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

exporter/exporter_xlsx.py

@@ -188,8 +188,6 @@ class ExcelExporter(ExporterBase):
                 logger.error(traceback.format_exc())
                 continue
             type_ = message.type
-            timestamp = message.timestamp
-            msgSvrId = message.server_id
             if type_ == MessageType.Image:
                 message.set_file_name()
                 image_index[message.server_id] = self.row
@@ -256,6 +254,8 @@ class ExcelExporter(ExporterBase):
         if MessageType.Image in self.message_types:
             for index, message in enumerate(messages):
                 if message.type == MessageType.Image:
+                    if not self.is_selected(message):
+                        continue
                     row = image_index[message.server_id]
                     img_path = find_image_with_known_extensions(os.path.join(self.origin_path, message.path))
                     if not img_path:

wxManager/db_v3/open_im_msg.py

@@ -21,6 +21,7 @@ from concurrent.futures import ThreadPoolExecutor
 from datetime import datetime, date
 from typing import Tuple
 
+from wxManager import MessageType
 from wxManager.merge import increase_data, increase_update_data
 from wxManager.log import logger
 from wxManager.model import DataBaseBase
@@ -61,6 +62,45 @@ def convert_to_timestamp(time_range) -> Tuple[int, int]:
         return convert_to_timestamp_(time_range[0]), convert_to_timestamp_(time_range[1])
 
 
+def get_local_type(type_: MessageType):
+    type_name_dict = {
+        MessageType.Text: (1, 0),
+        MessageType.Image: (3, 0),
+        MessageType.Audio: (34, 0),
+        MessageType.Video: (43, 0),
+        MessageType.Emoji: (47, 0),
+        MessageType.BusinessCard: (42, 0),
+        MessageType.OpenIMBCard: (66, 0),
+        MessageType.Position: (48, 0),
+        MessageType.FavNote: (49, 40),
+        MessageType.FavNote: (49, 24),
+        (49, 53): "接龙",
+        MessageType.File: (49, 0),
+        MessageType.Text2: (49, 1),
+        MessageType.Music: (49, 3),
+        MessageType.Music: (49, 76),
+        MessageType.LinkMessage: (49, 5),
+        MessageType.File: (49, 6),
+        (49, 8): "用户上传的GIF表情",
+        MessageType.System: (49, 17),  # 发起了位置共享
+        MessageType.MergedMessages: (49, 19),
+        MessageType.Applet: (49, 33),
+        MessageType.Applet2: (49, 36),
+        MessageType.WeChatVideo: (49, 51),
+        (49, 57): MessageType.Quote,
+        (49, 63): "视频号直播或直播回放等",
+        (49, 87): "群公告",
+        (49, 88): "视频号直播或直播回放等",
+        (49, 2000): MessageType.Transfer,
+        (49, 2003): "赠送红包封面",
+        (50, 0): MessageType.Voip,
+        (10000, 0): MessageType.System,
+        (10000, 4): MessageType.Pat,
+        (10000, 8000): MessageType.System
+    }
+    return type_name_dict.get(type_, (0, 0))
+
+
 class OpenIMMsgDB(DataBaseBase):
 
     def _get_messages_by_num(self, cursor, username_, start_sort_seq, msg_num):
@@ -134,6 +174,29 @@ class OpenIMMsgDB(DataBaseBase):
 
         return None
 
+    def _get_messages_by_type(self, cursor, username: str, type_: MessageType,
+                              time_range: Tuple[int | float | str | date, int | float | str | date] = None, ):
+        if time_range:
+            start_time, end_time = convert_to_timestamp(time_range)
+        local_type, sub_type = get_local_type(type_)
+        sql = f'''
+            select localId,TalkerId,Type,SubType,IsSender,CreateTime,Status,StrContent,strftime('%Y-%m-%d %H:%M:%S',CreateTime,'unixepoch','localtime') as StrTime,MsgSvrID,BytesExtra,CompressContent,DisplayContent
+            from MSG
+            where StrTalker=? and Type=? and SubType = ?
+            {'AND CreateTime>' + str(start_time) + ' AND CreateTime<' + str(end_time) if time_range else ''}
+            order by CreateTime
+        '''
+        cursor.execute(sql, [username, local_type, sub_type])
+        result = cursor.fetchall()
+        if result:
+            return result
+        else:
+            return None
+
+    def get_messages_by_type(self, username: str, type_: MessageType,
+                             time_range: Tuple[int | float | str | date, int | float | str | date] = None, ):
+        return self.get_messages_by_type(self.DB.cursor, username, type_, time_range)
+
     def merge(self, db_path):
         if not (os.path.exists(db_path) or os.path.isfile(db_path)):
             print(f'{db_path} 不存在')
@@ -143,4 +206,4 @@ class OpenIMMsgDB(DataBaseBase):
             increase_data(db_path, self.cursor, self.DB, 'ChatCRMsg', 'MsgSvrID', 1, exclude_column='localId')
         except:
             print(f"数据库操作错误: {traceback.format_exc()}")
-            self.DB.rollback()
+            self.DB.rollback()

wxManager/db_v3/public_msg.py

@@ -8,79 +8,36 @@ from datetime import date
 from typing import Tuple
 from concurrent.futures import ThreadPoolExecutor
 
