_PER_CITY > 0 and page > MAX_PAGES_PER_CITY:
-                print(f"达到分页上限({MAX_PAGES_PER_CITY})，停止 {province}-{city}")
-                break
-
-            list_url = LIST_TEMPLATE.format(pinyin=pinyin, page=page)
-            print(f"采集 {province}-{city} 第 {page} 页: {list_url}")
-            html = self._get(list_url)
-            if not html:
-                break
-
-            detail_urls = self._extract_detail_urls(html)
-            if not detail_urls:
-                print("  列表为空，结束当前城市")
-                break
-
-            fingerprint = "|".join(detail_urls[:8])
-            if fingerprint and fingerprint == prev_fingerprint:
-                print("  列表页重复，提前停止当前城市")
-                break
-            prev_fingerprint = fingerprint
-
-            lawyers: List[Dict[str, str]] = []
-            for detail_url in detail_urls:
-                row = self._parse_detail(detail_url, province, city, list_url)
-                if row:
-                    lawyers.append(row)
-                time.sleep(0.25)
-
-            inserted, skipped = self._save_lawyers(lawyers)
-            total_inserted += inserted
-            total_parsed += len(lawyers)
-            print(
-                f"  第 {page} 页完成: 列表{len(detail_urls)}条, "
-                f"解析{len(lawyers)}条, 新增{inserted}条, 跳过{skipped}条"
-            )
-
-            page += 1
-            time.sleep(0.5)
-        return total_inserted, total_parsed
-
     def run(self):
         print("启动大律师采集...")
         if not self.areas:
             print("无地区数据")
             return
 
-        all_inserted = 0
-        all_parsed = 0
         for area in self.areas:
-            inserted, parsed = self._crawl_city(area)
-            all_inserted += inserted
-            all_parsed += parsed
-        print(f"大律师采集完成: 解析{all_parsed}条, 新增{all_inserted}条")
+            pinyin = area.get("pinyin")
+            province = area.get("province", "")
+            city = area.get("city", "")
+            if not pinyin:
+                continue
+            page = 1
+            while True:
+                list_url = LIST_TEMPLATE.format(pinyin=pinyin, page=page)
+                print(f"采集 {province}-{city} 第 {page} 页: {list_url}")
+                html = self._get(list_url)
+                if not html:
+                    break
+                inserted = self._parse_list(html, province, city, list_url)
+                if inserted == 0:
+                    break
+                page += 1
+        print("大律师采集完成")
 
 
 if __name__ == "__main__":

common_sites/dls_fresh.py

@@ -22,7 +22,7 @@ if project_root not in sys.path:
     sys.path.append(project_root)
 
 from request.requests_client import RequestClientError, RequestsClient
-from utils.rate_limiter import wait_for_request
+from utils.rate_limiter import request_slot
 from Db import Db
 
 urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
@@ -107,9 +107,9 @@ class DlsFreshCrawler:
     def _get_text(self, url: str, timeout: int = 20, max_retries: int = 3) -> str:
         last_error: Optional[Exception] = None
         for attempt in range(max_retries):
-            wait_for_request()
             try:
-                resp = self.client.get_text(url, timeout=timeout, verify=False)
+                with request_slot():
+                    resp = self.client.get_text(url, timeout=timeout, verify=False)
                 code = resp.status_code
                 if code == 403:
                     if attempt < max_retries - 1:

common_sites/dls_pc.py

@@ -0,0 +1,438 @@
+import json
+import os
+import random
+import re
+import sys
+import time
+from typing import Dict, List, Optional, Set, Tuple
+from urllib.parse import urljoin
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+project_root = os.path.dirname(current_dir)
+request_dir = os.path.join(project_root, "request")
+if request_dir not in sys.path:
+    sys.path.insert(0, request_dir)
+if project_root not in sys.path:
+    sys.path.append(project_root)
+
+import requests
+from requests.adapters import HTTPAdapter
+from urllib3.util.retry import Retry
+import urllib3
+from bs4 import BeautifulSoup
+
+from request.proxy_config import get_proxies, report_proxy_status
+from utils.rate_limiter import request_slot
+from Db import Db
+
+urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
+
+DOMAIN = "大律师"
+SITE_BASE = "https://www.maxlaw.cn"
+LIST_URL_TEMPLATE = SITE_BASE + "/law/{pinyin}?page={page}"
+PROVINCE_API = "https://js.maxlaw.cn/js/ajax/common/getprovice.js"
+CITY_API_TEMPLATE = "https://js.maxlaw.cn/js/ajax/common/getcity_{province_id}.js"
+
+PHONE_RE = re.compile(r"1[3-9]\d{9}")
+REPLY_RE = re.compile(r"已回复[:：]?\s*(\d+)")
+AREA_PREFIX_RE = re.compile(r"^[A-Za-z]\s*")
+
+
+def normalize_phone(text: str) -> str:
+    compact = re.sub(r"\D", "", text or "")
+    match = PHONE_RE.search(compact)
+    return match.group(0) if match else ""
+
+
+def clean_area_name(text: str) -> str:
+    value = AREA_PREFIX_RE.sub("", (text or "").strip())
+    return value.strip()
+
+
+def normalize_region_text(text: str) -> str:
+    value = (text or "").strip()
+    value = value.replace("\xa0", " ")
+    value = value.replace("－", "-").replace("—", "-").replace("–", "-")
+    value = re.sub(r"\s*-\s*", "-", value)
+    value = re.sub(r"\s+", "", value)
+    return value
+
+
+class DlsPcSpider:
+    def __init__(self, db_connection):
+        self.db = db_connection
+        self.session = self._build_session()
+        self.max_pages = int(os.getenv("MAXLAW_PC_MAX_PAGES", "100"))
+        self.areas = self._load_areas()
+
+    def _build_session(self) -> requests.Session:
+        report_proxy_status()
+        session = requests.Session()
+        session.trust_env = False
+        proxies = get_proxies()
+        if proxies:
+            session.proxies.update(proxies)
+        else:
+            session.proxies.clear()
+
+        retries = Retry(
+            total=3,
+            backoff_factor=1,
+            status_forcelist=(429, 500, 502, 503, 504),
+            allowed_methods=frozenset(["GET"]),
+            raise_on_status=False,
+        )
+        adapter = HTTPAdapter(max_retries=retries)
+        session.mount("https://", adapter)
+        session.mount("http://", adapter)
+        session.headers.update({
+            "User-Agent": (
+                "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+                "AppleWebKit/537.36 (KHTML, like Gecko) "
+                "Chrome/136.0.0.0 Safari/537.36"
+            ),
+            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
+            "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
+            "Connection": "close",
+        })
+        return session
+
+    def _refresh_session(self) -> None:
+        try:
+            self.session.close()
+        except Exception:
+            pass
+        self.session = self._build_session()
+
+    def _get(self, url: str, max_retries: int = 3, headers: Optional[Dict[str, str]] = None) -> Optional[str]:
+        for attempt in range(max_retries):
+            try:
+                with request_slot():
+                    resp = self.session.get(url, timeout=(10, 25), verify=False, headers=headers)
+                status_code = resp.status_code
+                text = resp.text
+                resp.close()
+                if status_code == 403:
+                    if attempt < max_retries - 1:
+                        wait_time = 2 ** attempt + random.uniform(0.3, 1.0)
+                        print(f"403被拦截，{wait_time:.1f}秒后重试 ({attempt + 1}/{max_retries}): {url}")
+                        self._refresh_session()
+                        time.sleep(wait_time)
+                        continue
+                    print(f"请求失败 {url}: 403 Forbidden")
+                    return None
+                if status_code >= 400:
+                    raise requests.exceptions.HTTPError(f"{status_code} Error: {url}")
+                return text
+            except requests.exceptions.RequestException as exc:
+                if attempt < max_retries - 1:
+                    wait_time = 2 ** attempt + random.uniform(0.3, 1.0)
+                    print(f"请求失败，{wait_time:.1f}秒后重试 ({attempt + 1}/{max_retries}): {url} -> {exc}")
+                    time.sleep(wait_time)
+                    continue
+                print(f"请求失败 {url}: {exc}")
+                return None
+        return None
+
+    def _get_json(self, url: str) -> Optional[Dict]:
+        text = self._get(url)
+        if not text:
+            return None
+        try:
+            return json.loads(text.strip().lstrip("\ufeff"))
+        except ValueError as exc:
+            print(f"解析JSON失败 {url}: {exc}")
+            return None
+
+    def _load_areas(self) -> List[Dict[str, str]]:
+        areas = self._load_areas_from_site()
+        if areas:
+            print(f"[大律师PC] 地区来源: site, 地区数: {len(areas)}")
+            return areas
+
+        areas = self._load_areas_from_db()
+        if areas:
+            print(f"[大律师PC] 地区来源: db, 地区数: {len(areas)}")
+            return areas
+
+        print("[大律师PC] 无地区数据")
+        return []
+
+    def _load_areas_from_site(self) -> List[Dict[str, str]]:
+        data = self._get_json(PROVINCE_API)
+        if not data or str(data.get("status")) != "1":
+            return []
+
+        result: List[Dict[str, str]] = []
+        seen_pinyin: Set[str] = set()
+
+        for province in data.get("ds", []) or []:
+            province_id = province.get("id")
+            province_name = clean_area_name(province.get("name", ""))
+            province_pinyin = (province.get("py_code") or "").strip()
+
+            city_rows = []
+            if province_id:
+                city_data = self._get_json(CITY_API_TEMPLATE.format(province_id=province_id))
+                if city_data and str(city_data.get("status")) == "1":
+                    city_rows = city_data.get("ds", []) or []
+
+            if not city_rows and province_pinyin and province_pinyin not in seen_pinyin:
+                seen_pinyin.add(province_pinyin)
+                result.append({
+                    "province": province_name,
+                    "city": province_name,
+                    "pinyin": province_pinyin,
+                })
+                continue
+
+            for city in city_rows:
+                city_name = clean_area_name(city.get("name", ""))
+                city_pinyin = (city.get("py_code") or "").strip()
+                if not city_pinyin or city_pinyin in seen_pinyin:
+                    continue
+                seen_pinyin.add(city_pinyin)
+                result.append({
+                    "province": province_name,
+                    "city": city_name or province_name,
+                    "pinyin": city_pinyin,
+                })
+
+        return result
+
+    def _load_areas_from_db(self) -> List[Dict[str, str]]:
+        tables = ("area_new", "area", "area2")
+        last_error = None
+        for table in tables:
+            try:
+                rows = self.db.select_data(
+                    table,
+                    "province, city, pinyin",
+                    "domain='maxlaw' AND level=2",
+                ) or []
+            except Exception as exc:
+                last_error = exc
+                continue
+
+            if rows:
+                return rows
+
+        if last_error:
+            print(f"[大律师PC] 加载数据库地区失败: {last_error}")
+        return []
+
+    def _existing_phones(self, phones: List[str]) -> Set[str]:
+        if not phones:
+            return set()
+        existing: Set[str] = set()
+        cur = self.db.db.cursor()
+        try:
+            chunk_size = 500
+            for i in range(0, len(phones), chunk_size):
+                chunk = phones[i:i + chunk_size]
+                placeholders = ",".join(["%s"] * len(chunk))
+                sql = f"SELECT phone FROM lawyer WHERE domain=%s AND phone IN ({placeholders})"
+                cur.execute(sql, [DOMAIN, *chunk])
+                for row in cur.fetchall():
+                    existing.add(row[0])
+        finally:
+            cur.close()
+        return existing
+
+    def _build_list_url(self, pinyin: str, page: int) -> str:
+        return LIST_URL_TEMPLATE.format(pinyin=pinyin, page=page)
+
+    def _parse_location_line(
+        self,
+        text: str,
+        fallback_province: str,
+        fallback_city: str,
+    ) -> Tuple[str, str, str]:
+        raw = (text or "").replace("\xa0", " ")
+        raw = re.sub(r"\s+", " ", raw).strip()
+        if not raw:
+            return fallback_province, fallback_city or fallback_province, ""
+
+        parts = raw.split(" ", 1)
+        area_text = parts[0].strip()
+        law_firm = parts[1].strip() if len(parts) > 1 else ""
+
+        province = fallback_province
+        city = fallback_city or fallback_province
+        if "-" in area_text:
+            area_parts = [item.strip() for item in area_text.split("-", 1)]
+            if area_parts[0]:
+                province = area_parts[0]
+            if len(area_parts) > 1 and area_parts[1]:
+                city = area_parts[1]
+        elif area_text:
+            province = area_text
+            city = area_text
+
+        return province, city, law_firm
+
+    def _extract_page_region(self, soup: BeautifulSoup) -> str:
+        button = soup.select_one(".filter .filter-btn")
+        if button:
+            return normalize_region_text(button.get_text(" ", strip=True))
+        title = soup.select_one(".findLawyer-title h1")
+        if title:
+            return normalize_region_text(title.get_text(strip=True).replace("律师", ""))
+        return ""
+
+    def _page_matches_area(self, soup: BeautifulSoup, province: str, city: str) -> Tuple[bool, str]:
+        current_region = self._extract_page_region(soup)
+        if not current_region:
+            return True, current_region
+        if "全国" in current_region:
+            return False, current_region
+
+        norm_province = normalize_region_text(province)
+        norm_city = normalize_region_text(city or province)
+
+        if norm_city and norm_city != norm_province:
+            matched = norm_province in current_region and norm_city in current_region
+        else:
+            matched = norm_province in current_region
+
+        if matched:
+            return True, current_region
+
+        title = soup.select_one(".findLawyer-title h1")
+        title_text = ""
+        if title:
+            title_text = normalize_region_text(title.get_text(strip=True).replace("律师", ""))
+
+        if norm_city and norm_city != norm_province:
+            matched = norm_city in title_text
+        else:
+            matched = norm_province in title_text
+
+        return matched, current_region or title_text
+
+    def _parse_list(self, html: str, province: str, city: str, list_url: str, area_pinyin: str) -> Tuple[bool, int, int]:
+        soup = BeautifulSoup(html, "html.parser")
+        matched, current_region = self._page_matches_area(soup, province, city)
+        if not matched:
+            print(f"  页面地区不匹配，停止分页: 目标={province}-{city} 当前={current_region or '未知'}")
+            return False, 0, 0
+
+        cards = []
+        seen_page_phone: Set[str] = set()
+
+        for item in soup.select("ul.findLawyer-list > li.clearfix"):
+            name_link = item.select_one(".findLawyer-list-detail-name a[href]")
+            phone_tag = item.select_one(".findLawyer-list-detail-name span")
+            if not name_link or not phone_tag:
+                continue
+
+            phone = normalize_phone(phone_tag.get_text(" ", strip=True))
+            if not phone or phone in seen_page_phone:
+                continue
+            seen_page_phone.add(phone)
+
+            name = name_link.get_text(strip=True)
+            detail_url = urljoin(SITE_BASE, name_link.get("href", "").strip())
+
+            location_tag = item.select_one(".findLawyer-list-detail-the")
+            card_province, card_city, law_firm = self._parse_location_line(
+                location_tag.get_text(" ", strip=True) if location_tag else "",
+                province,
+                city,
+            )
+
+            specialties = []
+            for dd in item.select(".findLawyer-list-detail-fields dd"):
+                text = dd.get_text(strip=True)
+                if text:
+                    specialties.append(text)
+
+            reply_count = None
+            reply_tag = item.select_one(".findLawyer-list-detail-other a")
+            if reply_tag:
+                match = REPLY_RE.search(reply_tag.get_text(" ", strip=True))
+                if match:
+                    reply_count = int(match.group(1))
+
+            cards.append({
+                "name": name,
+                "law_firm": law_firm,
+                "province": card_province or province,
+                "city": card_city or city or province,
+                "phone": phone,
+                "url": detail_url,
+                "domain": DOMAIN,
+                "create_time": int(time.time()),
+                "params": json.dumps({
+                    "area_pinyin": area_pinyin,
+                    "source": list_url,
+                    "specialties": specialties,
+                    "reply_count": reply_count,
+                }, ensure_ascii=False),
+            })
+
+        if not cards:
+            return True, 0, 0
+
+        phones = [item["phone"] for item in cards if item.get("phone")]
+        existing = self._existing_phones(phones)
+        inserted = 0
+
+        for item in cards:
+            phone = item.get("phone")
+            if not phone:
+                continue
+            if phone in existing:
+                print(f"    -- 已存在: {item['name']} ({phone})")
+                continue
+            try:
+                self.db.insert_data("lawyer", item)
+                inserted += 1
+                print(f"    -> 新增: {item['name']} ({phone})")
+            except Exception as exc:
+                print(f"    插入失败 {item.get('url')}: {exc}")
+
+        return True, inserted, len(cards)
+
+    def run(self):
+        print("启动大律师 PC 站采集...")
+        if not self.areas:
+            print("无地区数据")
+            return
+
+        for area in self.areas:
+            province = (area.get("province") or "").strip()
+            city = (area.get("city") or province).strip()
+            pinyin = (area.get("pinyin") or "").strip()
+            if not province or not pinyin:
+                continue
+
+            area_label = province if not city or city == province else f"{province}-{city}"
+            print(f"采集地区: {area_label} ({pinyin})")
+
+            for page in range(1, self.max_pages + 1):
+                list_url = self._build_list_url(pinyin, page)
+                print(f"  第 {page} 页: {list_url}")
+                html = self._get(list_url, headers={"Referer": SITE_BASE + "/law"})
+                if not html:
+                    break
+
+                page_ok, inserted, parsed_count = self._parse_list(html, province, city, list_url, pinyin)
+                if not page_ok:
+                    break
+                if parsed_count == 0:
+                    print("  当前页无律师卡片，停止")
+                    break
+
+                if inserted == 0:
+                    print("  当前页无新增数据")
+
+                time.sleep(0.5)
+
+        print("大律师 PC 站采集完成")
+
+
+if __name__ == "__main__":
+    with Db() as db:
+        spider = DlsPcSpider(db)
+        spider.run()

common_sites/export_lawyers_excel.py

@@ -19,6 +19,9 @@ if project_root not in sys.path:
 from Db import Db
 
 
+DEFAULT_EXPORT_START_TS = 1772932103
+
+
 def parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser(description="导出律师数据到 Excel")
     parser.add_argument(
@@ -30,7 +33,10 @@ def parse_args() -> argparse.Namespace:
         "--start-ts",
         type=int,
         default=None,
-        help="create_time 起始时间戳（含），不传时默认取最近7天",
+        help=(
+            "create_time 起始时间戳（含），"
+            f"不传时默认取 {DEFAULT_EXPORT_START_TS} 之后的数据"
+        ),
     )
     parser.add_argument(
         "--end-ts",
@@ -43,6 +49,11 @@ def parse_args() -> argparse.Namespace:
         default="",
         help="按 domain 过滤，例如：大律师 / 找法网 / 华律",
     )
+    parser.add_argument(
+        "--exclude-domain",
+        default="",
+        help="排除指定 domain，例如：高德地图",
+    )
     parser.add_argument(
         "--province",
         default="",
@@ -74,13 +85,18 @@ def parse_args() -> argparse.Namespace:
         action="store_true",
         help="关闭 params JSON 扩展信息解析（默认开启）",
     )
+    parser.add_argument(
+        "--douyin-only",
+        action="store_true",
+        help="仅导出抖音采集数据（domain=抖音），并追加抖音专用字段",
+    )
     return parser.parse_args()
 
 
 def apply_default_time_filter(args: argparse.Namespace) -> None:
-    # 未显式传时间范围时，默认导出最近7天的数据
+    # 未显式传时间范围时，默认导出指定时间戳之后的数据
     if args.start_ts is None and args.end_ts is None:
-        args.start_ts = int(time.time()) - 7 * 24 * 3600
+        args.start_ts = DEFAULT_EXPORT_START_TS
         args.end_ts = 0
         return
     if args.start_ts is None:
@@ -109,15 +125,23 @@ def build_query(args: argparse.Namespace) -> (str, List):
     where: List[str] = []
     params: List = []
 
+    if args.douyin_only:
+        target_domain = args.domain.strip() or "抖音"
+        where.append("domain = %s")
+        params.append(target_domain)
+
     if args.start_ts > 0:
         where.append("create_time >= %s")
         params.append(args.start_ts)
     if args.end_ts > 0:
         where.append("create_time <= %s")
         params.append(args.end_ts)
-    if args.domain.strip():
+    if args.domain.strip() and not args.douyin_only:
         where.append("domain = %s")
         params.append(args.domain.strip())
+    if args.exclude_domain.strip():
+        where.append("domain <> %s")
+        params.append(args.exclude_domain.strip())
     if args.province.strip():
         where.append("province = %s")
         params.append(args.province.strip())
@@ -161,6 +185,13 @@ def parse_params(params_text: str) -> Dict[str, str]:
     else:
         specialties_text = ""
 
+    user_info = data.get("user_info") or {}
+    if not isinstance(user_info, dict):
+        user_info = {}
+
+    sec_uid = str(data.get("sec_uid") or "")
+    douyin_url = f"https://www.douyin.com/user/{sec_uid}" if sec_uid else ""
+
     return {
         "email": str(profile.get("email") or ""),
         "address": str(profile.get("address") or ""),
@@ -170,19 +201,34 @@ def parse_params(params_text: str) -> Dict[str, str]:
         "source_site": str(source.get("site") or ""),
         "detail_url": str(source.get("detail_url") or ""),
         "list_url": str(source.get("list_url") or ""),
+        "api_source": str(data.get("api_source") or ""),
+        "api_url": str(data.get("api_url") or ""),
+        "city_index": str(data.get("city_index") or ""),
+        "captured_at": str(data.get("captured_at") or ""),
+        "sec_uid": sec_uid,
+        "douyin_uid": str(user_info.get("uid") or ""),
+        "douyin_unique_id": str(user_info.get("unique_id") or ""),
+        "douyin_signature": str(user_info.get("signature") or ""),
+        "douyin_nickname": str(user_info.get("nickname") or ""),
+        "douyin_url": douyin_url,
     }
 
 
-def export_to_excel(rows: List[Dict], output_path: str, include_extra: bool, parse_params_flag: bool) -> int:
+def export_to_excel(
+    rows: List[Dict],
+    output_path: str,
+    include_extra: bool,
+    parse_params_flag: bool,
+    douyin_only: bool,
+) -> int:
     wb = Workbook()
     ws = wb.active
     ws.title = "lawyers"
 
-    headers = ["手机号", "姓名", "律所", "省份", "市区", "站点名称", "domain"]
+    headers = ["手机号", "姓名", "律所", "省份", "市区", "站点名称", "domain", "URL"]
     if include_extra:
         headers.extend(
             [
-                "URL",
                 "站点",
                 "create_time",
                 "create_time_text",
@@ -204,6 +250,22 @@ def export_to_excel(rows: List[Dict], output_path: str, include_extra: bool, par
                 "list_url",
             ]
         )
+    if parse_params_flag and douyin_only:
+        headers.extend(
+            [
+                "sec_uid",
+                "抖音uid",
+                "抖音号",
+                "抖音昵称",
+                "抖音简介",
+                "抖音主页URL",
+                "api_source",
+                "api_url",
+                "city_index",
+                "captured_at",
+                "captured_at_text",
+            ]
+        )
 
     ws.append(headers)
     for cell in ws[1]:
@@ -221,12 +283,12 @@ def export_to_excel(rows: List[Dict], output_path: str, include_extra: bool, par
             row.get("city", "") or "",
             site_name,
             row.get("domain", "") or "",
+            row.get("url", "") or "",
         ]
 
         if include_extra:
             line.extend(
                 [
-                    row.get("url", "") or "",
                     row.get("domain", "") or "",
                     row.get("create_time", "") or "",
                     ts_to_text(row.get("create_time")),
@@ -250,6 +312,29 @@ def export_to_excel(rows: List[Dict], output_path: str, include_extra: bool, par
                 ]
             )
 
+        if parse_params_flag and douyin_only:
+            captured_at_text = ""
+            try:
+                captured_at_text = ts_to_text(int(info.get("captured_at", "") or 0))
+            except Exception:
+                captured_at_text = ""
+
+            line.extend(
+                [
+                    info.get("sec_uid", ""),
+                    info.get("douyin_uid", ""),
+                    info.get("douyin_unique_id", ""),
+                    info.get("douyin_nickname", ""),
+                    info.get("douyin_signature", ""),
+                    info.get("douyin_url", ""),
+                    info.get("api_source", ""),
+                    info.get("api_url", ""),
+                    info.get("city_index", ""),
+                    info.get("captured_at", ""),
+                    captured_at_text,
+                ]
+            )
+
         ws.append(line)
         exported += 1
 
@@ -277,6 +362,7 @@ def main() -> None:
         output_path=output_path,
         include_extra=args.include_extra,
         parse_params_flag=not args.no_parse_params,
+        douyin_only=args.douyin_only,
     )
 
     print(f"[export] 导出完成，共 {count} 条")

common_sites/findlaw.py

@@ -1,16 +1,9 @@
-import argparse
-import ast
-import hashlib
 import json
 import os
-import random
-import re
 import sys
 import time
-from dataclasses import dataclass
-from typing import Dict, Iterable, List, Optional, Set, Tuple
-
-import urllib3
+import random
+from typing import Dict, List, Set, Optional
 
 current_dir = os.path.dirname(os.path.abspath(__file__))
 project_root = os.path.dirname(current_dir)
@@ -20,460 +13,214 @@ if request_dir not in sys.path:
 if project_root not in sys.path:
     sys.path.append(project_root)
 
+import requests
+from request.proxy_config import get_proxies, report_proxy_status
 from Db import Db
-from request.requests_client import RequestClientError, RequestsClient
-from utils.rate_limiter import wait_for_request
+from utils.rate_limiter import request_slot
 
-urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
-
-SITE_NAME = "findlaw"
-LEGACY_DOMAIN = "找法网"
-SITE_BASE = "https://m.findlaw.cn"
-CITY_DATA_URL = "https://static.findlawimg.com/minify/?b=js&f=m/public/v1/areaDataTwo.js"
-LIST_URL_TEMPLATE = SITE_BASE + "/{city_py}/q_lawyer/p{page}?ajax=1&order=0&sex=-1"
-
-PHONE_RE = re.compile(r"1[3-9]\d{9}")
+DOMAIN = "找法网"
+LIST_TEMPLATE = "https://m.findlaw.cn/{pinyin}/q_lawyer/p{page}?ajax=1&order=0&sex=-1"
 
 
-@dataclass
-class CityTarget:
-    province_id: str
-    province_name: str
-    province_py: str
-    city_id: str
-    city_name: str
-    city_py: str
-
-
-def normalize_phone(text: str) -> str:
-    compact = re.sub(r"\D", "", text or "")
-    match = PHONE_RE.search(compact)
-    return match.group(0) if match else ""
-
-
-class FindlawCrawler:
-    def __init__(
-        self,
-        max_pages: int = 9999,
-        sleep_seconds: float = 0.1,
-        use_proxy: bool = True,
-        db_connection=None,
-    ):
-        self.max_pages = max_pages
-        self.sleep_seconds = max(0.0, sleep_seconds)
+class FindlawSpider:
+    def __init__(self, db_connection):
         self.db = db_connection
-        self.client = RequestsClient(
-            headers={
-                "User-Agent": (
-                    "Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) "
-                    "AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 "
-                    "Mobile/15E148 Safari/604.1"
-                ),
-                "Accept": "application/json, text/javascript, */*; q=0.01",
-                "X-Requested-With": "XMLHttpRequest",
-                "Connection": "close",
-            },
-            use_proxy=use_proxy,
-            retry_total=2,
-            retry_backoff_factor=1,
-            retry_status_forcelist=(429, 500, 502, 503, 504),
-            retry_allowed_methods=("GET",),
-        )
+        self.session = self._build_session()
+        self.cities = self._load_cities()
 
-    def _get_text(
-        self,
-        url: str,
-        timeout: int = 20,
-        max_retries: int = 3,
-        referer: str = SITE_BASE,
-    ) -> str:
-        headers = {"Referer": referer}
-        last_error: Optional[Exception] = None
+    def _build_session(self) -> requests.Session:
+        report_proxy_status()
+        session = requests.Session()
+        session.trust_env = False
+        proxies = get_proxies()
+        if proxies:
+            session.proxies.update(proxies)
+        else:
+            session.proxies.clear()
+        session.headers.update({
+            "User-Agent": (
+                "Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) "
+                "AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 "
+                "Mobile/15E148 Safari/604.1"
+            ),
+            "Accept": "application/json, text/javascript, */*; q=0.01",
+            "X-Requested-With": "XMLHttpRequest",
+            "Connection": "close",
+        })
+        return session
 
-        for attempt in range(max_retries):
-            wait_for_request()
-            try:
-                resp = self.client.get_text(url, timeout=timeout, verify=False, headers=headers)
-                code = resp.status_code
-                if code == 403:
-                    if attempt < max_retries - 1:
-                        self.client.refresh()
-                        time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                        continue
-                    raise RequestClientError(f"{code} Error: {url}")
-                if code >= 500 and attempt < max_retries - 1:
-                    time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                    continue
-                if code >= 400:
-                    raise RequestClientError(f"{code} Error: {url}")
-                return resp.text
-            except Exception as exc:
-                last_error = exc
-                if attempt < max_retries - 1:
-                    time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                    continue
-                raise
-
-        if last_error is not None:
-            raise last_error
-        raise RequestClientError(f"Unknown request error: {url}")
-
-    def _parse_city_js_array(self, script_text: str, var_name: str) -> List[Dict]:
-        pattern = rf"var\s+{re.escape(var_name)}\s*=\s*(\[[\s\S]*?\]);"
-        match = re.search(pattern, script_text)
-        if not match:
-            return []
-        raw = match.group(1)
+    def _refresh_session(self) -> None:
         try:
-            rows = ast.literal_eval(raw)
-            return rows if isinstance(rows, list) else []
+            self.session.close()
         except Exception:
-            return []
+            pass
+        self.session = self._build_session()
 
-    def discover_cities(self) -> List[CityTarget]:
-        js_text = self._get_text(CITY_DATA_URL, referer=SITE_BASE + "/findlawyers/")
-        provinces = self._parse_city_js_array(js_text, "iosProvinces")
-        cities = self._parse_city_js_array(js_text, "iosCitys")
-
-        province_map: Dict[str, Dict] = {}
-        for item in provinces:
-            pid = str(item.get("id") or "").strip()
-            if pid:
-                province_map[pid] = item
-
-        results: List[CityTarget] = []
-        seen_py: Set[str] = set()
-        for city in cities:
-            city_py = str(city.get("pinyin") or "").strip()
-            city_name = str(city.get("value") or "").strip()
-            city_id = str(city.get("id") or "").strip()
-            province_id = str(city.get("parentId") or "").strip()
-            if not city_py or not city_name or not city_id:
-                continue
-            if city_py in seen_py:
-                continue
-            seen_py.add(city_py)
-
-            province_row = province_map.get(province_id, {})
-            province_name = str(province_row.get("value") or city_name).strip()
-            province_py = str(province_row.get("pinyin") or city_py).strip()
-
-            results.append(
-                CityTarget(
-                    province_id=province_id,
-                    province_name=province_name,
-                    province_py=province_py,
-                    city_id=city_id,
-                    city_name=city_name,
-                    city_py=city_py,
-                )
-            )
-        return results
-
-    def _parse_list_payload(self, text: str) -> Dict:
-        cleaned = (text or "").strip().lstrip("\ufeff")
-        try:
-            return json.loads(cleaned)
-        except ValueError:
-            start = cleaned.find("{")
-            end = cleaned.rfind("}")
-            if start == -1 or end == -1:
-                return {}
-            return json.loads(cleaned[start:end + 1])
-
-    def fetch_list_page(self, city_py: str, page: int) -> Tuple[List[Dict], bool, str]:
-        list_url = LIST_URL_TEMPLATE.format(city_py=city_py, page=page)
-        referer = f"{SITE_BASE}/{city_py}/q_lawyer/"
-        text = self._get_text(list_url, referer=referer)
-        payload = self._parse_list_payload(text)
-        if payload.get("errcode") != 0:
-            return [], False, list_url
-
-        data = payload.get("data", {}) or {}
-        items = data.get("lawyer_list", []) or []
-        has_more = str(data.get("has_more", "0")) == "1"
-        return items, has_more, list_url
-
-    def crawl_city(self, target: CityTarget) -> Iterable[Dict]:
-        for page in range(1, self.max_pages + 1):
+    def _get(self, url: str, referer: str, verify: bool = True, max_retries: int = 3) -> Optional[str]:
+        headers = {"Referer": referer}
+        for attempt in range(max_retries):
             try:
-                items, has_more, list_url = self.fetch_list_page(target.city_py, page)
-            except Exception as exc:
-                print(f"[list] 失败 {target.city_py} p{page}: {exc}")
-                break
+                with request_slot():
+                    resp = self.session.get(url, timeout=15, verify=verify, headers=headers)
+                status_code = resp.status_code
+                text = resp.text
+                resp.close()
+                if status_code == 403:
+                    if attempt < max_retries - 1:
+                        wait_time = 2 ** attempt + random.uniform(0.3, 1.0)
+                        print(f"403被拦截，{wait_time}秒后重试 ({attempt + 1}/{max_retries}): {url}")
+                        self._refresh_session()
+                        time.sleep(wait_time)
+                        continue
+                    print(f"请求失败 {url}: 403 Forbidden")
+                    return None
+                if status_code >= 400:
+                    raise requests.exceptions.HTTPError(f"{status_code} Error: {url}")
+                return text
+            except requests.exceptions.SSLError:
+                if verify:
+                    return self._get(url, referer, verify=False, max_retries=max_retries)
+                print(f"SSL错误 {url}")
+                return None
+            except requests.exceptions.RequestException as exc:
+                print(f"请求失败 {url}: {exc}")
+                return None
+        return None
 
-            if not items:
-                break
-
-            for item in items:
-                detail_url = item.get("siteask_m") or item.get("site_url") or ""
-                detail_url = str(detail_url).strip()
-                if not detail_url.startswith("http"):
-                    detail_url = list_url
-
-                phone = normalize_phone(item.get("mobile", ""))
-                profile = {
-                    "uid": str(item.get("uid") or ""),
-                    "name": str(item.get("username") or "").strip(),
-                    "law_firm": str(item.get("lawyer_lawroom") or "").strip(),
-                    "phone": phone,
-                    "lawyer_year": item.get("lawyer_year"),
-                    "service_area": str(item.get("service_area") or "").strip(),
-                    "address": str(item.get("addr") or "").strip(),
-                    "specialties": item.get("professionArr") or [],
-                    "answer_count": item.get("ansnum"),
-                    "comment_count": item.get("askcommentnum"),
-                }
-
-                now = int(time.time())
-                uid = profile.get("uid", "")
-                record_key = uid or detail_url
-                record_id = hashlib.md5(record_key.encode("utf-8")).hexdigest()
-
-                area = item.get("areaInfo", {}) or {}
-                yield {
-                    "record_id": record_id,
-                    "collected_at": now,
-                    "source": {
-                        "site": SITE_NAME,
-                        "list_url": list_url,
-                        "detail_url": detail_url,
-                        "province": str(area.get("province") or target.province_name),
-                        "province_py": target.province_py,
-                        "city": str(area.get("city") or target.city_name),
-                        "city_py": target.city_py,
-                        "page": page,
-                    },
-                    "list_snapshot": {
-                        "uid": uid,
-                        "name": profile["name"],
-                        "law_firm": profile["law_firm"],
-                        "answer_count": profile["answer_count"],
-                        "comment_count": profile["comment_count"],
-                    },
-                    "profile": profile,
-                    "raw": item,
-                }
-                if self.sleep_seconds:
-                    time.sleep(self.sleep_seconds)
-
-            if not has_more:
-                break
-
-    def _to_legacy_lawyer_row(self, record: Dict) -> Optional[Dict[str, str]]:
-        source = record.get("source", {}) or {}
-        profile = record.get("profile", {}) or {}
-        phone = normalize_phone(profile.get("phone", ""))
-        if not phone:
-            return None
-
-        province = (source.get("province") or "").strip()
-        city = (source.get("city") or province).strip()
-        return {
-            "name": (profile.get("name") or "").strip(),
-            "law_firm": (profile.get("law_firm") or "").strip(),
-            "province": province,
-            "city": city,
-            "phone": phone,
-            "url": (source.get("detail_url") or source.get("list_url") or "").strip(),
-            "domain": LEGACY_DOMAIN,
-            "create_time": int(record.get("collected_at") or time.time()),
-            "params": json.dumps(record, ensure_ascii=False),
-        }
-
-    def _existing_phones_in_db(self, phones: List[str]) -> Set[str]:
-        if not self.db or not phones:
+    def _existing_phones(self, phones: List[str]) -> Set[str]:
+        if not phones:
             return set()
-        deduped = sorted({p for p in phones if p})
-        if not deduped:
-            return set()
-
         existing: Set[str] = set()
         cur = self.db.db.cursor()
         try:
             chunk_size = 500
-            for i in range(0, len(deduped), chunk_size):
-                chunk = deduped[i:i + chunk_size]
+            for i in range(0, len(phones), chunk_size):
+                chunk = phones[i:i + chunk_size]
                 placeholders = ",".join(["%s"] * len(chunk))
                 sql = f"SELECT phone FROM lawyer WHERE domain=%s AND phone IN ({placeholders})"
-                cur.execute(sql, [LEGACY_DOMAIN, *chunk])
+                cur.execute(sql, [DOMAIN, *chunk])
                 for row in cur.fetchall():
                     existing.add(row[0])
         finally:
             cur.close()
         return existing
 
