fixing

extensions/sync_ehr_to_oa/job.py

@@ -79,23 +79,6 @@ def _normalize_job_no(v: Any) -> str:
     return s.upper()
 
 
-def _to_int_candidate(v: Any) -> int | None:
-    if v is None:
-        return None
-    if isinstance(v, dict):
-        v = _cell_value(v)
-    s = str(v).strip()
-    if not s:
-        return None
-    try:
-        # 兼容 "123.0"
-        if "." in s and s.endswith(".0"):
-            return int(float(s))
-        return int(s)
-    except Exception:
-        return None
-
-
 def _extract_oa_row_id_and_fields(row: dict[str, Any]) -> tuple[int | None, dict[str, Any]]:
     """
     兼容不同 OA export 返回结构，提取：
@@ -111,20 +94,14 @@ def _extract_oa_row_id_and_fields(row: dict[str, Any]) -> tuple[int | None, dict
         for k, v in master.items():
             if isinstance(k, str) and k.startswith("field"):
                 field_map[k] = v
-        # 常见主键位置优先尝试
+        for candidate in (row.get("id"), row.get("masterDataId"), master.get("id")):
-        for candidate in (
+            if candidate is None:
-            row.get("id"),
+                continue
-            row.get("masterDataId"),
+            try:
-            master.get("id"),
+                row_id = int(str(candidate))
-            master.get("ID"),
-            master.get("recordId"),
-            master.get("dataId"),
-            master.get("_id"),
-        ):
-            rid = _to_int_candidate(candidate)
-            if rid is not None:
-                row_id = rid
                 break
+            except Exception:
+                continue
 
     # 结构 B：masterTable.record.fields = [{name,value,showValue}, ...]
     master_table = row.get("masterTable")
@@ -140,7 +117,12 @@ def _extract_oa_row_id_and_fields(row: dict[str, Any]) -> tuple[int | None, dict
                     if name:
                         field_map[name] = fld
             if row_id is None:
-                row_id = _to_int_candidate(record.get("id"))
+                rid = record.get("id")
+                if rid is not None:
+                    try:
+                        row_id = int(str(rid))
+                    except Exception:
+                        pass
 
     # 结构 C：行级 fields 列表
     row_fields = row.get("fields")
@@ -226,6 +208,12 @@ class SyncEhrToOaFormJob(BaseJob):
         verbose_trace = _to_bool_or_none(params.get("verbose_trace"))
         if verbose_trace is None:
             verbose_trace = True
+        preview_ehr_data = _to_bool_or_none(params.get("preview_ehr_data"))
+        if preview_ehr_data is None:
+            preview_ehr_data = True
+        preview_limit = int(params.get("preview_limit") or 20)
+        if preview_limit <= 0:
+            preview_limit = 20
 
         seeyon = SeeyonClient(base_url=oa_base_url, rest_user=rest_user, rest_password=rest_password, loginName=login_name)
         ehr = SyncEhrToOaApi(secret_params={"app_key": app_key, "app_secret": app_secret})
@@ -277,6 +265,51 @@ class SyncEhrToOaFormJob(BaseJob):
             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))
+            if preview_ehr_data:
+                logger.info("EHR 字段预览开始：limit=%s", preview_limit)
+                count = 0
+                for job_no, item in ehr_by_job_no.items():
+                    emp = item.get("employeeInfo") or {}
+                    rec = item.get("recordInfo") or {}
+                    if not isinstance(emp, dict):
+                        emp = {}
+                    if not isinstance(rec, dict):
+                        rec = {}
+
+                    org_oid = str(rec.get("oIdOrganization") or rec.get("oIdDepartment") or "").strip()
+                    org = org_by_oid.get(org_oid, {})
+                    company = str((org or {}).get("name") or "")
+                    name = str(emp.get("name") or "")
+                    rd_attr = _custom_prop_value(rec.get("customProperties"), rd_attr_custom_key) or _custom_prop_value(
+                        emp.get("customProperties"), rd_attr_custom_key
+                    )
+                    place = str(rec.get("place") or "")
+                    entry_date = _date_only(rec.get("entryDate"))
+                    leave_date = _date_only(rec.get("lastWorkDate")) or "2099-12-31"
+                    id_number = str(emp.get("iDNumber") or "")
+                    hrbp = str((org or {}).get("hRBP") or "")
+                    manager = str(rec.get("pOIdEmpAdmin") or "")
+                    is_leaving = "是" if _date_only(rec.get("lastWorkDate")) else "否"
+                    domain_account = _custom_prop_value(emp.get("customProperties"), domain_custom_key) or str(emp.get("_Name") or "")
+                    logger.info(
+                        "EHR 字段预览：job_no=%s company=%s name=%s rd_attr=%s place=%s entry_date=%s leave_date=%s id_number=%s hrbp=%s manager=%s is_leaving=%s domain_account=%s",
+                        job_no,
+                        company,
+                        name,
+                        rd_attr,
+                        place,
+                        entry_date,
+                        leave_date,
+                        id_number,
+                        hrbp,
+                        manager,
+                        is_leaving,
+                        domain_account,
+                    )
+                    count += 1
+                    if count >= preview_limit:
+                        break
+                logger.info("EHR 字段预览结束：printed=%s", count)
 
             # 4) 导出 OA 表单，建立字段映射 + 工号到记录ID映射
             exp_resp = seeyon.export_cap4_form_soap(
@@ -369,8 +402,6 @@ class SyncEhrToOaFormJob(BaseJob):
             oa_id_by_job_no: dict[str, int] = {}
             oa_id_by_job_no_norm: dict[str, int] = {}
             row_parse_miss = 0
-            row_id_fallback_count = 0
-            row_id_fallback_samples: list[str] = []
             for row in rows:
                 if not isinstance(row, dict):
                     continue
@@ -388,30 +419,24 @@ class SyncEhrToOaFormJob(BaseJob):
                     continue
 
                 if row_id is None:
-                    # 某些 OA export 不返回主键 id；此时依赖 uniqueFiled(工号) 进行更新匹配。
+                    row_parse_miss += 1
-                    row_id = 0
+                    if verbose_trace and row_parse_miss <= 20:
-                    row_id_fallback_count += 1
-                    if len(row_id_fallback_samples) < 20:
-                        row_id_fallback_samples.append(job_no)
-                    if verbose_trace and row_id_fallback_count <= 20:
                         logger.info(
-                            "OA 行解析未取到记录ID，使用回退ID=0：job_no=%s row_keys=%s",
+                            "OA 行解析未取到记录ID：job_no=%s row_keys=%s",
                             job_no,
                             list(row.keys())[:20],
                         )
+                    continue
                 oa_id_by_job_no[job_no] = row_id
                 job_no_norm = _normalize_job_no(job_no)
                 if job_no_norm:
                     oa_id_by_job_no_norm[job_no_norm] = row_id
             logger.info(
-                "OA 工号索引完成：indexed_job_numbers=%s indexed_job_numbers_norm=%s parse_miss=%s id_fallback=%s",
+                "OA 工号索引完成：indexed_job_numbers=%s indexed_job_numbers_norm=%s parse_miss=%s",
                 len(oa_id_by_job_no),
                 len(oa_id_by_job_no_norm),
                 row_parse_miss,
-                row_id_fallback_count,
             )
-            if row_id_fallback_samples:
-                logger.info("OA 记录ID回退样本（前20）：%s", row_id_fallback_samples)
             if verbose_trace:
                 for job_no, row_id in list(oa_id_by_job_no.items()):
                     logger.info("OA 工号索引明细：raw=%s norm=%s row_id=%s", job_no, _normalize_job_no(job_no), row_id)