fix excel file message parsing and dingtalk fileType reply

remote_sync/bot.py

@@ -1446,16 +1446,37 @@ class OpenClawDingTalkBot(ChatbotHandler):
         return codes
 
     def _normalize_file_name(self, raw_name: str) -> str:
-        name = str(raw_name or "").strip().replace("\\", "_").replace("/", "_")
+        raw = str(raw_name or "").strip().replace("\\", "_").replace("/", "_")
-        name = re.sub(r"\s+", "_", name)
+        raw = re.sub(r"\s+", "_", raw)
-        name = re.sub(r"[^0-9A-Za-z._-]+", "_", name).strip("._")
+        stem, ext = os.path.splitext(raw)
-        if not name:
+        safe_stem = re.sub(r"[^0-9A-Za-z._-]+", "_", stem).strip("._-")
-            return f"upload_{int(time.time())}.xlsx"
+        safe_ext = re.sub(r"[^0-9A-Za-z.]+", "", ext).lower()
-        return name[:120]
+        if safe_ext and not safe_ext.startswith("."):
+            safe_ext = f".{safe_ext}"
+        if not safe_stem:
+            safe_stem = f"upload_{int(time.time())}"
+        normalized = f"{safe_stem}{safe_ext}" if safe_ext else safe_stem
+        return normalized[:120]
 
     def _is_supported_excel_file(self, file_name: str) -> bool:
         return str(file_name or "").strip().lower().endswith(".xlsx")
 
+    def _is_supported_excel_payload(self, file_name: str, file_type: str) -> bool:
+        if self._is_supported_excel_file(file_name):
+            return True
+        normalized = str(file_type or "").strip().lower()
+        if not normalized:
+            return False
+        normalized = normalized.split(";")[0].strip()
+        if normalized in {
+            "xlsx",
+            "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+        }:
+            return True
+        if "spreadsheetml.sheet" in normalized:
+            return True
+        return False
+
     def _is_excel_link_prompt(self, prompt: str) -> bool:
         raw = str(prompt or "").strip().lower()
         if not raw:
@@ -1480,16 +1501,45 @@ class OpenClawDingTalkBot(ChatbotHandler):
                         continue
             if isinstance(current, dict):
                 lowered = {str(k).lower(): k for k in current.keys()}
+                code = ""
                 if "downloadcode" in lowered:
                     raw_code = current.get(lowered["downloadcode"])
                     code = str(raw_code or "").strip()
-                    if code:
+
-                        file_name = ""
+                download_url = ""
-                        for candidate in ("fileName", "filename", "file_name", "name", "title"):
+                for key_name in (
-                            if candidate in current and str(current.get(candidate) or "").strip():
+                    "downloadurl",
-                                file_name = str(current.get(candidate)).strip()
+                    "download_url",
-                                break
+                    "privatedownloadurl",
-                        return {"download_code": code, "file_name": file_name}
+                    "private_download_url",
+                    "privateurl",
+                ):
+                    if key_name in lowered:
+                        raw_url = current.get(lowered[key_name])
+                        candidate_url = str(raw_url or "").strip()
+                        if candidate_url.lower().startswith(("http://", "https://")):
+                            download_url = candidate_url
+                            break
+
+                file_name = ""
+                for candidate in ("fileName", "filename", "file_name", "name", "title"):
+                    if candidate in current and str(current.get(candidate) or "").strip():
+                        file_name = str(current.get(candidate)).strip()
+                        break
+
+                file_type = ""
+                for candidate in ("fileType", "filetype", "mimeType", "mimetype", "suffix", "ext", "extension"):
+                    if candidate in current and str(current.get(candidate) or "").strip():
+                        file_type = str(current.get(candidate)).strip()
+                        break
+
+                if code or download_url:
+                    return {
+                        "download_code": code,
+                        "download_url": download_url,
+                        "file_name": file_name,
+                        "file_type": file_type,
+                    }
                 for child in current.values():
                     stack.append(child)
                 continue
@@ -1509,7 +1559,16 @@ class OpenClawDingTalkBot(ChatbotHandler):
                 payload = None
         if not payload:
             return {}
+
+        download_code = str(payload.get("download_code") or "").strip()
+        download_url = str(payload.get("download_url") or "").strip()
+        if download_url and not download_url.lower().startswith(("http://", "https://")):
+            download_url = ""
+        if not download_code and not download_url:
+            return {}
+
         file_name = payload.get("file_name") or ""
+        file_type = payload.get("file_type") or ""
         if not file_name and isinstance(callback_data, dict):
             for key in ("fileName", "filename", "name", "title"):
                 raw = callback_data.get(key)
@@ -1517,9 +1576,21 @@ class OpenClawDingTalkBot(ChatbotHandler):
                     file_name = str(raw).strip()
                     if file_name:
                         break
-        payload["file_name"] = self._normalize_file_name(file_name or "upload.xlsx")
+        if not file_type and isinstance(callback_data, dict):
-        if not self._is_supported_excel_file(payload["file_name"]):
+            for key in ("fileType", "filetype", "mimeType", "mimetype", "suffix", "ext", "extension"):
+                raw = callback_data.get(key)
+                if raw:
+                    file_type = str(raw).strip()
+                    if file_type:
+                        break
+        normalized_file_name = self._normalize_file_name(file_name or "upload.xlsx")
+        normalized_file_type = str(file_type or "").strip().lower()
+        if not self._is_supported_excel_payload(normalized_file_name, normalized_file_type):
             return {}
+        payload["download_code"] = download_code
+        payload["download_url"] = download_url
+        payload["file_name"] = normalized_file_name
+        payload["file_type"] = normalized_file_type
         return payload
 
