init

app/util/exporter/exporter.py

@@ -0,0 +1,176 @@
+import csv
+import html
+import os
+import shutil
+import sys
+
+import filecmp
+
+from PyQt5.QtCore import pyqtSignal, QThread
+
+from app.config import OUTPUT_DIR
+from app.person import Me, Contact
+
+os.makedirs(os.path.join(OUTPUT_DIR, '聊天记录'), exist_ok=True)
+
+
+def set_global_font(doc, font_name):
+    # 创建一个新样式
+    style = doc.styles['Normal']
+
+    # 设置字体名称
+    style.font.name = font_name
+    # 遍历文档中的所有段落，将样式应用到每个段落
+    for paragraph in doc.paragraphs:
+        for run in paragraph.runs:
+            run.font.name = font_name
+
+
+def makedirs(path):
+    os.makedirs(path, exist_ok=True)
+    os.makedirs(os.path.join(path, 'image'), exist_ok=True)
+    os.makedirs(os.path.join(path, 'emoji'), exist_ok=True)
+    os.makedirs(os.path.join(path, 'video'), exist_ok=True)
+    os.makedirs(os.path.join(path, 'voice'), exist_ok=True)
+    os.makedirs(os.path.join(path, 'file'), exist_ok=True)
+    os.makedirs(os.path.join(path, 'avatar'), exist_ok=True)
+    os.makedirs(os.path.join(path, 'music'), exist_ok=True)
+    os.makedirs(os.path.join(path, 'icon'), exist_ok=True)
+    resource_dir = os.path.join('app', 'resources', 'data', 'icons')
+    if not os.path.exists(resource_dir):
+        # 获取打包后的资源目录
+        resource_dir = getattr(sys, '_MEIPASS', os.path.abspath(os.path.dirname(__file__)))
+        # 构建 FFmpeg 可执行文件的路径
+        resource_dir = os.path.join(resource_dir, 'app', 'resources', 'data', 'icons')
+    target_folder = os.path.join(path, 'icon')
+    # 拷贝一些必备的图标
+    for root, dirs, files in os.walk(resource_dir):
+        relative_path = os.path.relpath(root, resource_dir)
+        target_path = os.path.join(target_folder, relative_path)
+
+        # 遍历文件夹中的文件
+        for file in files:
+            source_file_path = os.path.join(root, file)
+            target_file_path = os.path.join(target_path, file)
+            if not os.path.exists(target_file_path):
+                shutil.copy(source_file_path, target_file_path)
+            else:
+                # 比较文件内容
+                if not filecmp.cmp(source_file_path, target_file_path, shallow=False):
+                    # 文件内容不一致，进行覆盖拷贝
+                    shutil.copy(source_file_path, target_file_path)
+
+
+def escape_js_and_html(input_str):
+    if not input_str:
+        return ''
+    # 转义HTML特殊字符
+    html_escaped = html.escape(input_str, quote=False)
+
+    # 手动处理JavaScript转义字符
+    js_escaped = (
+        html_escaped
+        .replace("\\", "\\\\")
+        .replace("'", r"\'")
+        .replace('"', r'\"')
+        .replace("\n", r'\n')
+        .replace("\r", r'\r')
+        .replace("\t", r'\t')
+    )
+
+    return js_escaped
+
+
+class ExporterBase(QThread):
+    progressSignal = pyqtSignal(int)
+    rangeSignal = pyqtSignal(int)
+    okSignal = pyqtSignal(int)
+    i = 1
+    CSV = 0
+    DOCX = 1
+    HTML = 2
+    CSV_ALL = 3
+    CONTACT_CSV = 4
+    TXT = 5
+
+    def __init__(self, contact, type_=DOCX, message_types={}, time_range=None, messages=None,index=0, parent=None):
+        super().__init__(parent)
+        self.message_types = message_types  # 导出的消息类型
+        self.contact: Contact = contact  # 联系人
+        self.output_type = type_  # 导出文件类型
+        self.total_num = 1  # 总的消息数量
+        self.num = 0  # 当前处理的消息数量
+        self.index = index #
+        self.last_timestamp = 0
+        self.time_range = time_range
+        self.messages = messages
+        self.origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        makedirs(self.origin_path)
+
+    def run(self):
+        self.export()
+
+    def export(self):
+        raise NotImplementedError("export method must be implemented in subclasses")
+
+    def cancel(self):
+        self.requestInterruption()
+
+    def is_5_min(self, timestamp) -> bool:
+        if abs(timestamp - self.last_timestamp) > 300:
+            self.last_timestamp = timestamp
+            return True
+        return False
+
+    def get_avatar_path(self, is_send, message, is_absolute_path=False) -> str:
+        if is_absolute_path:
+            if self.contact.is_chatroom:
+                avatar = message[13].avatar_path
+            else:
+                avatar = Me().avatar_path if is_send else self.contact.avatar_path
+        else:
+            if self.contact.is_chatroom:
+                avatar = message[13].smallHeadImgUrl
+            else:
+                avatar = Me().smallHeadImgUrl if is_send else self.contact.smallHeadImgUrl
+        return avatar
+
+    def get_display_name(self, is_send, message) -> str:
+        if self.contact.is_chatroom:
+            if is_send:
+                display_name = Me().name
+            else:
+                display_name = message[13].remark
+        else:
+            display_name = Me().name if is_send else self.contact.remark
+        return escape_js_and_html(display_name)
+
+    def text(self, doc, message):
+        return
+
+    def image(self, doc, message):
+        return
+
+    def audio(self, doc, message):
+        return
+
+    def emoji(self, doc, message):
+        return
+
+    def file(self, doc, message):
+        return
+
+    def refermsg(self, doc, message):
+        return
+
+    def system_msg(self, doc, message):
+        return
+
+    def video(self, doc, message):
+        return
+
+    def music_share(self, doc, message):
+        return
+
+    def share_card(self, doc, message):
+        return

