update

extensions/sync_ehr_to_oa/api.py

@@ -491,28 +491,55 @@ class SyncEhrToOaApi:
         for i in range(0, len(clean_ids), chunk_size):
             chunk = clean_ids[i : i + chunk_size]
             for uid in chunk:
+                profile: dict[str, Any] | None = None
+                for params in ({"userId": str(uid)}, {"userid": str(uid)}):
+                    try:
                         resp = self._client.request(
                             "GET",
                             "/UserFrameworkApiV3/api/v1/staffs/Get",
-                    params={"userId": uid},
+                            params=params,
                         )
-                payload = resp.json() if resp.content else {}
-                # 兼容多种返回结构
-                data = payload.get("data", payload)
-                if isinstance(data, list):
-                    items = [x for x in data if isinstance(x, dict)]
-                    if items:
-                        out[uid] = items[0]
+                    except Exception:
                         continue
-                if isinstance(data, dict):
-                    # 可能是单条，也可能是 map
-                    if "userId" in data or "UserId" in data:
-                        out[uid] = data
-                    else:
-                        # map 场景：key=uid
-                        d2 = data.get(str(uid))
-                        if isinstance(d2, dict):
-                            out[uid] = d2
-                # 非 200 业务码场景不硬失败，避免单个用户影响全量
+                    payload = resp.json() if resp.content else {}
+                    profile = self._find_staff_profile_by_uid(payload, uid)
+                    if profile is not None:
+                        break
+                if profile is not None:
+                    out[uid] = profile
         logger.info("EHR 员工详情查询完成：input_user_ids=%s matched_profiles=%s", len(clean_ids), len(out))
         return out
+
+    @staticmethod
+    def _find_staff_profile_by_uid(payload: Any, uid: int) -> dict[str, Any] | None:
+        def _iter_dicts(node: Any):
+            if isinstance(node, dict):
+                yield node
+                for v in node.values():
+                    yield from _iter_dicts(v)
+            elif isinstance(node, list):
+                for it in node:
+                    yield from _iter_dicts(it)
+
+        def _uid_from_dict(d: dict[str, Any]) -> int:
+            for k in ("userId", "UserId", "userid", "UserID", "id", "Id", "ID"):
+                if k in d:
+                    try:
+                        return int(str(d.get(k)).strip())
+                    except Exception:
+                        continue
+            return 0
+
+        best: dict[str, Any] | None = None
+        for d in _iter_dicts(payload):
+            if not isinstance(d, dict):
+                continue
+            duid = _uid_from_dict(d)
+            if duid != uid:
+                continue
+            # 优先返回带工号字段的对象
+            if any(k in d for k in ("staffCode", "StaffCode", "code", "Code", "jobNumber", "JobNumber", "employeeNo", "EmployeeNo")):
+                return d
+            if best is None:
+                best = d
+        return best

extensions/sync_ehr_to_oa/job.py

@@ -355,10 +355,12 @@ class SyncEhrToOaFormJob(BaseJob):
             # 3) 员工按工号归并（同工号保留“最新”记录）
             ehr_by_job_no: dict[str, dict[str, Any]] = {}
             ehr_by_job_no_norm: dict[str, dict[str, Any]] = {}
+            ehr_user_id_to_job_no: dict[int, str] = {}
             for item in emp_rows:
                 if not isinstance(item, dict):
                     continue
                 record = item.get("recordInfo") or {}
+                emp_info = item.get("employeeInfo") or {}
                 if not isinstance(record, dict):
                     continue
                 job_no = str(record.get("jobNumber") or "").strip()
@@ -370,12 +372,19 @@ class SyncEhrToOaFormJob(BaseJob):
                 if job_no_norm:
                     ex2 = ehr_by_job_no_norm.get(job_no_norm)
                     ehr_by_job_no_norm[job_no_norm] = item if ex2 is None else _choose_better_record(ex2, item)
+                try:
+                    uid = int((emp_info or {}).get("userID"))
+                except Exception:
+                    uid = 0
+                if uid > 0 and uid not in ehr_user_id_to_job_no:
+                    ehr_user_id_to_job_no[uid] = job_no
             logger.info(
-                "EHR 数据准备完成：employee_rows=%s organization_rows=%s distinct_job_numbers=%s distinct_job_numbers_norm=%s",
+                "EHR 数据准备完成：employee_rows=%s organization_rows=%s distinct_job_numbers=%s distinct_job_numbers_norm=%s userid_jobno_map=%s",
                 len(emp_rows),
                 len(org_rows),
                 len(ehr_by_job_no),
                 len(ehr_by_job_no_norm),
+                len(ehr_user_id_to_job_no),
             )
             # 3.1) 按员工 UserID 查询合同主体公司（firstParty），作为“所属公司”的优先来源
             contract_user_ids: list[int] = []
@@ -424,6 +433,10 @@ class SyncEhrToOaFormJob(BaseJob):
                 code = _extract_staff_code(profile)
                 if code:
                     user_id_to_staff_code[int(uid)] = code
+            # 兜底：若 staffs/Get 未返回工号，尝试从本次员工数据中反查
+            for uid in resolve_user_ids:
+                if int(uid) not in user_id_to_staff_code and int(uid) in ehr_user_id_to_job_no:
+                    user_id_to_staff_code[int(uid)] = ehr_user_id_to_job_no[int(uid)]
             # 3.4) 将 HRBP/汇报人工号转换为 OA 人员ID（member 字段要求）
             need_member_codes = list({c for c in user_id_to_staff_code.values() if str(c or "").strip()})
             code_to_member: dict[str, dict[str, str]] = {}
@@ -448,6 +461,9 @@ class SyncEhrToOaFormJob(BaseJob):
                 len(user_id_to_staff_code),
                 len(code_to_member),
             )
+            if verbose_trace:
+                sample_codes = list(user_id_to_staff_code.items())[:30]
+                logger.info("人员工号反查样本(uid->code)=%s", sample_codes)
             if verbose_trace:
                 for job_no in list(ehr_by_job_no.keys()):
                     logger.info("EHR 工号明细：raw=%s norm=%s", job_no, _normalize_job_no(job_no))