-    def _write_records_to_db(self, records: List[Dict]) -> Tuple[int, int]:
-        if not self.db:
-            return 0, 0
-
-        rows: List[Dict[str, str]] = []
-        for record in records:
-            row = self._to_legacy_lawyer_row(record)
-            if row:
-                rows.append(row)
-        if not rows:
-            return 0, 0
-
-        existing = self._existing_phones_in_db([row["phone"] for row in rows])
-        inserted = 0
-        skipped = 0
-        for row in rows:
-            phone = row.get("phone", "")
-            if not phone or phone in existing:
-                skipped += 1
-                continue
+    def _load_cities(self):
+        condition = "domain='findlaw' AND level=2"
+        tables = ("area_new", "area2", "area")
+        last_error = None
+        for table in tables:
             try:
-                self.db.insert_data("lawyer", row)
-                existing.add(phone)
-                inserted += 1
+                rows = self.db.select_data(table, "city, province, pinyin", condition) or []
             except Exception as exc:
-                skipped += 1
-                print(f"[db] 插入失败 phone={phone} url={row.get('url', '')}: {exc}")
-        return inserted, skipped
+                last_error = exc
+                continue
+            if rows:
+                missing_pinyin = sum(1 for r in rows if not (r.get("pinyin") or "").strip())
+                print(f"[找法网] 城市来源表: {table}, 城市数: {len(rows)}, 缺少pinyin: {missing_pinyin}")
+                return rows
 
-    def crawl(
-        self,
-        output_path: str,
-        max_cities: int = 0,
-        city_filter: Optional[str] = None,
-    ) -> None:
-        cities = self.discover_cities()
-        print(f"[discover] 共发现城市 {len(cities)} 个")
-        if city_filter:
-            key = city_filter.strip().lower()
-            cities = [c for c in cities if key in c.city_py.lower() or key in c.city_name.lower()]
-            print(f"[discover] 过滤后城市 {len(cities)} 个, filter={city_filter}")
-        if max_cities > 0:
-            cities = cities[:max_cities]
-            print(f"[discover] 截断城市数 {len(cities)}")
+        if last_error:
+            print(f"[找法网] 加载地区数据失败: {last_error}")
+        print("[找法网] 无城市数据（已尝试 area_new/area2/area）")
+        for table in tables:
+            try:
+                cnt = self.db.select_data(table, "COUNT(*) AS cnt", condition)
+                c = (cnt[0].get("cnt") if cnt else 0) if isinstance(cnt, (list, tuple)) else 0
+                print(f"[找法网] 校验: {table} 满足条件记录数: {c}")
+            except Exception:
+                pass
+        return []
 
-        os.makedirs(os.path.dirname(output_path) or ".", exist_ok=True)
+    def _fetch_page(self, url: str, referer: str) -> List[Dict]:
+        text = self._get(url, referer, verify=True)
+        if not text:
+            return []
 
-        seen_ids: Set[str] = set()
-        if os.path.exists(output_path):
-            with open(output_path, "r", encoding="utf-8") as old_file:
-                for line in old_file:
-                    line = line.strip()
-                    if not line:
+        try:
+            # 某些返回体前会携带 BOM 或包装脚本，此处做兼容
+            text = text.strip().lstrip("\ufeff")
+            try:
+                data = json.loads(text)
+            except ValueError:
+                json_start = text.find('{')
+                json_end = text.rfind('}')
+                if json_start == -1 or json_end == -1:
+                    print(f"解析JSON失败 {url}: 返回内容开头: {text[:80]!r}")
+                    return []
+                cleaned = text[json_start:json_end + 1]
+                data = json.loads(cleaned)
+            if isinstance(data, str):
+                try:
+                    data = json.loads(data)
+                except ValueError:
+                    print(f"解析JSON失败 {url}: 二次解析仍为字符串，开头: {str(data)[:80]!r}")
+                    return []
+        except ValueError as exc:
+            print(f"解析JSON失败 {url}: {exc}")
+            return []
+
+        items = data.get("data", {}).get("lawyer_list", [])
+        parsed = []
+        for item in items:
+            phone = (item.get("mobile") or "").replace("-", "")
+            parsed.append({
+                "name": item.get("username", ""),
+                "law_firm": item.get("lawyer_lawroom", ""),
+                "province": item.get("areaInfo", {}).get("province", ""),
+                "city": item.get("areaInfo", {}).get("city", ""),
+                "phone": phone,
+                "url": url,
+                "domain": DOMAIN,
+                "create_time": int(time.time()),
+                "params": json.dumps(item, ensure_ascii=False)
+            })
+        return parsed
+
+    def run(self):
+        print("启动找法网采集...")
+        if not self.cities:
+            print("无城市数据")
+            return
+
+        for city in self.cities:
+            pinyin = city.get("pinyin")
+            province = city.get("province", "")
+            city_name = city.get("city", "")
+            if not pinyin:
+                continue
+            print(f"采集 {province}-{city_name}")
+            page = 1
+            while True:
+                url = LIST_TEMPLATE.format(pinyin=pinyin, page=page)
+                referer = f"https://m.findlaw.cn/{pinyin}/q_lawyer/"
+                print(f"  第 {page} 页: {url}")
+                items = self._fetch_page(url, referer)
+                if not items:
+                    break
+
+                phones = [it.get("phone") for it in items if (it.get("phone") or "").strip()]
+                existing = self._existing_phones(phones)
+
+                for entry in items:
+                    phone = entry.get("phone")
+                    if not phone:
+                        continue
+                    if phone in existing:
+                        print(f"    -- 已存在: {entry['name']} ({phone})")
                         continue
                     try:
-                        item = json.loads(line)
-                    except Exception:
-                        continue
-                    rid = item.get("record_id")
-                    if rid:
-                        seen_ids.add(rid)
-            print(f"[resume] 已有记录 {len(seen_ids)} 条")
+                        self.db.insert_data("lawyer", entry)
+                        print(f"    -> 新增: {entry['name']} ({phone})")
+                    except Exception as exc:
+                        print(f"    插入失败: {exc}")
 
-        total_new_json = 0
-        total_new_db = 0
-        total_skip_db = 0
+                page += 1
 
-        with open(output_path, "a", encoding="utf-8") as out:
-            for idx, target in enumerate(cities, start=1):
-                print(
-                    f"[city {idx}/{len(cities)}] {target.province_name}-{target.city_name} "
-                    f"({target.city_py})"
-                )
-                city_records = list(self.crawl_city(target))
-
-                city_new_json = 0
-                for record in city_records:
-                    rid = record["record_id"]
-                    if rid in seen_ids:
-                        continue
-                    out.write(json.dumps(record, ensure_ascii=False) + "\n")
-                    seen_ids.add(rid)
-                    city_new_json += 1
-                    total_new_json += 1
-
-                city_new_db, city_skip_db = self._write_records_to_db(city_records)
-                total_new_db += city_new_db
-                total_skip_db += city_skip_db
-                print(
-                    f"[city] 采集{len(city_records)}条, JSON新增{city_new_json}条, "
-                    f"DB新增{city_new_db}条, DB跳过{city_skip_db}条"
-                )
-
-        print(
-            f"[done] JSON新增{total_new_json}条, DB新增{total_new_db}条, "
-            f"DB跳过{total_skip_db}条, 输出: {output_path}"
-        )
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description="找法网全新采集脚本（重写版）")
-    parser.add_argument(
-        "--output",
-        default="/www/wwwroot/lawyers/data/findlaw_records_all.jsonl",
-        help="输出 jsonl 文件路径",
-    )
-    parser.add_argument(
-        "--max-cities",
-        type=int,
-        default=0,
-        help="最多采集多少个城市，0 表示不限",
-    )
-    parser.add_argument(
-        "--max-pages",
-        type=int,
-        default=9999,
-        help="每个城市最多采集多少页",
-    )
-    parser.add_argument(
-        "--city-filter",
-        default="",
-        help="按城市拼音或城市名过滤，如 beijing",
-    )
-    parser.add_argument(
-        "--sleep",
-        type=float,
-        default=0.1,
-        help="每条记录采集间隔秒数",
-    )
-    parser.add_argument(
-        "--direct",
-        action="store_true",
-        help="直连模式，不使用 proxy_settings.json 代理",
-    )
-    parser.add_argument(
-        "--no-db",
-        action="store_true",
-        help="只输出 JSONL，不写入数据库",
-    )
-    return parser.parse_args()
-
-
-def main():
-    args = parse_args()
-    if args.no_db:
-        crawler = FindlawCrawler(
-            max_pages=args.max_pages,
-            sleep_seconds=args.sleep,
-            use_proxy=not args.direct,
-            db_connection=None,
-        )
-        crawler.crawl(
-            output_path=args.output,
-            max_cities=args.max_cities,
-            city_filter=args.city_filter or None,
-        )
-        return
-
-    with Db() as db:
-        crawler = FindlawCrawler(
-            max_pages=args.max_pages,
-            sleep_seconds=args.sleep,
-            use_proxy=not args.direct,
-            db_connection=db,
-        )
-        crawler.crawl(
-            output_path=args.output,
-            max_cities=args.max_cities,
-            city_filter=args.city_filter or None,
-        )
+        print("找法网采集完成")
 
 
 if __name__ == "__main__":
-    main()
+    with Db() as db:
+        spider = FindlawSpider(db)
+        spider.run()

common_sites/hualv.py

@@ -1,18 +1,10 @@
-import argparse
-import ast
-import hashlib
 import json
 import os
-import random
 import re
 import sys
 import time
-from dataclasses import dataclass
-from typing import Dict, Iterable, List, Optional, Set, Tuple
-from urllib.parse import urljoin
-
-import urllib3
-from bs4 import BeautifulSoup
+import random
+from typing import Dict, Optional
 
 current_dir = os.path.dirname(os.path.abspath(__file__))
 project_root = os.path.dirname(current_dir)
@@ -22,638 +14,331 @@ if request_dir not in sys.path:
 if project_root not in sys.path:
     sys.path.append(project_root)
 
+import requests
+from bs4 import BeautifulSoup
+from request.proxy_config import get_proxies, report_proxy_status
+
 from Db import Db
-from request.requests_client import RequestClientError, RequestsClient
-from utils.rate_limiter import wait_for_request
+from config import HEADERS
+from utils.rate_limiter import request_slot
 
-urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
-
-SITE_NAME = "hualv"
-LEGACY_DOMAIN = "华律"
-SITE_BASE = "https://m.66law.cn"
-CITY_DATA_URL = "https://cache.66law.cn/dist/main-v2.0.js"
-LIST_API_URL = "https://m.66law.cn/findlawyer/rpc/loadlawyerlist/"
-
-PHONE_RE = re.compile(r"1[3-9]\d{9}")
-EMAIL_RE = re.compile(r"[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Za-z]{2,}")
-YEAR_RE = re.compile(r"执业\s*(\d+)\s*年")
+LIST_URL = "https://m.66law.cn/findlawyer/rpc/loadlawyerlist/"
+DOMAIN = "华律"
 
 
-@dataclass
-class CityTarget:
-    province_id: int
-    province_name: str
-    city_id: int
-    city_name: str
-
-
-def normalize_phone(text: str) -> str:
-    compact = re.sub(r"\D", "", text or "")
-    match = PHONE_RE.search(compact)
-    return match.group(0) if match else ""
-
-
-def strip_html_tags(text: str) -> str:
-    return re.sub(r"<[^>]+>", "", text or "").strip()
-
-
-class HualvCrawler:
-    def __init__(
-        self,
-        max_pages: int = 9999,
-        sleep_seconds: float = 0.15,
-        use_proxy: bool = True,
-        db_connection=None,
-    ):
-        self.max_pages = max_pages
-        self.sleep_seconds = max(0.0, sleep_seconds)
+class HualvSpider:
+    def __init__(self, db_connection):
         self.db = db_connection
-        self.client = RequestsClient(
-            headers={
-                "User-Agent": (
-                    "Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) "
-                    "AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 "
-                    "Mobile/15E148 Safari/604.1"
-                ),
-                "Accept": "application/json, text/javascript, */*; q=0.01",
-                "X-Requested-With": "XMLHttpRequest",
-                "Connection": "close",
-            },
-            use_proxy=use_proxy,
-            retry_total=2,
-            retry_backoff_factor=1,
-            retry_status_forcelist=(429, 500, 502, 503, 504),
-            retry_allowed_methods=("GET", "POST"),
+        self.session = self._build_session()
+        self.areas = self._load_areas()
+
+    def _build_session(self) -> requests.Session:
+        report_proxy_status()
+        session = requests.Session()
+        session.trust_env = False
+        proxies = get_proxies()
+        if proxies:
+            session.proxies.update(proxies)
+        else:
+            session.proxies.clear()
+        custom_headers = HEADERS.copy()
+        custom_headers['User-Agent'] = (
+            'Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) '
+            'AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 '
+            'Mobile/15E148 Safari/604.1'
         )
+        custom_headers["Connection"] = "close"
+        session.headers.update(custom_headers)
+        return session
 
-    def _request_text(
-        self,
-        method: str,
-        url: str,
-        *,
-        timeout: int = 20,
-        max_retries: int = 3,
-        referer: str = SITE_BASE,
-        data: Optional[Dict] = None,
-    ) -> str:
-        headers = {"Referer": referer}
-        last_error: Optional[Exception] = None
+    def _refresh_session(self) -> None:
+        try:
+            self.session.close()
+        except Exception:
+            pass
+        self.session = self._build_session()
 
-        for attempt in range(max_retries):
-            wait_for_request()
+    def _load_areas(self):
+        tables = ("area_new", "area2", "area")
+        last_error = None
+        for table in tables:
             try:
-                if method.upper() == "POST":
-                    resp = self.client.post_text(
-                        url,
-                        timeout=timeout,
-                        verify=False,
-                        headers=headers,
-                        data=data,
-                    )
-                else:
-                    resp = self.client.get_text(
-                        url,
-                        timeout=timeout,
-                        verify=False,
-                        headers=headers,
-                    )
-
-                code = resp.status_code
-                if code == 403:
-                    if attempt < max_retries - 1:
-                        self.client.refresh()
-                        time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                        continue
-                    raise RequestClientError(f"{code} Error: {url}")
-                if code >= 500 and attempt < max_retries - 1:
-                    time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                    continue
-                if code >= 400:
-                    raise RequestClientError(f"{code} Error: {url}")
-                return resp.text
+                provinces = self.db.select_data(
+                    table,
+                    "code, province, pinyin, id",
+                    "domain='66law' AND level=1"
+                ) or []
+                cities = self.db.select_data(
+                    table,
+                    "code, city, province, pid",
+                    "domain='66law' AND level=2"
+                ) or []
             except Exception as exc:
                 last_error = exc
-                if attempt < max_retries - 1:
-                    time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                    continue
-                raise
-
-        if last_error is not None:
-            raise last_error
-        raise RequestClientError(f"Unknown request error: {url}")
-
-    def _get_text(self, url: str, *, timeout: int = 20, max_retries: int = 3, referer: str = SITE_BASE) -> str:
-        return self._request_text(
-            "GET",
-            url,
-            timeout=timeout,
-            max_retries=max_retries,
-            referer=referer,
-        )
-
-    def _post_text(
-        self,
-        url: str,
-        *,
-        data: Dict,
-        timeout: int = 20,
-        max_retries: int = 3,
-        referer: str = SITE_BASE,
-    ) -> str:
-        return self._request_text(
-            "POST",
-            url,
-            timeout=timeout,
-            max_retries=max_retries,
-            referer=referer,
-            data=data,
-        )
-
-    def _extract_spc_location(self, script_text: str) -> List:
-        # main-v2.js 内置了 sPCLocation=new Array(...)，后面紧跟 cateinfo 数组
-        marker = "sPCLocation = new Array("
-        start = script_text.find(marker)
-        if start == -1:
-            marker = "sPCLocation=new Array("
-            start = script_text.find(marker)
-        if start == -1:
-            return []
-        start += len(marker)
-
-        next_marker = script_text.find("cateinfo = new Array(", start)
-        if next_marker == -1:
-            next_marker = script_text.find("cateinfo=new Array(", start)
-
-        if next_marker != -1:
-            end = script_text.rfind(");", start, next_marker)
-        else:
-            end = script_text.find(");", start)
-
-        if end == -1 or end <= start:
-            return []
-
-        raw = "[" + script_text[start:end] + "]"
-        try:
-            data = ast.literal_eval(raw)
-        except Exception:
-            return []
-        return data if isinstance(data, list) else []
-
-    def discover_cities(self) -> List[CityTarget]:
-        script_text = self._get_text(CITY_DATA_URL, referer=SITE_BASE + "/findlawyer/")
-        rows = self._extract_spc_location(script_text)
-
-        targets: List[CityTarget] = []
-        seen: Set[Tuple[int, int]] = set()
-
-        for province in rows:
-            if not isinstance(province, list) or len(province) < 3:
                 continue
-            try:
-                province_id = int(province[0])
-            except Exception:
+
+            if not cities:
                 continue
-            province_name = str(province[1] or "").strip()
-            city_rows = province[2] if isinstance(province[2], list) else []
 
-            for city in city_rows:
-                if not isinstance(city, list) or len(city) < 2:
-                    continue
-                try:
-                    city_id = int(city[0])
-                except Exception:
-                    continue
-                city_name = str(city[1] or "").strip()
-                if city_id <= 0 or not city_name:
-                    continue
-
-                key = (province_id, city_id)
-                if key in seen:
-                    continue
-                seen.add(key)
-
-                targets.append(
-                    CityTarget(
-                        province_id=province_id,
-                        province_name=province_name,
-                        city_id=city_id,
-                        city_name=city_name,
-                    )
-                )
-        return targets
-
-    def fetch_list_page(self, target: CityTarget, page: int) -> Tuple[List[Dict], int]:
-        payload = {
-            "pid": str(target.province_id),
-            "cid": str(target.city_id),
-            "page": str(page),
-        }
-        text = self._post_text(
-            LIST_API_URL,
-            data=payload,
-            referer=SITE_BASE + "/findlawyer/",
-        )
-        data = json.loads((text or "").strip().lstrip("\ufeff") or "{}")
-        items = data.get("lawyerList") or data.get("queryLawyerList") or []
-        if not isinstance(items, list):
-            items = []
-
-        page_count = 0
-        try:
-            page_count = int((data.get("lawyerItems") or {}).get("pageCount") or 0)
-        except Exception:
-            page_count = 0
-        return items, page_count
-
-    def parse_detail(self, detail_url: str) -> Dict:
-        contact_url = detail_url.rstrip("/") + "/lawyer_contact.aspx"
-        html = self._get_text(contact_url, referer=detail_url)
-        soup = BeautifulSoup(html, "html.parser")
-        full_text = soup.get_text(" ", strip=True)
-
-        name = ""
-        law_firm = ""
-        phone = ""
-        email = ""
-        address = ""
-        license_no = ""
-        practice_years: Optional[int] = None
-
-        name_tag = soup.select_one(".logo-box .title b")
-        if name_tag:
-            name = name_tag.get_text(strip=True).replace("律师", "").strip()
-        if not name and soup.title:
-            match = re.search(r"([^\s,，。_]+?)律师", soup.title.get_text(" ", strip=True))
-            if match:
-                name = match.group(1).strip()
-
-        phone_candidates = [
-            soup.select_one(".logo-box .r-bar .tel").get_text(" ", strip=True)
-            if soup.select_one(".logo-box .r-bar .tel")
-            else "",
-            soup.select_one(".lawyer-show ul.info").get_text(" ", strip=True)
-            if soup.select_one(".lawyer-show ul.info")
-            else "",
-            full_text,
-        ]
-        for candidate in phone_candidates:
-            phone = normalize_phone(candidate)
-            if phone:
-                break
-
-        for li in soup.select(".lawyer-show ul.info li"):
-            li_text = li.get_text(" ", strip=True)
-            if ("事务所" in li_text or "法律服务所" in li_text) and not law_firm:
-                law_firm = li_text
-
-        if not law_firm:
-            match = re.search(r'"affiliation":\{"@type":"Organization","name":"([^"]+)"', html)
-            if match:
-                law_firm = match.group(1).strip()
-
-        match = re.search(r'"identifier":"([^"]+)"', html)
-        if match:
-            license_no = match.group(1).strip()
-
-        match = re.search(r'"streetAddress":"([^"]+)"', html)
-        if match:
-            address = match.group(1).strip()
-
-        email_match = EMAIL_RE.search(html)
-        if email_match:
-            email = email_match.group(0).strip()
-
-        year_match = YEAR_RE.search(full_text)
-        if year_match:
-            try:
-                practice_years = int(year_match.group(1))
-            except Exception:
-                practice_years = None
-
-        specialties = [node.get_text(strip=True) for node in soup.select(".tag-h38 span")]
-        specialties = [x for x in specialties if x]
-
-        return {
-            "name": name,
-            "law_firm": law_firm,
-            "phone": phone,
-            "email": email,
-            "address": address,
-            "license_no": license_no,
-            "practice_years": practice_years,
-            "specialties": specialties,
-            "detail_url": detail_url,
-            "contact_url": contact_url,
-        }
-
-    def crawl_city(self, target: CityTarget) -> Iterable[Dict]:
-        seen_details: Set[str] = set()
-
-        for page in range(1, self.max_pages + 1):
-            try:
-                items, page_count = self.fetch_list_page(target, page)
-            except Exception as exc:
-                print(f"[list] 失败 pid={target.province_id} cid={target.city_id} p{page}: {exc}")
-                break
-
-            if not items:
-                break
-
-            for item in items:
-                detail_url = str(item.get("lawyerUrl") or "").strip()
-                if not detail_url:
-                    continue
-                if detail_url.startswith("//"):
-                    detail_url = "https:" + detail_url
-                if not detail_url.startswith("http"):
-                    detail_url = urljoin(SITE_BASE, detail_url)
-
-                if detail_url in seen_details:
-                    continue
-                seen_details.add(detail_url)
-
-                try:
-                    detail = self.parse_detail(detail_url)
-                except Exception as exc:
-                    print(f"[detail] 失败 {detail_url}: {exc}")
-                    continue
-
-                now = int(time.time())
-                uid = str(item.get("lawyerId") or item.get("globalUserId") or detail_url)
-                record_id = hashlib.md5(uid.encode("utf-8")).hexdigest()
-
-                list_name = str(item.get("name") or "").replace("律师", "").strip()
-                category_text = str(item.get("categoryNames") or "").strip()
-                category_arr = [x.strip() for x in re.split(r"[、,，]", category_text) if x.strip()]
-
-                yield {
-                    "record_id": record_id,
-                    "collected_at": now,
-                    "source": {
-                        "site": SITE_NAME,
-                        "province_id": target.province_id,
-                        "province": target.province_name,
-                        "city_id": target.city_id,
-                        "city": target.city_name,
-                        "page": page,
-                        "detail_url": detail_url,
-                        "contact_url": detail.get("contact_url", ""),
-                    },
-                    "list_snapshot": {
-                        "lawyer_id": item.get("lawyerId"),
-                        "name": list_name,
-                        "category_names": category_arr,
-                        "help_count": strip_html_tags(str(item.get("helpCount") or "")),
-                        "comment_score": strip_html_tags(str(item.get("commentScore") or "")),
-                        "response_time": str(item.get("responseTime") or "").strip(),
-                        "year": item.get("year"),
-                        "is_adv": bool(item.get("isAdv")),
-                    },
-                    "profile": {
-                        "name": detail.get("name") or list_name,
-                        "law_firm": detail.get("law_firm") or "",
-                        "phone": detail.get("phone") or "",
-                        "email": detail.get("email") or "",
-                        "address": detail.get("address") or "",
-                        "license_no": detail.get("license_no") or "",
-                        "practice_years": detail.get("practice_years"),
-                        "specialties": detail.get("specialties") or category_arr,
-                    },
-                    "raw": item,
+            province_map = {p.get('id'): {"code": p.get('code'), "name": p.get('province')} for p in provinces}
+            city_map = {}
+            for city in cities:
+                province_info = province_map.get(city.get('pid'), {}) or {}
+                province_code = province_info.get('code')
+                city_map[city.get('code')] = {
+                    "name": city.get('city'),
+                    "province": city.get('province'),
+                    "province_code": province_code,
                 }
+            print(f"[华律] 城市来源表: {table}, 城市数: {len(cities)}")
+            return city_map
 
-                if self.sleep_seconds:
-                    time.sleep(self.sleep_seconds)
+        if last_error:
+            print(f"[华律] 加载地区数据失败: {last_error}")
+        print("[华律] 无城市数据（已尝试 area_new/area2/area）")
+        return {}
 
-            if page_count > 0 and page >= page_count:
-                break
+    def _post(self, data: Dict[str, str], max_retries: int = 3) -> Optional[Dict]:
+        for attempt in range(max_retries):
+            try:
+                with request_slot():
+                    resp = self.session.post(LIST_URL, data=data, timeout=20, verify=False)
+                status_code = resp.status_code
+                text = resp.text
+                resp.close()
+                if status_code == 403:
+                    if attempt < max_retries - 1:
+                        wait_time = 2 ** attempt + random.uniform(0.3, 1.0)
+                        print(f"403被拦截，{wait_time}秒后重试 ({attempt + 1}/{max_retries})")
+                        self._refresh_session()
+                        time.sleep(wait_time)
+                        continue
+                    print("请求失败: 403 Forbidden")
+                    return None
+                if status_code >= 400:
+                    raise requests.exceptions.HTTPError(f"{status_code} Error")
+                try:
+                    return json.loads(text)
+                except ValueError as exc:
+                    print(f"解析JSON失败: {exc}")
+                    return None
+            except requests.exceptions.RequestException as exc:
+                print(f"请求失败: {exc}")
+                return None
+        return None
 
-    def _to_legacy_lawyer_row(self, record: Dict) -> Optional[Dict[str, str]]:
-        source = record.get("source", {}) or {}
-        profile = record.get("profile", {}) or {}
+    def _parse_detail(self, url: str, province: str, city: str) -> Optional[Dict[str, str]]:
+        contact_url = f"{url}lawyer_contact.aspx"
+        print(f"  详情: {contact_url}")
+        existing = self.db.select_data(
+            "lawyer",
+            "id, avatar_url",
+            f"domain='{DOMAIN}' AND url='{contact_url}'"
+        )
+        existing_id = None
+        if existing:
+            existing_id = existing[0].get("id")
+            avatar = (existing[0].get("avatar_url") or "").strip()
+            if avatar:
+                print("    -- 已存在且头像已补全，跳过")
+                return None
 
-        phone = normalize_phone(profile.get("phone", ""))
-        if not phone:
+        html = self._get_detail(contact_url)
+        if not html:
             return None
 
-        province = (source.get("province") or "").strip()
-        city = (source.get("city") or province).strip()
-        return {
-            "name": (profile.get("name") or "").strip(),
-            "law_firm": (profile.get("law_firm") or "").strip(),
+        soup = BeautifulSoup(html, "html.parser")
+        info_list = soup.find("ul", class_="information-list")
+        if not info_list:
+            return None
+
+        phone = ""
+        law_firm = ""
+        for li in info_list.find_all("li"):
+            text = li.get_text(strip=True)
+            if "手机号" in text:
+                cleaned = text.replace("手机号", "").replace("(咨询请说明来自 华律网)", "").strip()
+                match = re.search(r"1\d{10}", cleaned.replace('-', '').replace(' ', ''))
+                if match:
+                    phone = match.group(0)
+            if "执业单位" in text:
+                law_firm = text.replace("执业单位", "").strip()
+
+        name = ""
+        breadcrumb = soup.find("div", class_="weizhi")
+        if breadcrumb:
+            links = breadcrumb.find_all("a")
+            if len(links) > 2:
+                name = links[2].get_text(strip=True)
+
+        phone = phone.replace('-', '').strip()
+        if not phone or not re.fullmatch(r"1\d{10}", phone):
+            print("    无手机号，跳过")
+            return None
+
+        avatar_url, site_time = self._extract_avatar_and_time(soup)
+        data = {
+            "phone": phone,
             "province": province,
             "city": city,
-            "phone": phone,
-            "url": (source.get("contact_url") or source.get("detail_url") or "").strip(),
-            "domain": LEGACY_DOMAIN,
-            "create_time": int(record.get("collected_at") or time.time()),
-            "params": json.dumps(record, ensure_ascii=False),
+            "law_firm": law_firm,
+            "url": contact_url,
+            "avatar_url": avatar_url,
+            "create_time": int(time.time()),
+            "site_time": site_time,
+            "domain": DOMAIN,
+            "name": name,
+            "params": json.dumps({"source": url}, ensure_ascii=False)
         }
-
-    def _existing_phones_in_db(self, phones: List[str]) -> Set[str]:
-        if not self.db or not phones:
-            return set()
-
-        deduped = sorted({p for p in phones if p})
-        if not deduped:
-            return set()
-
-        existing: Set[str] = set()
-        cur = self.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 = f"SELECT phone FROM lawyer WHERE domain=%s AND phone IN ({placeholders})"
-                cur.execute(sql, [LEGACY_DOMAIN, *chunk])
-                for row in cur.fetchall():
-                    existing.add(row[0])
-        finally:
-            cur.close()
-
-        return existing
-
-    def _write_records_to_db(self, records: List[Dict]) -> Tuple[int, int]:
-        if not self.db:
-            return 0, 0
-
-        rows: List[Dict[str, str]] = []
-        for record in records:
-            row = self._to_legacy_lawyer_row(record)
-            if row:
-                rows.append(row)
-        if not rows:
-            return 0, 0
-
-        existing = self._existing_phones_in_db([row["phone"] for row in rows])
-        inserted = 0
-        skipped = 0
-
-        for row in rows:
-            phone = row.get("phone", "")
-            if not phone or phone in existing:
-                skipped += 1
-                continue
+        if existing_id:
+            update_data = {
+                "avatar_url": avatar_url,
+                "site_time": site_time,
+            }
+            if name:
+                update_data["name"] = name
+            if law_firm:
+                update_data["law_firm"] = law_firm
+            if province:
+                update_data["province"] = province
+            if city:
+                update_data["city"] = city
+            if phone:
+                update_data["phone"] = phone
+            update_data["params"] = json.dumps({"source": url}, ensure_ascii=False)
             try:
-                self.db.insert_data("lawyer", row)
-                existing.add(phone)
-                inserted += 1
+                self.db.update_data("lawyer", update_data, f"id={existing_id}")
+                print("    -- 已存在，已补全头像/时间")
             except Exception as exc:
-                skipped += 1
-                print(f"[db] 插入失败 phone={phone} url={row.get('url', '')}: {exc}")
+                print(f"    更新失败: {exc}")
+            return None
+        # 若手机号已存在，则更新头像/时间，不再插入新记录
+        existing_phone = self.db.select_data(
+            "lawyer",
+            "id, avatar_url, url",
+            f"domain='{DOMAIN}' AND phone='{phone}'"
+        )
+        if existing_phone:
+            existing_row = existing_phone[0]
+            avatar = (existing_row.get("avatar_url") or "").strip()
+            if avatar:
+                print("    -- 已存在手机号且头像已补全，跳过")
+                return None
+            update_data = {
+                "avatar_url": avatar_url,
+                "site_time": site_time,
+            }
+            if name:
+                update_data["name"] = name
+            if law_firm:
+                update_data["law_firm"] = law_firm
+            if province:
+                update_data["province"] = province
+            if city:
+                update_data["city"] = city
+            if phone:
+                update_data["phone"] = phone
+            if not existing_row.get("url"):
+                update_data["url"] = contact_url
+            update_data["params"] = json.dumps({"source": url}, ensure_ascii=False)
+            try:
+                self.db.update_data("lawyer", update_data, f"id={existing_row.get('id')}")
+                print("    -- 已存在手机号，已补全头像/时间")
+            except Exception as exc:
+                print(f"    更新失败: {exc}")
+            return None
+        return data
 
-        return inserted, skipped
+    def _extract_avatar_and_time(self, soup: BeautifulSoup) -> (str, Optional[int]):
+        avatar_url = ""
+        site_time = None
+        img_tag = soup.select_one(
+            "div.fixed-bottom-bar div.contact-lawye a.lr-photo img"
+        )
+        if img_tag:
+            src = (img_tag.get("src") or "").strip()
+            if src:
+                if src.startswith("//"):
+                    avatar_url = f"https:{src}"
+                else:
+                    avatar_url = src
+                match = re.search(r"/(20\d{2})(\d{2})/", avatar_url)
+                if match:
+                    site_time = int(f"{match.group(1)}{match.group(2)}")
+                else:
+                    match = re.search(r"(20\d{2})(\d{2})\d{2}", avatar_url)
+                    if match:
+                        site_time = int(f"{match.group(1)}{match.group(2)}")
+        return avatar_url, site_time
 
-    def crawl(
-        self,
-        output_path: str,
-        max_cities: int = 0,
-        city_filter: Optional[str] = None,
-    ) -> None:
-        cities = self.discover_cities()
-        print(f"[discover] 共发现城市 {len(cities)} 个")
+    def _get_detail(self, url: str, max_retries: int = 3) -> Optional[str]:
+        for attempt in range(max_retries):
+            try:
+                with request_slot():
+                    resp = self.session.get(url, timeout=15, verify=False)
+                status_code = resp.status_code
+                text = resp.text
+                resp.close()
+                if status_code == 403:
+                    if attempt < max_retries - 1:
+                        wait_time = 2 ** attempt + random.uniform(0.3, 1.0)
+                        print(f"    403被拦截，{wait_time}秒后重试 ({attempt + 1}/{max_retries})")
+                        self._refresh_session()
+                        time.sleep(wait_time)
+                        continue
+                    print("    请求失败: 403 Forbidden")
+                    return None
+                if status_code >= 400:
+                    raise requests.exceptions.HTTPError(f"{status_code} Error")
+                return text
+            except requests.exceptions.RequestException as exc:
+                print(f"    请求失败: {exc}")
+                return None
+        return None
 
-        if city_filter:
-            key = city_filter.strip().lower()
-            cities = [
-                c for c in cities
-                if key in c.city_name.lower() or key in str(c.city_id).lower()
-            ]
-            print(f"[discover] 过滤后城市 {len(cities)} 个, filter={city_filter}")
+    def run(self):
+        print("启动华律网采集...")
+        if not self.areas:
+            print("无城市数据")
+            return
 
-        if max_cities > 0:
-            cities = cities[:max_cities]
-            print(f"[discover] 截断城市数 {len(cities)}")
+        for city_code, city_info in self.areas.items():
+            province_code = city_info.get("province_code")
+            if not province_code:
+                continue
+            province_name = city_info.get("province", "")
+            city_name = city_info.get("name", "")
+            print(f"采集 {province_name}-{city_name}")
 
-        os.makedirs(os.path.dirname(output_path) or ".", exist_ok=True)
+            page = 1
+            while True:
+                payload = {"pid": province_code, "cid": city_code, "page": str(page)}
+                data = self._post(payload)
+                if not data or not data.get("lawyerList"):
+                    break
 
-        seen_ids: Set[str] = set()
-        if os.path.exists(output_path):
-            with open(output_path, "r", encoding="utf-8") as old_file:
-                for line in old_file:
-                    line = line.strip()
-                    if not line:
+                for item in data["lawyerList"]:
+                    result = self._parse_detail(item.get("lawyerUrl", ""), province_name, city_name)
+                    if not result:
                         continue
                     try:
-                        item = json.loads(line)
-                    except Exception:
-                        continue
-                    rid = item.get("record_id")
-                    if rid:
-                        seen_ids.add(rid)
-            print(f"[resume] 已有记录 {len(seen_ids)} 条")
+                        self.db.insert_data("lawyer", result)
+                        print(f"  -> 新增: {result['name']} ({result['phone']})")
+                    except Exception as exc:
+                        print(f"  插入失败: {exc}")
+                    time.sleep(1)
 
-        total_new_json = 0
-        total_new_db = 0
-        total_skip_db = 0
+                page_count = data.get("lawyerItems", {}).get("pageCount", page)
+                if page >= page_count:
+                    break
+                page += 1
+                time.sleep(2)
 
-        with open(output_path, "a", encoding="utf-8") as out:
-            for idx, target in enumerate(cities, start=1):
-                print(
-                    f"[city {idx}/{len(cities)}] {target.province_name}-{target.city_name} "
-                    f"(pid={target.province_id}, cid={target.city_id})"
-                )
-                city_records = list(self.crawl_city(target))
-
-                city_new_json = 0
-                for record in city_records:
-                    rid = record["record_id"]
-                    if rid in seen_ids:
-                        continue
-                    out.write(json.dumps(record, ensure_ascii=False) + "\n")
-                    seen_ids.add(rid)
-                    city_new_json += 1
-                    total_new_json += 1
-
-                city_new_db, city_skip_db = self._write_records_to_db(city_records)
-                total_new_db += city_new_db
-                total_skip_db += city_skip_db
-
-                print(
-                    f"[city] 采集{len(city_records)}条, JSON新增{city_new_json}条, "
-                    f"DB新增{city_new_db}条, DB跳过{city_skip_db}条"
-                )
-
-        print(
-            f"[done] JSON新增{total_new_json}条, DB新增{total_new_db}条, "
-            f"DB跳过{total_skip_db}条, 输出: {output_path}"
-        )
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description="华律网全新采集脚本（站点数据直采）")
-    parser.add_argument(
-        "--output",
-        default="/www/wwwroot/lawyers/data/hualv_records_all.jsonl",
-        help="输出 jsonl 文件路径",
-    )
-    parser.add_argument(
-        "--max-cities",
-        type=int,
-        default=0,
-        help="最多采集多少个城市，0 表示不限",
-    )
-    parser.add_argument(
-        "--max-pages",
-        type=int,
-        default=9999,
-        help="每个城市最多采集多少页",
-    )
-    parser.add_argument(
-        "--city-filter",
-        default="",
-        help="按城市名称或城市编码过滤，如 beijing / 110100",
-    )
-    parser.add_argument(
-        "--sleep",
-        type=float,
-        default=0.15,
-        help="详情页请求间隔秒数",
-    )
-    parser.add_argument(
-        "--direct",
-        action="store_true",
-        help="直连模式，不使用 proxy_settings.json 代理",
-    )
-    parser.add_argument(
-        "--no-db",
-        action="store_true",
-        help="只输出 JSONL，不写入数据库",
-    )
-    return parser.parse_args()
-
-
-def main():
-    args = parse_args()
-
-    if args.no_db:
-        crawler = HualvCrawler(
-            max_pages=args.max_pages,
-            sleep_seconds=args.sleep,
-            use_proxy=not args.direct,
-            db_connection=None,
-        )
-        crawler.crawl(
-            output_path=args.output,
-            max_cities=args.max_cities,
-            city_filter=args.city_filter or None,
-        )
-        return
-
-    with Db() as db:
-        crawler = HualvCrawler(
-            max_pages=args.max_pages,
-            sleep_seconds=args.sleep,
-            use_proxy=not args.direct,
-            db_connection=db,
-        )
-        crawler.crawl(
-            output_path=args.output,
-            max_cities=args.max_cities,
-            city_filter=args.city_filter or None,
-        )
+            time.sleep(1)
+        print("华律网采集完成")
 
 
 if __name__ == "__main__":
-    main()
+    with Db() as db:
+        spider = HualvSpider(db)
+        spider.run()

common_sites/lawtime.py

@@ -1,16 +1,13 @@
-import argparse
-import hashlib
 import json
 import os
-import random
 import re
 import sys
 import time
-from dataclasses import dataclass, field
-from typing import Dict, Iterable, List, Optional, Set, Tuple
-
-import urllib3
-from bs4 import BeautifulSoup
+import random
+from typing import Dict, Optional, List, Set
+from urllib.parse import urljoin
+from concurrent.futures import ThreadPoolExecutor, as_completed
+import threading
 
 current_dir = os.path.dirname(os.path.abspath(__file__))
 project_root = os.path.dirname(current_dir)
@@ -20,628 +17,283 @@ if request_dir not in sys.path:
 if project_root not in sys.path:
     sys.path.append(project_root)
 
