update

extensions/sync_oa_to_didi/job.py

@@ -17,6 +17,31 @@ def _mask_token(token: str) -> str:
     return f"{token[:6]}***{token[-4:]}"
 
 
+def _log_text_in_chunks(*, prefix: str, text: str, chunk_bytes: int = 8_000) -> None:
+    """
+    将大文本尽可能写入 run_log：
+    - 按 UTF-8 字节切分，避免单条日志过大导致整条无法写入（capture_logs 会在超过 max_bytes 时丢弃整条并标记截断）
+    - 由上层 capture_logs(max_bytes=200_000) 负责总量截断
+    """
+    try:
+        if not text:
+            logger.info("%s <empty>", prefix)
+            return
+
+        if chunk_bytes <= 0:
+            chunk_bytes = 8_000
+
+        raw_bytes = text.encode("utf-8", errors="replace")
+        total = (len(raw_bytes) + chunk_bytes - 1) // chunk_bytes
+        for i in range(total):
+            b = raw_bytes[i * chunk_bytes : (i + 1) * chunk_bytes]
+            chunk = b.decode("utf-8", errors="replace")
+            logger.info("%s chunk %s/%s: %s", prefix, i + 1, total, chunk)
+    except Exception:
+        # run_log 捕获属于“尽力而为”，任何异常都不应影响任务执行
+        return
+
+
 class SyncOAToDidiTokenJob(BaseJob):
     """
     示例 Job：演示致远 OA 的 token 获取与日志记录
@@ -133,6 +158,7 @@ class SyncOAToDidiExportFormJob(BaseJob):
             content_type,
             base_url,
         )
+        _log_text_in_chunks(prefix="Seeyon export_form_soap raw", text=raw_text, chunk_bytes=8_000)
 
         return {
             "raw": raw_text,