add excel failure alert log with error digest

remote_sync/bot.py

@@ -14,6 +14,7 @@ import re
 import subprocess
 import sys
 import tempfile
+import threading
 import time
 import urllib.parse
 import uuid
@@ -328,6 +329,16 @@ class OpenClawDingTalkBot(ChatbotHandler):
             10,
             min(int(os.getenv("EXCEL_TEXT_PROMPT_COOLDOWN_SECONDS", "120")), 600),
         )
+        self.excel_alert_log_enabled = bool_env("EXCEL_ALERT_LOG_ENABLED", True)
+        self.excel_alert_log_path = (
+            os.getenv(
+                "EXCEL_ALERT_LOG_PATH",
+                "/opt/openclaw-dingtalk-bridge/logs/excel_converter_alerts.log",
+            ).strip()
+            or "/opt/openclaw-dingtalk-bridge/logs/excel_converter_alerts.log"
+        )
+        self.excel_alert_detail_limit = max(200, min(int(os.getenv("EXCEL_ALERT_DETAIL_LIMIT", "1600")), 10000))
+        self._excel_alert_log_lock = threading.Lock()
         self._recent_excel_success_ts = {}
 
     def _should_handle(self, msg: ChatbotMessage, text: str) -> bool:
@@ -2253,6 +2264,71 @@ class OpenClawDingTalkBot(ChatbotHandler):
             self.logger.exception("upload excel output to s3 failed: %s", exc)
             return ""
 
+    def _excel_error_summary(self, detail: str) -> str:
+        raw = str(detail or "").strip()
+        if not raw:
+            return "unknown_error"
+        lines = [line.strip() for line in raw.splitlines() if line.strip()]
+        if not lines:
+            return raw[:300]
+        for line in reversed(lines):
+            lower_line = line.lower()
+            if lower_line.startswith("traceback"):
+                continue
+            if lower_line.startswith("file "):
+                continue
+            if lower_line in {"^^^^^^", "^^^^^"}:
+                continue
+            return line[:300]
+        return lines[-1][:300]
+
+    def _record_excel_failure_alert(
+        self,
+        msg: ChatbotMessage,
+        file_name: str,
+        stage: str,
+        detail: str,
+        return_code: int | None = None,
+        stdout_text: str = "",
+        stderr_text: str = "",
+    ) -> None:
+        normalized_detail = str(detail or "").strip() or "unknown_error"
+        summary = self._excel_error_summary(normalized_detail)
+        digest = hashlib.sha1(normalized_detail.encode("utf-8", errors="ignore")).hexdigest()[:12]
+        log_record = {
+            "ts": dt.datetime.now().isoformat(timespec="seconds"),
+            "event": "excel_convert_failed",
+            "stage": stage,
+            "digest": digest,
+            "summary": summary,
+            "return_code": return_code,
+            "file_name": str(file_name or ""),
+            "conversation_id": str(getattr(msg, "conversation_id", "") or ""),
+            "session_webhook": str(getattr(msg, "session_webhook", "") or ""),
+            "sender_staff_id": str(getattr(msg, "sender_staff_id", "") or ""),
+            "detail": normalized_detail[: self.excel_alert_detail_limit],
+            "stderr": str(stderr_text or "")[: self.excel_alert_detail_limit],
+            "stdout": str(stdout_text or "")[: self.excel_alert_detail_limit],
+        }
+        self.logger.warning(
+            "excel failure alert stage=%s digest=%s file=%s summary=%s",
+            stage,
+            digest,
+            file_name,
+            summary,
+        )
+        if not self.excel_alert_log_enabled:
+            return
+        try:
+            parent_dir = os.path.dirname(self.excel_alert_log_path) or "."
+            os.makedirs(parent_dir, exist_ok=True)
+            line = json.dumps(log_record, ensure_ascii=False)
+            with self._excel_alert_log_lock:
+                with open(self.excel_alert_log_path, "a", encoding="utf-8") as file_obj:
+                    file_obj.write(line + "\n")
+        except Exception:
+            self.logger.exception("write excel alert log failed")
+
     def _run_excel_image_link_job(self, msg: ChatbotMessage, file_payload: dict) -> tuple[str, bool]:
         if not self.excel_link_feature_enabled:
             return "Excel 图片链接转图片功能未启用。", False
@@ -2282,6 +2358,12 @@ class OpenClawDingTalkBot(ChatbotHandler):
                     fh.write(response.content)
             except Exception as exc:
                 self.logger.exception("download incoming excel failed")
+                self._record_excel_failure_alert(
+                    msg,
+                    file_name,
+                    stage="download_source_excel",
+                    detail=f"{type(exc).__name__}: {exc}",
+                )
                 return f"下载源文件失败：{exc}", False
 
             cmd = [
@@ -2312,6 +2394,12 @@ class OpenClawDingTalkBot(ChatbotHandler):
                     timeout=self.excel_converter_timeout,
                 )
             except subprocess.TimeoutExpired:
+                self._record_excel_failure_alert(
+                    msg,
+                    file_name,
+                    stage="excel_converter_timeout",
+                    detail=f"timeout>{self.excel_converter_timeout}s",
+                )
                 return f"Excel 转换超时（>{self.excel_converter_timeout}秒），请稍后重试。", False
 
             stdout_text = (proc.stdout or "").strip()
@@ -2319,6 +2407,15 @@ class OpenClawDingTalkBot(ChatbotHandler):
             if proc.returncode != 0 or not os.path.exists(output_path):
                 detail = stderr_text or stdout_text or "未知错误"
                 self.logger.error("excel converter failed code=%s detail=%s", proc.returncode, detail[:2500])
+                self._record_excel_failure_alert(
+                    msg,
+                    file_name,
+                    stage="excel_converter_failed",
+                    detail=detail,
+                    return_code=proc.returncode,
+                    stdout_text=stdout_text,
+                    stderr_text=stderr_text,
+                )
                 return f"Excel 转换失败：{detail[:1200]}", False
 
             summary_line = ""