初始化项目 v2.0：中国邮政揽收系统自动填单工具

基于 HTTP API + Node.js token bridge 方案，不依赖浏览器自动化。

pickup_api.py

@@ -0,0 +1,604 @@
+"""揽收系统 HTTP API 封装 — 直接调用后端接口，不依赖浏览器。
+
+依赖 Node.js token bridge 计算 $_ts 安全框架的 c93l5Fya 动态 token。
+"""
+
+from __future__ import annotations
+
+import json
+import os
+import re
+import subprocess
+import sys
+from typing import Any
+from urllib.parse import urlparse, urljoin
+
+from log import debug
+
+
+# ─── Token Bridge 管理 ────────────────────────────────────────────────
+
+def _get_bridge_dir() -> str:
+    """获取 token_bridge 目录路径（支持 PyInstaller 打包）。"""
+    if getattr(sys, "frozen", False):
+        base = sys._MEIPASS
+    else:
+        base = os.path.dirname(os.path.abspath(__file__))
+    return os.path.join(base, "token_bridge")
+
+
+class TokenBridge:
+    """管理 Node.js token 计算子进程（HTTP 通信）。"""
+
+    def __init__(self):
+        self._proc: subprocess.Popen | None = None
+        self._port: int = 0
+        self._base_url: str = ""
+
+    def start(self):
+        bridge_dir = _get_bridge_dir()
+        kwargs = dict(
+            cwd=bridge_dir,
+            stderr=subprocess.PIPE,
+            text=True, encoding="utf-8", errors="replace",
+        )
+        if sys.platform == "win32":
+            kwargs["creationflags"] = subprocess.CREATE_NO_WINDOW
+        self._proc = subprocess.Popen(["node", "server.js", "0"], **kwargs)
+
+        # 进程退出时强制杀子进程，防止 _MEIPASS 目录无法删除
+        import atexit as _atexit
+        _atexit.register(self.stop)
+        # 读取就绪信号（包含端口号）
+        line = self._proc.stderr.readline()
+        try:
+            msg = json.loads(line.strip())
+            if not msg.get("ready"):
+                raise RuntimeError(f"Bridge startup failed: {line}")
+            self._port = msg["port"]
+            self._base_url = f"http://127.0.0.1:{self._port}"
+        except json.JSONDecodeError:
+            raise RuntimeError(f"Bridge unexpected stderr: {line}")
+
+        # 后台线程持续读取 stderr 并输出到 log
+        import threading as _threading
+        def _read_stderr():
+            from log import trace
+            for line in self._proc.stderr:
+                trace(f"bridge stderr: {line.strip()}")
+        _threading.Thread(target=_read_stderr, daemon=True).start()
+
+    def stop(self):
+        if self._proc:
+            try:
+                self._proc.kill()
+                self._proc.wait(timeout=3)
+            except Exception:
+                pass
+            self._proc = None
+            self._port = 0
+
+    def _request(self, path: str, body: dict = None) -> dict:
+        """向 bridge HTTP 服务器发送请求。"""
+        from log import debug, trace
+        import requests as _requests
+        url = f"{self._base_url}{path}"
+        trace(f"bridge -> {path} {str(body)[:100] if body else ''}")
+        try:
+            if body:
+                resp = _requests.post(url, json=body, timeout=30)
+            else:
+                resp = _requests.get(url, timeout=30)
+            resp.raise_for_status()
+            result = resp.json()
+            trace(f"bridge <- {str(result)[:100]}")
+            if "error" in result:
+                raise RuntimeError(f"Bridge error: {result['error']}")
+            return result
+        except _requests.RequestException as e:
+            # 读取 stderr
+            err = ""
+            try:
+                while True:
+                    chunk = self._proc.stderr.readline()
+                    if not chunk:
+                        break
+                    err += chunk
+            except Exception:
+                pass
+            if err:
+                raise RuntimeError(f"Bridge request failed: {e}, stderr={err[:300]}")
+            raise
+
+    def initialize(self, html: str, scripts: list[dict], base_url: str):
+        return self._request("/init", {
+            "html": html,
+            "scripts": scripts,
+            "baseUrl": base_url,
+        })
+
+    def compute_token(self, method: str, path: str) -> str:
+        result = self._request("/token", {
+            "method": method,
+            "path": path,
+        })
+        return result["token"]
+
+    def ping(self) -> bool:
+        try:
+            self._request("/ping")
+            return True
+        except Exception:
+            return False
+
+
+# ─── HTTP 会话 & 页面资源管理 ─────────────────────────────────────────
+