-from Db import Db
-from request.requests_client import RequestClientError, RequestsClient
-from utils.rate_limiter import wait_for_request
+import requests
+import urllib3
+from bs4 import BeautifulSoup
+from request.proxy_config import get_proxies, report_proxy_status
 
 urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
 
-SITE_NAME = "lawtime"
-LEGACY_DOMAIN = "法律快车"
-SITE_BASE = "https://www.lawtime.cn"
-PROVINCE_API = "https://www.lawtime.cn/public/stationIndex/getAreaList?areacode=0"
-CITY_API_TEMPLATE = "https://www.lawtime.cn/public/stationIndex/getAreaList?areacode={province_id}"
-LIST_URL_TEMPLATE = SITE_BASE + "/{city_py}/lawyer"
+from Db import Db
+from config import LAWTIME_CONFIG
+from utils.rate_limiter import request_slot
 
-PHONE_RE = re.compile(r"1[3-9]\d{9}")
-YEAR_RE = re.compile(r"执业\s*(\d+)\s*年")
+LIST_BASE = "https://m.lawtime.cn/{pinyin}/lawyer/?page={page}"
+DETAIL_BASE = "https://m.lawtime.cn"
+DOMAIN = "法律快车"
 
 
-@dataclass
-class CityTarget:
-    province_id: str
-    province_name: str
-    province_py: str
-    city_id: str
-    city_name: str
-    city_py: str
-
-
-@dataclass
-class ListCard:
-    detail_url: str
-    name: str
-    phone: str
-    address: str = ""
-    specialties: List[str] = field(default_factory=list)
-    metric_text: str = ""
-
-
-def normalize_phone(text: str) -> str:
-    compact = re.sub(r"\D", "", text or "")
-    match = PHONE_RE.search(compact)
-    return match.group(0) if match else ""
-
-
-class LawtimeCrawler:
-    def __init__(
-        self,
-        max_pages: int = 9999,
-        sleep_seconds: float = 0.1,
-        use_proxy: bool = True,
-        db_connection=None,
-    ):
-        self.max_pages = max_pages
-        self.sleep_seconds = max(0.0, sleep_seconds)
+class LawtimeSpider:
+    def __init__(self, db_connection):
         self.db = db_connection
-        self.client = RequestsClient(
-            headers={
-                "User-Agent": (
-                    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
-                    "AppleWebKit/537.36 (KHTML, like Gecko) "
-                    "Chrome/122.0.0.0 Safari/537.36"
-                ),
-                "Accept": "text/html,application/json,*/*;q=0.8",
-                "Connection": "close",
-            },
-            use_proxy=use_proxy,
-            retry_total=2,
-            retry_backoff_factor=1,
-            retry_status_forcelist=(429, 500, 502, 503, 504),
-            retry_allowed_methods=("GET",),
-        )
+        self.session = self._build_session()
+        self.max_workers = int(os.getenv("SPIDER_WORKERS", "8"))
+        self._tls = threading.local()
 
-    def _get_text(
-        self,
-        url: str,
-        *,
-        timeout: int = 20,
-        max_retries: int = 3,
-        referer: str = SITE_BASE,
-    ) -> str:
-        headers = {"Referer": referer}
-        last_error: Optional[Exception] = None
+    def _build_session(self) -> requests.Session:
+        report_proxy_status()
+        session = requests.Session()
+        session.trust_env = False
+        proxies = get_proxies()
+        if proxies:
+            session.proxies.update(proxies)
+        else:
+            session.proxies.clear()
+        headers = LAWTIME_CONFIG.get("HEADERS", {})
+        if headers:
+            session.headers.update(headers)
+        session.headers.setdefault("Connection", "close")
+        return session
 
-        for attempt in range(max_retries):
-            wait_for_request()
-            try:
-                resp = self.client.get_text(
-                    url,
-                    timeout=timeout,
-                    verify=False,
-                    headers=headers,
-                )
-                code = resp.status_code
-                if code == 403:
-                    if attempt < max_retries - 1:
-                        self.client.refresh()
-                        time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                        continue
-                    raise RequestClientError(f"{code} Error: {url}")
-                if code >= 500 and attempt < max_retries - 1:
-                    time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                    continue
-                if code >= 400:
-                    raise RequestClientError(f"{code} Error: {url}")
-                return resp.text
-            except Exception as exc:
-                last_error = exc
-                if attempt < max_retries - 1:
-                    time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                    continue
-                raise
-
-        if last_error is not None:
-            raise last_error
-        raise RequestClientError(f"Unknown request error: {url}")
-
-    def _get_json(self, url: str, *, referer: str) -> List[Dict]:
-        text = self._get_text(url, referer=referer)
-        cleaned = (text or "").strip().lstrip("\ufeff")
-        if not cleaned or cleaned.startswith("<"):
-            return []
+    def _refresh_session(self) -> None:
         try:
-            data = json.loads(cleaned)
-        except ValueError:
-            return []
-        return data if isinstance(data, list) else []
+            self.session.close()
+        except Exception:
+            pass
+        self.session = self._build_session()
 
-    def discover_cities(self) -> List[CityTarget]:
-        provinces = self._get_json(PROVINCE_API, referer=SITE_BASE)
-        if not provinces:
-            print("[discover] 地区接口未返回有效数据")
-            return []
+    def _get_thread_session(self) -> requests.Session:
+        s = getattr(self._tls, "session", None)
+        if s is not None:
+            return s
+        s = self._build_session()
+        s.headers.update(dict(self.session.headers))
+        self._tls.session = s
+        return s
 
-        results: List[CityTarget] = []
-        seen_py: Set[str] = set()
-
-        for province in provinces:
-            province_id = str(province.get("id") or "").strip()
-            province_name = str(province.get("province") or province.get("city") or "").strip()
-            province_py = str(province.get("pinyin") or "").strip()
-            if not province_id or not province_name:
-                continue
-
-            city_api = CITY_API_TEMPLATE.format(province_id=province_id)
+    def _refresh_thread_session(self) -> None:
+        s = getattr(self._tls, "session", None)
+        if s is not None:
             try:
-                cities = self._get_json(city_api, referer=LIST_URL_TEMPLATE.format(city_py=province_py or ""))
-            except Exception as exc:
-                print(f"[city] 获取失败 province={province_id}: {exc}")
-                continue
-
-            if not cities:
-                cities = [
-                    {
-                        "id": province_id,
-                        "province": province_name,
-                        "city": province_name,
-                        "pinyin": province_py,
-                    }
-                ]
-
-            for city in cities:
-                city_id = str(city.get("id") or "").strip()
-                city_name = str(city.get("city") or city.get("province") or "").strip()
-                city_py = str(city.get("pinyin") or "").strip()
-                if not city_id or not city_name or not city_py:
-                    continue
-                if city_py in seen_py:
-                    continue
-                seen_py.add(city_py)
-
-                results.append(
-                    CityTarget(
-                        province_id=province_id,
-                        province_name=province_name,
-                        province_py=province_py,
-                        city_id=city_id,
-                        city_name=city_name,
-                        city_py=city_py,
-                    )
-                )
-
-        return results
-
-    def _build_list_url(self, city_py: str, page: int) -> str:
-        base = LIST_URL_TEMPLATE.format(city_py=city_py)
-        if page <= 1:
-            return base
-        return f"{base}?page={page}"
-
-    def fetch_list_page(self, target: CityTarget, page: int) -> Tuple[List[ListCard], bool, str]:
-        list_url = self._build_list_url(target.city_py, page)
-        html = self._get_text(list_url, referer=SITE_BASE + "/")
-
-        cards = self.parse_list_cards(html)
-
-        soup = BeautifulSoup(html, "html.parser")
-        next_link = soup.select_one(f"div.page a[href*='page={page + 1}']")
-        has_next = next_link is not None
-
-        return cards, has_next, list_url
-
-    def parse_list_cards(self, html: str) -> List[ListCard]:
-        soup = BeautifulSoup(html, "html.parser")
-        cards: List[ListCard] = []
-        seen: Set[str] = set()
-
-        for item in soup.select("li.lawyer-item-card"):
-            link_tag = item.select_one("a.name[href]") or item.select_one("a.lawyer-img-box[href]")
-            if not link_tag:
-                continue
-            detail_url = (link_tag.get("href") or "").strip()
-            if not detail_url.startswith("http"):
-                continue
-            if detail_url in seen:
-                continue
-            seen.add(detail_url)
-
-            name = link_tag.get_text(strip=True)
-            phone = ""
-            phone_tag = item.select_one("div.phone")
-            if phone_tag:
-                phone = normalize_phone(phone_tag.get_text(" ", strip=True))
-
-            address = ""
-            addr_tag = item.select_one("div.location .txt")
-            if addr_tag:
-                address = addr_tag.get_text(" ", strip=True)
-
-            specialties: List[str] = []
-            prof_tag = item.select_one("div.prof .txt")
-            if prof_tag:
-                specialties = [
-                    x.strip() for x in re.split(r"[、,，]", prof_tag.get_text(" ", strip=True)) if x.strip()
-                ]
-
-            metric_text = ""
-            metric_tag = item.select_one("div.num-msg")
-            if metric_tag:
-                metric_text = metric_tag.get_text(" ", strip=True)
-
-            cards.append(
-                ListCard(
-                    detail_url=detail_url,
-                    name=name,
-                    phone=phone,
-                    address=address,
-                    specialties=specialties,
-                    metric_text=metric_text,
-                )
-            )
-
-        return cards
-
-    def parse_detail(self, detail_url: str) -> Dict:
-        html = self._get_text(detail_url, referer=SITE_BASE)
-        if "网站防火墙" in html or "访问被拒绝" in html or "/ipfilter/verify" in html:
-            raise RequestClientError(f"firewall blocked: {detail_url}")
-
-        soup = BeautifulSoup(html, "html.parser")
-        text = soup.get_text(" ", strip=True)
-
-        name = ""
-        law_firm = ""
-        phone = ""
-        address = ""
-        practice_years: Optional[int] = None
-        specialties: List[str] = []
-
-        if soup.title:
-            title = soup.title.get_text(" ", strip=True)
-            match = re.search(r"([^\s_，,。]+?)律师", title)
-            if match:
-                name = match.group(1).strip()
-
-        phone_candidates = [
-            soup.select_one(".data-w .tel-b b").get_text(" ", strip=True)
-            if soup.select_one(".data-w .tel-b b")
-            else "",
-            soup.select_one(".law-info-b .item .two-r.b").get_text(" ", strip=True)
-            if soup.select_one(".law-info-b .item .two-r.b")
-            else "",
-            text,
-        ]
-        for candidate in phone_candidates:
-            phone = normalize_phone(candidate)
-            if phone:
-                break
-
-        law_firm_tag = soup.select_one(".law-info-b .item .two-nowrap")
-        if law_firm_tag:
-            law_firm = law_firm_tag.get_text(" ", strip=True)
-
-        for li in soup.select(".law-info-b .item"):
-            li_text = li.get_text(" ", strip=True)
-            if ("事务所" in li_text or "法律服务所" in li_text) and not law_firm:
-                law_firm = li_text
-
-        addr_tag = soup.select_one(".law-info-b .item .two-r[title]")
-        if addr_tag:
-            addr_value = (addr_tag.get("title") or "").strip()
-            if len(addr_value) > 8:
-                address = addr_value
-
-        if not address:
-            addr_tag = soup.select_one(".law-info-b .item .two-r")
-            if addr_tag:
-                addr_value = addr_tag.get_text(" ", strip=True)
-                if len(addr_value) > 8 and "律师" not in addr_value:
-                    address = addr_value
-
-        year_match = YEAR_RE.search(text)
-        if year_match:
-            try:
-                practice_years = int(year_match.group(1))
+                s.close()
             except Exception:
-                practice_years = None
+                pass
+        self._tls.session = None
 
-        specialties = [x.get_text(strip=True) for x in soup.select(".profession-b .item") if x.get_text(strip=True)]
-
-        return {
-            "name": name,
-            "law_firm": law_firm,
-            "phone": phone,
-            "address": address,
-            "practice_years": practice_years,
-            "specialties": specialties,
-            "detail_url": detail_url,
-        }
-
-    def crawl_city(self, target: CityTarget) -> Iterable[Dict]:
-        seen_details: Set[str] = set()
-
-        for page in range(1, self.max_pages + 1):
-            try:
-                cards, has_next, list_url = self.fetch_list_page(target, page)
-            except Exception as exc:
-                print(f"[list] 失败 {target.city_py} p{page}: {exc}")
-                break
-
-            if not cards:
-                break
-
-            for card in cards:
-                if card.detail_url in seen_details:
-                    continue
-                seen_details.add(card.detail_url)
-
-                detail: Dict = {}
-                try:
-                    detail = self.parse_detail(card.detail_url)
-                except Exception as exc:
-                    print(f"[detail] 失败 {card.detail_url}: {exc}")
-
-                phone = normalize_phone(detail.get("phone") or card.phone)
-                profile_name = (detail.get("name") or card.name).replace("律师", "").strip()
-
-                now = int(time.time())
-                record_id = hashlib.md5(card.detail_url.encode("utf-8")).hexdigest()
-
-                yield {
-                    "record_id": record_id,
-                    "collected_at": now,
-                    "source": {
-                        "site": SITE_NAME,
-                        "province_id": target.province_id,
-                        "province": target.province_name,
-                        "province_py": target.province_py,
-                        "city_id": target.city_id,
-                        "city": target.city_name,
-                        "city_py": target.city_py,
-                        "page": page,
-                        "list_url": list_url,
-                        "detail_url": card.detail_url,
-                    },
-                    "list_snapshot": {
-                        "name": card.name,
-                        "phone": card.phone,
-                        "address": card.address,
-                        "specialties": card.specialties,
-                        "metric_text": card.metric_text,
-                    },
-                    "profile": {
-                        "name": profile_name,
-                        "law_firm": (detail.get("law_firm") or "").strip(),
-                        "phone": phone,
-                        "address": (detail.get("address") or card.address or "").strip(),
-                        "practice_years": detail.get("practice_years"),
-                        "specialties": detail.get("specialties") or card.specialties,
-                    },
-                }
-
-                if self.sleep_seconds:
-                    time.sleep(self.sleep_seconds)
-
-            if not has_next:
-                break
-
-    def _to_legacy_lawyer_row(self, record: Dict) -> Optional[Dict[str, str]]:
-        source = record.get("source", {}) or {}
-        profile = record.get("profile", {}) or {}
-
-        phone = normalize_phone(profile.get("phone", ""))
-        if not phone:
-            return None
-
-        province = (source.get("province") or "").strip()
-        city = (source.get("city") or province).strip()
-        return {
-            "name": (profile.get("name") or "").strip(),
-            "law_firm": (profile.get("law_firm") or "").strip(),
-            "province": province,
-            "city": city,
-            "phone": phone,
-            "url": (source.get("detail_url") or source.get("list_url") or "").strip(),
-            "domain": LEGACY_DOMAIN,
-            "create_time": int(record.get("collected_at") or time.time()),
-            "params": json.dumps(record, ensure_ascii=False),
-        }
-
-    def _existing_phones_in_db(self, phones: List[str]) -> Set[str]:
-        if not self.db or not phones:
+    def _existing_phones(self, phones: List[str]) -> Set[str]:
+        if not phones:
             return set()
-
-        deduped = sorted({p for p in phones if p})
-        if not deduped:
-            return set()
-
         existing: Set[str] = set()
         cur = self.db.db.cursor()
         try:
             chunk_size = 500
-            for i in range(0, len(deduped), chunk_size):
-                chunk = deduped[i:i + chunk_size]
+            for i in range(0, len(phones), chunk_size):
+                chunk = phones[i:i + chunk_size]
                 placeholders = ",".join(["%s"] * len(chunk))
                 sql = f"SELECT phone FROM lawyer WHERE domain=%s AND phone IN ({placeholders})"
-                cur.execute(sql, [LEGACY_DOMAIN, *chunk])
+                cur.execute(sql, [DOMAIN, *chunk])
                 for row in cur.fetchall():
                     existing.add(row[0])
         finally:
             cur.close()
-
         return existing
 
-    def _write_records_to_db(self, records: List[Dict]) -> Tuple[int, int]:
-        if not self.db:
-            return 0, 0
+    def _load_areas(self):
+        condition = "level = 2 and domain='法律快车'"
+        tables = ("area_new", "area", "area2")
+        last_error = None
+        for table in tables:
+            try:
+                rows = self.db.select_data(table, "pinyin, province, city", condition) or []
+            except Exception as exc:
+                last_error = exc
+                continue
+            if rows:
+                missing_pinyin = sum(1 for r in rows if not (r.get("pinyin") or "").strip())
+                print(f"[法律快车] 城市来源表: {table}, 城市数: {len(rows)}, 缺少pinyin: {missing_pinyin}")
+                return rows
 
-        rows: List[Dict[str, str]] = []
-        for record in records:
-            row = self._to_legacy_lawyer_row(record)
-            if row:
-                rows.append(row)
-        if not rows:
-            return 0, 0
+        if last_error:
+            print(f"[法律快车] 加载地区数据失败: {last_error}")
+        print("[法律快车] 无城市数据（已尝试 area_new/area/area2）")
+        return []
 
-        existing = self._existing_phones_in_db([row["phone"] for row in rows])
-        inserted = 0
-        skipped = 0
+    def _get(self, url: str, max_retries: int = 3) -> Optional[str]:
+        return self._get_with_session(self.session, url, max_retries=max_retries, is_thread=False)
 
-        for row in rows:
-            phone = row.get("phone", "")
-            if not phone or phone in existing:
-                skipped += 1
+    def _get_with_session(self, session: requests.Session, url: str, max_retries: int = 3, is_thread: bool = False) -> Optional[str]:
+        for attempt in range(max_retries):
+            try:
+                with request_slot():
+                    resp = session.get(url, timeout=15, verify=False)
+                status_code = resp.status_code
+                text = resp.text
+                resp.close()
+                if status_code == 403:
+                    if attempt < max_retries - 1:
+                        wait_time = 2 ** attempt + random.uniform(0.3, 1.0)
+                        print(f"请求失败 {url}: 403，{wait_time}秒后重试 ({attempt + 1}/{max_retries})")
+                        if is_thread:
+                            self._refresh_thread_session()
+                            session = self._get_thread_session()
+                        else:
+                            self._refresh_session()
+                            session = self.session
+                        time.sleep(wait_time)
+                        continue
+                    print(f"请求失败 {url}: 403 Forbidden")
+                    return None
+                if status_code >= 400:
+                    raise requests.exceptions.HTTPError(f"{status_code} Error: {url}")
+                return text
+            except requests.exceptions.RequestException as exc:
+                print(f"请求失败 {url}: {exc}")
+                return None
+        return None
+
+    def _parse_list(self, html: str, province: str, city: str) -> int:
+        soup = BeautifulSoup(html, "html.parser")
+        links = [a.get("href", "") for a in soup.select("a.hide_link")]
+        links = [link.replace("lll", "int") for link in links if link]
+        if not links:
+            return 0
+
+        detail_urls = [urljoin(DETAIL_BASE, link) for link in links]
+
+        results: List[Dict[str, str]] = []
+        with ThreadPoolExecutor(max_workers=self.max_workers) as ex:
+            futs = [ex.submit(self._parse_detail, u, province, city) for u in detail_urls]
+            for fut in as_completed(futs):
+                try:
+                    data = fut.result()
+                except Exception as exc:
+                    print(f"  详情解析异常: {exc}")
+                    continue
+                if data and data.get("phone"):
+                    results.append(data)
+
+        if not results:
+            return len(detail_urls)
+
+        phones = [d["phone"] for d in results if d.get("phone")]
+        existing = self._existing_phones(phones)
+
+        for data in results:
+            phone = data.get("phone")
+            if not phone:
+                continue
+            if phone in existing:
+                print(f"  -- 已存在: {data['name']} ({phone})")
                 continue
             try:
-                self.db.insert_data("lawyer", row)
-                existing.add(phone)
-                inserted += 1
+                self.db.insert_data("lawyer", data)
+                print(f"  -> 新增: {data['name']} ({phone})")
             except Exception as exc:
-                skipped += 1
-                print(f"[db] 插入失败 phone={phone} url={row.get('url', '')}: {exc}")
+                print(f"  插入失败 {data.get('url')}: {exc}")
 
-        return inserted, skipped
+        return len(detail_urls)
 
-    def crawl(
-        self,
-        output_path: str,
-        max_cities: int = 0,
-        city_filter: Optional[str] = None,
-    ) -> None:
-        cities = self.discover_cities()
-        print(f"[discover] 共发现城市 {len(cities)} 个")
+    def _parse_detail(self, url: str, province: str, city: str) -> Optional[Dict[str, str]]:
+        html = None
+        sess = self._get_thread_session()
+        html = self._get_with_session(sess, url, max_retries=3, is_thread=True)
+        if not html:
+            return None
 
-        if city_filter:
-            key = city_filter.strip().lower()
-            cities = [
-                c for c in cities
-                if key in c.city_py.lower() or key in c.city_name.lower()
-            ]
-            print(f"[discover] 过滤后城市 {len(cities)} 个, filter={city_filter}")
+        soup = BeautifulSoup(html, "html.parser")
+        text = soup.get_text(" ")
 
-        if max_cities > 0:
-            cities = cities[:max_cities]
-            print(f"[discover] 截断城市数 {len(cities)}")
+        name = ""
+        title_tag = soup.find("title")
+        if title_tag:
+            match = re.search(r"(\S+)律师", title_tag.get_text())
+            if match:
+                name = match.group(1)
+        if not name:
+            intl_div = soup.find("div", class_="intl")
+            if intl_div:
+                match = re.search(r"(\S+)律师", intl_div.get_text())
+                if match:
+                    name = match.group(1)
 
-        os.makedirs(os.path.dirname(output_path) or ".", exist_ok=True)
+        phone = ""
+        phone_pattern = r"1[3-9]\d{9}"
+        for item in soup.select("div.item.flex"):
+            label = item.find("div", class_="label")
+            desc = item.find("div", class_="desc")
+            if not label or not desc:
+                continue
+            label_text = label.get_text()
+            desc_text = desc.get_text().replace("-", "")
+            if "联系电话" in label_text or "电话" in label_text:
+                matches = re.findall(phone_pattern, desc_text)
+                if matches:
+                    phone = matches[0]
+                    break
+        if not phone:
+            matches = re.findall(phone_pattern, text.replace("-", ""))
+            if matches:
+                phone = matches[0]
+        if not phone:
+            print(f"  无手机号: {url}")
+            return None
 
-        seen_ids: Set[str] = set()
-        if os.path.exists(output_path):
-            with open(output_path, "r", encoding="utf-8") as old_file:
-                for line in old_file:
-                    line = line.strip()
-                    if not line:
-                        continue
-                    try:
-                        item = json.loads(line)
-                    except Exception:
-                        continue
-                    rid = item.get("record_id")
-                    if rid:
-                        seen_ids.add(rid)
-            print(f"[resume] 已有记录 {len(seen_ids)} 条")
+        law_firm = ""
+        for item in soup.select("div.item.flex"):
+            label = item.find("div", class_="label")
+            desc = item.find("div", class_="desc")
+            if not label or not desc:
+                continue
+            if "执业律所" in label.get_text() or "律所" in label.get_text():
+                law_firm = desc.get_text(strip=True).replace("已认证", "")
+                break
 
-        total_new_json = 0
-        total_new_db = 0
-        total_skip_db = 0
+        params = {
+            "list_url": url,
+            "province": province,
+            "city": city,
+        }
 
-        with open(output_path, "a", encoding="utf-8") as out:
-            for idx, target in enumerate(cities, start=1):
-                print(
-                    f"[city {idx}/{len(cities)}] {target.province_name}-{target.city_name} "
-                    f"({target.city_py})"
-                )
-                city_records = list(self.crawl_city(target))
+        return {
+            "name": name or "",
+            "law_firm": law_firm,
+            "province": province,
+            "city": city,
+            "phone": phone,
+            "url": url,
+            "domain": DOMAIN,
+            "create_time": int(time.time()),
+            "params": json.dumps(params, ensure_ascii=False)
+        }
 
-                city_new_json = 0
-                for record in city_records:
-                    rid = record["record_id"]
-                    if rid in seen_ids:
-                        continue
-                    out.write(json.dumps(record, ensure_ascii=False) + "\n")
-                    seen_ids.add(rid)
-                    city_new_json += 1
-                    total_new_json += 1
+    def run(self):
+        print("启动法律快车采集...")
+        areas = self._load_areas()
+        if not areas:
+            print("无地区数据")
+            return
 
-                city_new_db, city_skip_db = self._write_records_to_db(city_records)
-                total_new_db += city_new_db
-                total_skip_db += city_skip_db
-
-                print(
-                    f"[city] 采集{len(city_records)}条, JSON新增{city_new_json}条, "
-                    f"DB新增{city_new_db}条, DB跳过{city_skip_db}条"
-                )
-
-        print(
-            f"[done] JSON新增{total_new_json}条, DB新增{total_new_db}条, "
-            f"DB跳过{total_skip_db}条, 输出: {output_path}"
-        )
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description="法律快车全新采集脚本（站点数据直采）")
-    parser.add_argument(
-        "--output",
-        default="/www/wwwroot/lawyers/data/lawtime_records_all.jsonl",
-        help="输出 jsonl 文件路径",
-    )
-    parser.add_argument(
-        "--max-cities",
-        type=int,
-        default=0,
-        help="最多采集多少个城市，0 表示不限",
-    )
-    parser.add_argument(
-        "--max-pages",
-        type=int,
-        default=9999,
-        help="每个城市最多采集多少页",
-    )
-    parser.add_argument(
-        "--city-filter",
-        default="",
-        help="按城市拼音或城市名过滤，如 beijing",
-    )
-    parser.add_argument(
-        "--sleep",
-        type=float,
-        default=0.1,
-        help="详情页请求间隔秒数",
-    )
-    parser.add_argument(
-        "--direct",
-        action="store_true",
-        help="直连模式，不使用 proxy_settings.json 代理",
-    )
-    parser.add_argument(
-        "--no-db",
-        action="store_true",
-        help="只输出 JSONL，不写入数据库",
-    )
-    return parser.parse_args()
-
-
-def main():
-    args = parse_args()
-
-    if args.no_db:
-        crawler = LawtimeCrawler(
-            max_pages=args.max_pages,
-            sleep_seconds=args.sleep,
-            use_proxy=not args.direct,
-            db_connection=None,
-        )
-        crawler.crawl(
-            output_path=args.output,
-            max_cities=args.max_cities,
-            city_filter=args.city_filter or None,
-        )
-        return
-
-    with Db() as db:
-        crawler = LawtimeCrawler(
-            max_pages=args.max_pages,
-            sleep_seconds=args.sleep,
-            use_proxy=not args.direct,
-            db_connection=db,
-        )
-        crawler.crawl(
-            output_path=args.output,
-            max_cities=args.max_cities,
-            city_filter=args.city_filter or None,
-        )
+        for area in areas:
+            pinyin = area.get("pinyin")
+            province = area.get("province", "")
+            city = area.get("city", "")
+            if not pinyin:
+                continue
+            page = 1
+            while True:
+                list_url = LIST_BASE.format(pinyin=pinyin, page=page)
+                print(f"采集 {province}-{city} 第 {page} 页: {list_url}")
+                html = self._get(list_url)
+                if not html:
+                    break
+                link_count = self._parse_list(html, province, city)
+                if link_count == 0:
+                    break
+                page += 1
+        print("法律快车采集完成")
 
 
 if __name__ == "__main__":
-    main()
+    with Db() as db:
+        spider = LawtimeSpider(db)
+        spider.run()

common_sites/six4365.py

@@ -1,17 +1,11 @@
-import argparse
-import hashlib
 import json
 import os
-import random
-import re
 import sys
 import time
-from dataclasses import dataclass
-from typing import Dict, Iterable, List, Optional, Set, Tuple
-from urllib.parse import urljoin
-
-import urllib3
-from bs4 import BeautifulSoup
+import random
+from typing import Dict, Optional, List, Set
+from concurrent.futures import ThreadPoolExecutor, as_completed
+import threading
 
 current_dir = os.path.dirname(os.path.abspath(__file__))
 project_root = os.path.dirname(current_dir)
@@ -21,237 +15,167 @@ if request_dir not in sys.path:
 if project_root not in sys.path:
     sys.path.append(project_root)
 
-from Db import Db
-from request.requests_client import RequestClientError, RequestsClient
-from utils.rate_limiter import wait_for_request
+import requests
+import urllib3
+from bs4 import BeautifulSoup
+from request.proxy_config import get_proxies, report_proxy_status
 
 urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
 
-SITE_NAME = "64365"
-LEGACY_DOMAIN = "律图"
-SITE_BASE = "https://m.64365.com"
-AREA_DATA_URL = "https://image.64365.com/ui_v3/m/js/public/area-cate-data.js"
-LIST_API_URL = "https://m.64365.com/findLawyer/rpc/FindLawyer/LawyerRecommend/"
+from Db import Db
+from utils.rate_limiter import request_slot
 
-PHONE_RE = re.compile(r"1[3-9]\d{9}")
-YEAR_RE = re.compile(r"(\d+)\s*年")
+DOMAIN = "律图"
+LIST_URL = "https://m.64365.com/findLawyer/rpc/FindLawyer/LawyerRecommend/"
 
 
-@dataclass
-class CityTarget:
-    area_id: str
-    province_id: str
-    province_name: str
-    province_py: str
-    city_name: str
-    city_py: str
-
-
-@dataclass
-class ListCard:
-    detail_url: str
-    name: str
-    specialties: List[str]
-    score_text: str
-    service_text: str
-
-
-def normalize_phone(text: str) -> str:
-    compact = re.sub(r"\D", "", text or "")
-    match = PHONE_RE.search(compact)
-    return match.group(0) if match else ""
-
-
-class Six4365Crawler:
-    def __init__(
-        self,
-        max_pages: int = 9999,
-        sleep_seconds: float = 0.1,
-        use_proxy: bool = True,
-        db_connection=None,
-    ):
-        self.max_pages = max_pages
-        self.sleep_seconds = max(0.0, sleep_seconds)
+class Six4365Spider:
+    def __init__(self, db_connection):
         self.db = db_connection
-        self.client = RequestsClient(
-            headers={
-                "User-Agent": (
-                    "Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) "
-                    "AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 "
-                    "Mobile/15E148 Safari/604.1"
-                ),
-                "Accept": "text/html, */*; q=0.01",
-                "Connection": "close",
-            },
-            use_proxy=use_proxy,
-            retry_total=2,
-            retry_backoff_factor=1,
-            retry_status_forcelist=(429, 500, 502, 503, 504),
-            retry_allowed_methods=("GET", "POST"),
-        )
+        self.session = self._build_session()
+        self.max_workers = int(os.getenv("SPIDER_WORKERS", "8"))
+        self._tls = threading.local()
+        self.cities = self._load_cities()
 
-    def _request_text(
-        self,
-        method: str,
-        url: str,
-        *,
-        timeout: int = 20,
-        max_retries: int = 3,
-        referer: str = SITE_BASE,
-        data: Optional[Dict] = None,
-    ) -> str:
-        headers = {"Referer": referer}
-        last_error: Optional[Exception] = None
+    def _build_session(self) -> requests.Session:
+        report_proxy_status()
+        session = requests.Session()
+        session.trust_env = False
+        proxies = get_proxies()
+        if proxies:
+            session.proxies.update(proxies)
+        else:
+            session.proxies.clear()
+        session.headers.update({
+            "User-Agent": (
+                "Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) "
+                "AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 "
+                "Mobile/15E148 Safari/604.1"
+            ),
+            "Connection": "close",
+        })
+        return session
 
-        for attempt in range(max_retries):
-            wait_for_request()
+    def _refresh_session(self) -> None:
+        try:
+            self.session.close()
+        except Exception:
+            pass
+        self.session = self._build_session()
+
+    def _get_thread_session(self) -> requests.Session:
+        """requests.Session 不是严格线程安全：每个线程用独立 session（但共享同样代理/headers）"""
+        s = getattr(self._tls, "session", None)
+        if s is not None:
+            return s
+        s = self._build_session()
+        s.headers.update(dict(self.session.headers))
+        self._tls.session = s
+        return s
+
+    def _refresh_thread_session(self) -> None:
+        s = getattr(self._tls, "session", None)
+        if s is not None:
             try:
-                if method.upper() == "POST":
-                    resp = self.client.post_text(
-                        url,
-                        timeout=timeout,
-                        verify=False,
-                        headers=headers,
-                        data=data,
-                    )
-                else:
-                    resp = self.client.get_text(
-                        url,
-                        timeout=timeout,
-                        verify=False,
-                        headers=headers,
-                    )
+                s.close()
+            except Exception:
+                pass
+        self._tls.session = None
 
-                code = resp.status_code
-                if code == 403:
-                    if attempt < max_retries - 1:
-                        self.client.refresh()
-                        time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                        continue
-                    raise RequestClientError(f"{code} Error: {url}")
-                if code >= 500 and attempt < max_retries - 1:
-                    time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                    continue
-                if code >= 400:
-                    raise RequestClientError(f"{code} Error: {url}")
-                return resp.text
+    def _existing_urls(self, urls: List[str]) -> Set[str]:
+        """批量查重，减少 N 次 is_data_exist"""
+        if not urls:
+            return set()
+        existing: Set[str] = set()
+        cur = self.db.db.cursor()
+        try:
+            # IN 参数过多会失败，分批
+            chunk_size = 500
+            for i in range(0, len(urls), chunk_size):
+                chunk = urls[i:i + chunk_size]
+                placeholders = ",".join(["%s"] * len(chunk))
+                sql = f"SELECT url FROM lawyer WHERE url IN ({placeholders})"
+                cur.execute(sql, chunk)
+                for row in cur.fetchall():
+                    # pymysql 默认返回 tuple
+                    existing.add(row[0])
+        finally:
+            cur.close()
+        return existing
+
+    def _load_cities(self):
+        tables = ("area_new", "area2", "area")
+        last_error = None
+        for table in tables:
+            try:
+                provinces = self.db.select_data(
+                    table,
+                    "id, code, province",
+                    "domain='64365' AND level=1"
+                ) or []
+                cities = self.db.select_data(
+                    table,
+                    "code, city, province, pid",
+                    "domain='64365' AND level=2"
+                ) or []
             except Exception as exc:
                 last_error = exc
-                if attempt < max_retries - 1:
-                    time.sleep((2 ** attempt) + random.uniform(0.2, 0.8))
-                    continue
-                raise
+                continue
 
-        if last_error is not None:
-            raise last_error
-        raise RequestClientError(f"Unknown request error: {url}")
+            if not cities:
+                continue
 
-    def _get_text(self, url: str, *, timeout: int = 20, max_retries: int = 3, referer: str = SITE_BASE) -> str:
-        return self._request_text(
-            "GET",
-            url,
-            timeout=timeout,
-            max_retries=max_retries,
-            referer=referer,
-        )
+            province_map = {row.get('id'): row for row in provinces}
+            data = {}
+            for city in cities:
+                province_row = province_map.get(city.get('pid'), {}) or {}
+                data[str(city.get('code'))] = {
+                    "name": city.get('city'),
+                    "province": city.get('province'),
+                    "province_name": province_row.get('province', city.get('province')),
+                }
+            print(f"[律图] 城市来源表: {table}, 城市数: {len(cities)}")
+            return data
 
-    def _post_text(
-        self,
-        url: str,
-        *,
-        data: Dict,
-        timeout: int = 20,
-        max_retries: int = 3,
-        referer: str = SITE_BASE,
-    ) -> str:
-        return self._request_text(
-            "POST",
-            url,
-            timeout=timeout,
-            max_retries=max_retries,
-            referer=referer,
-            data=data,
-        )
+        if last_error:
+            print(f"[律图] 加载地区数据失败: {last_error}")
+        print("[律图] 无城市数据（已尝试 area_new/area2/area）")
+        return {}
 