+from wxManager import MessageType
 from wxManager.merge import increase_data
-from wxManager.db_v3.msg import convert_to_timestamp
+from wxManager.db_v3.msg import convert_to_timestamp,get_local_type
 from wxManager.model import DataBaseBase
 
 
 class PublicMsg(DataBaseBase):
 
-    def get_messages(
-            self,
-            username_: str,
-            time_range: Tuple[int | float | str | date, int | float | str | date] = None,
-    ):
-        """
-        return list
-            a[0]: localId,
-            a[1]: talkerId, （和strtalker对应的，不是群聊信息发送人）
-            a[2]: type,
-            a[3]: subType,
-            a[4]: is_sender,
-            a[5]: timestamp,
-            a[6]: status, （没啥用）
-            a[7]: str_content,
-            a[8]: str_time, （格式化的时间）
-            a[9]: msgSvrId,
-            a[10]: BytesExtra,
-            a[11]: CompressContent,
-            a[12]: DisplayContent,
-            a[13]: 联系人的类（如果是群聊就有，不是的话没有这个字段）
-        """
-        if not self.open_flag:
-            return []
+    def _get_messages_by_type(self, cursor, username: str, type_: MessageType,
+                              time_range: Tuple[int | float | str | date, int | float | str | date] = None, ):
         if time_range:
             start_time, end_time = convert_to_timestamp(time_range)
+        local_type, sub_type = get_local_type(type_)
         sql = f'''
             select localId,TalkerId,Type,SubType,IsSender,CreateTime,Status,StrContent,strftime('%Y-%m-%d %H:%M:%S',CreateTime,'unixepoch','localtime') as StrTime,MsgSvrID,BytesExtra,CompressContent,DisplayContent
-            from PublicMsg
-            where StrTalker=?
+            from MSG
+            where StrTalker=? and Type=? and SubType = ?
             {'AND CreateTime>' + str(start_time) + ' AND CreateTime<' + str(end_time) if time_range else ''}
             order by CreateTime
         '''
-        try:
-            lock.acquire(True)
-            self.cursor.execute(sql, [username_])
-            result = self.cursor.fetchall()
-        finally:
-            lock.release()
-        return result
+        cursor.execute(sql, [username, local_type, sub_type])
+        result = cursor.fetchall()
+        if result:
+            return result
+        else:
+            return None
 
-    def get_messages_by_type(
-            self,
-            username_: str,
-            type_,
-            sub_type=None,
-            time_range: Tuple[int | float | str | date, int | float | str | date] = None,
-    ):
-        if not self.open_flag:
-            return []
-        if time_range:
-            start_time, end_time = convert_to_timestamp(time_range)
-        sql = f'''
-            select localId,TalkerId,Type,SubType,IsSender,CreateTime,Status,StrContent,strftime('%Y-%m-%d %H:%M:%S',CreateTime,'unixepoch','localtime') as StrTime,MsgSvrID,BytesExtra,CompressContent,DisplayContent
-            from PublicMsg
-            where StrTalker=? AND Type=? {'AND SubType=' + str(sub_type) if sub_type else ''}
-            {'AND CreateTime>' + str(start_time) + ' AND CreateTime<' + str(end_time) if time_range else ''}
-            order by CreateTime
-        '''
-        try:
-            lock.acquire(True)
-            self.cursor.execute(sql, [username_, type_])
-            result = self.cursor.fetchall()
-        finally:
-            lock.release()
-        return result
+    def get_messages_by_type(self, username: str, type_: MessageType,
+                             time_range: Tuple[int | float | str | date, int | float | str | date] = None, ):
+        return self.get_messages_by_type(self.DB.cursor, username, type_, time_range)
 
     def get_sport_score_by_name(self, username,
                                 time_range: Tuple[int | float | str | date, int | float | str | date] = None, ):
@@ -126,7 +83,7 @@ class PublicMsg(DataBaseBase):
 
     def get_messages_by_username(self, username: str,
                                  time_range: Tuple[int | float | str | date, int | float | str | date] = None, ):
-        return self._get_messages_by_username(self.DB.cursor(),username,time_range)
+        return self._get_messages_by_username(self.DB.cursor(), username, time_range)
 
     def get_message_by_server_id(self, username, server_id):
         """

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