+class PickupSession:
+    """管理 HTTP 会话（cookies）和页面资源加载。"""
+
+    def __init__(self, base_url: str):
+        self.base_url = base_url.rstrip("/")
+        parsed = urlparse(self.base_url)
+        self._origin = f"{parsed.scheme}://{parsed.netloc}"
+        # 使用 requests 保持 cookie 会话（同步，足够简单可靠）
+        import requests as _requests
+        self._http: _requests.Session = _requests.Session()
+        self._http.verify = False  # 内部自签证书
+        import urllib3
+        urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
+        self._http.headers.update({
+            "User-Agent": (
+                "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+                "AppleWebKit/537.36 (KHTML, like Gecko) "
+                "Chrome/109.0.0.0 Safari/537.36"
+            ),
+            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
+            "Cache-Control": "no-cache",
+            "Pragma": "no-cache",
+        })
+        self._scripts: list[dict] = []
+        self._html: str = ""
+        self._pickup_path: str = ""
+        self._ts_cd: str = ""
+        self._ts_nsd: int = 0
+
+    @property
+    def cookies_str(self) -> str:
+        """将 requests cookies 转为 Cookie 请求头字符串。"""
+        pairs = []
+        for cookie in self._http.cookies:
+            pairs.append(f"{cookie.name}={cookie.value}")
+        return "; ".join(pairs)
+
+    def load_session(self, session_path: str) -> bool:
+        """从文件加载之前保存的 cookies。返回是否加载成功。"""
+        import requests as _requests
+        try:
+            with open(session_path, "r") as f:
+                data = json.load(f)
+            seen = set()
+            for cookie_data in data.get("cookies", []):
+                key = (cookie_data["name"], cookie_data.get("domain", ""), cookie_data.get("path", "/"))
+                if key in seen:
+                    continue
+                seen.add(key)
+                self._http.cookies.set(
+                    cookie_data["name"],
+                    cookie_data["value"],
+                    domain=cookie_data.get("domain", ""),
+                    path=cookie_data.get("path", "/"),
+                )
+            return len(seen) > 0
+        except (FileNotFoundError, json.JSONDecodeError, KeyError):
+            return False
+
+    def save_session(self, session_path: str):
+        """保存 cookies 到文件。"""
+        data = {
+            "cookies": [
+                {"name": c.name, "value": c.value, "domain": c.domain, "path": c.path}
+                for c in self._http.cookies
+            ]
+        }
+        with open(session_path, "w") as f:
+            json.dump(data, f, ensure_ascii=False, indent=2)
+
+    def _abs_url(self, path_or_url: str) -> str:
+        if path_or_url.startswith("http"):
+            return path_or_url
+        return urljoin(self._origin, path_or_url)
+
+    def fetch_page_resources(self) -> dict:
+        """加载 pickup 页面及所需 JS 文件。返回 {html, scripts, base_url, nsd}。"""
+        import requests as _requests
+
+        pickup_path = urlparse(self.base_url).path
+        self._pickup_path = pickup_path
+
+        # 加载 pickup 页面（需要 Referer 满足安全框架要求）
+        portal_url = f"{self._origin}/portal/a"
+        debug(f"fetch_page: loading {self.base_url}, cookies={list(self._http.cookies.keys())}")
+        resp = self._http.get(
+            self.base_url,
+            headers={
+                "Referer": portal_url,
+                "Sec-Fetch-Dest": "iframe",
+                "Sec-Fetch-Mode": "navigate",
+                "Sec-Fetch-Site": "same-origin",
+                "Upgrade-Insecure-Requests": "1",
+            },
+            timeout=30,
+        )
+        debug(f"fetch_page: status={resp.status_code}")
+        if resp.status_code != 200:
+            debug(f"fetch_page: resp_headers={dict(resp.headers)}")
+            debug(f"fetch_page: resp_body={resp.text[:500]}")
+        resp.raise_for_status()
+        self._html = resp.text
+
+        # 从页面 HTML 中提取 $_ts 安全参数
+        self._ts_cd = ""
+        self._ts_nsd = 0
+        m = re.search(r'\$_ts\.cd="([^"]*)"', self._html)
+        if m:
+            self._ts_cd = m.group(1)
+        m = re.search(r'\$_ts\.nsd=(\d+)', self._html)
+        if m:
+            self._ts_nsd = int(m.group(1))
+        if not self._ts_cd or not self._ts_nsd:
+            raise RuntimeError("Failed to extract $_ts.cd/nsd from page")