-    def _extract_area_data(self, text: str) -> List[Dict]:
-        match = re.search(
-            r"lvtuData\.areaData\s*=\s*(\[[\s\S]*?\])\s*;\s*lvtuData\.categroyData",
-            text,
-            re.S,
-        )
-        if not match:
-            return []
-
-        raw = match.group(1)
-        try:
-            data = json.loads(raw)
-        except Exception:
-            return []
-        return data if isinstance(data, list) else []
-
-    def discover_cities(self) -> List[CityTarget]:
-        text = self._get_text(AREA_DATA_URL, referer=SITE_BASE + "/findlawyer/")
-        provinces = self._extract_area_data(text)
-
-        targets: List[CityTarget] = []
-        seen_area: Set[str] = set()
-
-        for province in provinces:
-            province_id = str(province.get("id") or "").strip()
-            province_name = str(province.get("name") or "").strip()
-            province_py = str(province.get("py") or "").strip()
-            child_rows = province.get("child") or []
-
-            # 常规省份 child 是地级市；直辖市 child 是区县，此时使用省级 id 抓取
-            if child_rows and any((row.get("child") or []) for row in child_rows):
-                for city in child_rows:
-                    area_id = str(city.get("id") or "").strip()
-                    city_name = str(city.get("name") or "").strip()
-                    city_py = str(city.get("py") or "").strip()
-                    if not area_id or not city_name:
+    def _post(self, payload: Dict[str, str], max_retries: int = 3) -> Optional[str]:
+        for attempt in range(max_retries):
+            try:
+                with request_slot():
+                    resp = self.session.post(LIST_URL, data=payload, timeout=10, verify=False)
+                status_code = resp.status_code
+                text = resp.text
+                resp.close()
+                if status_code == 403:
+                    if attempt < max_retries - 1:
+                        wait_time = 2 ** attempt + random.uniform(0.3, 1.0)
+                        print(f"403被拦截，{wait_time}秒后重试 ({attempt + 1}/{max_retries})")
+                        self._refresh_session()
+                        time.sleep(wait_time)
                         continue
-                    if area_id in seen_area:
-                        continue
-                    seen_area.add(area_id)
-                    targets.append(
-                        CityTarget(
-                            area_id=area_id,
-                            province_id=province_id,
-                            province_name=province_name,
-                            province_py=province_py,
-                            city_name=city_name,
-                            city_py=city_py,
-                        )
-                    )
-            else:
-                if not province_id or not province_name:
-                    continue
-                if province_id in seen_area:
-                    continue
-                seen_area.add(province_id)
-                targets.append(
-                    CityTarget(
-                        area_id=province_id,
-                        province_id=province_id,
-                        province_name=province_name,
-                        province_py=province_py,
-                        city_name=province_name,
-                        city_py=province_py,
-                    )
-                )
+                    print("请求失败: 403 Forbidden")
+                    return None
+                if status_code >= 400:
+                    raise requests.exceptions.HTTPError(f"{status_code} Error")
+                return text
+            except requests.exceptions.RequestException as exc:
+                print(f"请求失败: {exc}")
+                return None
+        return None
 
-        return targets
-
-    def _build_payload(self, area_id: str, page: int) -> Dict[str, str]:
-        ua = self.client.headers.get("User-Agent", "")
+    def _build_payload(self, city_code: str, page: int) -> Dict[str, str]:
         return {
             "AdCode": "",
-            "RegionId": str(area_id),
+            "RegionId": str(city_code),
             "CategoryId": "",
             "MaxNumber": "",
             "OnlyData": "true",
             "IgnoreButton": "",
-            "LawyerRecommendRequest[AreaId]": str(area_id),
+            "LawyerRecommendRequest[AreaId]": str(city_code),
             "LawyerRecommendRequest[LawCategoryIds]": "",
             "LawyerRecommendRequest[LawFirmPersonCount]": "",
             "LawyerRecommendRequest[LawFirmScale]": "",
@@ -268,429 +192,164 @@ class Six4365Crawler:
             "LawyerRecommendRequest[RefferUrl]": "",
             "LawyerRecommendRequest[RequestUrl]": "https://m.64365.com/findlawyer/",
             "LawyerRecommendRequest[resource_type_name]": "",
-            "LawyerRecommendRequest[UserAgent]": ua,
+            "LawyerRecommendRequest[UserAgent]": self.session.headers["User-Agent"],
             "LawyerRecommendRequest[AddLawyerWithNoData]": "false",
             "ShowCaseButton": "true",
         }
 
-    def fetch_list_html(self, target: CityTarget, page: int) -> str:
-        payload = self._build_payload(target.area_id, page)
-        return self._post_text(
-            LIST_API_URL,
-            data=payload,
-            referer=SITE_BASE + "/findlawyer/",
-        )
-
-    def parse_list_cards(self, html: str) -> List[ListCard]:
+    def _parse_list(self, html: str, province: str, city: str) -> int:
         soup = BeautifulSoup(html, "html.parser")
-        cards: List[ListCard] = []
-        seen: Set[str] = set()
+        lawyers = soup.find_all("a", class_="lawyer")
+        if not lawyers:
+            return 0
 
-        for anchor in soup.select("a.lawyer[href]"):
-            href = (anchor.get("href") or "").strip()
+        detail_urls: List[str] = []
+        for lawyer in lawyers:
+            href = lawyer.get("href")
             if not href:
                 continue
-            detail_url = urljoin(SITE_BASE, href)
-            if detail_url in seen:
+            detail_urls.append(f"{href.rstrip('/')}/info/")
+
+        if not detail_urls:
+            return 0
+
+        results: List[Dict[str, str]] = []
+        with ThreadPoolExecutor(max_workers=self.max_workers) as ex:
+            futs = [ex.submit(self._parse_detail, u, province, city) for u in detail_urls]
+            for fut in as_completed(futs):
+                try:
+                    data = fut.result()
+                except Exception as exc:
+                    print(f"    详情解析异常: {exc}")
+                    continue
+                if data:
+                    results.append(data)
+
+        if not results:
+            return len(detail_urls)
+
+        existing = self._existing_urls([r.get("url", "") for r in results if r.get("url")])
+        for data in results:
+            if not data:
                 continue
-            seen.add(detail_url)
+            url = data.get("url", "")
+            if not url:
+                continue
+            if url in existing:
+                print(f"  -- 已存在URL: {url}")
+                continue
+            try:
+                self.db.insert_data("lawyer", data)
+                print(f"  -> 新增: {data['name']} ({data['phone']})")
+            except Exception as exc:
+                print(f"  插入失败 {url}: {exc}")
 
-            name = ""
-            name_tag = anchor.select_one("b.name")
-            if name_tag:
-                name = name_tag.get_text(strip=True)
+        return len(detail_urls)
 
-            specialties: List[str] = []
-            skill_tag = anchor.select_one("div.skill")
-            if skill_tag:
-                raw = skill_tag.get_text(" ", strip=True).replace("擅长：", "")
-                specialties = [x.strip() for x in re.split(r"[、,，]", raw) if x.strip()]
+    def _parse_detail(self, url: str, province: str, city: str) -> Optional[Dict[str, str]]:
+        html = self._get_detail(url)
+        if not html:
+            return None
 
-            score_text = ""
-            score_tag = anchor.select_one("div.info span[title='评分'] em")
-            if score_tag:
-                score_text = score_tag.get_text(strip=True)
-
-            service_text = ""
-            service_tag = anchor.select_one("div.info")
-            if service_tag:
-                service_text = service_tag.get_text(" ", strip=True)
-
-            cards.append(
-                ListCard(
-                    detail_url=detail_url,
-                    name=name,
-                    specialties=specialties,
-                    score_text=score_text,
-                    service_text=service_text,
-                )
-            )
-
-        return cards
-
-    def parse_detail(self, detail_url: str) -> Dict:
-        info_url = detail_url.rstrip("/") + "/info/"
-        html = self._get_text(info_url, referer=detail_url)
         soup = BeautifulSoup(html, "html.parser")
+        base_info = soup.find("ul", class_="intro-basic-bar")
+        if not base_info:
+            return None
 
         name = ""
         law_firm = ""
         phone = ""
-        practice_years: Optional[int] = None
-        office_area = ""
-        address = ""
-        specialties: List[str] = []
 
-        for li in soup.select("ul.intro-basic-bar li"):
-            label_tag = li.select_one("span.label")
-            value_tag = li.select_one("div.txt")
-            if not label_tag or not value_tag:
+        for li in base_info.find_all("li"):
+            label = li.find("span", class_="label")
+            txt = li.find("div", class_="txt")
+            if not label or not txt:
                 continue
+            label_text = label.get_text(strip=True)
+            if "姓名" in label_text:
+                name = txt.get_text(strip=True)
+            if "执业律所" in label_text:
+                law_firm = txt.get_text(strip=True)
 
-            label = label_tag.get_text(" ", strip=True).replace("：", "")
-            value = value_tag.get_text(" ", strip=True)
+        more_section = soup.find("div", class_="more-intro-basic")
+        if more_section:
+            phone_ul = more_section.find("ul", class_="intro-basic-bar")
+            if phone_ul:
+                for li in phone_ul.find_all("li"):
+                    label = li.find("span", class_="label")
+                    txt = li.find("div", class_="txt")
+                    if label and txt and "联系电话" in label.get_text(strip=True):
+                        phone = txt.get_text(strip=True).replace(" ", "")
+                        break
 
-            if "姓名" in label and not name:
-                name = value
-            elif "执业律所" in label and not law_firm:
-                law_firm = value
-            elif "联系电话" in label and not phone:
-                phone = normalize_phone(value)
-            elif "执业年限" in label and practice_years is None:
-                year_match = YEAR_RE.search(value)
-                if year_match:
-                    try:
-                        practice_years = int(year_match.group(1))
-                    except Exception:
-                        practice_years = None
-            elif "办公地区" in label and not office_area:
-                office_area = value
-            elif "办公地址" in label and not address:
-                address = value
-
-        text = soup.get_text(" ", strip=True)
-        if not phone:
-            phone = normalize_phone(text)
-
-        if not name and soup.title:
-            title = soup.title.get_text(" ", strip=True)
-            match = re.search(r"([^\s_，,。]+?)律师", title)
-            if match:
-                name = match.group(1).strip()
-
-        skill_match = re.search(r"擅长：([^\n]+)", text)
-        if skill_match:
-            specialties = [x.strip() for x in re.split(r"[、,，]", skill_match.group(1)) if x.strip()]
-
-        return {
-            "name": name,
-            "law_firm": law_firm,
-            "phone": phone,
-            "practice_years": practice_years,
-            "office_area": office_area,
-            "address": address,
-            "specialties": specialties,
-            "detail_url": detail_url,
-            "info_url": info_url,
-        }
-
-    def crawl_city(self, target: CityTarget) -> Iterable[Dict]:
-        seen_detail_urls: Set[str] = set()
-        page_first_seen: Set[str] = set()
-
-        for page in range(1, self.max_pages + 1):
-            try:
-                html = self.fetch_list_html(target, page)
-            except Exception as exc:
-                print(f"[list] 失败 area={target.area_id} p{page}: {exc}")
-                break
-
-            cards = self.parse_list_cards(html)
-            if not cards:
-                break
-
-            first_url = cards[0].detail_url
-            if first_url in page_first_seen:
-                break
-            page_first_seen.add(first_url)
-
-            for card in cards:
-                if card.detail_url in seen_detail_urls:
-                    continue
-                seen_detail_urls.add(card.detail_url)
-
-                try:
-                    detail = self.parse_detail(card.detail_url)
-                except Exception as exc:
-                    print(f"[detail] 失败 {card.detail_url}: {exc}")
-                    continue
-
-                now = int(time.time())
-                uid_match = re.search(r"/lawyer/(\d+)/", card.detail_url)
-                uid = uid_match.group(1) if uid_match else card.detail_url
-                record_id = hashlib.md5(uid.encode("utf-8")).hexdigest()
-
-                yield {
-                    "record_id": record_id,
-                    "collected_at": now,
-                    "source": {
-                        "site": SITE_NAME,
-                        "province_id": target.province_id,
-                        "province": target.province_name,
-                        "province_py": target.province_py,
-                        "area_id": target.area_id,
-                        "city": target.city_name,
-                        "city_py": target.city_py,
-                        "page": page,
-                        "detail_url": card.detail_url,
-                        "info_url": detail.get("info_url", ""),
-                    },
-                    "list_snapshot": {
-                        "name": card.name,
-                        "specialties": card.specialties,
-                        "score_text": card.score_text,
-                        "service_text": card.service_text,
-                    },
-                    "profile": {
-                        "name": detail.get("name") or card.name,
-                        "law_firm": detail.get("law_firm") or "",
-                        "phone": detail.get("phone") or "",
-                        "practice_years": detail.get("practice_years"),
-                        "office_area": detail.get("office_area") or "",
-                        "address": detail.get("address") or "",
-                        "specialties": detail.get("specialties") or card.specialties,
-                    },
-                }
-
-                if self.sleep_seconds:
-                    time.sleep(self.sleep_seconds)
-
-    def _to_legacy_lawyer_row(self, record: Dict) -> Optional[Dict[str, str]]:
-        source = record.get("source", {}) or {}
-        profile = record.get("profile", {}) or {}
-
-        phone = normalize_phone(profile.get("phone", ""))
-        if not phone:
+        phone = phone.replace('-', '').strip()
+        if not name or not phone:
             return None
 
-        province = (source.get("province") or "").strip()
-        city = (source.get("city") or province).strip()
-        return {
-            "name": (profile.get("name") or "").strip(),
-            "law_firm": (profile.get("law_firm") or "").strip(),
+        data = {
+            "phone": phone,
             "province": province,
             "city": city,
-            "phone": phone,
-            "url": (source.get("info_url") or source.get("detail_url") or "").strip(),
-            "domain": LEGACY_DOMAIN,
-            "create_time": int(record.get("collected_at") or time.time()),
-            "params": json.dumps(record, ensure_ascii=False),
+            "law_firm": law_firm,
+            "url": url,
+            "domain": DOMAIN,
+            "name": name,
+            "create_time": int(time.time()),
+            "params": json.dumps({"province": province, "city": city}, ensure_ascii=False)
         }
+        return data
 
-    def _existing_phones_in_db(self, phones: List[str]) -> Set[str]:
-        if not self.db or not phones:
-            return set()
-
-        deduped = sorted({p for p in phones if p})
-        if not deduped:
-            return set()
-
-        existing: Set[str] = set()
-        cur = self.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 = f"SELECT phone FROM lawyer WHERE domain=%s AND phone IN ({placeholders})"
-                cur.execute(sql, [LEGACY_DOMAIN, *chunk])
-                for row in cur.fetchall():
-                    existing.add(row[0])
-        finally:
-            cur.close()
-
-        return existing
-
-    def _write_records_to_db(self, records: List[Dict]) -> Tuple[int, int]:
-        if not self.db:
-            return 0, 0
-
-        rows: List[Dict[str, str]] = []
-        for record in records:
-            row = self._to_legacy_lawyer_row(record)
-            if row:
-                rows.append(row)
-        if not rows:
-            return 0, 0
-
-        existing = self._existing_phones_in_db([row["phone"] for row in rows])
-        inserted = 0
-        skipped = 0
-
-        for row in rows:
-            phone = row.get("phone", "")
-            if not phone or phone in existing:
-                skipped += 1
-                continue
+    def _get_detail(self, url: str, max_retries: int = 3) -> Optional[str]:
+        session = self._get_thread_session()
+        for attempt in range(max_retries):
             try:
-                self.db.insert_data("lawyer", row)
-                existing.add(phone)
-                inserted += 1
-            except Exception as exc:
-                skipped += 1
-                print(f"[db] 插入失败 phone={phone} url={row.get('url', '')}: {exc}")
-
-        return inserted, skipped
-
-    def crawl(
-        self,
-        output_path: str,
-        max_cities: int = 0,
-        city_filter: Optional[str] = None,
-    ) -> None:
-        cities = self.discover_cities()
-        print(f"[discover] 共发现地区 {len(cities)} 个")
-
-        if city_filter:
-            key = city_filter.strip().lower()
-            cities = [
-                c for c in cities
-                if key in c.city_name.lower() or key in c.city_py.lower() or key in c.area_id
-            ]
-            print(f"[discover] 过滤后地区 {len(cities)} 个, filter={city_filter}")
-
-        if max_cities > 0:
-            cities = cities[:max_cities]
-            print(f"[discover] 截断地区数 {len(cities)}")
-
-        os.makedirs(os.path.dirname(output_path) or ".", exist_ok=True)
-
-        seen_ids: Set[str] = set()
-        if os.path.exists(output_path):
-            with open(output_path, "r", encoding="utf-8") as old_file:
-                for line in old_file:
-                    line = line.strip()
-                    if not line:
+                with request_slot():
+                    resp = session.get(url, timeout=10, verify=False)
+                status_code = resp.status_code
+                text = resp.text
+                resp.close()
+                if status_code == 403:
+                    if attempt < max_retries - 1:
+                        wait_time = 2 ** attempt + random.uniform(0.3, 1.0)
+                        print(f"    403被拦截，{wait_time}秒后重试 ({attempt + 1}/{max_retries})")
+                        self._refresh_thread_session()
+                        session = self._get_thread_session()
+                        time.sleep(wait_time)
                         continue
-                    try:
-                        item = json.loads(line)
-                    except Exception:
-                        continue
-                    rid = item.get("record_id")
-                    if rid:
-                        seen_ids.add(rid)
-            print(f"[resume] 已有记录 {len(seen_ids)} 条")
+                    print("    请求失败: 403 Forbidden")
+                    return None
+                if status_code >= 400:
+                    raise requests.exceptions.HTTPError(f"{status_code} Error")
+                return text
+            except requests.exceptions.RequestException as exc:
+                print(f"    请求失败: {exc}")
+                return None
+        return None
 
-        total_new_json = 0
-        total_new_db = 0
-        total_skip_db = 0
+    def run(self):
+        print("启动律图采集...")
+        if not self.cities:
+            print("无城市数据")
+            return
 
-        with open(output_path, "a", encoding="utf-8") as out:
-            for idx, target in enumerate(cities, start=1):
-                print(
-                    f"[city {idx}/{len(cities)}] {target.province_name}-{target.city_name} "
-                    f"(area={target.area_id})"
-                )
-                city_records = list(self.crawl_city(target))
-
-                city_new_json = 0
-                for record in city_records:
-                    rid = record["record_id"]
-                    if rid in seen_ids:
-                        continue
-                    out.write(json.dumps(record, ensure_ascii=False) + "\n")
-                    seen_ids.add(rid)
-                    city_new_json += 1
-                    total_new_json += 1
-
-                city_new_db, city_skip_db = self._write_records_to_db(city_records)
-                total_new_db += city_new_db
-                total_skip_db += city_skip_db
-
-                print(
-                    f"[city] 采集{len(city_records)}条, JSON新增{city_new_json}条, "
-                    f"DB新增{city_new_db}条, DB跳过{city_skip_db}条"
-                )
-
-        print(
-            f"[done] JSON新增{total_new_json}条, DB新增{total_new_db}条, "
-            f"DB跳过{total_skip_db}条, 输出: {output_path}"
-        )
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description="律图全新采集脚本（站点数据直采）")
-    parser.add_argument(
-        "--output",
-        default="/www/wwwroot/lawyers/data/six4365_records_all.jsonl",
-        help="输出 jsonl 文件路径",
-    )
-    parser.add_argument(
-        "--max-cities",
-        type=int,
-        default=0,
-        help="最多采集多少个地区，0 表示不限",
-    )
-    parser.add_argument(
-        "--max-pages",
-        type=int,
-        default=9999,
-        help="每个地区最多采集多少页",
-    )
-    parser.add_argument(
-        "--city-filter",
-        default="",
-        help="按城市名称/拼音/编码过滤",
-    )
-    parser.add_argument(
-        "--sleep",
-        type=float,
-        default=0.1,
-        help="详情页请求间隔秒数",
-    )
-    parser.add_argument(
-        "--direct",
-        action="store_true",
-        help="直连模式，不使用 proxy_settings.json 代理",
-    )
-    parser.add_argument(
-        "--no-db",
-        action="store_true",
-        help="只输出 JSONL，不写入数据库",
-    )
-    return parser.parse_args()
-
-
-def main():
-    args = parse_args()
-
-    if args.no_db:
-        crawler = Six4365Crawler(
-            max_pages=args.max_pages,
-            sleep_seconds=args.sleep,
-            use_proxy=not args.direct,
-            db_connection=None,
-        )
-        crawler.crawl(
-            output_path=args.output,
-            max_cities=args.max_cities,
-            city_filter=args.city_filter or None,
-        )
-        return
-
-    with Db() as db:
-        crawler = Six4365Crawler(
-            max_pages=args.max_pages,
-            sleep_seconds=args.sleep,
-            use_proxy=not args.direct,
-            db_connection=db,
-        )
-        crawler.crawl(
-            output_path=args.output,
-            max_cities=args.max_cities,
-            city_filter=args.city_filter or None,
-        )
+        for city_code, info in self.cities.items():
+            province = info.get("province_name", "")
+            city = info.get("name", "")
+            print(f"采集 {province}-{city}")
+            page = 1
+            while True:
+                payload = self._build_payload(city_code, page)
+                html = self._post(payload)
+                if not html:
+                    break
+                link_count = self._parse_list(html, province, city)
+                if link_count == 0:
+                    break
+                page += 1
+        print("律图采集完成")
 
 
 if __name__ == "__main__":
-    main()
+    with Db() as db:
+        spider = Six4365Spider(db)
+        spider.run()

common_sites/smoke_test_sites.py

@@ -0,0 +1,220 @@
+import json
+import os
+import sys
+import time
+from dataclasses import dataclass
+from typing import Any, Dict, List, Optional, Tuple
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+project_root = os.path.dirname(current_dir)
+request_dir = os.path.join(project_root, "request")
+if request_dir not in sys.path:
+    sys.path.insert(0, request_dir)
+if project_root not in sys.path:
+    sys.path.append(project_root)
+
+import requests
+
+from request.proxy_config import get_proxies, report_proxy_status
+
+
+@dataclass
+class CheckResult:
+    site: str
+    url: str
+    method: str
+    ok: bool
+    status_code: Optional[int]
+    error: str
+    hint: str
+    elapsed_ms: int
+
+
+def _now_ms() -> int:
+    return int(time.time() * 1000)
+
+
+def _short_hint(text: str) -> str:
+    s = (text or "").strip().lower()
+    flags = []
+    for key, label in [
+        ("403", "403"),
+        ("429", "429"),
+        ("captcha", "captcha"),
+        ("验证码", "captcha_cn"),
+        ("人机", "bot_check_cn"),
+        ("access denied", "access_denied"),
+        ("forbidden", "forbidden"),
+        ("too many requests", "rate_limited"),
+        ("cloudflare", "cloudflare"),
+        ("challenge", "challenge"),
+    ]:
+        if key in s:
+            flags.append(label)
+    return ",".join(flags)[:120]
+
+
+def _build_session() -> requests.Session:
+    report_proxy_status()
+    s = requests.Session()
+    s.trust_env = False
+    proxies = get_proxies()
+    if proxies:
+        s.proxies.update(proxies)
+    else:
+        s.proxies.clear()
+    s.headers.update(
+        {
+            "User-Agent": (
+                "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+                "AppleWebKit/537.36 (KHTML, like Gecko) "
+                "Chrome/136.0.0.0 Safari/537.36"
+            ),
+            "Accept": "*/*",
+            "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
+            "Connection": "close",
+        }
+    )
+    return s
+
+
+def _check(
+    session: requests.Session,
+    *,
+    site: str,
+    method: str,
+    url: str,
+    timeout: Tuple[float, float] = (10.0, 15.0),
+    headers: Optional[Dict[str, str]] = None,
+    data: Optional[Dict[str, Any]] = None,
+) -> CheckResult:
+    start = _now_ms()
+    try:
+        resp = session.request(
+            method=method,
+            url=url,
+            timeout=timeout,
+            verify=False,
+            headers=headers,
+            data=data,
+        )
+        text = resp.text or ""
+        status = resp.status_code
+        hint = _short_hint(text[:1200])
+        ok = 200 <= status < 400
+        return CheckResult(
+            site=site,
+            url=url,
+            method=method,
+            ok=ok,
+            status_code=status,
+            error="",
+            hint=hint,
+            elapsed_ms=_now_ms() - start,
+        )
+    except Exception as exc:
+        return CheckResult(
+            site=site,
+            url=url,
+            method=method,
+            ok=False,
+            status_code=None,
+            error=str(exc)[:200],
+            hint="",
+            elapsed_ms=_now_ms() - start,
+        )
+    finally:
+        try:
+            resp.close()  # type: ignore[name-defined]
+        except Exception:
+            pass
+
+
+def _tests() -> List[Dict[str, Any]]:
+    # 每个站点选一个“代表性列表/API”作为冒烟：能快速暴露 403/验证码/限频。
+    return [
+        {
+            "site": "大律师(m站)",
+            "method": "GET",
+            "url": "https://m.maxlaw.cn/",
+        },
+        {
+            "site": "大律师(PC站)",
+            "method": "GET",
+            "url": "https://www.maxlaw.cn/law/beijing?page=1",
+            "headers": {"Referer": "https://www.maxlaw.cn/"},
+        },
+        {
+            "site": "找法网(m站)",
+            "method": "GET",
+            "url": "https://m.findlaw.cn/beijing/q_lawyer/p1?ajax=1&order=0&sex=-1",
+            "headers": {
+                "Referer": "https://m.findlaw.cn/beijing/q_lawyer/",
+                "X-Requested-With": "XMLHttpRequest",
+                "Accept": "application/json, text/javascript, */*; q=0.01",
+            },
+        },
+        {
+            "site": "法律快车(m站)",
+            "method": "GET",
+            "url": "https://m.lawtime.cn/beijing/lawyer/?page=1",
+        },
+        {
+            "site": "律图(m站)",
+            "method": "POST",
+            "url": "https://m.64365.com/findLawyer/rpc/FindLawyer/LawyerRecommend/",
+            "data": {
+                "RegionId": "110100",  # 北京市
+                "OnlyData": "true",
+                "LawyerRecommendRequest[AreaId]": "110100",
+                "LawyerRecommendRequest[PageIndex]": "1",
+                "LawyerRecommendRequest[PageSize]": "10",
+                "LawyerRecommendRequest[OrderType]": "0",
+                "LawyerRecommendRequest[Type]": "1",
+            },
+        },
+        {
+            "site": "华律(m站)",
+            "method": "POST",
+            "url": "https://m.66law.cn/findlawyer/rpc/loadlawyerlist/",
+            "data": {
+                "pid": "110000",  # 北京
+                "cid": "110100",  # 北京市
+                "page": "1",
+            },
+        },
+    ]
+
+
+def main() -> int:
+    mode = os.getenv("PROXY_ENABLED")
+    print(f"[smoke] PROXY_ENABLED={mode!r}")
+    s = _build_session()
+    results: List[CheckResult] = []
+    for item in _tests():
+        res = _check(
+            s,
+            site=item["site"],
+            method=item["method"],
+            url=item["url"],
+            headers=item.get("headers"),
+            data=item.get("data"),
+        )
+        results.append(res)
+        print(
+            f"[smoke] {res.site} {res.method} {res.status_code} ok={res.ok} "
+            f"{res.elapsed_ms}ms hint={res.hint or '-'} err={res.error or '-'}"
+        )
+        time.sleep(0.3)
+
+    summary = {
+        "proxy_enabled": mode,
+        "results": [res.__dict__ for res in results],
+    }
+    print("[smoke] summary_json=" + json.dumps(summary, ensure_ascii=False))
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())
+

common_sites/start.sh

@@ -1,80 +1,40 @@
 #!/usr/bin/env bash
 set -euo pipefail
 
-SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
-PROJECT_ROOT="$(cd "${SCRIPT_DIR}/.." && pwd)"
-LOG_DIR="${PROJECT_ROOT}/logs"
-DATA_DIR="${PROJECT_ROOT}/data"
+# 切换到脚本所在目录，确保相对路径正确
+cd "$(dirname "$0")"
 
-mkdir -p "${LOG_DIR}" "${DATA_DIR}"
+echo "使用 request/proxy_settings.json 读取代理配置"
+export PROXY_MAX_REQUESTS_PER_SECOND="${PROXY_MAX_REQUESTS_PER_SECOND:-5}"
+export PROXY_MAX_CONCURRENT_REQUESTS="${PROXY_MAX_CONCURRENT_REQUESTS:-5}"
 
-if [[ -x "${PROJECT_ROOT}/.venv/bin/python" ]]; then
-  PYTHON_BIN="${PROJECT_ROOT}/.venv/bin/python"
-else
-  PYTHON_BIN="python3"
-fi
-
-RUN_MODE="${RUN_MODE:-parallel}" # parallel | sequential
-
-echo "[start] project=${PROJECT_ROOT}"
-echo "[start] python=${PYTHON_BIN}"
-echo "[start] mode=${RUN_MODE}"
-echo "[start] proxy=request/proxy_settings.json"
-
-# 大律师（新结构采集 + 写库）可通过环境变量控制
-DLS_OUTPUT="${DLS_OUTPUT:-${DATA_DIR}/dls_records.jsonl}"
-DLS_MAX_CITIES="${DLS_MAX_CITIES:-0}"
-DLS_MAX_PAGES="${DLS_MAX_PAGES:-0}"
-DLS_SLEEP="${DLS_SLEEP:-0.2}"
-DLS_CITY_FILTER="${DLS_CITY_FILTER:-}"
-DLS_EXTRA_ARGS=()
-
-if [[ "${DLS_MAX_CITIES}" != "0" ]]; then
-  DLS_EXTRA_ARGS+=(--max-cities "${DLS_MAX_CITIES}")
-fi
-if [[ "${DLS_MAX_PAGES}" != "0" ]]; then
-  DLS_EXTRA_ARGS+=(--max-pages "${DLS_MAX_PAGES}")
-fi
-if [[ -n "${DLS_CITY_FILTER}" ]]; then
-  DLS_EXTRA_ARGS+=(--city-filter "${DLS_CITY_FILTER}")
-fi
-DLS_EXTRA_ARGS+=(--sleep "${DLS_SLEEP}" --output "${DLS_OUTPUT}")
-
-if [[ "${DLS_DIRECT:-0}" == "1" ]]; then
-  DLS_EXTRA_ARGS+=(--direct)
-fi
-if [[ "${DLS_NO_DB:-0}" == "1" ]]; then
-  DLS_EXTRA_ARGS+=(--no-db)
-fi
-
-run_bg() {
-  local name="$1"
-  shift
-  local logfile="${LOG_DIR}/${name}.log"
-  nohup env PYTHONUNBUFFERED=1 "$@" > "${logfile}" 2>&1 &
-  echo "[start] ${name} pid=$! log=${logfile}"
+is_job_running() {
+  local script="$1"
+  local script_regex="${script//./\\.}"
+  pgrep -af "(^|[[:space:]/])${script_regex}([[:space:]]|$)" || true
 }
 
-run_fg() {
-  local name="$1"
-  shift
-  local logfile="${LOG_DIR}/${name}.log"
-  echo "[start] ${name} fg log=${logfile}"
-  env PYTHONUNBUFFERED=1 "$@" > "${logfile}" 2>&1
+start_job() {
+  local script="$1"
+  local log_file="$2"
+  local label="$3"
+  local existing
+
+  existing="$(is_job_running "${script}")"
+  if [[ -n "${existing}" ]]; then
+    echo "跳过 ${label}: ${script} 已在运行"
+    echo "${existing}" | head -n 1
+    return 0
+  fi
+
+  nohup python "../common_sites/${script}" > "${log_file}" 2>&1 &
+  echo "启动 ${label}: ${script} -> ${log_file}"
+  sleep 1
 }
 
-if [[ "${RUN_MODE}" == "sequential" ]]; then
-  run_fg "dls_fresh" "${PYTHON_BIN}" "${SCRIPT_DIR}/dls_fresh.py" "${DLS_EXTRA_ARGS[@]}"
-  run_fg "findlaw" "${PYTHON_BIN}" "${SCRIPT_DIR}/findlaw.py"
-  run_fg "lawtime" "${PYTHON_BIN}" "${SCRIPT_DIR}/lawtime.py"
-  run_fg "six4365" "${PYTHON_BIN}" "${SCRIPT_DIR}/six4365.py"
-  run_fg "hualv" "${PYTHON_BIN}" "${SCRIPT_DIR}/hualv.py"
-  echo "[done] sequential completed"
-else
-  run_bg "dls_fresh" "${PYTHON_BIN}" "${SCRIPT_DIR}/dls_fresh.py" "${DLS_EXTRA_ARGS[@]}"
-  run_bg "findlaw" "${PYTHON_BIN}" "${SCRIPT_DIR}/findlaw.py"
-  run_bg "lawtime" "${PYTHON_BIN}" "${SCRIPT_DIR}/lawtime.py"
-  run_bg "six4365" "${PYTHON_BIN}" "${SCRIPT_DIR}/six4365.py"
-  run_bg "hualv" "${PYTHON_BIN}" "${SCRIPT_DIR}/hualv.py"
-  echo "[done] all crawlers started in background"
-fi
+start_job "dls.py" "dls.log" "大律师"
+start_job "dls_pc.py" "dls_pc.log" "大律师PC站"
+start_job "findlaw.py" "findlaw.log" "找法网"
+start_job "lawtime.py" "lawtime.log" "法律快车"
+start_job "six4365.py" "six4365.log" "律图"
+start_job "hualv.py" "hualv.log" "华律"

common_sites/start_direct_twice_weekly.sh

@@ -0,0 +1,48 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 切换到脚本所在目录，确保相对路径正确
+cd "$(dirname "$0")"
+
+# 强制直连：不使用代理 IP
+export PROXY_ENABLED=0
+
+# 直连模式建议更保守一些，降低被临时风控的概率
+export PROXY_MAX_REQUESTS_PER_SECOND="${PROXY_MAX_REQUESTS_PER_SECOND:-5}"
+export PROXY_MAX_CONCURRENT_REQUESTS="${PROXY_MAX_CONCURRENT_REQUESTS:-5}"
+
+is_job_running() {
+  local script="$1"
+  local script_regex="${script//./\\.}"
+  pgrep -af "(^|[[:space:]/])${script_regex}([[:space:]]|$)" || true
+}
+
+start_job() {
+  local script="$1"
+  local log_file="$2"
+  local label="$3"
+  local existing
+
+  existing="$(is_job_running "${script}")"
+  if [[ -n "${existing}" ]]; then
+    echo "跳过 ${label}: ${script} 已在运行"
+    echo "${existing}" | head -n 1
+    return 0
+  fi
+
+  nohup python "../common_sites/${script}" > "${log_file}" 2>&1 &
+  echo "启动 ${label}: ${script} -> ${log_file}"
+  sleep 1
+}
+
+echo "直连模式（PROXY_ENABLED=0），每周两次建议用 cron 调度"
+echo "当前归入直连组：大律师(m/PC)、华律、律图"
+
+# 直连优先站点：
+# - 大律师(m站/PC站)：当前可直接访问，未见明显强风控
+# - 华律：当前网页可直接访问，未见明显强风控
+# - 律图：当前网页可直接访问，未见明显强风控
+start_job "dls.py" "direct_dls.log" "大律师(直连)"
+start_job "dls_pc.py" "direct_dls_pc.log" "大律师PC站(直连)"
+start_job "hualv.py" "direct_hualv.log" "华律(直连)"
+start_job "six4365.py" "direct_six4365.log" "律图(直连)"

common_sites/start_proxy_weekly.sh

@@ -0,0 +1,53 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 切换到脚本所在目录，确保相对路径正确
+cd "$(dirname "$0")"
+
+# 强制开启代理：用于容易被限频/拦截的站点
+export PROXY_ENABLED=1
+
+# 代理模式下默认更保守一点，避免冲爆代理与触发风控
+export PROXY_MAX_REQUESTS_PER_SECOND="${PROXY_MAX_REQUESTS_PER_SECOND:-5}"
+export PROXY_MAX_CONCURRENT_REQUESTS="${PROXY_MAX_CONCURRENT_REQUESTS:-5}"
+
+# 可选：开启代理连通性测试输出（部分脚本会打印测试信息/代理状态）
+export PROXY_TEST="${PROXY_TEST:-0}"
+
+is_job_running() {
+  local script="$1"
+  local script_regex="${script//./\\.}"
+  pgrep -af "(^|[[:space:]/])${script_regex}([[:space:]]|$)" || true
+}
+
+start_job() {
+  local script="$1"
+  local log_file="$2"
+  local label="$3"
+  local existing
+
+  existing="$(is_job_running "${script}")"
+  if [[ -n "${existing}" ]]; then
+    echo "跳过 ${label}: ${script} 已在运行"
+    echo "${existing}" | head -n 1
+    return 0
+  fi
+
+  nohup python "../common_sites/${script}" > "${log_file}" 2>&1 &
+  echo "启动 ${label}: ${script} -> ${log_file}"
+  sleep 1
+}
+
+echo "代理模式（PROXY_ENABLED=1），每周一次建议用 cron 调度"
+echo "代理配置读取自 request/proxy_settings.json"
+echo "每周一次代理任务 = 全量采集所有站点"
+
+# 每周一次代理任务做全量采集：
+# - 强风控/更敏感站点：找法网、法律快车
+# - 其余站点也一并跑，保证每周至少有一次“全量最新数据”刷新
+start_job "dls.py" "proxy_dls.log" "大律师(代理全量)"
+start_job "dls_pc.py" "proxy_dls_pc.log" "大律师PC站(代理全量)"
+start_job "findlaw.py" "proxy_findlaw.log" "找法网(代理)"
+start_job "lawtime.py" "proxy_lawtime.log" "法律快车(代理)"
+start_job "hualv.py" "proxy_hualv.log" "华律(代理全量)"
+start_job "six4365.py" "proxy_six4365.log" "律图(代理全量)"

compare_firm_phones.py

@@ -0,0 +1,565 @@
+#!/usr/bin/env python3
+import argparse
+import re
+from collections import defaultdict
+from dataclasses import dataclass
+from typing import Dict, Iterable, List, Optional, Sequence, Tuple
+
+import pymysql
+from openpyxl import Workbook, load_workbook
+from openpyxl.styles import Font
+
+from config import DB_CONFIG
+
+
+@dataclass(frozen=True)
+class LawyerRecord:
+    id: int
+    name: str
+    phone: str
+    law_firm: str
+    province: str
+    city: str
+    domain: str
+    create_time: int
+
+
+@dataclass(frozen=True)
+class PhoneBackfill:
+    matched_phones: List[str]
+    records: List[LawyerRecord]
+    best_name: str
+    best_law_firm: str
+    best_domain: str
+    candidate_names: List[str]
+    candidate_firms: List[str]
+    candidate_domains: List[str]
+
+
+DOMAIN_PRIORITY = {
+    "华律": 90,
+    "大律师": 85,
+    "找法网": 82,
+    "法律快车": 80,
+    "律图": 72,
+    "众法利单页": 68,
+    "众法利": 66,
+    "六四三六五": 64,
+    "智飞律师在线": 40,
+    "高德地图": 10,
+}
+
+GENERIC_FIRMS = {"高德搜索"}
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="按律所名从数据库补手机号并导出对比表")
+    parser.add_argument("--input", default="man.xlsx", help="原始 xlsx 文件路径")
+    parser.add_argument(
+        "--output",
+        default="man_firm_phone_compare.xlsx",
+        help="输出 xlsx 文件路径",
+    )
+    return parser.parse_args()
+
+
+def normalize_text(value: object) -> str:
+    text = str(value or "").strip()
+    text = text.replace("（", "(").replace("）", ")")
+    text = re.sub(r"\s+", "", text)
+    return text
+
+
+def normalize_firm(value: object) -> str:
+    text = normalize_text(value)
+    text = text.replace("本地大所", "").replace("特色律所", "")
+    return text
+
+
+def normalize_name(value: object) -> str:
+    text = normalize_text(value)
+    return text.replace("律师", "")
+
+
+def normalize_province(value: object) -> str:
+    text = str(value or "").strip()
+    mapping = {
+        "北京市": "北京",
+        "天津市": "天津",
+        "上海市": "上海",
+        "重庆市": "重庆",
+        "内蒙古自治区": "内蒙古",
+        "广西壮族自治区": "广西",
+        "宁夏回族自治区": "宁夏",
+        "新疆维吾尔自治区": "新疆",
+        "西藏自治区": "西藏",
+        "香港特别行政区": "香港",
+        "澳门特别行政区": "澳门",
+        "新疆生产建设兵团": "新疆",
+    }
+    if text in mapping:
+        return mapping[text]
+    if text.endswith("省") and len(text) > 1:
+        return text[:-1]
+    return text
+
+
+def normalize_city(value: object) -> str:
+    text = str(value or "").strip()
+    for suffix in ("市", "地区", "盟"):
+        if text.endswith(suffix) and len(text) > len(suffix):
+            return text[: -len(suffix)]
+    return text
+
+
+def split_phones(value: object) -> List[str]:
+    return re.findall(r"1\d{10}", str(value or ""))
+
+
+def unique_phones(records: Sequence[LawyerRecord]) -> List[str]:
+    output: List[str] = []
+    seen = set()
+    for record in sorted(records, key=lambda item: (item.create_time, item.id), reverse=True):
+        if record.phone and record.phone not in seen:
+            seen.add(record.phone)
+            output.append(record.phone)
+    return output
+
+
+def unique_values(records: Sequence[LawyerRecord], attr: str) -> List[str]:
+    output: List[str] = []
+    seen = set()
+    for record in sorted(records, key=lambda item: (item.create_time, item.id), reverse=True):
+        value = getattr(record, attr, "")
+        if value and value not in seen:
+            seen.add(value)
+            output.append(value)
+    return output
+
+
+def phone_record_sort_key(
+    record: LawyerRecord,
+    target_name: object,
+    target_province: object,
+    target_city: object,
+) -> Tuple[int, int, int]:
+    score = 0
+    normalized_target_name = normalize_name(target_name)
+    normalized_target_province = normalize_province(target_province)
+    normalized_target_city = normalize_city(target_city)
+
+    if normalized_target_name:
+        if normalize_name(record.name) == normalized_target_name:
+            score += 400
+        elif record.name:
+            score -= 40
+
+    if record.law_firm and record.law_firm not in GENERIC_FIRMS:
+        score += 220
+    elif record.law_firm:
+        score += 40
+
+    if record.name:
+        score += 100
+
+    if normalized_target_city:
+        if normalize_city(record.city) == normalized_target_city:
+            score += 45
+        elif record.city:
+            score -= 10
+
+    if normalized_target_province:
+        if normalize_province(record.province) == normalized_target_province:
+            score += 25
+        elif record.province:
+            score -= 5
+
+    score += DOMAIN_PRIORITY.get(record.domain, 50)
+    return score, record.create_time, record.id
+
+
+def compare_result(original_phones: Sequence[str], candidate_phones: Sequence[str]) -> str:
+    if not candidate_phones:
+        return "未匹配"
+    if not original_phones:
+        return "原手机号为空"
+
+    original_set = set(original_phones)
+    candidate_set = set(candidate_phones)
+    if original_set == candidate_set:
+        return "完全一致"
+    if original_set & candidate_set:
+        return "候选包含原手机号"
+    return "不包含原手机号"
+
+
+def infer_firm_from_address(address: object, ordered_firms: Sequence[str]) -> str:
+    normalized_address = normalize_text(address)
+    if not normalized_address:
+        return ""
+    for firm in ordered_firms:
+        if len(firm) < 4:
+            continue
+        if firm in normalized_address:
+            return firm
+    return ""
+
+
+def load_db_indexes() -> Tuple[Dict[str, List[LawyerRecord]], List[str], Dict[str, List[LawyerRecord]]]:
+    conn = pymysql.connect(**DB_CONFIG)
+    firm_index: Dict[str, List[LawyerRecord]] = defaultdict(list)
+    phone_index: Dict[str, List[LawyerRecord]] = defaultdict(list)
+    try:
+        with conn.cursor() as cur:
+            cur.execute(
+                """
+                SELECT id, name, phone, law_firm, province, city, domain, create_time
+                FROM lawyer
+                WHERE phone IS NOT NULL
+                  AND phone <> ''
+                """
+            )
+            for row in cur.fetchall():
+                record = LawyerRecord(
+                    id=int(row[0]),
+                    name=str(row[1] or "").strip(),
+                    phone=str(row[2] or "").strip(),
+                    law_firm=str(row[3] or "").strip(),
+                    province=str(row[4] or "").strip(),
+                    city=str(row[5] or "").strip(),
+                    domain=str(row[6] or "").strip(),
+                    create_time=int(row[7] or 0),
+                )
+                phone_index[record.phone].append(record)
+                normalized_firm = normalize_firm(record.law_firm)
+                if normalized_firm:
+                    firm_index[normalized_firm].append(record)
+    finally:
+        conn.close()
+
+    ordered_firms = sorted(firm_index.keys(), key=len, reverse=True)
+    return firm_index, ordered_firms, phone_index
+
+
+def build_phone_backfill(
+    original_phone: object,
+    name: object,
+    province: object,
+    city: object,
+    phone_index: Dict[str, List[LawyerRecord]],
+) -> PhoneBackfill:
+    def pick_best_name(records: Sequence[LawyerRecord], target_name: object) -> str:
+        normalized_target_name = normalize_name(target_name)
+        if normalized_target_name:
+            for item in records:
+                if item.name and normalize_name(item.name) == normalized_target_name:
+                    return item.name
+        for item in records:
+            if item.name:
+                return item.name
+        return ""
+
+    records: List[LawyerRecord] = []
+    seen_ids = set()
+    for phone in split_phones(original_phone):
+        for record in phone_index.get(phone, []):
+            if record.id in seen_ids:
+                continue
+            seen_ids.add(record.id)
+            records.append(record)
+
+    sorted_records = sorted(
+        records,
+        key=lambda item: phone_record_sort_key(item, name, province, city),
+        reverse=True,
+    )
+    candidate_names = unique_values(sorted_records, "name")
+    candidate_firms = unique_values(
+        [item for item in sorted_records if item.law_firm and item.law_firm not in GENERIC_FIRMS],
+        "law_firm",
+    )
+    if not candidate_firms:
+        candidate_firms = unique_values(
+            [item for item in sorted_records if item.law_firm],
+            "law_firm",
+        )
+    candidate_domains = unique_values(sorted_records, "domain")
+    matched_phones = unique_values(sorted_records, "phone")
+
+    best_name = pick_best_name(sorted_records, name)
+    best_law_firm = ""
+    best_domain = ""
+    preferred_name = normalize_name(name) or normalize_name(best_name)
+
+    for record in sorted_records:
+        if not record.law_firm or record.law_firm in GENERIC_FIRMS:
+            continue
+        if preferred_name and normalize_name(record.name) != preferred_name:
+            continue
+        best_law_firm = record.law_firm
+        best_domain = record.domain
+        break
+
+    if not best_law_firm:
+        for record in sorted_records:
+            if record.law_firm and record.law_firm not in GENERIC_FIRMS:
+                best_law_firm = record.law_firm
+                best_domain = record.domain
+                break
+
+    if not best_domain and sorted_records:
+        best_domain = sorted_records[0].domain
+
+    return PhoneBackfill(
+        matched_phones=matched_phones,
+        records=sorted_records,
+        best_name=best_name,
+        best_law_firm=best_law_firm,
+        best_domain=best_domain,
+        candidate_names=candidate_names,
+        candidate_firms=candidate_firms,
+        candidate_domains=candidate_domains,
+    )
+
+
+def match_row(
+    name: object,
+    original_phone: object,
+    law_firm: object,
+    province: object,
+    city: object,
+    address: object,
+    phone_backfill: PhoneBackfill,
+    firm_index: Dict[str, List[LawyerRecord]],
+    ordered_firms: Sequence[str],
+) -> Tuple[str, str, List[LawyerRecord]]:
+    def add_method(part: str, method_parts: List[str]) -> None:
+        if part and part not in method_parts:
+            method_parts.append(part)
+
+    matched_firm = normalize_firm(law_firm)
+    used_phone_backfill_firm = False
+    inferred_from_address = False
+    if not matched_firm:
+        matched_firm = normalize_firm(phone_backfill.best_law_firm)
+        used_phone_backfill_firm = bool(matched_firm)
+    if not matched_firm:
+        matched_firm = infer_firm_from_address(address, ordered_firms)
+        inferred_from_address = bool(matched_firm)
+    if not matched_firm:
+        return "", "无可用律所名", []
+
+    candidates = firm_index.get(matched_firm, [])
+    if not candidates:
+        return matched_firm, "数据库无此律所", []
+
+    method_parts = ["律所"]
+    chosen = list(candidates)
+
+    normalized_name = normalize_name(name)
+    if not normalized_name:
+        normalized_name = normalize_name(phone_backfill.best_name)
+    if normalized_name:
+        name_filtered = [item for item in chosen if normalize_name(item.name) == normalized_name]
+        if name_filtered:
+            chosen = name_filtered
+            add_method("姓名", method_parts)
+
+    if len(unique_phones(chosen)) != 1:
+        normalized_province = normalize_province(province)
+        normalized_city = normalize_city(city)
+
+        if normalized_province and normalized_city:
+            province_city_filtered = [
+                item
+                for item in chosen
+                if normalize_province(item.province) == normalized_province
+                and normalize_city(item.city) == normalized_city
+            ]
+            if province_city_filtered:
+                chosen = province_city_filtered
+                add_method("省份", method_parts)
+                add_method("城市", method_parts)
+
+        if len(unique_phones(chosen)) != 1 and normalized_city:
+            city_filtered = [
+                item for item in chosen if normalize_city(item.city) == normalized_city
+            ]
+            if city_filtered:
+                chosen = city_filtered
+                add_method("城市", method_parts)
+
+        if len(unique_phones(chosen)) != 1 and normalized_province:
+            province_filtered = [
+                item
+                for item in chosen
+                if normalize_province(item.province) == normalized_province
+            ]
+            if province_filtered:
+                chosen = province_filtered
+                add_method("省份", method_parts)
+
+    method = "+".join(method_parts)
+    if used_phone_backfill_firm:
+        method = "手机号回填律所|" + method
+    elif inferred_from_address:
+        method = "地址推断律所|" + method
+    return matched_firm, method, chosen
+
+
+def autosize_columns(ws) -> None:
+    for column_cells in ws.columns:
+        values = [str(cell.value or "") for cell in column_cells]
+        max_length = min(max((len(value) for value in values), default=0), 60)
+        column_letter = column_cells[0].column_letter
+        ws.column_dimensions[column_letter].width = max_length + 2
+
+
+def iter_input_rows(ws) -> Iterable[Tuple[int, List[object]]]:
+    for row_idx in range(1, ws.max_row + 1):
+        yield row_idx, [ws.cell(row_idx, col_idx).value for col_idx in range(1, 8)]
+
+
+def build_output(input_path: str, output_path: str) -> Dict[str, int]:
+    workbook = load_workbook(input_path)
+    source_ws = workbook.active
+
+    firm_index, ordered_firms, phone_index = load_db_indexes()
+
+    out_wb = Workbook()
+    out_ws = out_wb.active
+    out_ws.title = "firm_phone_compare"
+    headers = [
+        "原始行号",
+        "原姓名",
+        "原手机号",
+        "原律所",
+        "原省份",
+        "原城市",
+        "原地址",
+        "原备注",
+        "手机号命中记录数",
+        "手机号命中手机号",
+        "手机号补全姓名",
+        "手机号补全律所",
+        "手机号补全来源",
+        "手机号候选姓名",
+        "手机号候选律所",
+        "用于匹配的律所",
+        "匹配方式",
+        "数据库候选手机号",
+        "候选数量",
+        "原手机号对比",
+        "数据库候选姓名",
+        "数据库候选省市",
+        "数据库来源",
+    ]
+    out_ws.append(headers)
+    for cell in out_ws[1]:
+        cell.font = Font(bold=True)
+
+    stats = defaultdict(int)
+    for row_idx, row in iter_input_rows(source_ws):
+        name, original_phone, law_firm, province, city, address, remark = row
+        needs_phone_completion = not normalize_firm(law_firm)
+        phone_backfill = build_phone_backfill(
+            original_phone=original_phone,
+            name=name,
+            province=province,
+            city=city,
+            phone_index=phone_index,
+        )
+        matched_firm, method, matched_records = match_row(
+            name=name,
+            original_phone=original_phone,
+            law_firm=law_firm,
+            province=province,
+            city=city,
+            address=address,
+            phone_backfill=phone_backfill,
+            firm_index=firm_index,
+            ordered_firms=ordered_firms,
+        )
+        candidate_phones = unique_phones(matched_records)
+        compare = compare_result(split_phones(original_phone), candidate_phones)
+        candidate_names = unique_values(matched_records, "name")
+        candidate_domains = unique_values(matched_records, "domain")
+        city_province_pairs = []
+        seen_pairs = set()
+        for record in matched_records:
+            pair = f"{record.province}-{record.city}".strip("-")
+            if pair and pair not in seen_pairs:
+                seen_pairs.add(pair)
+                city_province_pairs.append(pair)
+
+        out_ws.append(
+            [
+                row_idx,
+                name or "",
+                original_phone or "",
+                law_firm or "",
+                province or "",
+                city or "",
+                address or "",
+                remark or "",
+                len(phone_backfill.records) if needs_phone_completion else "",
+                " / ".join(phone_backfill.matched_phones) if needs_phone_completion else "",
+                phone_backfill.best_name if needs_phone_completion else "",
+                phone_backfill.best_law_firm if needs_phone_completion else "",
+                phone_backfill.best_domain if needs_phone_completion else "",
+                " / ".join(phone_backfill.candidate_names) if needs_phone_completion else "",
+                " / ".join(phone_backfill.candidate_firms) if needs_phone_completion else "",
+                matched_firm or "",
+                method or "",
+                " / ".join(candidate_phones) or "",
+                len(candidate_phones),
+                compare,
+                " / ".join(candidate_names) or "",
+                " / ".join(city_province_pairs) or "",
+                " / ".join(candidate_domains) or "",
+            ]
+        )
+
+        if needs_phone_completion and phone_backfill.records:
+            stats["phone_backfill_hit_rows"] += 1
+        if needs_phone_completion and phone_backfill.best_name:
+            stats["phone_backfill_name_rows"] += 1
+        if needs_phone_completion and phone_backfill.best_law_firm:
+            stats["phone_backfill_firm_rows"] += 1
+        if needs_phone_completion and method.startswith("手机号回填律所|"):
+            stats["phone_backfill_used_for_match_rows"] += 1
+
+        if candidate_phones:
+            stats["matched_rows"] += 1
+            if len(candidate_phones) == 1:
+                stats["unique_rows"] += 1
+            else:
+                stats["multi_rows"] += 1
+        else:
+            stats["unmatched_rows"] += 1
+
+        if compare == "完全一致":
+            stats["same_rows"] += 1
+        elif compare == "候选包含原手机号":
+            stats["contains_rows"] += 1
+        elif compare == "不包含原手机号":
+            stats["diff_rows"] += 1
+        elif compare == "原手机号为空":
+            stats["blank_phone_rows"] += 1
+
+    out_ws.freeze_panes = "A2"
+    autosize_columns(out_ws)
+    out_wb.save(output_path)
+    return dict(stats)
+
+
+def main() -> None:
+    args = parse_args()
+    stats = build_output(args.input, args.output)
+    print(f"已生成: {args.output}")
+    for key in sorted(stats):
+        print(f"{key}={stats[key]}")
+
+
+if __name__ == "__main__":
+    main()

config.py

@@ -1,22 +1,114 @@
-# common_sites 独立项目配置
-
+# 数据库连接配置
 DB_CONFIG = {
-    "host": "8.134.219.222",
-    "user": "lawyer",
-    "password": "CTxr8yGwsSX3NdfJ",
-    "database": "lawyer",
+   "host": "8.134.219.222",  # 数据库地址
+    "user": "lawyer",  # 数据库用户名
+    "password": "CTxr8yGwsSX3NdfJ",  # 数据库密码
+    "database": "lawyer",  # 数据库名称
     "charset": "utf8mb4",
 }
 
-HEADERS = {
-    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/138.0.0.0 Safari/537.36",
-    "Accept": "*/*",
-    "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8,zh-TW;q=0.7",
-    "X-Requested-With": "XMLHttpRequest",
+# 高德地图 API 配置
+GAODE_CONFIG = {
+    "API_KEY": "f261575fb28003761c433f6c9379e89d",
 }
 
+# 微信爬虫特定的配置
+WEIXIN_CONFIG = {
+        "TOKEN": "553117235", # 您的Token
+        "FINGERPRINT": "3c02c35093184e9a9a668ac3c81e53f9",
+        "COOKIE": {
+            "appmsglist_action_3258147150": "card",
+            "_qimei_uuid42": "1a302160d051008226aec905b63f99ff3989f30009",
+            "_qimei_i_3": "63b22b84c15204dfc595ac6452d722b1f0bdf0f6145b568ae68a7c0e70947438686637943989e2a1d792",
+            "_qimei_h38": "215986ce26aec905b63f99ff0200000e81a302",
+            "ua_id": "S7gglu0eZh9NkAzLAAAAADH8dynpnFZVN29lxm7BQo0=",
+            "wxuin": "73074968761097",
+            "mm_lang": "zh_CN",
+            "eas_sid": "91X7I7K4K5k364U2z3k2I980F5",
+            "_qimei_q36": "",
+            "_qimei_fingerprint": "d895c46d5fda98cab67d9daec00068ed",
+            "_qimei_i_1": "4dc76680945f59d3c7c4ab325dd526b3feeea1a31458558bbdd97e582493206c6163629d39d8e1dcd4b2c28f",
+            "pgv_pvid": "6923507145",
+            "ts_uid": "9585717820",
+            "_t_qbtool_uid": "aaaa2vn5byd280l00iglw701zci788cb",
+            "_ga": "GA1.1.1323926288.1775838938",
+            "_ga_TPFW0KPXC1": "GS2.1.s1775841484$o2$g1$t1775841485$j59$l0$h0",
+            "uuid": "20d1cfb540221c6e7b6d665ab1d4a8f7",
+            "rand_info": "CAESIA8LYV6dvWh5dYrgQLPhZb8TXwUJoWdcdDzN0TTdztSj",
+            "slave_bizuin": "3258147150",
+            "data_bizuin": "3258147150",
+            "bizuin": "3258147150",
+            "data_ticket": "dgLFmSrI8f1q6JnYOd2Y/sKJIWjh6YlLSau1n1+Mv5iOTR5hgsm1qjNLypWflGd6",
+            "slave_sid": "VGVnNmM5NmFpV19ESElmVlZOTGZfVVJfWE5HanlHNjN0WEswZVkxVk9vc2FTenQzVGRsWUxDT0xGQVBJRVZzU0JNVV9RckRJVE9jSVUwbjl4Z2VHaEZKSzE5WVc3THRCRW96T0Z1V1VwbnBLSnkxSWdKaHdaN1dYdzI1SmdpZ0IyOFJtUE45OTR2Q2NvM1FB",
+            "slave_user": "gh_fe76760560d0",
+            "xid": "4893c62dc8518b6a1628fd34bc9aa276",
+            "_clck": "3258147150|1|g5g|0",
+            "_clsk": "1p4oo3h|1776957001796|5|1|mp.weixin.qq.com/weheat-agent/payload/record"
+            },
+        "COUNT": 20,  # 单页条数
+        "REQUESTS_PER_SECOND": 8,  # 每秒最大请求数（调高更快，但有风控风险）
+        "PAGE_DELAY": 0.8,  # 每页采集后的等待秒数
+        "CITY_DELAY": 0.3,  # 每城市采集后的等待秒数
+}
+
+# 通用请求头
+HEADERS = {
+    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/138.0.0.0 Safari/537.36',
+    'Accept': '*/*',
+    'Accept-Language': 'zh-CN,zh;q=0.9,en;q=0.8,zh-TW;q=0.7',
+    'X-Requested-With': 'XMLHttpRequest',
+}
+
+# 法律快车爬虫配置
 LAWTIME_CONFIG = {
     "HEADERS": {
         "User-Agent": "Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 Mobile/15E148 Safari/604.1"
     }
 }
+
+# Redis配置 - 用于采集索引和断点恢复
+REDIS_CONFIG = {
+    "host": "127.0.0.1",
+    "port": 6379,
+    "password": "",
+    "db": 0,  # 使用数据库0
+    "decode_responses": True,  # 自动解码响应
+    "socket_timeout": 5,  # 连接超时时间
+    "socket_connect_timeout": 5,  # 连接建立超时时间
+    "health_check_interval": 30,  # 健康检查间隔
+    "retry_on_timeout": True,  # 超时重试
+    "max_connections": 20,  # 最大连接数
+}
+
+# Redis键名配置
+REDIS_KEYS = {
+    "spider_progress": "lawyer:spider:progress:{spider_name}",  # 爬虫进度
+    "url_processed": "lawyer:url:processed:{spider_name}",      # 已处理URL集合
+    "url_failed": "lawyer:url:failed:{spider_name}",            # 失败URL集合
+    "spider_stats": "lawyer:stats:{spider_name}",               # 爬虫统计信息
+    "global_stats": "lawyer:global:stats",                      # 全局统计
+    "session_info": "lawyer:session:{session_id}",              # 会话信息
+    "url_queue": "lawyer:queue:{spider_name}",                  # URL队列
+    "duplicate_filter": "lawyer:duplicate:{spider_name}",       # 去重过滤器
+}
+
+# MongoDB配置 - 用于日志存储
+MONGO_CONFIG = {
+    "uri": "mongodb://127.0.0.1:27017/",
+    "database": "lawyer",
+    "collections": {
+        "logs": "logs",                    # 通用日志
+        "spider_logs": "spider_logs",      # 爬虫专用日志
+        "error_logs": "error_logs",        # 错误日志
+        "system_logs": "system_logs",      # 系统日志
+        "performance_logs": "performance_logs"  # 性能日志
+    },
+    "options": {
+        "maxPoolSize": 10,                 # 连接池最大连接数
+        "minPoolSize": 1,                  # 连接池最小连接数
+        "maxIdleTimeMS": 30000,            # 最大空闲时间
+        "serverSelectionTimeoutMS": 5000,  # 服务器选择超时
+        "connectTimeoutMS": 10000,         # 连接超时
+        "socketTimeoutMS": 30000,          # Socket超时
+    }
+}

docker-compose.yml

@@ -0,0 +1,14 @@
+services:
+  mongodb:
+    image: mongo:7
+    container_name: lawyers_mongodb
+    restart: always
+    ports:
+      - "27017:27017"
+    volumes:
+      - mongodb_data:/data/db
+    environment:
+      MONGO_INITDB_DATABASE: lawyer
+
+volumes:
+  mongodb_data:

gaode.py

@@ -0,0 +1,401 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+import os
+import sys
+import time
+import json
+import math
+import logging
+import re
+from typing import Dict, List, Optional
+from urllib.parse import urlencode
+
+import requests
+from requests.adapters import HTTPAdapter
+from requests.exceptions import RequestException, HTTPError, ConnectionError, Timeout
+from urllib3.util.retry import Retry
+
+# 添加项目根目录到系统路径（保留你的原逻辑）
+current_dir = os.path.dirname(os.path.abspath(__file__))
+project_root = os.path.dirname(current_dir)
+if project_root not in sys.path:
+    sys.path.append(project_root)
+
+from Db import Db  # 你的 DB 封装
+import config as project_config
+
+# logging 配置
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s %(levelname)s %(message)s",
+)
+logger = logging.getLogger(__name__)
+
+
+class GaodeSpider:
+    """高德地图 API 商户手机号采集 - 重构版"""
+
+    def __init__(
+        self,
+        db_connection,
+        api_key: Optional[str] = None,
+        offset: int = 20,
+        max_pages_per_city: int = 10,
+        sleep_between_pages: float = 2.0,
+        sleep_between_cities: float = 3.0,
+    ):
+        self.db = db_connection
+        config_api_key = ""
+        gaode_config = getattr(project_config, "GAODE_CONFIG", None)
+        if isinstance(gaode_config, dict):
+            config_api_key = str(gaode_config.get("API_KEY", "")).strip()
+
+        self.api_key = (api_key or os.environ.get("AMAP_API_KEY", "") or config_api_key).strip()
+        if not self.api_key:
+            raise ValueError("高德 API Key 未配置，请在 config.py 的 GAODE_CONFIG.API_KEY 或环境变量 AMAP_API_KEY 中填写")
+        self.api_base = "https://restapi.amap.com/v3/place/text"
+        self.offset = offset
+        self.session = self._build_session()
+        self.max_pages_per_city = max_pages_per_city
+        self.sleep_between_pages = sleep_between_pages
+        self.sleep_between_cities = sleep_between_cities
+
+        # 加载地区数据
+        self.cities = self._load_area_data()
+
+    def _build_session(self) -> requests.Session:
+        s = requests.Session()
+        # Retry for idempotent errors (GET) and some server errors
+        retries = Retry(
+            total=3,
+            backoff_factor=1,
+            status_forcelist=(429, 500, 502, 503, 504),
+            allowed_methods=frozenset(["GET", "POST"])
+        )
+        adapter = HTTPAdapter(max_retries=retries)
+        s.mount("https://", adapter)
+        s.mount("http://", adapter)
+        s.headers.update({
+            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+        })
+        return s
+
+    def _load_area_data(self) -> Dict[int, Dict]:
+        """从数据库加载地区数据（保持与你原来表结构兼容）。
+        要求 area_new 表含：id, code, city, province, pid, pinyin, domain, level
+        仅加载 domain='findlaw' 且 level=2 的城市
+        返回字典：{ city_id: {code, name, province, pid, pinyin} }
+        """
+        try:
+            rows = self.db.select_data("area_new", "id, code, city, province, pid, pinyin", "domain='maxlaw' AND level=2")
+            result = {}
+            for r in rows:
+                cid = r.get("id")
+                result[cid] = {
+                    "code": r.get("code") or "",
+                    "name": r.get("city") or "",
+                    "province": r.get("province") or "",
+                    "pid": r.get("pid"),
+                    "pinyin": r.get("pinyin") or ""
+                }
+            logger.info("加载城市数量: %d", len(result))
+            return result
+        except Exception as e:
+            logger.exception("从数据库加载地区数据失败: %s", e)
+            return {}
+
+    def _search_gaode_api(self, keywords: str, city: str, page: int = 1) -> Dict:
+        """调用高德 API 搜索，返回整个 JSON 响应（或空 dict）"""
+        params = {
+            "keywords": keywords,
+            "city": city,
+            "offset": self.offset,
+            "page": page,
+            "key": self.api_key,
+            "extensions": "all"
+        }
+        print(params)
+        try:
+            resp = self.session.get(self.api_base, params=params, timeout=15)
+            resp.raise_for_status()
+            data = resp.json()
+            return data
+        except (HTTPError, ConnectionError, Timeout) as e:
+            logger.warning("高德 API 请求失败(%s %s page=%s): %s", keywords, city, page, e)
+            return {}
+        except ValueError as e:
+            logger.error("高德 API 返回非 JSON 数据: %s", e)
+            return {}
+
+    def _split_and_clean_phones(self, raw_tel: str) -> List[str]:
+        """把 raw tel 拆成候选号码并清洗"""
+        if not raw_tel:
+            return []
+        
+        logger.debug("原始电话号码: %s", raw_tel)
+        
+        # 常见分隔符 ; / ， 、 | 空格
+        parts = re.split(r"[;,/，、\|]+|\s+", raw_tel.strip())
+        cleaned = []
+        
+        for p in parts:
+            if not p:
+                continue
+                
+            original_p = p
+            # 移除括号内内容以及非数字和连字符
+            p = re.sub(r"（.*?）|\(.*?\)|[^\d\-+]", "", p)
+            # 有的号码含国际码 +86
+            p = p.lstrip("+")
+            # 移除前导 86（如果之后是11位）
+            if p.startswith("86") and len(p) > 11:
+                p = p[2:]
+            # 最后移除短横线
+            p = p.replace("-", "")
+            
+            if p:
+                cleaned.append(p)
+                logger.debug("清洗后号码: %s -> %s", original_p, p)
+        
+        logger.debug("清洗后共 %d 个号码: %s", len(cleaned), cleaned)
+        return cleaned
+
+    def _is_valid_phone(self, phone: str) -> bool:
+        """验证手机号码，必须为11位且以1开头的手机号。"""
+        if not phone:
+            return False
+        # 强制要求：11位且以1开头的手机号
+        if re.fullmatch(r"1\d{10}", phone):
+            return True
+        return False
+
+    def _extract_phones_from_poi(self, poi: Dict) -> List[str]:
+        """
+        从 POI 数据中提取所有候选电话号码。
+        优先查找 poi['business']['tel']（你的示例），并兼容早期可能的字段如 tel/phone。
+        返回去重后且通过校验的号码列表。
+        """
+        candidates = []
+
+        # 1) 优先查找 business.tel 字段（高德API的主要电话字段）
+        business = poi.get("business") or {}
+        tel = business.get("tel")
+        if tel:
+            logger.debug("从 business.tel 提取: %s", tel)
+            candidates.extend(self._split_and_clean_phones(str(tel)))
+
+        # 2) 兼容旧结构：顶层 tel/phone/contact 等
+        for key in ("tel", "phone", "contact", "business_area"):
+            v = poi.get(key)
+            if v:
+                logger.debug("从 %s 提取: %s", key, v)
+                candidates.extend(self._split_and_clean_phones(str(v)))
+
+        # 3) 审慎兼容：一些扩展字段可能也包含电话
+        for nested_key in ("biz_ext", "ext", "attributes"):
+            nested = poi.get(nested_key) or {}
+            if isinstance(nested, dict):
+                for subkey in ("tel", "phone", "contact"):
+                    if nested.get(subkey):
+                        logger.debug("从 %s.%s 提取: %s", nested_key, subkey, nested.get(subkey))
+                        candidates.extend(self._split_and_clean_phones(str(nested.get(subkey))))
+
+        # 去重并只保留合法号码（按 _is_valid_phone）
+        unique = []
+        for c in candidates:
+            if c not in unique and self._is_valid_phone(c):
+                unique.append(c)
+                logger.debug("有效电话号码: %s", c)
+        
+        logger.debug("POI %s 提取到 %d 个有效电话号码", poi.get("name", ""), len(unique))
+        return unique
+
+
+    def _is_duplicate(self, phone: str) -> bool:
+        """检查某个 phone 是否已存在（domain='高德地图'）"""
+        try:
+            condition = f"phone='{phone}' AND domain='高德地图'"
+            exists = self.db.is_data_exist("lawyer", condition)
+            if exists:
+                logger.debug("手机号已存在: %s (domain=高德地图)", phone)
+            return exists
+        except Exception as e:
+            logger.exception("去重检查失败: %s", e)
+            # 若出错，返回 True 以避免重复插入或脏数据
+            return True
+
+    def _parse_poi_to_record(self, poi: Dict, city_info: Dict, province_info: Dict, used_phone: str) -> Dict:
+        """把单条 poi 转为数据库记录（针对某个已选的号码 used_phone）"""
+        # 安全处理 shopinfo 字段，可能是字符串或字典
+        shopinfo = poi.get("shopinfo")
+        if isinstance(shopinfo, dict):
+            law_firm = shopinfo.get("shop_name", "高德搜索")
+        else:
+            law_firm = "高德搜索"
+        
+        record = {
+            "name": poi.get("name", "").strip(),
+            "phone": used_phone,
+            "law_firm": law_firm,
+            "province": province_info.get("name", ""),
+            "city": city_info.get("name", ""),
+            "url": poi.get("website", "") or "",
+            "domain": "高德地图",
+            "create_time": int(time.time()),
+            "params": json.dumps({
+                "address": poi.get("address", ""),
+                "location": poi.get("location", ""),
+                "type": poi.get("type", ""),
+                "business_area": poi.get("business_area", ""),
+                "raw_tel": poi.get("tel", "") or "",
+                "raw_poi": poi
+            }, ensure_ascii=False)
+        }
+        return record
+
+    def _save_lawyer(self, record: Dict) -> bool:
+        """存储律师信息到数据库"""
+        try:
+            self.db.insert_data("lawyer", record)
+            logger.info("新增商户: %s (%s)", record.get("name"), record.get("phone"))
+            return True
+        except Exception as e:
+            logger.exception("存储失败: %s %s", record.get("name"), record.get("phone"))
+            return False
+
+    def _search_city(self, keywords: str, city_info: Dict, province_info: Dict) -> int:
+        """在指定城市搜索并存储；返回新增条数"""
+        # city 参数可以是 city code 或 city name；使用你存表里的 code 优先
+        # city_code = city_info.get("code") or city_info.get("name")
+        city_code = city_info.get("name")
+        total_added = 0
+
+        # 先请求第一页拿到 count
+        page = 1
+        first_resp = self._search_gaode_api(keywords, city_code, page)
+        if not first_resp:
+            logger.info("  未获取到第一页数据: %s", keywords)
+            return 0
+
+        status = first_resp.get("status")
+        if str(status) != "1":
+            logger.warning("  高德返回错误: %s", first_resp.get("info"))
+            return 0
+
+        try:
+            count = int(first_resp.get("count", 0))
+        except Exception:
+            count = 0
+        # 计算总页数
+        total_pages = math.ceil(count / self.offset) if count else 1
+        total_pages = min(total_pages, self.max_pages_per_city)
+
+        logger.info("  城市 %s 搜索到 count=%s, pages=%s (限制 %s)", city_code, count, total_pages, self.max_pages_per_city)
+
+        # 处理第一页 POIs
+        def process_page(page_num: int, page_data: Dict) -> int:
+            """处理单页数据，返回新增条数"""
+            nonlocal total_added
+            if not page_data:
+                logger.info("  page %s 未返回数据", page_num)
+                return 0
+            if str(page_data.get("status")) != "1":
+                logger.warning("  page %s 返回状态非1: %s", page_num, page_data.get("info"))
+                return 0
+
+            pois = page_data.get("pois") or []
+            page_added = 0
+            for poi in pois:
+                name = (poi.get("name") or "").strip()
+                if not name:
+                    continue
+                phones = self._extract_phones_from_poi(poi)
+                if not phones:
+                    logger.debug("    跳过无电话: %s", name)
+                    continue
+                for ph in phones:
+                    # 如果已存在跳过该号码
+                    if self._is_duplicate(ph):
+                        logger.debug("    跳过已存在号码: %s (%s)", name, ph)
+                        continue
+                    rec = self._parse_poi_to_record(poi, city_info, province_info, ph)
+                    ok = self._save_lawyer(rec)
+                    if ok:
+                        page_added += 1
+                        total_added += 1
+                        # 如果该 POI 有多个号码，我们仍尝试插入其它号码（有用时）
+                # end for phones
+            # end for pois
+            return page_added
+
+        # 先处理第一页
+        first_page_added = process_page(1, first_resp)
+        logger.info("  城市 %s 第 1 页新增 %d 条", city_code, first_page_added)
+        
+        # 记录已处理的页面，避免重复处理
+        processed_pages = {1}
+
+        # 依次请求剩余页面，直到读完或无数据
+        for page_num in range(2, total_pages + 1):
+            if page_num in processed_pages:
+                continue
+
+            time.sleep(self.sleep_between_pages)
+            page_data = self._search_gaode_api(keywords, city_code, page_num)
+
+            if not page_data:
+                logger.info("  第 %s 页无响应数据，停止翻页", page_num)
+                break
+
+            if str(page_data.get("status")) != "1":
+                logger.info("  第 %s 页状态异常(%s)，停止翻页", page_num, page_data.get("info"))
+                break
+
+            pois = page_data.get("pois") or []
+            if not pois:
+                logger.info("  第 %s 页返回空pois，提前结束", page_num)
+                break
+
+            page_added = process_page(page_num, page_data)
+            logger.info("  城市 %s 第 %s 页新增 %d 条", city_code, page_num, page_added)
+            processed_pages.add(page_num)
+
+            # 如果结果数量不足一页，说明已经接近尾部
+            if len(pois) < self.offset:
+                logger.info("  第 %s 页结果不足一页，推测已到尾页，提前结束", page_num)
+                break
+
+        return total_added
+
+    def run(self):
+        logger.info("启动高德地图律师信息采集...")
+        if not self.cities:
+            logger.error("未加载城市列表，退出")
+            return
+
+        total_stored = 0
+        keywords_suffix = "律师"
+
+        for city_id, city_info in self.cities.items():
+            try:
+                province_info = {"name": city_info.get("province", "")}
+                city_name = city_info.get("name", "")
+                if not city_name:
+                    continue
+                search_keywords = f"{keywords_suffix}"
+                added = self._search_city(search_keywords, city_info, province_info)
+                total_stored += added
+                logger.info("城市 %s 完成，新增 %d 条，总计 %d", city_name, added, total_stored)
+                time.sleep(self.sleep_between_cities)
+            except Exception as e:
+                logger.exception("处理城市 %s 时出错: %s", city_info.get("name", ""), e)
+
+        logger.info("采集完成，共新增 %d 条商户信息。", total_stored)
+
+
+if __name__ == "__main__":
+    # 运行示例
+    with Db() as db:
+        spider = GaodeSpider(db)
+        spider.run()