     def _guess_image_ext(self, content_type: str, source_url: str) -> str:
@@ -2200,9 +2271,11 @@ class OpenClawDingTalkBot(ChatbotHandler):
             raw = (proc.stdout or "").strip()
             return f"结果解析失败: {exc}.\n原始输出: {raw[:500]}"
 
-    def _reply_file_by_media_id(self, msg: ChatbotMessage, media_id: str) -> bool:
+    def _reply_file_by_media_id(self, msg: ChatbotMessage, media_id: str, file_name: str = "") -> bool:
         if not media_id:
             return False
+        safe_name = self._normalize_file_name(file_name or f"excel_{int(time.time())}.xlsx")
+        file_ext = os.path.splitext(safe_name)[1].lower().lstrip(".") or "file"
         request_headers = {
             "Content-Type": "application/json",
             "Accept": "*/*",
@@ -2211,6 +2284,8 @@ class OpenClawDingTalkBot(ChatbotHandler):
             "msgtype": "file",
             "file": {
                 "mediaId": media_id,
+                "fileName": safe_name,
+                "fileType": file_ext,
             },
             "at": {
                 "atUserIds": [msg.sender_staff_id] if msg.sender_staff_id else [],
@@ -2337,15 +2412,17 @@ class OpenClawDingTalkBot(ChatbotHandler):
             return f"转换脚本不存在：{script_path}", False
 
         download_code = str(file_payload.get("download_code") or "").strip()
+        download_url = str(file_payload.get("download_url") or "").strip()
         file_name = self._normalize_file_name(file_payload.get("file_name") or "upload.xlsx")
-        if not download_code:
+        if not download_code and not download_url:
-            return "未获取到文件下载码，请重试上传。", False
+            return "未获取到文件下载地址，请重试上传。", False
         if not self._is_supported_excel_file(file_name):
             return "当前仅支持 .xlsx 文件，请上传Excel后重试。", False
 
-        download_url = self.get_image_download_url(download_code)
         if not download_url:
-            return "文件下载地址获取失败，请重试。", False
+            download_url = self.get_image_download_url(download_code)
+            if not download_url:
+                return "文件下载地址获取失败，请重试。", False
 
         with tempfile.TemporaryDirectory(prefix="dingtalk_excel_link_") as tmpdir:
             input_path = os.path.join(tmpdir, file_name)
@@ -2437,7 +2514,7 @@ class OpenClawDingTalkBot(ChatbotHandler):
             except Exception as exc:
                 self.logger.exception("upload converted excel failed")
 
-            sent = self._reply_file_by_media_id(msg, media_id) if media_id else False
+            sent = self._reply_file_by_media_id(msg, media_id, output_name) if media_id else False
             if not sent:
                 s3_url = self._upload_excel_output_to_s3(output_path, output_name)
                 if s3_url:
@@ -2588,6 +2665,15 @@ class OpenClawDingTalkBot(ChatbotHandler):
             bool(incoming_image_codes),
             len(text or ""),
         )
+        if str(getattr(msg, "message_type", "")).lower() == "file" and not has_excel_file:
+            callback_keys = []
+            if isinstance(callback.data, dict):
+                callback_keys = list(callback.data.keys())[:15]
+            self.logger.warning(
+                "file message without excel payload: callback_keys=%s extensions_keys=%s",
+                callback_keys,
+                list((getattr(msg, "extensions", {}) or {}).keys())[:15],
+            )
 
         if not text and not incoming_image_codes and not has_excel_file:
             return AckMessage.STATUS_OK, "ignored"