+
+        # 2. 提取外部 JS 引用，只保留 security JS（cfd114a.js）
+        #    页面业务脚本（jQuery、main.js 等）会导致 jsdom 卡死，不需要
+        script_urls = re.findall(
+            r'<script[^>]+src="([^"]+)"[^>]*>', self._html, re.IGNORECASE
+        )
+        self._scripts = []
+        for url in script_urls:
+            if "cfd114a.js" not in url:
+                continue
+            abs_url = self._abs_url(url)
+            try:
+                script_resp = self._http.get(abs_url, timeout=15)
+                if script_resp.status_code == 200:
+                    self._scripts.append({"url": url, "content": script_resp.text})
+                    debug(f"fetch_page: loaded security JS {url}")
+            except _requests.RequestException:
+                debug(f"Failed to fetch script: {abs_url}")
+
+        if not self._scripts:
+            raise RuntimeError("Security JS (cfd114a.js) not found on page")
+
+        return {
+            "html": self._html,
+            "scripts": self._scripts,
+            "base_url": self.base_url,
+            "ts_cd": self._ts_cd,
+            "ts_nsd": self._ts_nsd,
+        }
+
+    def request(self, method: str, path: str, token: str,
+                data: dict[str, str] = None) -> dict[str, Any]:
+        """向 pickup-web 后端发送 API 请求。
+
+        Args:
+            method: HTTP 方法 (GET/POST)
+            path: API 路径 (如 /pickup-web/a/common/customerautocomplete)
+            token: c93l5Fya 安全 token
+            data: POST 表单数据
+
+        Returns:
+            JSON 响应（如果是 JSON），否则返回 {"_text": response_text}
+        """
+        full_url = self._abs_url(path)
+        if "?" in full_url:
+            full_url += f"&c93l5Fya={token}"
+        else:
+            full_url += f"?c93l5Fya={token}"
+
+        headers = {
+            "X-Requested-With": "XMLHttpRequest",
+            "Accept": "application/json, text/javascript, */*; q=0.01",
+            "Referer": self.base_url,
+        }
+        if data:
+            headers["Content-Type"] = "application/x-www-form-urlencoded; charset=UTF-8"
+
+        debug(f"API {method} {path}")
+        resp = self._http.request(
+            method, full_url,
+            headers=headers,
+            data=data or None,
+            timeout=30,
+        )
+        debug(f"API {method} {path} → {resp.status_code}")
+        if resp.status_code >= 400:
+            debug(f"API error body: {resp.text[:500]}")
+        resp.raise_for_status()
+
+        text = resp.text
+        if text and ("<html" in text[:300].lower() or "<!doctype" in text[:300].lower()):
+            raise RuntimeError("Session 已过期，返回了登录页面，请重新登录并获取 cookies")
+
+        content_type = resp.headers.get("Content-Type", "")
+        if "json" in content_type or text.strip().startswith(("[", "{")):
+            try:
+                return resp.json()
+            except ValueError:
+                pass
+        return {"_text": text}
+
+
+# ─── 业务 API ─────────────────────────────────────────────────────────
+
+class PickupAPI:
+    """揽收系统业务 API 封装。
+
+    每个方法对应页面上的一个 AJAX 操作。
+    所有需要 token 的方法都通过 token_bridge 计算 c93l5Fya 参数。
+    """
+
+    def __init__(self, session: PickupSession, bridge: TokenBridge):
+        self._sess = session
+        self._bridge = bridge
+
+    def _token(self, method: str, path: str) -> str:
+        return self._bridge.compute_token(method, path)
+
+    def _post(self, path: str, data: dict[str, str] = None) -> dict:
+        token = self._token("POST", path)
+        return self._sess.request("POST", path, token, data)
+
+    def _get(self, path: str) -> dict:
+        token = self._token("GET", path)
+        return self._sess.request("GET", path, token)
+
+    # ── 产品 ──────────────────────────────────────────────────
+
+    def query_product(self, keyword: str) -> list[dict]:
+        """搜索产品名称，返回候选列表（data=id|bizProductNo, value=显示名）。"""
+        result = self._post(
+            "/pickup-web/a/postwaybilledit/queryproductmanager",
+            data={"no": "on", "query": keyword},
+        )
+        from log import trace
+        trace(f"query_product raw: {str(result)[:300]}")
+        suggestions = result.get("suggestions", []) if isinstance(result, dict) else []
+        # 解析 data 字段: "productId|bizProductNo"
+        parsed = []
+        for s in suggestions:
+            data_parts = s.get("data", "").split("|")