js/douyin.js

@@ -0,0 +1,832 @@
+// ==UserScript==
+// @name         Douyin Batch City Search + AutoScroll + Capture
+// @namespace    http://tampermonkey.net/
+// @version      1.1
+// @description  从 Python 服务获取地区列表，按 city + "律师" 搜索并自动下滑，拦截 /aweme/v1/web/discover/search/ 返回并转发到入库接口。
+// @author       You
+// @match        https://www.douyin.com/*
+// @grant        GM_xmlhttpRequest
+// @connect      *
+// @run-at       document-idle
+// ==/UserScript==
+
+(function () {
+    'use strict';
+  
+    /********************* 配置区（按需修改） *********************/
+    const API_BASE = 'http://127.0.0.1:9002'; // 改成你部署 Python 服务的地址，例如 http://nas.nepiedg.site:9002
+    const AREA_API = `${API_BASE}/api/layer/get_area?server=1`; // 获取城市列表的接口
+    const SEND_TARGETS = [
+      `${API_BASE}/api/layer/index?server=1&save_only=0`
+    ];
+  
+    // 搜索框与按钮选择器（根据页面更新）
+    const SEARCH_INPUT_SELECTORS = [
+      'input[data-e2e="search-input"]',
+      'input[data-e2e="searchbar-input"]',
+      'form[data-e2e="searchbar"] input',
+      'input[placeholder*="搜索"]'
+    ];
+    const SEARCH_BTN_SELECTORS = [
+      '[data-e2e="search-button"]',
+      'button[data-e2e="search-button"]',
+      'span[data-e2e="search-button"]',
+      'button[data-e2e="searchbar-button"]',
+      'span.btn-title'
+    ];
+  
+    // 每个城市搜索时的自动下滑配置
+    const SCROLL_INTERVAL_MS = 2000;
+    const MAX_STABLE_COUNT = 6;
+    const MAX_SCROLLS_PER_CITY = 120;
+    const SCROLL_BY = 2200;
+    const WAIT_AFTER_SEARCH_MS = 1000;
+    const DELAY_BETWEEN_CITIES_MS = 1500;
+
+    // 断点续跑配置
+    const PROGRESS_STORAGE_KEY = 'dm_batch_progress_v1';
+    const DEVICE_ID_STORAGE_KEY = 'dm_batch_device_id_v1';
+    const PROGRESS_SYNC_ENABLED = true;
+    const PROGRESS_KEY = 'douyin_batch_default';
+    const PROGRESS_API = `${API_BASE}/api/layer/progress?server=1`;
+  
+    // 可选：如果希望只发送包含手机号的条目，可在此启用并调整正则
+    const ONLY_SEND_IF_HAS_PHONE = false;
+    const PHONE_REGEX = /(?:\+?86)?1[3-9]\d{9}/g;
+  
+    /********************* 运行时状态 *********************/
+    let areaList = [];
+    let stopFlag = false; // 由 UI 控制，true 表示停止整个任务
+    let skipCurrentCityFlag = false; // 由 UI 控制，true 表示跳过当前城市
+    let currentCityIndex = -1;
+    let currentAreaSignature = '';
+    let isLoopRunning = false;
+    let inputEl = null;
+    let btnEl = null;
+    const DEVICE_ID = getOrCreateDeviceId();
+  
+    // 节流/去重发送
+    let lastSentHash = null;
+    let lastSentAt = 0;
+    const SEND_MIN_INTERVAL_MS = 800;
+    let progressSyncInFlight = false;
+    let progressSyncPendingPayload = null;
+  
+    /********************* 工具函数 *********************/
+    function log(...args) { console.log('[DouyinBatch] ', ...args); }
+    function err(...args) { console.error('[DouyinBatch] ', ...args); }
+  
+    function hashString(str) {
+      let h = 2166136261 >>> 0;
+      for (let i = 0; i < str.length; i++) {
+        h ^= str.charCodeAt(i);
+        h = Math.imul(h, 16777619) >>> 0;
+      }
+      return h.toString(16);
+    }
+
+    function sleep(ms) {
+      return new Promise(r => setTimeout(r, ms));
+    }
+
+    function getOrCreateDeviceId() {
+      try {
+        const old = localStorage.getItem(DEVICE_ID_STORAGE_KEY);
+        if (old) return old;
+        const generated = (window.crypto && typeof window.crypto.randomUUID === 'function')
+          ? window.crypto.randomUUID()
+          : `dm-${Date.now()}-${Math.random().toString(16).slice(2, 10)}`;
+        localStorage.setItem(DEVICE_ID_STORAGE_KEY, generated);
+        return generated;
+      } catch (_) {
+        return `dm-${Date.now()}-${Math.random().toString(16).slice(2, 10)}`;
+      }
+    }
+
+    function getAreaRowName(row) {
+      if (!row || typeof row !== 'object') return '';
+      return String(row.city || row.province || row.name || '').trim();
+    }
+
+    function buildAreaSignature(list) {
+      try {
+        if (!Array.isArray(list) || list.length === 0) return 'empty';
+        const names = list.map(getAreaRowName).filter(Boolean);
+        return hashString(`${list.length}|${names.join('|')}`);
+      } catch (e) {
+        return 'unknown';
+      }
+    }
+
+    function readProgress() {
+      try {
+        const raw = localStorage.getItem(PROGRESS_STORAGE_KEY);
+        if (!raw) return null;
+        const parsed = JSON.parse(raw);
+        if (!parsed || typeof parsed !== 'object') return null;
+        return parsed;
+      } catch (_) {
+        return null;
+      }
+    }
+
+    function buildProgressPayload(nextCityIndex, reason = '') {
+      const safeIndex = Number.isFinite(nextCityIndex) ? Math.max(0, Math.floor(nextCityIndex)) : 0;
+      const currentArea = areaList[safeIndex] || areaList[Math.max(0, currentCityIndex)] || {};
+      return {
+        progress_key: PROGRESS_KEY,
+        device_id: DEVICE_ID,
+        next_city_index: safeIndex,
+        area_signature: currentAreaSignature || '',
+        area_total: Array.isArray(areaList) ? areaList.length : 0,
+        current_city: getAreaRowName(currentArea),
+        reason,
+        status: stopFlag ? 'paused' : 'running',
+        extra: {
+          path: location.pathname || '',
+          href: location.href || '',
+        },
+      };
+    }
+
+    function persistProgress(nextCityIndex, reason = '') {
+      try {
+        const payload = buildProgressPayload(nextCityIndex, reason);
+        localStorage.setItem(PROGRESS_STORAGE_KEY, JSON.stringify({
+          nextCityIndex: payload.next_city_index,
+          areaSignature: payload.area_signature,
+          reason: payload.reason,
+          updatedAt: Date.now(),
+          progressKey: payload.progress_key,
+          deviceId: payload.device_id,
+        }));
+
+        enqueueRemoteProgressSync(payload);
+      } catch (e) {
+        err('保存进度失败', e);
+      }
+    }
+
+    function restoreProgress(areaSignature, listLength) {
+      const progress = readProgress();
+      if (!progress) return 0;
+      if (!progress.areaSignature || progress.areaSignature !== areaSignature) return 0;
+      const idx = Number.isFinite(progress.nextCityIndex) ? Math.floor(progress.nextCityIndex) : 0;
+      if (idx < 0 || idx >= listLength) return 0;
+      return idx;
+    }
+
+    function clearProgress() {
+      try { localStorage.removeItem(PROGRESS_STORAGE_KEY); } catch (_) {}
+      enqueueRemoteProgressSync({
+        action: 'clear',
+        progress_key: PROGRESS_KEY,
+        device_id: DEVICE_ID,
+      });
+    }
+  
+    function gmGetJson(url) {
+      return new Promise((resolve, reject) => {
+        GM_xmlhttpRequest({
+          method: 'GET',
+          url,
+          onload(res) {
+            try {
+              const json = JSON.parse(res.responseText);
+              resolve(json);
+            } catch (e) {
+              reject(e);
+            }
+          },
+          onerror(err) { reject(err); }
+        });
+      });
+    }
+
+    function gmPostJson(url, data) {
+      return new Promise((resolve, reject) => {
+        GM_xmlhttpRequest({
+          method: 'POST',
+          url,
+          headers: { 'Content-Type': 'application/json' },
+          data: JSON.stringify(data || {}),
+          onload(res) {
+            try {
+              const json = JSON.parse(res.responseText || '{}');
+              resolve(json);
+            } catch (e) {
+              reject(e);
+            }
+          },
+          onerror(err) { reject(err); }
+        });
+      });
+    }
+
+    function enqueueRemoteProgressSync(payload) {
+      if (!PROGRESS_SYNC_ENABLED) return;
+      if (!payload || typeof payload !== 'object') return;
+      progressSyncPendingPayload = payload;
+      if (progressSyncInFlight) return;
+      flushRemoteProgressSync();
+    }
+
+    async function flushRemoteProgressSync() {
+      if (!PROGRESS_SYNC_ENABLED) return;
+      if (progressSyncInFlight) return;
+
+      progressSyncInFlight = true;
+      try {
+        while (progressSyncPendingPayload) {
+          const payload = progressSyncPendingPayload;
+          progressSyncPendingPayload = null;
+          try {
+            await gmPostJson(PROGRESS_API, payload);
+          } catch (e) {
+            err('同步远端进度失败', e);
+            break;
+          }
+        }
+      } finally {
+        progressSyncInFlight = false;
+      }
+    }
+
+    async function restoreRemoteProgress(areaSignature, listLength) {
+      if (!PROGRESS_SYNC_ENABLED) return 0;
+      try {
+        const url = `${PROGRESS_API}&progress_key=${encodeURIComponent(PROGRESS_KEY)}`;
+        const response = await gmGetJson(url);
+        const data = response && response.data ? response.data : null;
+        if (!data || typeof data !== 'object') return 0;
+
+        const remoteSignature = String(data.area_signature || '');
+        if (!remoteSignature || remoteSignature !== areaSignature) return 0;
+
+        const idxRaw = data.next_city_index;
+        const idx = Number.isFinite(idxRaw) ? Math.floor(idxRaw) : Math.floor(Number(idxRaw || 0));
+        if (!Number.isFinite(idx) || idx < 0 || idx >= listLength) return 0;
+        return idx;
+      } catch (e) {
+        err('读取远端进度失败', e);
+        return 0;
+      }
+    }
+  
+    function setNativeValue(el, value) {
+      if (!el) return;
+      const prototype = el.constructor && el.constructor.prototype ? el.constructor.prototype : window.HTMLInputElement && window.HTMLInputElement.prototype;
+      const descriptor = prototype ? Object.getOwnPropertyDescriptor(prototype, 'value') : null;
+      if (descriptor && descriptor.set) {
+        descriptor.set.call(el, value);
+      } else {
+        el.value = value;
+      }
+    }
+  
+    async function simulateSearchInput(keyword) {
+      if (!inputEl) return;
+      try {
+        inputEl.focus();
+        inputEl.dispatchEvent(new Event('focus', { bubbles: false }));
+  
+        // 清空旧值并触发事件
+        if (inputEl.value) {
+          setNativeValue(inputEl, '');
+          if (typeof InputEvent === 'function') {
+            inputEl.dispatchEvent(new InputEvent('input', { bubbles: true, inputType: 'deleteContentBackward', data: '' }));
+          } else {
+            inputEl.dispatchEvent(new Event('input', { bubbles: true }));
+          }
+        }
+  
+        setNativeValue(inputEl, keyword);
+        if (typeof InputEvent === 'function') {
+          inputEl.dispatchEvent(new InputEvent('beforeinput', { bubbles: true, inputType: 'insertText', data: keyword }));
+          inputEl.dispatchEvent(new InputEvent('input', { bubbles: true, inputType: 'insertText', data: keyword }));
+        } else {
+          inputEl.dispatchEvent(new Event('input', { bubbles: true }));
+        }
+        inputEl.dispatchEvent(new Event('change', { bubbles: true }));
+        inputEl.dispatchEvent(new Event('blur', { bubbles: false }));
+      } catch (e) {
+        err('simulateSearchInput error', e);
+      }
+      await new Promise(r => setTimeout(r, 80));
+    }
+  
+    function simulateSearchTrigger() {
+      let triggered = false;
+      if (btnEl && btnEl.isConnected) {
+        try {
+          btnEl.focus();
+          btnEl.dispatchEvent(new MouseEvent('mousedown', { bubbles: true, cancelable: true, view: window }));
+          btnEl.dispatchEvent(new MouseEvent('mouseup', { bubbles: true, cancelable: true, view: window }));
+          btnEl.dispatchEvent(new MouseEvent('click', { bubbles: true, cancelable: true, view: window }));
+          triggered = true;
+        } catch (e) {
+          err('simulateSearchTrigger click error', e);
+        }
+      }
+  
+      if (!triggered && inputEl) {
+        try {
+          const opts = { bubbles: true, cancelable: true, key: 'Enter', code: 'Enter', keyCode: 13, which: 13 };
+          inputEl.dispatchEvent(new KeyboardEvent('keydown', opts));
+          inputEl.dispatchEvent(new KeyboardEvent('keypress', opts));
+          inputEl.dispatchEvent(new KeyboardEvent('keyup', opts));
+          triggered = true;
+        } catch (e) {
+          err('Enter 触发搜索失败', e);
+        }
+      }
+  
+      return triggered;
+    }
+  
+    function sendToTargets(data) {
+      try {
+        const body = typeof data === 'string' ? data : JSON.stringify(data);
+        if (ONLY_SEND_IF_HAS_PHONE) {
+          if (!PHONE_REGEX.test(body)) {
+            // 未匹配手机号则跳过发送
+            return;
+          }
+        }
+        const hash = hashString(body);
+        const now = Date.now();
+        if (hash === lastSentHash && now - lastSentAt < SEND_MIN_INTERVAL_MS) {
+          return;
+        }
+        lastSentHash = hash;
+        lastSentAt = now;
+  
+        for (const target of SEND_TARGETS) {
+          GM_xmlhttpRequest({
+            method: 'POST',
+            url: target,
+            headers: { 'Content-Type': 'application/json' },
+            data: body,
+            onload(res) { log(`sent -> ${target}, status: ${res.status}`); },
+            onerror(e) { err(`send error to ${target}`, e); }
+          });
+        }
+      } catch (e) {
+        err('sendToTargets error', e);
+      }
+    }
+  
+    /********************* 拦截 fetch 与 XHR（捕获目标接口返回） *********************/
+    const TARGET_PATH = '/aweme/v1/web/discover/search/';
+  
+    (function interceptFetch() {
+      if (!window.fetch) return;
+      const orig = window.fetch.bind(window);
+      window.fetch = function (...args) {
+        try {
+          const resource = args[0];
+          const url = (typeof resource === 'string') ? resource : (resource && resource.url) ? resource.url : '';
+          if (url && url.includes(TARGET_PATH)) {
+            return orig(...args).then((response) => {
+              try {
+                const cloned = response.clone();
+                cloned.json().then((json) => {
+                  if (json && typeof json === 'object') {
+                    sendToTargets({ source: 'fetch', url, data: json, ts: Date.now(), cityIndex: currentCityIndex });
+                  }
+                }).catch(()=>{});
+              } catch (e) { /* ignore */ }
+              return response;
+            });
+          }
+        } catch (e) { err('fetch wrapper error', e); }
+        return orig(...args);
+      };
+    })();
+  
+    (function interceptXHR() {
+      const XHR = window.XMLHttpRequest;
+      if (!XHR) return;
+      const origOpen = XHR.prototype.open;
+      const origSend = XHR.prototype.send;
+  
+      XHR.prototype.open = function (method, url, ...rest) {
+        try { this.__dm_url = (typeof url === 'string') ? url : ''; } catch(e){}
+        return origOpen.apply(this, [method, url, ...rest]);
+      };
+  
+      XHR.prototype.send = function (body) {
+        try {
+          const targetUrl = this.__dm_url || '';
+          if (targetUrl && targetUrl.includes(TARGET_PATH)) {
+            this.addEventListener('readystatechange', function () {
+              if (this.readyState === 4) {
+                try {
+                  const text = this.responseText;
+                  if (!text) return;
+                  try {
+                    const json = JSON.parse(text);
+                    sendToTargets({ source: 'xhr', url: targetUrl, data: json, ts: Date.now(), cityIndex: currentCityIndex });
+                  } catch (err) {
+                    // 非 json 忽略
+                  }
+                } catch (e) { /* ignore */ }
+              }
+            });
+          }
+        } catch (e) { err('XHR wrapper error', e); }
+        return origSend.apply(this, [body]);
+      };
+    })();
+  
+    /********************* 自动下滑函数（单次搜索） *********************/
+    async function autoScrollUntilStable(statusNode, maxScrolls = MAX_SCROLLS_PER_CITY) {
+      let lastHeight = -1;
+      let stableCount = 0;
+      let scrolls = 0;
+  
+      while (!stopFlag) {
+        if (skipCurrentCityFlag) {
+          statusNode.textContent = '收到跳过指令，结束当前地区滚动。';
+          break;
+        }
+
+        scrolls++;
+        if (scrolls > maxScrolls) {
+          statusNode.textContent = `达到单次搜索最大滚动 ${maxScrolls}，停止本次自动下滑。`;
+          break;
+        }
+  
+        // 执行滚动
+        try {
+          window.scrollBy({ top: SCROLL_BY, left: 0, behavior: 'smooth' });
+        } catch (e) {
+          window.scrollTo(0, (document.body.scrollHeight || document.documentElement.scrollHeight));
+        }
+  
+        await sleep(SCROLL_INTERVAL_MS);
+
+        if (skipCurrentCityFlag) {
+          statusNode.textContent = '收到跳过指令，结束当前地区滚动。';
+          break;
+        }
+  
+        const curHeight = document.body.scrollHeight || document.documentElement.scrollHeight || 0;
+        if (curHeight === lastHeight) {
+          stableCount++;
+        } else {
+          stableCount = 0;
+          lastHeight = curHeight;
+        }
+  
+        statusNode.textContent = `滚动次数: ${scrolls}, 稳定计数: ${stableCount}/${MAX_STABLE_COUNT}`;
+  
+        if (stableCount >= MAX_STABLE_COUNT) {
+          statusNode.textContent = `页面高度稳定 (${stableCount}), 本次搜索加载结束。`;
+          break;
+        }
+      }
+    }
+  
+    /********************* 页面元素辅助：等待元素出现 *********************/
+    function waitForSelector(selector, timeout = 10000) {
+      const selectors = Array.isArray(selector) ? selector.filter(Boolean) : [selector];
+      return new Promise((resolve, reject) => {
+        let timer;
+        const root = document.documentElement || document.body;
+  
+        const cleanup = (observer) => {
+          try { observer && observer.disconnect(); } catch (_) {}
+          if (timer) clearTimeout(timer);
+        };
+  
+        const pick = () => {
+          for (const sel of selectors) {
+            if (!sel) continue;
+            try {
+              const found = document.querySelector(sel);
+              if (found) {
+                return found;
+              }
+            } catch (e) {
+              err('query selector error', sel, e);
+            }
+          }
+          return null;
+        };
+  
+        const immediate = pick();
+        if (immediate) {
+          return resolve(immediate);
+        }
+  
+        const observer = new MutationObserver(() => {
+          const node = pick();
+          if (node) {
+            cleanup(observer);
+            resolve(node);
+          }
+        });
+  
+        if (root) {
+          observer.observe(root, { childList: true, subtree: true });
+        }
+  
+        timer = setTimeout(() => {
+          cleanup(observer);
+          reject(new Error('timeout waiting for ' + selectors.join(', ')));
+        }, timeout);
+      });
+    }
+  
+    async function ensureSearchControls(statusNode) {
+      const isConnected = (node) => {
+        if (!node) return false;
+        try {
+          if (node.isConnected !== undefined) return node.isConnected;
+          return document.contains(node);
+        } catch (_) {
+          return false;
+        }
+      };
+  
+      if (!isConnected(inputEl)) inputEl = null;
+      if (!isConnected(btnEl)) btnEl = null;
+  
+      if (!inputEl) {
+        statusNode && (statusNode.textContent = '等待搜索输入框可用...');
+        inputEl = await waitForSelector(SEARCH_INPUT_SELECTORS, 10000);
+      }
+  
+      if (!btnEl) {
+        try {
+          statusNode && (statusNode.textContent = '等待搜索按钮可用...');
+          btnEl = await waitForSelector(SEARCH_BTN_SELECTORS, 8000);
+          if (btnEl && btnEl.tagName !== 'BUTTON') {
+            const maybeButton = btnEl.closest('button');
+            if (maybeButton) btnEl = maybeButton;
+          }
+        } catch (e) {
+          btnEl = null;
+          err('未找到搜索按钮，将使用 Enter 键进行触发。');
+        }
+      }
+  
+      if (!inputEl) {
+        throw new Error('未定位到搜索输入框');
+      }
+  
+      return { inputEl, btnEl };
+    }
+  
+    /********************* UI 控制（右下角） *********************/
+    function createUI() {
+      const css = `
+        #dm-batch-btn { position: fixed; right: 12px; bottom: 12px; z-index:999999; background: rgba(0,0,0,0.65); color:#fff;
+          padding:8px 10px; border-radius:8px; font-size:13px; cursor:pointer; user-select:none;}
+        #dm-batch-skip { position: fixed; right:12px; bottom:50px; z-index:999999; background: rgba(30,30,30,0.72); color:#fff;
+          padding:7px 10px; border-radius:8px; font-size:12px; cursor:pointer; user-select:none;}
+        #dm-batch-status { position: fixed; right:12px; bottom:88px; z-index:999999; background: rgba(0,0,0,0.45); color:#fff;
+          padding:6px 8px; border-radius:6px; font-size:12px; max-width:320px; word-break:break-word;}
+      `;
+      const s = document.createElement('style'); s.textContent = css; document.head && document.head.appendChild(s);
+  
+      const btn = document.createElement('div');
+      btn.id = 'dm-batch-btn';
+      btn.textContent = 'BatchSearch：停止';
+      btn.dataset.running = '1';
+      document.body.appendChild(btn);
+
+      const skipBtn = document.createElement('div');
+      skipBtn.id = 'dm-batch-skip';
+      skipBtn.textContent = 'BatchSearch：跳过当前';
+      document.body.appendChild(skipBtn);
+  
+      const status = document.createElement('div');
+      status.id = 'dm-batch-status';
+      status.textContent = '准备中...';
+      document.body.appendChild(status);
+  
+      btn.addEventListener('click', () => {
+        const running = btn.dataset.running === '1';
+        btn.dataset.running = running ? '0' : '1';
+        btn.textContent = running ? 'BatchSearch：已停止' : 'BatchSearch：停止';
+        status.textContent = running ? '已手动停止（已保存断点）' : '已启动';
+        stopFlag = running; // if was running and clicked -> set stopFlag true; if restarting, set false
+        if (running) {
+          skipCurrentCityFlag = false;
+          persistProgress(Math.max(currentCityIndex, 0), 'manual_pause');
+        }
+        if (!stopFlag) {
+          // restart loop if needed
+          runBatchSearchLoop(status).catch(e => err(e));
+        }
+      });
+
+      skipBtn.addEventListener('click', () => {
+        if (currentCityIndex < 0) {
+          status.textContent = '当前还未开始处理城市，稍后再跳过。';
+          return;
+        }
+        skipCurrentCityFlag = true;
+        const areaName = getAreaRowName(areaList[currentCityIndex] || {});
+        status.textContent = `收到跳过指令：${areaName || `索引${currentCityIndex}`}`;
+      });
+
+      skipBtn.addEventListener('contextmenu', (event) => {
+        event.preventDefault();
+        clearProgress();
+        currentCityIndex = 0;
+        status.textContent = '已清除断点。下次将从第 1 个地区开始。';
+      });
+  
+      return { btn, skipBtn, status };
+    }
+  
+    /********************* 主流程：获取城市并循环搜索 *********************/
+    async function runBatchSearchLoop(statusNode) {
+      if (isLoopRunning) {
+        statusNode.textContent = '批量任务已在运行中，请勿重复启动。';
+        return;
+      }
+
+      isLoopRunning = true;
+      try {
+        stopFlag = (document.getElementById('dm-batch-btn') && document.getElementById('dm-batch-btn').dataset.running === '0');
+        skipCurrentCityFlag = false;
+
+        if (stopFlag) {
+          statusNode.textContent = '当前是暂停状态，点击“BatchSearch：停止”可继续。';
+          return;
+        }
+
+        // 获取 area list（仅在内存为空时获取）
+        if (!areaList || !Array.isArray(areaList) || areaList.length === 0) {
+          statusNode.textContent = '正在获取城市列表...';
+          try {
+            const data = await gmGetJson(AREA_API);
+            const normalizedAreaList = Array.isArray(data)
+              ? data
+              : (data && Array.isArray(data.data) ? data.data : []);
+
+            if (normalizedAreaList.length > 0) {
+              areaList = normalizedAreaList;
+              log('获取城市列表数量:', areaList.length);
+              statusNode.textContent = `获取到 ${areaList.length} 个城市，准备开始循环。`;
+            } else {
+              err('area API returned not array', data);
+              statusNode.textContent = '获取城市列表失败（返回格式异常）';
+              return;
+            }
+          } catch (e) {
+            err('获取城市列表失败', e);
+            statusNode.textContent = '获取城市列表失败: ' + e.message;
+            return;
+          }
+        }
+
+        currentAreaSignature = buildAreaSignature(areaList);
+        const restoredIndexLocal = restoreProgress(currentAreaSignature, areaList.length);
+        const restoredIndexRemote = await restoreRemoteProgress(currentAreaSignature, areaList.length);
+        const restoredIndex = Math.max(restoredIndexLocal, restoredIndexRemote);
+        const startIndex = (currentCityIndex >= 0 && currentCityIndex < areaList.length)
+          ? currentCityIndex
+          : restoredIndex;
+        currentCityIndex = startIndex;
+
+        if (startIndex > 0) {
+          statusNode.textContent = `检测到断点（本地:${restoredIndexLocal + 1} 远端:${restoredIndexRemote + 1}），将从第 ${startIndex + 1}/${areaList.length} 个地区继续。`;
+          await sleep(500);
+        }
+  
+        // 等待搜索输入与按钮可用
+        try {
+          await ensureSearchControls(statusNode);
+        } catch (e) {
+          err('未找到搜索输入或按钮', e);
+          statusNode.textContent = '未找到搜索输入或按钮，脚本仍会监听接口，但无法自动搜索。';
+          return;
+        }
+
+        let completedAll = true;
+  
+        // 主循环：对每个 city 执行搜索 -> 下滑 -> 发送结果 -> 下一 city
+        for (let i = startIndex; i < areaList.length; i++) {
+          if (stopFlag) {
+            completedAll = false;
+            persistProgress(i, 'manual_stop');
+            statusNode.textContent = '已停止（断点已保存）。';
+            break;
+          }
+
+          currentCityIndex = i;
+          skipCurrentCityFlag = false;
+          persistProgress(i, 'start_city');
+
+          const city = (areaList[i].city || areaList[i].province || '').trim();
+          if (!city) {
+            persistProgress(i + 1, 'empty_city');
+            continue;
+          }
+
+          const keyword = `${city}律师`;
+          statusNode.textContent = `正在搜索：${keyword} （${i+1}/${areaList.length}）`;
+          log(`开始城市[${i+1}/${areaList.length}] 搜索:`, keyword);
+  
+          // 将搜索词放入输入框 (触发 input 事件)
+          try {
+            await ensureSearchControls(statusNode);
+          } catch (e) {
+            err('刷新搜索控件失败', e);
+            statusNode.textContent = '刷新搜索控件失败，终止批量搜索。';
+            completedAll = false;
+            persistProgress(i, 'search_control_error');
+            break;
+          }
+  
+          await simulateSearchInput(keyword);
+  
+          const triggered = simulateSearchTrigger();
+          if (!triggered) {
+            statusNode.textContent = '搜索触发失败，尝试刷新控件...';
+            btnEl = null;
+            await ensureSearchControls(statusNode);
+            if (!simulateSearchTrigger()) {
+              statusNode.textContent = '搜索触发失败，终止批量搜索。';
+              completedAll = false;
+              persistProgress(i, 'search_trigger_error');
+              break;
+            }
+          }
+  
+          // 等待搜索结果开始加载
+          await new Promise(r => setTimeout(r, WAIT_AFTER_SEARCH_MS));
+  
+          // 自动下滑直到稳定或达到上限
+          await autoScrollUntilStable(statusNode, MAX_SCROLLS_PER_CITY);
+
+          if (skipCurrentCityFlag) {
+            skipCurrentCityFlag = false;
+            persistProgress(i + 1, 'skip_city');
+            statusNode.textContent = `已跳过 ${keyword}，继续下一个地区...`;
+            await sleep(Math.min(DELAY_BETWEEN_CITIES_MS, 800));
+            continue;
+          }
+  
+          if (stopFlag) {
+            completedAll = false;
+            persistProgress(i, 'manual_stop_after_scroll');
+            statusNode.textContent = '已停止（断点已保存）。';
+            break;
+          }
+
+          persistProgress(i + 1, 'city_done');
+  
+          // 等待短暂间隔再进行下一个城市
+          statusNode.textContent = `完成 ${keyword} 的加载，等待 ${DELAY_BETWEEN_CITIES_MS} ms 后继续...`;
+          await sleep(DELAY_BETWEEN_CITIES_MS);
+        }
+
+        if (completedAll && !stopFlag) {
+          clearProgress();
+          currentCityIndex = -1;
+          statusNode.textContent = '批量搜索完成，已清除断点进度。';
+          log('批量搜索循环结束: completed');
+        } else {
+          log('批量搜索循环结束: paused/broken');
+        }
+      } catch (e) {
+        err('runBatchSearchLoop error', e);
+        persistProgress(Math.max(currentCityIndex, 0), 'loop_exception');
+      } finally {
+        isLoopRunning = false;
+      }
+    }
+  
+    /********************* 启动脚本 *********************/
+    (function init() {
+      window.addEventListener('beforeunload', () => {
+        if (currentCityIndex >= 0) {
+          persistProgress(Math.max(currentCityIndex, 0), 'page_unload');
+        }
+      });
+
+      const ui = createUI();
+      ui.status.textContent = '就绪 - 可暂停/跳过，自动保存断点（右键跳过按钮可清除断点）';
+      console.log(location.pathname)
+      // 如果当前为目标页面（/jingxuan/search/），则自动启动；否则仍可在任何页面打开并手动启动。
+      const isAutoPage = location.pathname && location.pathname.indexOf('/search/') !== -1;
+      if (isAutoPage) {
+        ui.status.textContent = '检测到 /jingxuan/search/ 页面，准备开始批量搜索...';
+        // 给页面一点时间加载必要脚本与 dom
+        setTimeout(() => {
+          runBatchSearchLoop(ui.status).catch(e => err(e));
+        }, 800);
+      } else {
+        // 非目标页面，仍可手动点击按钮（按钮初始化为运行状态，点击色变为已停止）
+        ui.status.textContent = '非 /jingxuan/search/ 页面。导航至该页面或手动控制开始。';
+      }
+    })();
+  
+  })();
+  

