feat: improve quoted message rendering in dashboard

utils/message_formatter.py

@@ -3,6 +3,93 @@ import html
 import re
 
 
+def _clean_text(value: str) -> str:
+    if not value:
+        return ""
+    value = html.unescape(value)
+    value = re.sub(r"<br\s*/?>", "\n", value, flags=re.IGNORECASE)
+    value = re.sub(r"<[^>]+>", "", value)
+    return value.strip()
+
+
+def _extract_first(pattern: str, text: str, default: str = "") -> str:
+    match = re.search(pattern, text, re.DOTALL | re.IGNORECASE)
+    return match.group(1) if match else default
+
+
+def _format_referenced_content(ref_type: str, quoted_content: str, xml_content: str) -> str:
+    cleaned = _clean_text(quoted_content)
+    lower_xml = (quoted_content or "") + (xml_content or "")
+    lower_xml = lower_xml.lower()
+
+    if ref_type in {"3"} or "<img" in lower_xml or "cdnthumburl" in lower_xml:
+        return "[图片]"
+    if ref_type in {"43", "62"} or "<videomsg" in lower_xml or "cdnvideourl" in lower_xml:
+        return "[视频]"
+    if ref_type in {"47", "1048625", "1090519089"} or "<emoji" in lower_xml or "<emoticonmd5>" in lower_xml:
+        return "[表情]"
+    if ref_type in {"34"} or "<voicemsg" in lower_xml:
+        return "[语音]"
+    if ref_type in {"48"} or "<location" in lower_xml:
+        return "[位置]"
+    if ref_type in {"49"}:
+        title = _extract_first(r"<title>(.*?)</title>", quoted_content) or _extract_first(r"<title>(.*?)</title>", xml_content)
+        title = _clean_text(title)
+        return f"[链接] {title}" if title else "[链接]"
+
+    if cleaned:
+        return cleaned
+    return "[消息]"
+
+
+def _extract_media_preview(ref_type: str, quoted_content: str) -> dict:
+    payload = html.unescape(quoted_content or "")
+    preview = {"reference_type": "text", "preview_image": "", "preview_video_thumb": ""}
+
+    if ref_type in {"3"} or "<img" in payload.lower():
+        preview["reference_type"] = "image"
+        preview["preview_image"] = (
+            _extract_first(r'cdnthumburl="(.*?)"', payload)
+            or _extract_first(r"<cdnthumburl><!\[CDATA\[(.*?)\]\]></cdnthumburl>", payload)
+            or _extract_first(r"<cdnmidimgurl><!\[CDATA\[(.*?)\]\]></cdnmidimgurl>", payload)
+        )
+        return preview
+
+    if ref_type in {"43", "62"} or "<videomsg" in payload.lower():
+        preview["reference_type"] = "video"
+        preview["preview_video_thumb"] = (
+            _extract_first(r'cdnthumburl="(.*?)"', payload)
+            or _extract_first(r"<cdnthumburl><!\[CDATA\[(.*?)\]\]></cdnthumburl>", payload)
+        )
+        return preview
+
+    if ref_type in {"47", "1048625", "1090519089"} or "<emoji" in payload.lower():
+        preview["reference_type"] = "emoji"
+        return preview
+
+    return preview
+
+
+def parse_quote_message(xml_content: str) -> dict:
+    xml_content = xml_content.replace('&lt;', '<').replace('&gt;', '>')
+    main_content = _clean_text(_extract_first(r'<title>(.*?)</title>', xml_content, "[无标题]")) or "[无标题]"
+    display_name = _clean_text(_extract_first(r'<displayname>(.*?)</displayname>', xml_content, "未知用户")) or "未知用户"
+    quoted_content = _extract_first(r'<refermsg>.*?<content>(.*?)</content>', xml_content)
+    ref_type = _extract_first(r'<refermsg>.*?<type>(.*?)</type>', xml_content)
+    pretty_reference = _format_referenced_content(ref_type, quoted_content, xml_content)
+    media_preview = _extract_media_preview(ref_type, quoted_content)
+
+    return {
+        "main_content": main_content,
+        "display_name": display_name,
+        "quoted_content": pretty_reference,
+        "reference_type": media_preview.get("reference_type", "text"),
+        "preview_image": media_preview.get("preview_image", ""),
+        "preview_video_thumb": media_preview.get("preview_video_thumb", ""),
+        "formatted_message": f"{main_content}\n引用 {display_name}：{pretty_reference}" if display_name and pretty_reference else main_content
+    }
+
+
 def format_quote_message(xml_content):
     """
     格式化引用消息
@@ -14,31 +101,7 @@ def format_quote_message(xml_content):
         格式化后的消息文本
     """
     try:
-        
-        xml_content = xml_content.replace('&lt;', '<').replace('&gt;', '>')
-        # 使用正则表达式直接提取关键信息，避免XML解析问题
-        title_match = re.search(r'<title>(.*?)</title>', xml_content)
-        main_content = title_match.group(1) if title_match else "[无标题]"
-        
-        # 提取引用消息的发送者和内容
-        display_name_match = re.search(r'<displayname>(.*?)</displayname>', xml_content)
-        display_name = display_name_match.group(1) if display_name_match else "未知用户"
-        
-        quoted_content_match = re.search(r'<refermsg>.*?<content>(.*?)</content>', xml_content, re.DOTALL)
-        quoted_content = quoted_content_match.group(1) if quoted_content_match else ""
-        
-        # 解码HTML实体
-        try:
-            quoted_content = html.unescape(quoted_content)
-        except:
-            pass  # 如果解码失败，使用原始内容
-        
-        # 构建格式化的引用消息
-        if display_name and quoted_content:
-            formatted_message = f"{main_content}\n引用 {display_name}：{quoted_content}"
-            return formatted_message
-        
-        return main_content
+        return parse_quote_message(xml_content)["formatted_message"]
     except Exception as e:
         # 如果解析失败，尝试提取title标签内容
         try: