适配微信4.0.3正式版，修复表情包和合并转发的聊天记录解析失败的问题

wxManager/parser/emoji_parser.py

@@ -26,7 +26,7 @@ def parser_emoji(xml_content):
         'height': 0,
         'desc': ''
     }
-    xml_content = xml_content.strip()
+    xml_content = xml_content.strip().replace('&', '&')
     try:
         xml_dict = xmltodict.parse(xml_content)
         emoji_dic = xml_dict.get('msg', {}).get('emoji', {})

wxManager/parser/link_parser.py

@@ -9,6 +9,7 @@
 @Description : 
 """
 import html
+import re
 import traceback
 from datetime import datetime, timedelta
 import xml.etree.ElementTree as ET
@@ -206,13 +207,27 @@ def parser_business(xml_content):
         return result
 
 
+def replace_entity(match):
+    # 获取匹配的数字
+    return ''
+
+
+def process_xml(xml_string):
+    # 使用正则表达式替换所有十进制转义字符
+    processed_xml = re.sub(r'&#(\d+);', replace_entity, xml_string)
+    return processed_xml
+
+
 def parser_record_item(recorditem, output_dir, wxid, msg_time, level=0):
     xml_string = recorditem
     if isinstance(xml_string, dict):
         recorditem_dic = xml_string
     else:
-        recorditem_dic = xmltodict.parse(xml_string)
-
+        try:
+            recorditem_dic = xmltodict.parse(xml_string)
+        except:
+            xml_string = process_xml(xml_string)
+            recorditem_dic = xmltodict.parse(xml_string)
     # logger.error(recorditem_dic)
     datalist = recorditem_dic.get('recordinfo', {}).get('datalist', {})
     count = datalist.get('@count', 0)
@@ -522,7 +537,7 @@ def parser_record_item(recorditem, output_dir, wxid, msg_time, level=0):
     return result
 
 
-def parser_merged_messages(xml, output_dir, wxid, msg_time, level=0):
+def parser_merged_messages(xml: str, output_dir, wxid, msg_time, level=0):
     try:
         try:
             data_dic = xmltodict.parse(xml).get('msg', {})
@@ -543,8 +558,8 @@ def parser_merged_messages(xml, output_dir, wxid, msg_time, level=0):
         }
     except:
         logger.error(xml)
-        logger.error(new_xml1)
-        logger.error(new_xml2)
+        # logger.error(new_xml1)
+        # logger.error(new_xml2)
         logger.error(traceback.format_exc())
         # raise ValueError('合并转发的消息解析失败')
         return {

wxManager/parser/util/protocbuf/packed_info_data_img2.proto

@@ -3,9 +3,11 @@ syntax = "proto3";
 message PackedInfoDataImg2 {
   int32 field1 = 1;
   int32 field2 = 2;
-  ImageInfo imageInfo = 3;
-  VideoInfo videoInfo = 4;
-  FileInfo fileInfo = 7;
+  ImageInfo imageInfo = 3;  // 图片
+  VideoInfo videoInfo = 4;  // 视频
+  AudioInfo audioInfo = 5;  // 语音
+  FileInfo fileInfo = 7;    // 文件
+  MergeInfo mergeInfo = 9;  // 合并转发的聊天记录
 }
 
 message ImageInfo {
@@ -35,4 +37,13 @@ message FileSubMessage2 {
   string field1 = 1;
   string field2 = 2;
   string field3 = 3;
+}
+
+message MergeInfo {
+  string dir = 1;
+}
+
+message AudioInfo {
+  uint32 field1 = 1;
+  string audioTxt = 2; // 语音转文字结果
 }

wxManager/parser/wechat_v4.py

@@ -190,7 +190,7 @@ class Singleton:
             self.contacts[wxid] = manager.get_contact_by_username(wxid)
         if isinstance(message[12], bytes):
             message_content = decompress(message[12])
-            message_content = message_content.replace('', '').replace(' ', ' ')
+            message_content = message_content.replace('&amp#x01;', '').replace(' ', ' ')
             # logger.error(message_content)
         else:
             message_content = message[12]
@@ -198,7 +198,9 @@ class Singleton:
             2] != MessageType.Pat:
             # 群聊文字消息格式：<wxid>:<content>
             message_content = ':'.join(message_content.split(':')[1:]).strip()
-
+        if message_content and message_content.startswith(username):
+            # md 微信不知道在搞什么，弄一些乱七八糟的东西 4.0.3.22
+            message_content = message_content.strip(f'{username}:').replace('<?xml version="1.0"?>', '')
         return is_sender, wxid, message_content
 
 
@@ -876,7 +878,7 @@ class FileMessageFactory(MessageFactory, Singleton):
         is_sender, wxid, message_content = self.common_attribute(message, username, manager)
         info = parser_file(message_content)
         md5 = info.get('md5', '')
-        filename = info.get('filename','')
+        filename = info.get('filename', '')
         if not filename:
             try:
                 # 2025年3月微信4.0.3正式版修改了img命名方式才有了这个东西