one_off_sites/zhongfali_group80_basic.py

@@ -0,0 +1,411 @@
+#!/usr/bin/env python3
+import argparse
+from concurrent.futures import ThreadPoolExecutor, as_completed
+import hashlib
+import json
+import os
+import re
+import subprocess
+import sys
+import time
+import urllib.request
+from typing import Dict, List, Optional, Set, Tuple
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+project_root = os.path.dirname(current_dir)
+if project_root not in sys.path:
+    sys.path.append(project_root)
+
+from Db import Db
+
+SITE_NAME = "zhongfali_group80"
+LEGACY_DOMAIN = "众法利单页"
+START_URL = "http://m.zhongfali.com/pg.jsp?groupId=80&pgt=0&pgs=1"
+DEFAULT_OUTPUT = "/www/wwwroot/lawyers/data/one_off_sites/zhongfali_records_all.jsonl"
+
+SOCKS_PROXY = "127.0.0.1:7891"
+CLASH_CONTROLLER = os.environ.get("CLASH_CONTROLLER", "http://127.0.0.1:9090")
+CLASH_SECRET = os.environ.get("CLASH_SECRET", "")
+PHONE_RE = re.compile(r"1[3-9]\d{9}")
+INITIAL_STATE_RE = re.compile(r"window\.__INITIAL_STATE__\s*=\s*(\{.*?\})</script>", re.S)
+
+
+class ProxyRotator:
+    def __init__(self, controller: str, secret: str):
+        self.controller = controller.rstrip("/")
+        self.secret = secret.strip()
+        self.nodes: List[str] = []
+        self.index = 0
+
+    def _api(self, path: str, method: str = "GET", payload: Optional[Dict] = None) -> Dict:
+        headers = {}
+        if self.secret:
+            headers["Authorization"] = f"Bearer {self.secret}"
+        body = None
+        if payload is not None:
+            headers["Content-Type"] = "application/json"
+            body = json.dumps(payload).encode("utf-8")
+        req = urllib.request.Request(
+            f"{self.controller}{path}",
+            data=body,
+            headers=headers,
+            method=method,
+        )
+        with urllib.request.urlopen(req, timeout=10) as resp:
+            raw = resp.read().decode("utf-8", errors="ignore")
+            return json.loads(raw) if raw else {}
+
+    def initialize(self) -> None:
+        if not self.secret:
+            return
+        try:
+            self._api("/configs", method="PATCH", payload={"mode": "global"})
+            proxy_data = self._api("/proxies")
+            proxies = proxy_data.get("proxies", {}) or {}
+            skip = {
+                "GLOBAL",
+                "DIRECT",
+                "REJECT",
+                "REJECT-DROP",
+                "PASS",
+                "COMPATIBLE",
+                "🔰 选择节点",
+                "☁️ OneDrive",
+                "🐟 漏网之鱼",
+                "🎯 全球直连",
+                "🛑 拦截广告",
+                "🌍 爱奇艺&哔哩哔哩",
+                "🎮 Steam 登录/下载",
+                "🎮 Steam 商店/社区",
+                "🌩️ Cloudflare",
+                "🎬 动画疯",
+                "🎓学术网站",
+                "🇨🇳 国内网站",
+            }
+            self.nodes = [
+                name
+                for name, info in proxies.items()
+                if name not in skip and isinstance(info, dict)
+                and info.get("type") not in {"Selector", "URLTest", "Fallback", "LoadBalance"}
+            ]
+            if self.nodes:
+                self.switch_to(self.nodes[0])
+        except Exception as exc:
+            print(f"[proxy] rotator init failed: {exc}")
+            self.nodes = []
+
+    def switch_to(self, node_name: str) -> None:
+        self._api("/proxies/GLOBAL", method="PUT", payload={"name": node_name})
+
+    def rotate(self) -> None:
+        if not self.nodes:
+            return
+        self.index = (self.index + 1) % len(self.nodes)
+        node = self.nodes[self.index]
+        self.switch_to(node)
+
+
+def normalize_phone(value: str) -> str:
+    compact = "".join(ch for ch in str(value or "") if ch.isdigit())
+    match = PHONE_RE.search(compact)
+    return match.group(0) if match else ""
+
+
+def fetch_html(
+    url: str,
+    rotator: Optional[ProxyRotator] = None,
+    max_retries: int = 6,
+    timeout_seconds: int = 18,
+) -> str:
+    last_error = ""
+    for attempt in range(max_retries):
+        cmd = [
+            "curl",
+            "-sS",
+            "--socks5-hostname",
+            SOCKS_PROXY,
+            "-L",
+            "--compressed",
+            "--max-time",
+            str(timeout_seconds),
+            "-w",
+            "\n__CODE__:%{http_code}",
+            url,
+        ]
+        proc = subprocess.run(cmd, capture_output=True)
+        if proc.returncode == 0:
+            raw = proc.stdout.decode("utf-8", errors="ignore")
+            marker = "\n__CODE__:"
+            split_at = raw.rfind(marker)
+            if split_at != -1:
+                text = raw[:split_at]
+                code_text = raw[split_at + len(marker):].strip()
+            else:
+                text = raw
+                code_text = ""
+            code_ok = code_text == "200" if code_text else bool(text)
+            if text and code_ok:
+                return text
+            last_error = "empty body"
+        else:
+            last_error = proc.stderr.decode("utf-8", errors="ignore").strip() or f"exit={proc.returncode}"
+        if rotator and rotator.nodes:
+            try:
+                rotator.rotate()
+            except Exception as exc:
+                last_error = f"{last_error}; rotate failed: {exc}"
+        if attempt < max_retries - 1:
+            time.sleep(0.6 * (attempt + 1))
+    raise RuntimeError(f"fetch failed: {url}, reason={last_error}")
+
+
+def parse_initial_state(html: str) -> Dict:
+    match = INITIAL_STATE_RE.search(html)
+    if not match:
+        raise ValueError("window.__INITIAL_STATE__ not found")
+    return json.loads(match.group(1))
+
+
+def extract_group_urls_from_group80(state: Dict) -> List[str]:
+    module = (state.get("currentPageModuleIdMap") or {}).get("21") or {}
+    ext_info = module.get("extInfo", {}) or {}
+    second_group_map = ext_info.get("secondGroupMap", {}) or {}
+    rows = second_group_map.get("80") or []
+
+    urls: Set[str] = set()
+    for row in rows:
+        url = str(row.get("url") or "").strip()
+        if url:
+            urls.add(url)
+        for city in row.get("thirdGroupList") or []:
+            city_url = str(city.get("url") or "").strip()
+            if city_url:
+                urls.add(city_url)
+    return sorted(urls)
+
+
+def extract_detail_urls_from_group_html(html: str) -> Set[str]:
+    detail_ids = set(re.findall(r"h-pd-(\d+)\.html", html))
+    return {f"http://m.zhongfali.com/h-pd-{pid}.html" for pid in detail_ids}
+
+
+def parse_location_and_name(product_name: str) -> Tuple[str, str, str]:
+    text = re.sub(r"\s+", " ", str(product_name or "")).strip()
+    province = ""
+    city = ""
+    name = ""
+
+    province_match = re.search(r"([\u4e00-\u9fa5]{2,}省)", text)
+    if province_match:
+        province = province_match.group(1)
+
+    city_match = re.search(r"(?:[\u4e00-\u9fa5]+省)?([\u4e00-\u9fa5]+(?:市|区|县|州|盟))", text)
+    if city_match:
+        city = city_match.group(1)
+
+    name_match = re.search(r"([\u4e00-\u9fa5]{2,4})\s*律师", text)
+    if name_match:
+        name = name_match.group(1)
+
+    return province, city, name
+
+
+def parse_detail_record(detail_url: str, html: str, source_list_url: str) -> Optional[Dict]:
+    state = parse_initial_state(html)
+
+    module = None
+    for mod in (state.get("currentPageModuleIdMap") or {}).values():
+        if isinstance(mod, dict) and (mod.get("extInfo") or {}).get("productInfo"):
+            module = mod
+            break
+    if not module:
+        return None
+
+    ext_info = module.get("extInfo", {}) or {}
+    product_info = ext_info.get("productInfo", {}) or {}
+
+    phone = normalize_phone(product_info.get("material", ""))
+    if not phone:
+        return None
+
+    product_name = str(product_info.get("name") or "").strip()
+    province, city, lawyer_name = parse_location_and_name(product_name)
+    law_firm = str(product_info.get("prop0") or "").strip()
+
+    if not lawyer_name:
+        lawyer_name = product_name
+
+    now = int(time.time())
+    record_id = hashlib.md5(detail_url.encode("utf-8")).hexdigest()
+    return {
+        "record_id": record_id,
+        "collected_at": now,
+        "source": {
+            "site": SITE_NAME,
+            "list_url": source_list_url,
+            "detail_url": detail_url,
+            "province": province,
+            "province_py": "",
+            "city": city,
+            "city_py": "",
+            "page": 1,
+        },
+        "list_snapshot": {
+            "name": lawyer_name,
+            "law_firm": law_firm,
+            "specialties": [],
+            "answer_count": None,
+        },
+        "profile": {
+            "name": lawyer_name,
+            "law_firm": law_firm,
+            "phone": phone,
+            "license_no": str(product_info.get("prop1") or "").strip(),
+            "practice_years": None,
+            "email": "",
+            "address": str(product_info.get("prop3") or "").strip(),
+            "specialties": [],
+        },
+        "raw": {
+            "product_name": product_name,
+            "group_ids": product_info.get("groupIdList") or [],
+        },
+    }
+
+
+def to_legacy_row(record: Dict) -> Optional[Dict[str, str]]:
+    profile = record.get("profile", {}) or {}
+    source = record.get("source", {}) or {}
+    phone = normalize_phone(profile.get("phone", ""))
+    if not phone:
+        return None
+
+    province = str(source.get("province") or "").strip()
+    city = str(source.get("city") or province).strip()
+    return {
+        "name": str(profile.get("name") or "").strip(),
+        "law_firm": str(profile.get("law_firm") or "").strip(),
+        "province": province,
+        "city": city,
+        "phone": phone,
+        "url": str(source.get("detail_url") or "").strip(),
+        "domain": LEGACY_DOMAIN,
+        "create_time": int(record.get("collected_at") or time.time()),
+        "params": json.dumps(record, ensure_ascii=False),
+    }
+
+
+def delete_old_domain_data(db: Db, domain: str) -> int:
+    cur = db.db.cursor()
+    try:
+        cur.execute("DELETE FROM lawyer WHERE domain=%s", (domain,))
+        affected = cur.rowcount
+        db.db.commit()
+        return affected
+    finally:
+        cur.close()
+
+
+def write_records_to_db(db: Db, records: List[Dict]) -> int:
+    inserted = 0
+    for record in records:
+        row = to_legacy_row(record)
+        if not row:
+            continue
+        try:
+            db.insert_data("lawyer", row)
+            inserted += 1
+        except Exception as exc:
+            print(f"[db] insert failed phone={row.get('phone', '')}: {exc}")
+    return inserted
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="众法利 groupId=80 基础字段采集（姓名/手机号/地区）")
+    parser.add_argument("--start-url", default=START_URL, help="入口分组页 URL")
+    parser.add_argument("--output", default=DEFAULT_OUTPUT, help="JSONL 输出路径")
+    parser.add_argument("--no-db", action="store_true", help="只写 JSON，不写 DB")
+    parser.add_argument("--no-reset", action="store_true", help="不清理 domain 旧数据")
+    parser.add_argument("--workers", type=int, default=16, help="详情页并发数")
+    return parser.parse_args()
+
+
+def main() -> None:
+    args = parse_args()
+    os.makedirs(os.path.dirname(args.output) or ".", exist_ok=True)
+
+    rotator = ProxyRotator(CLASH_CONTROLLER, CLASH_SECRET)
+    rotator.initialize()
+    if rotator.nodes:
+        print(f"[proxy] rotator enabled, nodes={len(rotator.nodes)}")
+    else:
+        print("[proxy] rotator disabled, using current proxy route")
+
+    start_retries = max(8, len(rotator.nodes) + 2) if rotator.nodes else 8
+    group_html = fetch_html(args.start_url, rotator=rotator, max_retries=start_retries)
+    group_state = parse_initial_state(group_html)
+    group_urls = extract_group_urls_from_group80(group_state)
+    print(f"[group] found group urls: {len(group_urls)}")
+
+    detail_url_to_source: Dict[str, str] = {}
+    for idx, rel_url in enumerate(group_urls, start=1):
+        list_url = f"http://m.zhongfali.com/{rel_url.lstrip('/')}"
+        try:
+            html = fetch_html(list_url, rotator=rotator, max_retries=4, timeout_seconds=12)
+            detail_urls = extract_detail_urls_from_group_html(html)
+        except Exception as exc:
+            print(f"[group] failed {list_url}: {exc}")
+            continue
+
+        for detail_url in detail_urls:
+            detail_url_to_source.setdefault(detail_url, list_url)
+        if idx % 10 == 0:
+            print(f"[group] {idx}/{len(group_urls)} detail_urls={len(detail_url_to_source)}")
+
+    records: List[Dict] = []
+    seen_phones: Set[str] = set()
+    detail_urls = sorted(detail_url_to_source.keys())
+    print(f"[detail] total detail urls: {len(detail_urls)}")
+
+    def process_detail(detail_url: str) -> Optional[Dict]:
+        try:
+            html = fetch_html(detail_url, rotator=rotator, max_retries=2, timeout_seconds=8)
+            record = parse_detail_record(detail_url, html, detail_url_to_source[detail_url])
+            return record
+        except Exception as exc:
+            print(f"[detail] failed {detail_url}: {exc}")
+            return None
+
+    done = 0
+    with ThreadPoolExecutor(max_workers=max(1, int(args.workers))) as executor:
+        futures = [executor.submit(process_detail, detail_url) for detail_url in detail_urls]
+        for future in as_completed(futures):
+            done += 1
+            record = future.result()
+            if record:
+                phone = normalize_phone((record.get("profile", {}) or {}).get("phone", ""))
+                if phone and phone not in seen_phones:
+                    seen_phones.add(phone)
+                    records.append(record)
+            if done % 50 == 0:
+                print(f"[detail] {done}/{len(detail_urls)} valid_records={len(records)}")
+
+    with open(args.output, "w", encoding="utf-8") as out:
+        for record in records:
+            out.write(json.dumps(record, ensure_ascii=False) + "\n")
+
+    deleted = 0
+    inserted = 0
+    if not args.no_db:
+        with Db() as db:
+            if not args.no_reset:
+                deleted = delete_old_domain_data(db, LEGACY_DOMAIN)
+            inserted = write_records_to_db(db, records)
+
+    print(
+        f"[done] records={len(records)}, db_deleted={deleted}, db_inserted={inserted}, output={args.output}"
+    )
+
+
+if __name__ == "__main__":
+    main()

one_off_sites/zhongfali_single.py

@@ -0,0 +1,501 @@
+#!/usr/bin/env python3
+import argparse
+import hashlib
+import json
+import os
+import re
+import sys
+import time
+from typing import Dict, List, Optional, Set, Tuple
+
+import urllib3
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+project_root = os.path.dirname(current_dir)
+request_dir = os.path.join(project_root, "request")
+if request_dir not in sys.path:
+    sys.path.insert(0, request_dir)
+if project_root not in sys.path:
+    sys.path.append(project_root)
+
+from Db import Db
+from request.requests_client import RequestClientError, RequestsClient
+
+urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
+
+SITE_NAME = "zhongfali_single"
+LEGACY_DOMAIN = "众法利单页"
+DEFAULT_URL = "http://m.zhongfali.com/h-pd-552.html#mid=3&groupId=196&desc=false"
+DEFAULT_OUTPUT = "/www/wwwroot/lawyers/data/one_off_sites/zhongfali_records_all.jsonl"
+
+PHONE_RE = re.compile(r"1[3-9]\d{9}")
+INITIAL_STATE_RE = re.compile(r"window\.__INITIAL_STATE__\s*=\s*(\{.*?\})</script>", re.S)
+
+
+def normalize_phone(text: str) -> str:
+    compact = re.sub(r"\D", "", text or "")
+    match = PHONE_RE.search(compact)
+    return match.group(0) if match else ""
+
+
+def split_specialties(text: str) -> List[str]:
+    source = (text or "").strip()
+    if not source:
+        return []
+    parts = [item.strip() for item in re.split(r"[、,，;；\s]+", source) if item.strip()]
+    seen: Set[str] = set()
+    result: List[str] = []
+    for item in parts:
+        if item in seen:
+            continue
+        seen.add(item)
+        result.append(item)
+    return result
+
+
+def strip_html(text: str) -> str:
+    cleaned = re.sub(r"<[^>]+>", " ", text or "")
+    cleaned = cleaned.replace("&nbsp;", " ")
+    cleaned = re.sub(r"\s+", " ", cleaned)
+    return cleaned.strip()
+
+
+def extract_specialties_from_remark(remark: str) -> List[str]:
+    plain = strip_html(remark)
+    if not plain:
+        return []
+
+    match = re.search(r"专业领域[:：]\s*([^。；]+)", plain)
+    if match:
+        return split_specialties(match.group(1))
+    return []
+
+
+def value_at(values: List[str], index: int) -> str:
+    if index < 0 or index >= len(values):
+        return ""
+    return str(values[index] or "").strip()
+
+
+def parse_initial_state(html: str) -> Dict:
+    match = INITIAL_STATE_RE.search(html)
+    if not match:
+        raise ValueError("未找到 window.__INITIAL_STATE__")
+    return json.loads(match.group(1))
+
+
+def extract_location_and_name(product_name: str) -> Tuple[str, str, str]:
+    text = re.sub(r"\s+", " ", product_name or "").strip()
+    province = ""
+    city = ""
+    lawyer_name = ""
+
+    province_match = re.search(r"([\u4e00-\u9fa5]{2,}省)", text)
+    city_match = re.search(r"(?:[\u4e00-\u9fa5]+省)?([\u4e00-\u9fa5]+市)", text)
+    name_match = re.search(r"([\u4e00-\u9fa5]{2,4})\s*律师", text)
+
+    if province_match:
+        province = province_match.group(1)
+    if city_match:
+        city = city_match.group(1)
+    if name_match:
+        lawyer_name = name_match.group(1)
+
+    return province, city, lawyer_name
+
+
+def pick_product_module(state: Dict) -> Optional[Dict]:
+    module_map = state.get("currentPageModuleIdMap", {}) or {}
+    page_ids = state.get("currentPageModuleIds", []) or []
+
+    for module_id in page_ids:
+        module = module_map.get(str(module_id)) or module_map.get(module_id)
+        if not isinstance(module, dict):
+            continue
+        ext_info = module.get("extInfo", {}) or {}
+        if ext_info.get("productInfo"):
+            return module
+
+    for module in module_map.values():
+        if not isinstance(module, dict):
+            continue
+        ext_info = module.get("extInfo", {}) or {}
+        if ext_info.get("productInfo"):
+            return module
+
+    return None
+
+
+def parse_group_id_from_url(url: str) -> int:
+    match = re.search(r"(?:[?&#]|^)groupId=(\d+)", url)
+    if not match:
+        return 0
+    try:
+        return int(match.group(1))
+    except ValueError:
+        return 0
+
+
+def extract_records(url: str, state: Dict) -> List[Dict]:
+    module = pick_product_module(state)
+    if not module:
+        return []
+
+    ext_info = module.get("extInfo", {}) or {}
+    product_info = ext_info.get("productInfo", {}) or {}
+    product_name = str(product_info.get("name") or "").strip()
+
+    province, city, current_name = extract_location_and_name(product_name)
+    group_id = product_info.get("groupId")
+    if not group_id:
+        group_id = parse_group_id_from_url(url)
+    module_id = module.get("id")
+
+    prop_map: Dict[str, List[str]] = {}
+    for prop in ext_info.get("propList", []) or []:
+        name = str(prop.get("name") or "").strip()
+        values = [str(item or "").strip() for item in (prop.get("valueList") or [])]
+        if name:
+            prop_map[name] = values
+
+    result: List[Dict] = []
+    seen_phones: Set[str] = set()
+    now = int(time.time())
+
+    phone_values = prop_map.get("电话", [])
+    for idx, raw_phone in enumerate(phone_values):
+        phone = normalize_phone(raw_phone)
+        if not phone or phone in seen_phones:
+            continue
+        seen_phones.add(phone)
+
+        law_firm = value_at(prop_map.get("律师所", []), idx)
+        area = value_at(prop_map.get("所在地区", []), idx)
+        direction = value_at(prop_map.get("主攻方向", []), idx)
+        specialty_text = value_at(prop_map.get("专业特长", []), idx)
+        license_no = value_at(prop_map.get("执业证号", []), idx)
+        address = value_at(prop_map.get("地址", []), idx)
+        email = value_at(prop_map.get("电子邮箱", []), idx)
+        seat_phone = value_at(prop_map.get("座机", []), idx)
+        wechat = value_at(prop_map.get("微信", []), idx)
+        qq = value_at(prop_map.get("QQ", []), idx)
+        first_practice_date = value_at(prop_map.get("首次执业日期", []), idx)
+
+        specialties = split_specialties(direction)
+        if not specialties:
+            specialties = split_specialties(specialty_text)
+
+        record = {
+            "record_id": hashlib.md5(f"{url}|{phone}".encode("utf-8")).hexdigest(),
+            "collected_at": now,
+            "source": {
+                "site": SITE_NAME,
+                "list_url": url,
+                "detail_url": "",
+                "province": province,
+                "province_py": "",
+                "city": area or city,
+                "city_py": "",
+                "page": 1,
+                "group_id": group_id,
+                "module_id": module_id,
+                "detail_url_status": "unresolved_from_pool",
+            },
+            "list_snapshot": {
+                "name": "",
+                "law_firm": law_firm,
+                "specialties": specialties,
+                "answer_count": None,
+            },
+            "profile": {
+                "name": "",
+                "law_firm": law_firm,
+                "phone": phone,
+                "license_no": license_no,
+                "practice_years": None,
+                "email": email,
+                "address": address,
+                "specialties": specialties,
+            },
+            "raw": {
+                "source_index": idx,
+                "direction": direction,
+                "specialty_text": specialty_text,
+                "seat_phone": seat_phone,
+                "wechat": wechat,
+                "qq": qq,
+                "first_practice_date": first_practice_date,
+            },
+        }
+        result.append(record)
+
+    current_phone = normalize_phone(str(product_info.get("material") or ""))
+    if current_phone and current_phone not in seen_phones:
+        seen_phones.add(current_phone)
+        remark = str(product_info.get("remark") or "")
+        specialties = extract_specialties_from_remark(remark)
+        result.append(
+            {
+                "record_id": hashlib.md5(f"{url}|{current_phone}".encode("utf-8")).hexdigest(),
+                "collected_at": now,
+                "source": {
+                    "site": SITE_NAME,
+                    "list_url": url,
+                    "detail_url": url,
+                    "province": province,
+                    "province_py": "",
+                    "city": city,
+                    "city_py": "",
+                    "page": 1,
+                    "group_id": group_id,
+                    "module_id": module_id,
+                },
+                "list_snapshot": {
+                    "name": current_name,
+                    "law_firm": str(product_info.get("prop0") or "").strip(),
+                    "specialties": specialties,
+                    "answer_count": None,
+                },
+                "profile": {
+                    "name": current_name,
+                    "law_firm": str(product_info.get("prop0") or "").strip(),
+                    "phone": current_phone,
+                    "license_no": str(product_info.get("prop1") or "").strip(),
+                    "practice_years": None,
+                    "email": "",
+                    "address": str(product_info.get("prop3") or "").strip(),
+                    "specialties": specialties,
+                },
+                "raw": {
+                    "from_product_info": True,
+                    "product_name": product_name,
+                    "remark": remark,
+                },
+            }
+        )
+
+    return result
+
+
+def to_legacy_row(record: Dict) -> Optional[Dict[str, str]]:
+    source = record.get("source", {}) or {}
+    profile = record.get("profile", {}) or {}
+    phone = normalize_phone(profile.get("phone", ""))
+    if not phone:
+        return None
+
+    province = str(source.get("province") or "").strip()
+    city = str(source.get("city") or province).strip()
+    return {
+        "name": str(profile.get("name") or "").strip(),
+        "law_firm": str(profile.get("law_firm") or "").strip(),
+        "province": province,
+        "city": city,
+        "phone": phone,
+        "url": str(source.get("detail_url") or "").strip(),
+        "domain": LEGACY_DOMAIN,
+        "create_time": int(record.get("collected_at") or time.time()),
+        "params": json.dumps(record, ensure_ascii=False),
+    }
+
+
+def existing_phones_in_db(db: Db, phones: List[str]) -> Set[str]:
+    deduped = sorted({phone for phone in phones if phone})
+    if not deduped:
+        return set()
+
+    existing: Set[str] = set()
+    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 = f"SELECT phone FROM lawyer WHERE domain=%s AND phone IN ({placeholders})"
+            cur.execute(sql, [LEGACY_DOMAIN, *chunk])
+            for row in cur.fetchall():
+                existing.add(row[0])
+    finally:
+        cur.close()
+    return existing
+
+
+def write_records_to_db(db: Db, records: List[Dict]) -> Tuple[int, int]:
+    rows: List[Dict[str, str]] = []
+    for record in records:
+        row = to_legacy_row(record)
+        if row:
+            rows.append(row)
+    if not rows:
+        return 0, 0
+
+    existing = existing_phones_in_db(db, [row["phone"] for row in rows])
+    inserted = 0
+    skipped = 0
+
+    for row in rows:
+        phone = row.get("phone", "")
+        if not phone or phone in existing:
+            skipped += 1
+            continue
+        try:
+            db.insert_data("lawyer", row)
+            existing.add(phone)
+            inserted += 1
+        except Exception as exc:
+            skipped += 1
+            print(f"[db] 插入失败 phone={phone}: {exc}")
+    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()
+
+
+def main() -> None:
+    args = parse_args()
+    os.makedirs(os.path.dirname(args.output) or ".", exist_ok=True)
+
+    client = RequestsClient(
+        headers={
+            "User-Agent": (
+                "Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) "
+                "AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 "
+                "Mobile/15E148 Safari/604.1"
+            ),
+            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
+            "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8",
+            "Connection": "close",
+        },
+        use_proxy=not args.direct,
+        retry_total=2,
+        retry_backoff_factor=1,
+        retry_status_forcelist=(429, 500, 502, 503, 504),
+        retry_allowed_methods=("GET",),
+    )
+
+    try:
+        resp = client.get_text(args.url, timeout=30, verify=False)
+        if resp.status_code >= 400:
+            raise RequestClientError(f"{resp.status_code} Error: {args.url}")
+        state = parse_initial_state(resp.text)
+        records = extract_records(args.url, state)
+    finally:
+        client.close()
+
+    if not records:
+        print("[done] 未采集到有效手机号")
+        return
+
+    seen_ids: Set[str] = set()
+    if os.path.exists(args.output):
+        with open(args.output, "r", encoding="utf-8") as old_file:
+            for line in old_file:
+                line = line.strip()
+                if not line:
+                    continue
+                try:
+                    item = json.loads(line)
+                except Exception:
+                    continue
+                record_id = item.get("record_id")
+                if record_id:
+                    seen_ids.add(record_id)
+
+    json_new = 0
+    with open(args.output, "a", encoding="utf-8") as out:
+        for record in records:
+            record_id = record["record_id"]
+            if record_id in seen_ids:
+                continue
+            out.write(json.dumps(record, ensure_ascii=False) + "\n")
+            seen_ids.add(record_id)
+            json_new += 1
+
+    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, 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}"
+    )
+
+
+if __name__ == "__main__":
+    main()

request/__init__.py

@@ -1,19 +1 @@
-from request.requests_client import (
-    RequestClientError,
-    RequestConnectTimeout,
-    RequestConnectionError,
-    RequestSSLError,
-    RequestTimeout,
-    RequestsClient,
-    ResponseData,
-)
-
-__all__ = [
-    "RequestsClient",
-    "ResponseData",
-    "RequestClientError",
-    "RequestConnectTimeout",
-    "RequestTimeout",
-    "RequestConnectionError",
-    "RequestSSLError",
-]
+# Package marker for request utilities.

request/proxy_config.py

@@ -24,6 +24,19 @@ def _normalize_bool(value, default: bool = True) -> bool:
     return text not in ("0", "false", "no", "off", "")
 
 
+def _env_proxy_override() -> Optional[bool]:
+    """
+    环境变量覆盖代理开关：
+    - PROXY_ENABLED 未设置：返回 None（不覆盖，仍读取 proxy_settings.json）
+    - PROXY_ENABLED=0/false/off：强制关闭代理
+    - PROXY_ENABLED=1/true/on：强制开启代理（前提是配置字段齐全）
+    """
+    raw = os.getenv("PROXY_ENABLED")
+    if raw is None:
+        return None
+    return _normalize_bool(raw, True)
+
+
 def _load_config() -> Dict[str, str]:
     if not os.path.exists(CONFIG_PATH):
         return dict(DEFAULT_CONFIG)
@@ -48,7 +61,12 @@ def report_proxy_status() -> None:
     _PROXY_STATUS_REPORTED = True
 
     config = _load_config()
-    enabled = _normalize_bool(config.get("enabled"), True)
+    override = _env_proxy_override()
+    if override is False:
+        print("[proxy] disabled by env (PROXY_ENABLED=0)")
+        return
+
+    enabled = _normalize_bool(config.get("enabled"), True) if override is None else True
     if not enabled:
         print("[proxy] disabled by config")
         return
@@ -66,7 +84,10 @@ def get_proxies() -> Optional[Dict[str, str]]:
     代理配置从 proxy_settings.json 读取，不依赖环境变量。
     """
     config = _load_config()
-    if not _normalize_bool(config.get("enabled"), True):
+    override = _env_proxy_override()
+    if override is False:
+        return None
+    if override is None and not _normalize_bool(config.get("enabled"), True):
         return None
 
     tunnel = str(config.get("tunnel") or "").strip()
@@ -95,3 +116,22 @@ def apply_proxy(session) -> Optional[Dict[str, str]]:
 
 
 __all__ = ["get_proxies", "apply_proxy", "report_proxy_status"]
+
+
+def is_proxy_enabled() -> bool:
+    """
+    判断当前进程是否启用了代理。
+
+    优先遵循环境变量 PROXY_ENABLED；
+    未设置时回退到 proxy_settings.json 的 enabled 配置。
+    """
+    config = _load_config()
+    override = _env_proxy_override()
+    if override is False:
+        return False
+    if override is True:
+        return True
+    return _normalize_bool(config.get("enabled"), True)
+
+
+__all__ = ["get_proxies", "apply_proxy", "report_proxy_status", "is_proxy_enabled"]

requirements.txt

@@ -1,6 +1,18 @@
+# 数据库驱动
 pymysql>=1.0.2
+pymongo>=4.0.0
+
+# 调度器
+schedule>=1.2.0
+
+# 其他可能需要的依赖
 requests>=2.28.0
 beautifulsoup4>=4.11.0
-urllib3>=1.26.0
 lxml>=4.9.0
-openpyxl>=3.1.0
+redis>=4.0.0
+pyppeteer>=1.0.2
+# 可选：提升反检测能力
+pyppeteer-stealth>=2.7.4
+
+# 日志相关
+python-dateutil>=2.8.2