+            parsed.append({
+                "productId": data_parts[0] if len(data_parts) > 0 else "",
+                "bizProductNo": data_parts[1] if len(data_parts) > 1 else "",
+                "productName": s.get("value", ""),
+            })
+        return parsed
+
+    def select_product(self, biz_product_id: str, biz_product_no: str) -> dict:
+        """选择产品后获取产品详情（模板、材料等）。"""
+        product_info = {"bizProductId": biz_product_id, "bizProductNo": biz_product_no}
+
+        # 获取模板（返回数组，取第一个）
+        template = self._post(
+            "/pickup-web/a/common/templatecodemodel",
+            data={"bizProductNo": biz_product_no},
+        )
+        tmpl = template[0] if isinstance(template, list) and template else template
+        if isinstance(tmpl, dict):
+            product_info.update(tmpl)
+
+        # 查询产品详情
+        detail = self._post(
+            "/pickup-web/a/common/queryproductbymanager",
+            data={
+                "bizProductId": biz_product_id,
+                "bizProductNo": biz_product_no,
+            },
+        )
+        if isinstance(detail, dict):
+            product_info.update(detail)
+        elif isinstance(detail, list) and detail:
+            product_info.update(detail[0])
+
+        # 查询物料信息（返回数组或 dict）
+        material = self._post(
+            "/pickup-web/a/common/findppoductmaterialid",
+            data={"productCode": biz_product_no},
+        )
+        mat = material[0] if isinstance(material, list) and material else material
+        if isinstance(mat, dict):
+            product_info["material"] = mat
+
+        return product_info
+
+    # ── 客户 ──────────────────────────────────────────────────
+
+    def query_customer(self, keyword: str) -> list[dict]:
+        """搜索客户名称，返回候选列表（data=customerId, valueName=短名, value=全名）。"""
+        result = self._post(
+            "/pickup-web/a/common/customerautocomplete",
+            data={"query": keyword},
+        )
+        from log import trace
+        trace(f"query_customer raw: {str(result)[:300]}")
+        suggestions = result.get("suggestions", []) if isinstance(result, dict) else []
+        parsed = []
+        for s in suggestions:
+            parsed.append({
+                "customerId": s.get("data", ""),
+                "customerName": s.get("valueName", s.get("value", "")),
+                "customerFullName": s.get("value", ""),
+            })
+        return parsed
+
+    def select_customer(self, customer_id: str) -> dict:
+        """选择客户后获取客户详情。"""
+        result = self._post(
+            "/pickup-web/a/common/customerins",
+            data={"customerId": customer_id},
+        )
+        if isinstance(result, dict):
+            return result
+        return {}
+
+    def find_sender_history(self, sender_no: str) -> dict:
+        """根据寄件人编号查询实名历史。"""
+        result = self._post(
+            "/pickup-web/a/postwaybillargs/findrealnamehistorybysender",
+            data={"senderNo": sender_no},
+        )
+        if isinstance(result, dict):
+            return result
+        return {}
+
+    # ── 运单 ──────────────────────────────────────────────────
+
+    def query_waybill(self, waybill_no: str, ecommerce_no: str,
+                      biz_product_no: str, payment_mode: str,
+                      biz_product_id: str = "",
+                      post_blend: str = "0",
+                      lock_customer: str = "0") -> dict:
+        """查询运单信息。"""
+        return self._post(
+            "/pickup-web/a/postwaybill/querypostwaybillmanager",
+            data={
+                "waybillNo": waybill_no,
+                "ecommerceNo": ecommerce_no,
+                "bizProductNo": biz_product_no,
+                "postBlend": post_blend,
+                "lockCustomer": lock_customer,
+                "paymentMode": payment_mode,
+                "bizProductId": biz_product_id,
+            },
+        )
+
+    # ── 地址 ──────────────────────────────────────────────────
+
+    def address_match_raw(self, receiver_addr: str) -> dict:
+        """根据地址文本查询区划代码（addressMatch API）。"""
+        return self._post(
+            "/pickup-web/a/postwaybill/addressMatch",
+            data={"receiverAddr": receiver_addr},
+        )
+
+    def query_district(self, dist_code: str) -> dict:
+        """根据区划代码查询地址基本信息。"""
+        return self._post(
+            "/pickup-web/a/postwaybilledit/addressbasicno",
+            data={"dto": dist_code},
+        )
+
+    def match_address(self, receiver_arrive_org_no: str,
+                      base_product_no: str,
+                      waybill_no: str,
+                      receiver_addr: str,
+                      sender_no: str = "",
+                      transfer_type: str = "3",
+                      contents_attribute: str = "3",
+                      single_seal_flag: str = "0",
+                      inter_flag: str = "0",
+                      reserved29: str = "0",
+                      safety_type: str = "0",
+                      io_type: str = "10") -> dict:
+        """匹配收件地址，获取寄达地网格等信息。"""
+        return self._post(
+            "/pickup-web/a/common/findGridPrammer",
+            data={
+                "receiverArriveOrgNo": receiver_arrive_org_no,
+                "baseProductNo": base_product_no,
+                "transferType": transfer_type,
+                "contentsAttribute": contents_attribute,
+                "singleSealFlag": single_seal_flag,
+                "interFlag": inter_flag,
+                "reserved29": reserved29,
+                "waybillNo": waybill_no,
+                "receiverAddr": receiver_addr,
+                "senderNo": sender_no,
+                "safetyType": safety_type,
+                "ioType": io_type,
+            },
+        )
+
+    # ── 校验 ──────────────────────────────────────────────────
+
+    def check_product_material(self, cargo_names: str, waybill_no: str,
+                                product_code: str, dist_code: str) -> dict:
+        """校验产品物料。"""
+        return self._post(
+            "/pickup-web/a/common/checkppoductmaterialidnew",
+            data={
+                "cargoNames": cargo_names,
+                "cargoOptions": "",
+                "waybillNo": waybill_no,
+                "productCode": product_code,
+                "distCode": dist_code,
+            },
+        )
+
+    def check_reach_area(self, product_code: str, dist_code: str,
+                          cust_code: str, weight: str, waybill_no: str) -> dict:
+        """校验可达区域。"""
+        return self._post(
+            "/pickup-web/a/common/reachareaverification",
+            data={
+                "productCode": product_code,
+                "distCode": dist_code,
+                "custCode": cust_code,
+                "weight": weight,
+                "waybillNo": waybill_no,
+            },
+        )
+
+    def check_important_address(self, waybill_no: str, addr: str) -> dict:
+        """检查重要地址。"""
+        return self._post(
+            "/pickup-web/a/postwaybill/importantaddress",
+            data={"waybillNo": waybill_no, "queryAddr": addr},
+        )
+
+    def check_sensitive_customers(self, mobile_phones: str, waybill_no: str) -> dict:
+        """检查敏感客户。"""
+        return self._post(
+            "/pickup-web/a/postwaybillargs/findsensitivecustomers",
+            data={"mobilePhone": mobile_phones, "waybillNo": waybill_no},
+        )
+
+    # ── 杂项查询 ──────────────────────────────────────────────
+
+    def query_ecommerce(self, query: str = "B") -> list:
+        """查询电商平台列表。"""
+        return self._post(
+            "/pickup-web/a/postwaybill/queryecommerce",
+            data={"query": query},
+        )
+
+    def query_package_sort(self) -> list:
+        """查询包装分类。"""
+        return self._post("/pickup-web/a/common/querypackagesort")
+
+    def query_deliver_date(self) -> dict:
+        """查询可投递日期。"""
+        return self._post("/pickup-web/a/postwaybillargs/deliverdate")
+
+    def query_reason_temp_storage(self) -> dict:
+        """查询暂存原因列表。"""
+        return self._post("/pickup-web/a/postwaybill/queryreasonfortemporarystorage")
+
+    def check_customer_warning(self) -> dict:
+        """检查客户预警。"""
+        return self._post("/pickup-web/a/common/customerwarning")
+
+    def check_sheet_warning(self) -> dict:
+        """检查面单预警。"""
+        return self._post("/pickup-web/a/postwaybill/sheetWarning")
+
+    # ── 提交 ──────────────────────────────────────────────────
+
+    def submit(self, form_data: dict[str, str]) -> dict:
+        """提交揽收表单。返回 {"resCode": "", "resMess": "收寄成功", ...}。"""
+        return self._post(
+            "/pickup-web/a/postwaybill/isnotetosave",
+            data=form_data,
+        )