services/area_sync_service.py

@@ -0,0 +1,849 @@
+import json
+import os
+import re
+import sys
+import threading
+import time
+from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
+from typing import Any, Dict, Iterable, List, Optional, Set, Tuple
+from urllib.parse import parse_qs, urlparse
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+project_root = os.path.dirname(current_dir)
+if project_root not in sys.path:
+    sys.path.append(project_root)
+
+from Db import Db
+
+
+AREA_TABLE = os.getenv("AREA_TARGET_TABLE", "area_new")
+AREA_DOMAIN = os.getenv("AREA_DOMAIN", "maxlaw")
+DOUYIN_DOMAIN = os.getenv("DOUYIN_DOMAIN", "抖音")
+DOUYIN_RAW_DIR = os.getenv("DOUYIN_RAW_DIR", os.path.join(project_root, "data", "douyin_raw"))
+DOUYIN_SAVE_ONLY_ENV = os.getenv("DOUYIN_SAVE_ONLY", "1")
+LAWYER_KEYWORDS_ENV = os.getenv("DOUYIN_LAWYER_KEYWORDS", "律师,律所")
+PROGRESS_TABLE = os.getenv("LAYER_PROGRESS_TABLE", "layer_progress")
+PROGRESS_DEFAULT_KEY = os.getenv("LAYER_PROGRESS_DEFAULT_KEY", "douyin_batch_default")
+SERVICE_HOST = os.getenv("AREA_SERVICE_HOST", "0.0.0.0")
+SERVICE_PORT = int(os.getenv("AREA_SERVICE_PORT", "9002"))
+
+PHONE_REGEX = re.compile(r"(?:\+?86[-\s]?)?(1[3-9]\d{9})")
+WX_CONTEXT_REGEX = re.compile(r"(?i)(?:微信|微.?信|wx|vx|weixin|v信|v号|v)\s*[:：/\-\s]\s*([a-zA-Z0-9._-]{3,40})")
+LAW_FIRM_REGEX = re.compile(r"([\u4e00-\u9fa5A-Za-z·]{2,40}律师事务所)")
+RAW_WRITE_LOCK = threading.Lock()
+
+LAWYER_KEYWORDS: Tuple[str, ...] = tuple(
+    keyword.strip() for keyword in LAWYER_KEYWORDS_ENV.split(",") if keyword.strip()
+)
+
+
+def _is_safe_table_name(table_name: str) -> bool:
+    return bool(re.fullmatch(r"[A-Za-z0-9_]+", table_name or ""))
+
+
+def _parse_int(value: Any, default: int = 0) -> int:
+    try:
+        return int(str(value).strip())
+    except Exception:
+        return default
+
+
+def _parse_bool(value: Any, default: bool = False) -> bool:
+    if value is None:
+        return default
+    text = str(value).strip().lower()
+    if text in {"1", "true", "yes", "y", "on"}:
+        return True
+    if text in {"0", "false", "no", "n", "off"}:
+        return False
+    return default
+
+
+def _first_param(params: Dict[str, List[str]], key: str, default: str = "") -> str:
+    values = params.get(key) or []
+    if not values:
+        return default
+    return values[0]
+
+
+def _append_jsonl(file_path: str, payload: Dict[str, Any]) -> None:
+    os.makedirs(os.path.dirname(file_path) or ".", exist_ok=True)
+    line = json.dumps(payload, ensure_ascii=False)
+    with RAW_WRITE_LOCK:
+        with open(file_path, "a", encoding="utf-8") as out:
+            out.write(line)
+            out.write("\n")
+
+
+def _save_raw_index_payload(payload: Dict[str, Any], query: Dict[str, List[str]], client_ip: str) -> str:
+    now_ts = int(time.time())
+    day = time.strftime("%Y%m%d", time.localtime(now_ts))
+    file_path = os.path.join(DOUYIN_RAW_DIR, f"douyin_index_{day}.jsonl")
+
+    wrapped = {
+        "received_at": now_ts,
+        "client_ip": client_ip,
+        "query": query,
+        "payload": payload,
+    }
+    _append_jsonl(file_path, wrapped)
+    return file_path
+
+
+def _ensure_progress_table() -> None:
+    if not _is_safe_table_name(PROGRESS_TABLE):
+        raise ValueError("非法进度表名")
+
+    with Db() as db:
+        cursor = db.db.cursor()
+        sql = f"""
+        CREATE TABLE IF NOT EXISTS `{PROGRESS_TABLE}` (
+          `id` bigint(20) unsigned NOT NULL AUTO_INCREMENT,
+          `progress_key` varchar(128) NOT NULL,
+          `next_city_index` int(11) DEFAULT 0,
+          `area_signature` varchar(128) DEFAULT NULL,
+          `area_total` int(11) DEFAULT 0,
+          `current_city` varchar(128) DEFAULT NULL,
+          `reason` varchar(64) DEFAULT NULL,
+          `status` varchar(32) DEFAULT NULL,
+          `device_id` varchar(128) DEFAULT NULL,
+          `extra_json` longtext,
+          `updated_at` bigint(20) DEFAULT NULL,
+          `create_time` bigint(20) DEFAULT NULL,
+          PRIMARY KEY (`id`),
+          UNIQUE KEY `uk_progress_key` (`progress_key`),
+          KEY `idx_updated_at` (`updated_at`)
+        ) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4
+        """
+        cursor.execute(sql)
+        db.db.commit()
+        cursor.close()
+
+
+def _get_progress(progress_key: str) -> Optional[Dict[str, Any]]:
+    key = str(progress_key or "").strip()
+    if not key:
+        return None
+
+    _ensure_progress_table()
+    with Db() as db:
+        cursor = db.db.cursor()
+        sql = (
+            f"SELECT progress_key, next_city_index, area_signature, area_total, current_city, "
+            f"reason, status, device_id, extra_json, updated_at, create_time "
+            f"FROM `{PROGRESS_TABLE}` WHERE progress_key=%s LIMIT 1"
+        )
+        cursor.execute(sql, (key,))
+        row = cursor.fetchone()
+        cursor.close()
+
+    if not row:
+        return None
+
+    extra_json = row[8] or ""
+    extra_obj: Any = {}
+    if extra_json:
+        try:
+            extra_obj = json.loads(extra_json)
+        except Exception:
+            extra_obj = extra_json
+
+    return {
+        "progress_key": row[0] or "",
+        "next_city_index": _parse_int(row[1], 0),
+        "area_signature": row[2] or "",
+        "area_total": _parse_int(row[3], 0),
+        "current_city": row[4] or "",
+        "reason": row[5] or "",
+        "status": row[6] or "",
+        "device_id": row[7] or "",
+        "extra": extra_obj,
+        "updated_at": _parse_int(row[9], 0),
+        "create_time": _parse_int(row[10], 0),
+    }
+
+
+def _upsert_progress(progress_key: str, payload: Dict[str, Any]) -> Dict[str, Any]:
+    key = str(progress_key or "").strip()
+    if not key:
+        raise ValueError("progress_key 不能为空")
+
+    _ensure_progress_table()
+    now_ts = int(time.time())
+    next_city_index = _parse_int(payload.get("next_city_index"), 0)
+    area_signature = str(payload.get("area_signature") or "").strip()
+    area_total = _parse_int(payload.get("area_total"), 0)
+    current_city = str(payload.get("current_city") or "").strip()
+    reason = str(payload.get("reason") or "").strip()
+    status = str(payload.get("status") or "").strip()
+    device_id = str(payload.get("device_id") or "").strip()
+    extra = payload.get("extra")
+    if extra is None:
+        extra = payload.get("extra_json")
+
+    if isinstance(extra, str):
+        extra_json = extra
+    else:
+        try:
+            extra_json = json.dumps(extra or {}, ensure_ascii=False)
+        except Exception:
+            extra_json = "{}"
+
+    with Db() as db:
+        cursor = db.db.cursor()
+        sql = (
+            f"INSERT INTO `{PROGRESS_TABLE}` "
+            "(progress_key, next_city_index, area_signature, area_total, current_city, reason, status, "
+            "device_id, extra_json, updated_at, create_time) "
+            "VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s) "
+            "ON DUPLICATE KEY UPDATE "
+            "next_city_index=VALUES(next_city_index), "
+            "area_signature=VALUES(area_signature), "
+            "area_total=VALUES(area_total), "
+            "current_city=VALUES(current_city), "
+            "reason=VALUES(reason), "
+            "status=VALUES(status), "
+            "device_id=VALUES(device_id), "
+            "extra_json=VALUES(extra_json), "
+            "updated_at=VALUES(updated_at)"
+        )
+        cursor.execute(
+            sql,
+            (
+                key,
+                next_city_index,
+                area_signature,
+                area_total,
+                current_city,
+                reason,
+                status,
+                device_id,
+                extra_json,
+                now_ts,
+                now_ts,
+            ),
+        )
+        db.db.commit()
+        cursor.close()
+
+    return _get_progress(key) or {}
+
+
+def _clear_progress(progress_key: str) -> int:
+    key = str(progress_key or "").strip()
+    if not key:
+        return 0
+
+    _ensure_progress_table()
+    with Db() as db:
+        cursor = db.db.cursor()
+        sql = f"DELETE FROM `{PROGRESS_TABLE}` WHERE progress_key=%s"
+        cursor.execute(sql, (key,))
+        affected = cursor.rowcount
+        db.db.commit()
+        cursor.close()
+    return affected
+
+
+def _query_area_data(table_name: str, domain: str) -> List[Dict[str, Any]]:
+    if not _is_safe_table_name(table_name):
+        raise ValueError("非法表名")
+
+    with Db() as db:
+        cursor = db.db.cursor()
+        sql = (
+            f"SELECT province, city, name, pid, pinyin, code, domain, level, create_time "
+            f"FROM `{table_name}` WHERE domain=%s ORDER BY id ASC"
+        )
+        cursor.execute(sql, (domain,))
+        rows = cursor.fetchall()
+        cursor.close()
+
+    result: List[Dict[str, Any]] = []
+    for row in rows:
+        result.append(
+            {
+                "province": row[0] or "",
+                "city": row[1] or "",
+                "name": row[2] or "",
+                "pid": row[3] if row[3] is not None else 0,
+                "pinyin": row[4] or "",
+                "code": row[5] or "",
+                "domain": row[6] or "",
+                "level": row[7] if row[7] is not None else 0,
+                "create_time": row[8] if row[8] is not None else 0,
+            }
+        )
+    return result
+
+
+def _iter_dict_nodes(value: Any) -> Iterable[Dict[str, Any]]:
+    stack: List[Any] = [value]
+    while stack:
+        current = stack.pop()
+        if isinstance(current, dict):
+            yield current
+            stack.extend(current.values())
+        elif isinstance(current, list):
+            stack.extend(current)
+
+
+def _extract_phones_from_text(text: str) -> List[str]:
+    phones: List[str] = []
+    seen: Set[str] = set()
+    for match in PHONE_REGEX.finditer(text or ""):
+        phone = match.group(1)
+        if not phone or phone in seen:
+            continue
+        seen.add(phone)
+        phones.append(phone)
+    return phones
+
+
+def _extract_phones_from_user_info(user_info: Dict[str, Any]) -> List[str]:
+    signature = str(user_info.get("signature") or "")
+    unique_id = str(user_info.get("unique_id") or "")
+    versatile = str(user_info.get("versatile_display") or "")
+
+    # 1) 优先从简介直接匹配手机号
+    phones = set(_extract_phones_from_text(signature))
+    if phones:
+        return sorted(phones)
+
+    # 2) 从微信相关标记中提取，再从抖音号字段兜底
+    for text in (signature, unique_id, versatile):
+        for match in WX_CONTEXT_REGEX.finditer(text):
+            wx_value = match.group(1) or ""
+            for phone in _extract_phones_from_text(wx_value):
+                phones.add(phone)
+
+    for text in (unique_id, versatile):
+        for phone in _extract_phones_from_text(text):
+            phones.add(phone)
+
+    return sorted(phones)
+
+
+def _extract_law_firm_from_user_info(user_info: Dict[str, Any]) -> str:
+    candidates: List[str] = []
+
+    signature = str(user_info.get("signature") or "")
+    if signature:
+        candidates.append(signature)
+
+    verify_reason = str(user_info.get("enterprise_verify_reason") or "")
+    if verify_reason:
+        candidates.append(verify_reason)
+
+    cert_text = ""
+    account_cert_info = user_info.get("account_cert_info")
+    if isinstance(account_cert_info, str) and account_cert_info.strip():
+        try:
+            cert_obj = json.loads(account_cert_info)
+            if isinstance(cert_obj, dict):
+                cert_text = str(cert_obj.get("label_text") or "").strip()
+        except Exception:
+            cert_text = account_cert_info.strip()
+    if cert_text:
+        candidates.append(cert_text)
+
+    for text in candidates:
+        match = LAW_FIRM_REGEX.search(text)
+        if match:
+            return match.group(1)
+
+    return ""
+
+
+def _extract_account_cert_text(user_info: Dict[str, Any]) -> str:
+    account_cert_info = user_info.get("account_cert_info")
+    if isinstance(account_cert_info, str) and account_cert_info.strip():
+        try:
+            cert_obj = json.loads(account_cert_info)
+            if isinstance(cert_obj, dict):
+                return str(cert_obj.get("label_text") or "").strip()
+        except Exception:
+            return account_cert_info.strip()
+    return ""
+
+
+def _is_lawyer_related_user(user_info: Dict[str, Any], name: str, law_firm: str) -> bool:
+    texts = [
+        name,
+        str(user_info.get("nickname") or ""),
+        str(user_info.get("signature") or ""),
+        str(user_info.get("custom_verify") or ""),
+        str(user_info.get("enterprise_verify_reason") or ""),
+        str(user_info.get("versatile_display") or ""),
+        str(user_info.get("unique_id") or ""),
+        _extract_account_cert_text(user_info),
+        law_firm,
+    ]
+    merged = "\n".join(text for text in texts if text).strip()
+    if not merged:
+        return False
+    return any(keyword in merged for keyword in LAWYER_KEYWORDS)
+
+
+def _pick_first_str(node: Dict[str, Any], keys: Tuple[str, ...]) -> str:
+    for key in keys:
+        value = node.get(key)
+        if isinstance(value, str):
+            text = value.strip()
+            if text:
+                return text
+    return ""
+
+
+def _extract_name(node: Dict[str, Any]) -> str:
+    direct = _pick_first_str(node, ("name", "nickname", "nick_name", "author_name", "title", "account_name"))
+    if direct:
+        return direct
+
+    for nested_key in ("author", "user", "user_info", "profile", "account"):
+        nested = node.get(nested_key)
+        if isinstance(nested, dict):
+            nested_name = _pick_first_str(nested, ("name", "nickname", "nick_name", "author_name", "title"))
+            if nested_name:
+                return nested_name
+
+    return ""
+
+
+def _extract_law_firm(node: Dict[str, Any]) -> str:
+    direct = _pick_first_str(
+        node,
+        (
+            "law_firm",
+            "firm",
+            "lawFirm",
+            "office",
+            "org_name",
+            "organization",
+            "company",
+            "enterprise",
+        ),
+    )
+    if direct:
+        return direct
+
+    enterprise = node.get("enterprise")
+    if isinstance(enterprise, dict):
+        company_name = _pick_first_str(enterprise, ("name", "company_name", "enterprise_name"))
+        if company_name:
+            return company_name
+
+    return ""
+
+
+def _extract_detail_url(node: Dict[str, Any], fallback_api_url: str) -> str:
+    url = _pick_first_str(node, ("share_url", "url", "web_url", "detail_url", "jump_url"))
+    if url:
+        return url
+
+    aweme_id = node.get("aweme_id") or node.get("item_id")
+    if aweme_id:
+        aid = str(aweme_id).strip()
+        if aid:
+            return f"https://www.douyin.com/video/{aid}"
+
+    sec_uid = node.get("sec_uid")
+    if sec_uid:
+        sec_uid_text = str(sec_uid).strip()
+        if sec_uid_text:
+            return f"https://www.douyin.com/user/{sec_uid_text}"
+
+    return fallback_api_url
+
+
+def _city_from_index(city_index: int, table_name: str, domain: str) -> Tuple[str, str]:
+    if city_index < 0:
+        return "", ""
+    try:
+        areas = _query_area_data(table_name, domain)
+    except Exception:
+        return "", ""
+    if city_index >= len(areas):
+        return "", ""
+    area = areas[city_index]
+    province = str(area.get("province") or "").strip()
+    city = str(area.get("city") or province).strip()
+    return province, city
+
+
+def _existing_phones(domain: str, phones: List[str]) -> Set[str]:
+    if not phones:
+        return set()
+
+    deduped = sorted({p for p in phones if p})
+    if not deduped:
+        return set()
+
+    existing: Set[str] = set()
+    with Db() as db:
+        cursor = db.db.cursor()
+        chunk_size = 500
+        for i in range(0, len(deduped), chunk_size):
+            chunk = deduped[i:i + chunk_size]
+            placeholders = ",".join(["%s"] * len(chunk))
+            sql = f"SELECT phone FROM lawyer WHERE domain=%s AND phone IN ({placeholders})"
+            cursor.execute(sql, [domain, *chunk])
+            for row in cursor.fetchall():
+                existing.add(str(row[0]))
+        cursor.close()
+    return existing
+
+
+def _insert_lawyer_rows(rows: List[Dict[str, Any]], domain: str) -> Tuple[int, int]:
+    if not rows:
+        return 0, 0
+
+    def row_score(item: Dict[str, Any]) -> int:
+        score = 0
+        if str(item.get("name") or "").strip():
+            score += 5
+        if str(item.get("law_firm") or "").strip():
+            score += 3
+        if str(item.get("url") or "").strip():
+            score += 1
+        if str(item.get("province") or "").strip() or str(item.get("city") or "").strip():
+            score += 1
+        phone_count_in_node = _parse_int(item.get("phone_count_in_node"), 1)
+        if phone_count_in_node > 1:
+            score -= (phone_count_in_node - 1)
+        return score
+
+    deduped_by_phone: Dict[str, Dict[str, Any]] = {}
+    skipped = 0
+    for row in rows:
+        phone = str(row.get("phone") or "").strip()
+        if not phone:
+            skipped += 1
+            continue
+        old_row = deduped_by_phone.get(phone)
+        if old_row is not None:
+            if row_score(row) > row_score(old_row):
+                deduped_by_phone[phone] = row
+            skipped += 1
+            continue
+        deduped_by_phone[phone] = row
+
+    existing = _existing_phones(domain, list(deduped_by_phone.keys()))
+
+    inserted = 0
+    with Db() as db:
+        cursor = db.db.cursor()
+        sql = (
+            "INSERT INTO lawyer "
+            "(name, phone, law_firm, province, city, url, domain, create_time, site_time, params) "
+            "VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s, %s)"
+        )
+        for phone, row in deduped_by_phone.items():
+            if phone in existing:
+                skipped += 1
+                continue
+
+            cursor.execute(
+                sql,
+                (
+                    row.get("name") or "",
+                    phone,
+                    row.get("law_firm") or "",
+                    row.get("province") or "",
+                    row.get("city") or "",
+                    row.get("url") or "",
+                    domain,
+                    _parse_int(row.get("create_time"), int(time.time())),
+                    _parse_int(row.get("site_time"), int(time.time())),
+                    row.get("params") or "{}",
+                ),
+            )
+            inserted += 1
+            existing.add(phone)
+
+        db.db.commit()
+        cursor.close()
+
+    return inserted, skipped
+
+
+def _extract_lawyer_rows_from_payload(
+    payload: Dict[str, Any],
+    area_table: str,
+    area_domain: str,
+    save_domain: str,
+) -> List[Dict[str, Any]]:
+    now_ts = int(time.time())
+    api_url = str(payload.get("url") or "").strip()
+    city_index = _parse_int(payload.get("cityIndex"), -1)
+    city_province, city_name = _city_from_index(city_index, area_table, area_domain)
+
+    rows: List[Dict[str, Any]] = []
+    data = payload.get("data") if isinstance(payload, dict) else None
+    user_list = data.get("user_list") if isinstance(data, dict) else None
+    if not isinstance(user_list, list):
+        return rows
+
+    for user_item in user_list:
+        if not isinstance(user_item, dict):
+            continue
+        user_info = user_item.get("user_info")
+        if not isinstance(user_info, dict):
+            continue
+
+        name = str(user_info.get("nickname") or "").strip()
+        law_firm = _extract_law_firm_from_user_info(user_info)
+
+        # 强约束：必须出现“律师/律所”等关键词，避免非法律相关账号入库
+        if not _is_lawyer_related_user(user_info, name, law_firm):
+            continue
+
+        phones = _extract_phones_from_user_info(user_info)
+        if not phones:
+            continue
+
+        sec_uid = str(user_info.get("sec_uid") or "").strip()
+        if not sec_uid:
+            continue
+        url = f"https://www.douyin.com/user/{sec_uid}"
+
+        province = city_province
+        city = city_name or city_province
+
+        source_record = {
+            "source": "douyin",
+            "api_source": payload.get("source") or "",
+            "api_url": api_url,
+            "city_index": city_index,
+            "captured_at": now_ts,
+            "sec_uid": sec_uid,
+            "user_info": {
+                "uid": user_info.get("uid"),
+                "nickname": user_info.get("nickname"),
+                "signature": user_info.get("signature"),
+                "unique_id": user_info.get("unique_id"),
+                "versatile_display": user_info.get("versatile_display"),
+            },
+        }
+
+        for phone in phones:
+            rows.append(
+                {
+                    "name": name,
+                    "phone": phone,
+                    "law_firm": law_firm,
+                    "province": province,
+                    "city": city,
+                    "url": url,
+                    "domain": save_domain,
+                    "create_time": now_ts,
+                    "site_time": _parse_int(payload.get("ts"), now_ts),
+                    "phone_count_in_node": len(phones),
+                    "params": json.dumps(source_record, ensure_ascii=False),
+                }
+            )
+
+    return rows
+
+
+class AreaSyncHandler(BaseHTTPRequestHandler):
+    server_version = "AreaSyncService/2.0"
+
+    def _write_json(self, status: int, payload: Any) -> None:
+        body = json.dumps(payload, ensure_ascii=False).encode("utf-8")
+        self.send_response(status)
+        self.send_header("Content-Type", "application/json; charset=utf-8")
+        self.send_header("Content-Length", str(len(body)))
+        self.send_header("Access-Control-Allow-Origin", "*")
+        self.send_header("Access-Control-Allow-Methods", "GET, POST, OPTIONS")
+        self.send_header("Access-Control-Allow-Headers", "Content-Type")
+        self.end_headers()
+        self.wfile.write(body)
+
+    def _read_json_body(self) -> Any:
+        length = _parse_int(self.headers.get("Content-Length"), 0)
+        if length <= 0:
+            return {}
+
+        raw = self.rfile.read(length)
+        if not raw:
+            return {}
+
+        try:
+            return json.loads(raw.decode("utf-8"))
+        except Exception:
+            return {}
+
+    def do_OPTIONS(self) -> None:
+        self.send_response(204)
+        self.send_header("Access-Control-Allow-Origin", "*")
+        self.send_header("Access-Control-Allow-Methods", "GET, POST, OPTIONS")
+        self.send_header("Access-Control-Allow-Headers", "Content-Type")
+        self.end_headers()
+
+    def do_GET(self) -> None:
+        parsed = urlparse(self.path)
+        params = parse_qs(parsed.query)
+
+        if parsed.path == "/health":
+            self._write_json(200, {"ok": True, "service": "layer-service"})
+            return
+
+        if parsed.path == "/api/layer/get_area":
+            table_name = _first_param(params, "table", AREA_TABLE).strip() or AREA_TABLE
+            domain = _first_param(params, "domain", AREA_DOMAIN).strip() or AREA_DOMAIN
+            with_meta = _parse_bool(_first_param(params, "meta", "0"), False)
+
+            try:
+                rows = _query_area_data(table_name, domain)
+            except Exception as exc:
+                self._write_json(500, {"ok": False, "error": str(exc)})
+                return
+
+            if with_meta:
+                self._write_json(
+                    200,
+                    {
+                        "ok": True,
+                        "count": len(rows),
+                        "table": table_name,
+                        "domain": domain,
+                        "data": rows,
+                    },
+                )
+            else:
+                self._write_json(200, rows)
+            return
+
+        if parsed.path == "/api/layer/progress":
+            progress_key = _first_param(params, "progress_key", PROGRESS_DEFAULT_KEY).strip() or PROGRESS_DEFAULT_KEY
+            try:
+                row = _get_progress(progress_key)
+            except Exception as exc:
+                self._write_json(500, {"ok": False, "error": str(exc)})
+                return
+
+            self._write_json(
+                200,
+                {
+                    "ok": True,
+                    "progress_key": progress_key,
+                    "data": row,
+                },
+            )
+            return
+
+        self._write_json(404, {"ok": False, "error": "not found"})
+
+    def do_POST(self) -> None:
+        parsed = urlparse(self.path)
+        params = parse_qs(parsed.query)
+
+        if parsed.path == "/api/layer/progress":
+            body = self._read_json_body()
+            if not isinstance(body, dict):
+                body = {}
+
+            progress_key = str(body.get("progress_key") or _first_param(params, "progress_key", PROGRESS_DEFAULT_KEY)).strip() or PROGRESS_DEFAULT_KEY
+            action = str(body.get("action") or _first_param(params, "action", "upsert")).strip().lower() or "upsert"
+
+            try:
+                if action == "clear":
+                    deleted = _clear_progress(progress_key)
+                    self._write_json(
+                        200,
+                        {
+                            "ok": True,
+                            "action": "clear",
+                            "progress_key": progress_key,
+                            "deleted": deleted,
+                        },
+                    )
+                    return
+
+                saved = _upsert_progress(progress_key, body)
+                self._write_json(
+                    200,
+                    {
+                        "ok": True,
+                        "action": "upsert",
+                        "progress_key": progress_key,
+                        "data": saved,
+                    },
+                )
+                return
+            except Exception as exc:
+                self._write_json(500, {"ok": False, "error": str(exc)})
+                return
+
+        if parsed.path == "/api/layer/index":
+            body = self._read_json_body()
+            if not isinstance(body, dict) or not body:
+                self._write_json(400, {"ok": False, "error": "invalid json body"})
+                return
+
+            area_table = _first_param(params, "table", AREA_TABLE).strip() or AREA_TABLE
+            area_domain = _first_param(params, "area_domain", AREA_DOMAIN).strip() or AREA_DOMAIN
+            save_domain = _first_param(params, "save_domain", DOUYIN_DOMAIN).strip() or DOUYIN_DOMAIN
+            save_only_default = _parse_bool(DOUYIN_SAVE_ONLY_ENV, True)
+            save_only = _parse_bool(_first_param(params, "save_only", DOUYIN_SAVE_ONLY_ENV), save_only_default)
+
+            try:
+                saved_file = _save_raw_index_payload(body, params, self.client_address[0] if self.client_address else "")
+            except Exception as exc:
+                self._write_json(500, {"ok": False, "error": f"save raw payload failed: {exc}"})
+                return
+
+            if save_only:
+                self._write_json(
+                    200,
+                    {
+                        "ok": True,
+                        "message": "saved_only",
+                        "save_only": True,
+                        "saved_file": saved_file,
+                    },
+                )
+                return
+
+            try:
+                extracted = _extract_lawyer_rows_from_payload(body, area_table, area_domain, save_domain)
+                inserted, skipped = _insert_lawyer_rows(extracted, save_domain)
+            except Exception as exc:
+                self._write_json(500, {"ok": False, "error": str(exc)})
+                return
+
+            self._write_json(
+                200,
+                {
+                    "ok": True,
+                    "message": "received",
+                    "save_domain": save_domain,
+                    "save_only": False,
+                    "saved_file": saved_file,
+                    "extracted": len(extracted),
+                    "inserted": inserted,
+                    "skipped": skipped,
+                },
+            )
+            return
+
+        self._write_json(404, {"ok": False, "error": "not found"})
+
+
+def run() -> None:
+    try:
+        _ensure_progress_table()
+    except Exception as exc:
+        print(f"[layer-service] init progress table failed: {exc}")
+
+    server = ThreadingHTTPServer((SERVICE_HOST, SERVICE_PORT), AreaSyncHandler)
+    print(f"[layer-service] running on http://{SERVICE_HOST}:{SERVICE_PORT}")
+    print(f"[layer-service] get_area -> table/domain: {AREA_TABLE}/{AREA_DOMAIN}")
+    print(f"[layer-service] index -> save domain: {DOUYIN_DOMAIN}")
+    print(f"[layer-service] progress table/default key: {PROGRESS_TABLE}/{PROGRESS_DEFAULT_KEY}")
+    server.serve_forever()
+
+
+if __name__ == "__main__":
+    run()

utils/rate_limiter.py

@@ -1,76 +1,210 @@
 """
 全局请求速率限制器
-确保代理每秒不超过5次请求
+
+默认按“所有爬虫进程共享一个桶”来限流，避免 `bash start.sh`
+同时启动多个进程时，每个进程各自 5 次/秒，叠加后把代理冲爆。
 """
+from contextlib import contextmanager
+import json
+import os
+import tempfile
 import time
 import threading
-from collections import deque
+from pathlib import Path
+from uuid import uuid4
+
+import fcntl
+from request.proxy_config import is_proxy_enabled
 
 
 class RateLimiter:
     """
-    令牌桶算法实现的速率限制器
+    基于文件锁的跨进程滑动窗口限流器。
+
+    - 同一台机器上的多个 Python 进程会共享同一个状态文件
+    - 同一个进程内的多个线程也会一起走这个限流器
     """
-    def __init__(self, max_requests_per_second: int = 5):
-        """
-        初始化速率限制器
-        
-        Args:
-            max_requests_per_second: 每秒最大请求数
-        """
-        self.max_requests = max_requests_per_second
-        self.requests = deque()
-        self.lock = threading.RLock()
-    
-    def acquire(self):
-        """
-        获取请求权限，如果需要则等待
-        """
-        with self.lock:
-            now = time.time()
-            
-            # 清理超过1秒的请求记录
-            while self.requests and now - self.requests[0] >= 1.0:
-                self.requests.popleft()
-            
-            # 如果当前请求数已达上限，等待
-            if len(self.requests) >= self.max_requests:
-                # 计算需要等待的时间
-                wait_time = 1.0 - (now - self.requests[0])
-                if wait_time > 0:
-                    time.sleep(wait_time)
-                    return self.acquire()  # 递归调用以重新检查
-            
-            # 记录这次请求
-            self.requests.append(now)
-    
+
+    def __init__(
+        self,
+        max_requests_per_second: int = 5,
+        window_seconds: float = 1.0,
+        state_file: str | None = None,
+    ):
+        self.max_requests = max(1, int(max_requests_per_second))
+        self.max_concurrent = max(
+            1,
+            int(os.getenv("PROXY_MAX_CONCURRENT_REQUESTS", str(self.max_requests))),
+        )
+        self.window_seconds = max(0.1, float(window_seconds))
+        self.lease_seconds = max(
+            5.0,
+            float(os.getenv("PROXY_REQUEST_LEASE_SECONDS", "120")),
+        )
+        default_state = os.path.join(
+            tempfile.gettempdir(),
+            "lawyers_proxy_rate_limiter.json",
+        )
+        self.state_file = Path(
+            state_file or os.getenv("PROXY_RATE_LIMIT_FILE", default_state)
+        )
+        self.lock_file = self.state_file.with_suffix(self.state_file.suffix + ".lock")
+        self._thread_lock = threading.RLock()
+        self.state_file.parent.mkdir(parents=True, exist_ok=True)
+        self.lock_file.parent.mkdir(parents=True, exist_ok=True)
+
+    def _load_state(self) -> dict:
+        if not self.state_file.exists():
+            return {"timestamps": [], "leases": {}}
+        try:
+            raw = self.state_file.read_text(encoding="utf-8").strip()
+            if not raw:
+                return {"timestamps": [], "leases": {}}
+            data = json.loads(raw)
+            if isinstance(data, list):
+                return {
+                    "timestamps": [float(item) for item in data],
+                    "leases": {},
+                }
+            if not isinstance(data, dict):
+                return {"timestamps": [], "leases": {}}
+            timestamps = data.get("timestamps", []) or []
+            leases = data.get("leases", {}) or {}
+            return {
+                "timestamps": [float(item) for item in timestamps],
+                "leases": {str(key): float(value) for key, value in leases.items()},
+            }
+        except Exception:
+            return {"timestamps": [], "leases": {}}
+
+    def _save_state(self, state: dict) -> None:
+        payload = json.dumps(state, ensure_ascii=False)
+        self.state_file.write_text(payload, encoding="utf-8")
+
+    def _normalize_state(self, state: dict, now: float) -> dict:
+        timestamps = [
+            float(ts)
+            for ts in (state.get("timestamps", []) or [])
+            if now - float(ts) < self.window_seconds
+        ]
+        leases = {
+            str(key): float(value)
+            for key, value in (state.get("leases", {}) or {}).items()
+            if now - float(value) < self.lease_seconds
+        }
+        return {"timestamps": timestamps, "leases": leases}
+
+    def acquire(self) -> None:
+        token = None
+        while True:
+            token = self.try_acquire_slot()
+            if token:
+                self.release(token)
+                return
+            time.sleep(0.05)
+
+    def try_acquire_slot(self) -> str | None:
+        while True:
+            wait_time = 0.0
+            with self._thread_lock:
+                with open(self.lock_file, "a+", encoding="utf-8") as lock_fp:
+                    fcntl.flock(lock_fp.fileno(), fcntl.LOCK_EX)
+                    now = time.time()
+                    state = self._normalize_state(self._load_state(), now)
+                    timestamps = state["timestamps"]
+                    leases = state["leases"]
+
+                    if len(timestamps) < self.max_requests and len(leases) < self.max_concurrent:
+                        token = uuid4().hex
+                        timestamps.append(now)
+                        leases[token] = now
+                        self._save_state(state)
+                        fcntl.flock(lock_fp.fileno(), fcntl.LOCK_UN)
+                        return token
+
+                    wait_candidates = []
+                    if len(timestamps) >= self.max_requests and timestamps:
+                        wait_candidates.append(self.window_seconds - (now - timestamps[0]))
+                    if len(leases) >= self.max_concurrent:
+                        wait_candidates.append(0.05)
+                    wait_time = max(0.05, min([item for item in wait_candidates if item > 0] or [0.05]))
+                    fcntl.flock(lock_fp.fileno(), fcntl.LOCK_UN)
+
+            time.sleep(wait_time)
+
+    def release(self, token: str | None) -> None:
+        if not token:
+            return
+        with self._thread_lock:
+            with open(self.lock_file, "a+", encoding="utf-8") as lock_fp:
+                fcntl.flock(lock_fp.fileno(), fcntl.LOCK_EX)
+                now = time.time()
+                state = self._normalize_state(self._load_state(), now)
+                leases = state["leases"]
+                if token in leases:
+                    leases.pop(token, None)
+                    self._save_state(state)
+                else:
+                    self._save_state(state)
+                fcntl.flock(lock_fp.fileno(), fcntl.LOCK_UN)
+
     def can_make_request(self) -> bool:
-        """
-        检查是否可以立即发起请求（非阻塞）
-        """
-        with self.lock:
-            now = time.time()
-            
-            # 清理超过1秒的请求记录
-            while self.requests and now - self.requests[0] >= 1.0:
-                self.requests.popleft()
-            
-            return len(self.requests) < self.max_requests
+        with self._thread_lock:
+            with open(self.lock_file, "a+", encoding="utf-8") as lock_fp:
+                fcntl.flock(lock_fp.fileno(), fcntl.LOCK_EX)
+                now = time.time()
+                state = self._normalize_state(self._load_state(), now)
+                self._save_state(state)
+                allowed = (
+                    len(state["timestamps"]) < self.max_requests
+                    and len(state["leases"]) < self.max_concurrent
+                )
+                fcntl.flock(lock_fp.fileno(), fcntl.LOCK_UN)
+                return allowed
 
 
-# 全局速率限制器实例
-global_rate_limiter = RateLimiter(max_requests_per_second=5)
+global_rate_limiter = RateLimiter(
+    max_requests_per_second=int(os.getenv("PROXY_MAX_REQUESTS_PER_SECOND", "5"))
+)
+
+
+def _should_limit_proxy_requests() -> bool:
+    """
+    仅在当前进程实际启用代理时启用全局代理限流。
+    """
+    try:
+        return is_proxy_enabled()
+    except Exception:
+        return True
 
 
 def wait_for_request():
-    """
-    等待直到可以发起请求
-    """
+    """等待直到可以发起请求。"""
+    if not _should_limit_proxy_requests():
+        return
     global_rate_limiter.acquire()
 
 
 def can_request_now() -> bool:
-    """
-    检查是否可以立即发起请求
-    """
+    """检查是否可以立即发起请求。"""
+    if not _should_limit_proxy_requests():
+        return True
     return global_rate_limiter.can_make_request()
+
+
+@contextmanager
+def request_slot():
+    """
+    申请一个跨进程共享的请求槽位，请求结束后自动释放。
+
+    这样既能限制“每秒启动多少请求”，也能限制“同时在飞多少请求”。
+    """
+    if not _should_limit_proxy_requests():
+        yield
+        return
+
+    token = global_rate_limiter.try_acquire_slot()
+    try:
+        yield
+    finally:
+        global_rate_limiter.release(token)

weixin.py

@@ -0,0 +1,377 @@
+import copy
+import json
+import os
+import re
+import sys
+import time
+from html import unescape
+from http.cookies import SimpleCookie
+from typing import Dict, Optional
+from urllib.parse import urlencode
+
+import requests
+import urllib3
+
+current_dir = os.path.dirname(os.path.abspath(__file__))
+project_root = os.path.dirname(current_dir)
+for path in (current_dir, project_root):
+    if path not in sys.path:
+        sys.path.append(path)
+
+import config as project_config
+from utils.rate_limiter import wait_for_request, global_rate_limiter
+
+API_ENDPOINT = "https://mp.weixin.qq.com/cgi-bin/videosnap"
+DOMAIN = "mp.weixin.qq.com"
+DEFAULT_HEADERS = {
+    "User-Agent": (
+        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+        "AppleWebKit/537.36 (KHTML, like Gecko) "
+        "Chrome/146.0.0.0 Safari/537.36"
+    ),
+    "Accept": "*/*",
+    "Accept-Language": "zh-CN,zh;q=0.9,en;q=0.8,zh-TW;q=0.7",
+    "DNT": "1",
+    "Priority": "u=1, i",
+    "Sec-CH-UA": '"Chromium";v="146", "Not-A.Brand";v="24", "Google Chrome";v="146"',
+    "Sec-CH-UA-Mobile": "?0",
+    "Sec-CH-UA-Platform": '"Windows"',
+    "Sec-Fetch-Dest": "empty",
+    "Sec-Fetch-Mode": "cors",
+    "Sec-Fetch-Site": "same-origin",
+    "X-Requested-With": "XMLHttpRequest",
+}
+DEFAULT_WEIXIN_CONFIG = {
+    "TOKEN": "609153506",
+    "FINGERPRINT": "46a7e6ac6ccf205986adc0aa99127860",
+    "COOKIE": {
+        "appmsglist_action_3258147150": "card",
+        "_qimei_uuid42": "1a302160d051008226aec905b63f99ff3989f30009",
+        "_qimei_i_3": "63b22b84c15204dfc595ac6452d722b1f0bdf0f6145b568ae68a7c0e70947438686637943989e2a1d792",
+        "_qimei_h38": "215986ce26aec905b63f99ff0200000e81a302",
+        "ua_id": "S7gglu0eZh9NkAzLAAAAADH8dynpnFZVN29lxm7BQo0=",
+        "wxuin": "73074968761097",
+        "mm_lang": "zh_CN",
+        "eas_sid": "91X7I7K4K5k364U2z3k2I980F5",
+        "_qimei_q36": "",
+        "_qimei_fingerprint": "d895c46d5fda98cab67d9daec00068ed",
+        "_clck": "501quy|1|g4t|0",
+        "uuid": "210d1c199a63afd4c774eccd9a06a27f",
+        "rand_info": "CAESIE4WqrFFVVjqrrNflbCUM7wPD5NXjuGbjfHolAEsMmEm",
+        "slave_bizuin": "3258147150",
+        "data_bizuin": "3258147150",
+        "bizuin": "3258147150",
+        "data_ticket": "tpcLjRB7B7AlUY3rFe/ILEjtCKs7dEEGsn8kXnHVzdTb9dgIpSPN1aP8FlE6FDhj",
+        "slave_sid": "U3hfU1Z0UV91N0U5d0lkRDhyTzh3d3hmbnBHMjBnbmFNdzVJeGlJeTJ6OTVxRjJQVVE2VkNhejYzTkxETVVSZkF3eWRORmtRS01XWFBjdnFZZWFLNjR2ZGtwdUJ2MzByclg0NjF4SHlDeVJneEhsczdSYUJVNE45VEhNRWVTQXg1dlpGdWQ0bU5VM3pnRzJN",
+        "slave_user": "gh_fe76760560d0",
+        "xid": "ef503a6864cceaef225c615a45606e4a",
+        "_clsk": "12arnf1|1774975723874|4|1|mp.weixin.qq.com/weheat-agent/payload/record",
+        "_qimei_i_1": "2ddc6a80945f59d3c7c4ab325dd526b3feeea1a31458558bbdd97e582493206c6163629d39d8e1dcd49fddc7"
+    },
+    "COUNT": 21,
+    "REFERER": "https://mp.weixin.qq.com/",
+    "HEADERS": {},
+    "REQUEST_PARAMS": {
+        "action": "search",
+        "scene": "1",
+        "lang": "zh_CN",
+        "f": "json",
+        "ajax": "1",
+    },
+    "REQUESTS_PER_SECOND": 5,
+    "PAGE_DELAY": 5,
+    "CITY_DELAY": 2,
+}
+
+
+def _deep_merge_dict(base: Dict, incoming: Dict) -> Dict:
+    merged = copy.deepcopy(base)
+    for key, value in incoming.items():
+        if isinstance(value, dict) and isinstance(merged.get(key), dict):
+            merged[key] = _deep_merge_dict(merged[key], value)
+        else:
+            merged[key] = value
+    return merged
+
+
+def _parse_cookie_value(cookie_value) -> Dict[str, str]:
+    if isinstance(cookie_value, dict):
+        return {str(key): str(value) for key, value in cookie_value.items()}
+
+    if not cookie_value:
+        return {}
+
+    if isinstance(cookie_value, str):
+        text = cookie_value.strip()
+        if not text:
+            return {}
+        try:
+            parsed = json.loads(text)
+        except json.JSONDecodeError:
+            parsed = None
+        if isinstance(parsed, dict):
+            return {str(key): str(value) for key, value in parsed.items()}
+
+        cookie = SimpleCookie()
+        cookie.load(text)
+        return {key: morsel.value for key, morsel in cookie.items()}
+
+    return {}
+
+
+def _load_weixin_config() -> Dict:
+    config = copy.deepcopy(DEFAULT_WEIXIN_CONFIG)
+    module_config = getattr(project_config, "WEIXIN_CONFIG", None)
+    if isinstance(module_config, dict):
+        config = _deep_merge_dict(config, module_config)
+
+    env_mapping = {
+        "TOKEN": os.getenv("WEIXIN_TOKEN"),
+        "FINGERPRINT": os.getenv("WEIXIN_FINGERPRINT"),
+        "COOKIE": os.getenv("WEIXIN_COOKIE"),
+        "REFERER": os.getenv("WEIXIN_REFERER"),
+        "COUNT": os.getenv("WEIXIN_COUNT"),
+        "REQUESTS_PER_SECOND": os.getenv("WEIXIN_REQUESTS_PER_SECOND"),
+        "PAGE_DELAY": os.getenv("WEIXIN_PAGE_DELAY"),
+        "CITY_DELAY": os.getenv("WEIXIN_CITY_DELAY"),
+    }
+    for key, value in env_mapping.items():
+        if value not in (None, ""):
+            config[key] = value
+
+    config["COOKIE"] = _parse_cookie_value(config.get("COOKIE"))
+
+    for key in ("COUNT", "REQUESTS_PER_SECOND"):
+        try:
+            config[key] = int(config[key])
+        except (TypeError, ValueError):
+            config[key] = DEFAULT_WEIXIN_CONFIG[key]
+
+    for key in ("PAGE_DELAY", "CITY_DELAY"):
+        try:
+            config[key] = float(config[key])
+        except (TypeError, ValueError):
+            config[key] = DEFAULT_WEIXIN_CONFIG[key]
+
+    return config
+
+
+def _strip_html(text: str) -> str:
+    if not text:
+        return ""
+    return re.sub(r"<[^>]+>", "", unescape(text)).strip()
+
+
+class WeixinSpider:
+    """基于 requests 的微信视频号采集器"""
+
+    def __init__(self, db_connection):
+        self.db = db_connection
+        self.config = _load_weixin_config()
+        self.token = str(self.config.get("TOKEN", "")).strip()
+        self.fingerprint = str(self.config.get("FINGERPRINT", "")).strip()
+        self.cookies = self.config.get("COOKIE", {})
+        self.count = str(self.config.get("COUNT", DEFAULT_WEIXIN_CONFIG["COUNT"]))
+        self.referer = str(self.config.get("REFERER", DEFAULT_WEIXIN_CONFIG["REFERER"])).strip()
+        self.request_params = {
+            str(key): str(value)
+            for key, value in (self.config.get("REQUEST_PARAMS", {}) or {}).items()
+            if value is not None
+        }
+        self.page_delay = max(0.0, float(self.config.get("PAGE_DELAY", DEFAULT_WEIXIN_CONFIG["PAGE_DELAY"])))
+        self.city_delay = max(0.0, float(self.config.get("CITY_DELAY", DEFAULT_WEIXIN_CONFIG["CITY_DELAY"])))
+        max_rps = self.config.get("REQUESTS_PER_SECOND")
+        if max_rps:
+            global_rate_limiter.max_requests = int(max_rps)
+
+        headers = DEFAULT_HEADERS.copy()
+        project_headers = getattr(project_config, "HEADERS", None)
+        if isinstance(project_headers, dict):
+            headers.update(project_headers)
+        config_headers = self.config.get("HEADERS", {})
+        if isinstance(config_headers, dict):
+            headers.update({str(key): str(value) for key, value in config_headers.items()})
+        if self.referer:
+            headers["Referer"] = self.referer
+        self.session = requests.Session()
+        self.session.trust_env = False
+        self.session.headers.update(headers)
+        if self.cookies:
+            self.session.cookies.update(self.cookies)
+        urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
+
+    def _validate_runtime_config(self) -> bool:
+        missing = []
+        if not self.token:
+            missing.append("TOKEN")
+        if not self.fingerprint:
+            missing.append("FINGERPRINT")
+        if not self.cookies:
+            missing.append("COOKIE")
+
+        if not missing:
+            return True
+
+        print(
+            "[微信] 配置不完整，缺少: "
+            + ", ".join(missing)
+            + "。请在 config.py 的 WEIXIN_CONFIG 中补齐，"
+            + "或通过环境变量 WEIXIN_TOKEN / WEIXIN_FINGERPRINT / WEIXIN_COOKIE 提供。"
+        )
+        return False
+
+    def _load_areas(self):
+        condition = "domain='maxlaw' AND level=2"
+        tables = ("area_new", "area", "area2")
+        last_error = None
+        for table in tables:
+            try:
+                rows = self.db.select_data(table, "province, city", condition) or []
+            except Exception as exc:
+                last_error = exc
+                continue
+            if rows:
+                print(f"[微信] 城市来源表: {table}, 城市数: {len(rows)}")
+                return rows
+
+        if last_error:
+            print(f"[微信] 加载地区数据失败: {last_error}")
+        print("[微信] 无城市数据（已尝试 area_new/area/area2）")
+        return []
+
+    def _build_query_url(self, query: str, buffer: str) -> str:
+        params = self.request_params.copy()
+        params.update({
+            "query": query,
+            "count": self.count,
+            "buffer": buffer,
+            "fingerprint": self.fingerprint,
+            "token": self.token,
+        })
+        return f"{API_ENDPOINT}?{urlencode(params)}"
+
+    def _extract_phone(self, text: str) -> Optional[str]:
+        if not text:
+            return None
+        match = re.search(r"1[3-9]\d{9}", text)
+        return match.group(0) if match else None
+
+    def _parse_name(self, acct: Dict) -> str:
+        highlight = _strip_html(acct.get("highlight_nickname", ""))
+        if highlight:
+            return highlight
+        return _strip_html(acct.get("nickname", ""))
+
+    def _store_account(self, acct: Dict, province: str, city: str) -> None:
+        signature = acct.get("signature", "")
+        phone = self._extract_phone(signature)
+        if not phone:
+            return
+
+        if self.db.is_data_exist("lawyer", f"phone='{phone}' and domain='{DOMAIN}'"):
+            name = self._parse_name(acct)
+            print(f"  -- 已存在律师: {name} ({phone})")
+            return
+
+        params = json.dumps(acct, ensure_ascii=False)
+        lawyer_data = {
+            "phone": phone,
+            "province": province,
+            "city": city,
+            "law_firm": acct.get("auth_info", {}).get("auth_profession"),
+            "url": f"https://channels.weixin.qq.com/finder/creator/feeds?finder_username={acct.get('username', '')}",
+            "create_time": int(time.time()),
+            "domain": DOMAIN,
+            "name": self._parse_name(acct),
+            "params": params,
+        }
+
+        try:
+            inserted_id = self.db.insert_data("lawyer", lawyer_data)
+            print(f"  -> 新增律师: {lawyer_data['name']} ({phone}), ID: {inserted_id}")
+        except Exception as exc:
+            print(f"  插入失败 {lawyer_data['name']} ({phone}): {exc}")
+
+    def _search_city(self, province: str, city: str) -> None:
+        city_name = city.replace('市', '')
+        query = f"{city_name}律所"
+        print(f"--- [微信] 开始采集城市: {province} - {city_name} ---")
+
+        buffer = ""
+        has_more = True
+        page_no = 0
+
+        while has_more:
+            page_no += 1
+            url = self._build_query_url(query, buffer)
+            print(f"正在采集 '{query}' 第 {page_no} 页: {url}")
+
+            wait_for_request()
+            try:
+                response = self.session.get(
+                    url,
+                    timeout=15,
+                    cookies=self.cookies,
+                    proxies={},  # 明确禁用代理
+                    verify=False,
+                )
+                response.raise_for_status()
+                data = response.json()
+            except requests.exceptions.RequestException as exc:
+                print(f"网络请求失败: {exc}")
+                break
+            except json.JSONDecodeError:
+                print("解析返回的JSON失败。返回内容:", response.text[:200])
+                break
+
+            base_resp = data.get("base_resp", {})
+            if base_resp.get("ret") != 0:
+                print(f"API返回错误: {base_resp.get('err_msg')}")
+                if "invalid ticket" in (base_resp.get('err_msg') or ""):
+                    print("Token 或 Cookie 可能失效，请更新配置。")
+                break
+
+            accounts = data.get("acct_list", [])
+            if not accounts:
+                print("本页未找到更多律师信息。")
+                break
+
+            for acct in accounts:
+                self._store_account(acct, province, city_name)
+
+            has_more = bool(data.get("acct_continue_flag"))
+            buffer = data.get("last_buff", "")
+            time.sleep(self.page_delay)
+
+        print(f"--- [微信] 城市: {city_name} 采集完成 ---\n")
+
+    def run(self) -> None:
+        print("启动微信视频号律师信息采集...")
+        if not self._validate_runtime_config():
+            return
+
+        areas = self._load_areas()
+        if not areas:
+            print("[微信] 未能从 `area_new` 表获取到地区信息。")
+            return
+
+        for area in areas:
+            province = area.get("province", "")
+            city = area.get("city", "")
+            if not city:
+                continue
+            try:
+                self._search_city(province, city)
+            except Exception as exc:
+                print(f"采集 {province}-{city} 时发生错误: {exc}")
+            time.sleep(self.city_delay)
+
+        print("微信视频号律师信息采集完成。")
+
+
+if __name__ == "__main__":
+    from Db import Db
+
+    with Db() as db:
+        spider = WeixinSpider(db)
+        spider.run()