app/util/exporter/exporter_ai_txt.py

@@ -0,0 +1,96 @@
+import os
+import re
+
+from app.DataBase import msg_db
+from app.util.compress_content import parser_reply, share_card
+from app.util.exporter.exporter import ExporterBase
+
+
+def remove_privacy_info(text):
+    # 正则表达式模式
+    patterns = {
+        'phone': r'\b(\+?86[-\s]?)?1[3-9]\d{9}\b',  # 手机号
+        'email': r'\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Z|a-z]{2,}\b',  # 邮箱
+        'id_card': r'\b\d{15}|\d{18}|\d{17}X\b',  # 身份证号
+        'password': r'\b(?:password|pwd|pass|psw)[\s=:]*\S+\b',  # 密码
+        'account': r'\b(?:account|username|user|acct)[\s=:]*\S+\b'  # 账号
+    }
+
+    for key, pattern in patterns.items():
+        text = re.sub(pattern, f'[{key} xxx]', text)
+
+    return text
+
+
+class AiTxtExporter(ExporterBase):
+    last_is_send = -1
+
+    def title(self, message):
+        str_time = message[8]
+        is_send = message[4]
+        display_name = ''
+        if is_send != self.last_is_send:
+            display_name = '\n' + self.get_display_name(is_send, message) + ':'
+        self.last_is_send = is_send
+        return display_name
+
+    def text(self, doc, message):
+        str_content = remove_privacy_info(message[7])
+        doc.write(
+            f'''{self.title(message)}{str_content} '''
+        )
+
+    def image(self, doc, message):
+        doc.write(
+            f'''{self.title(message)}[图片]'''
+        )
+
+    def audio(self, doc, message):
+        doc.write(
+            f'''{self.title(message)}[语音]'''
+        )
+
+    def emoji(self, doc, message):
+        doc.write(
+            f'''{self.title(message)}[表情包]'''
+        )
+
+    def file(self, doc, message):
+        doc.write(
+            f'''{self.title(message)}[文件]'''
+        )
+
+    def system_msg(self, doc, message):
+        str_content = message[7]
+        str_time = message[8]
+        str_content = str_content.replace('<![CDATA[', "").replace(
+            ' <a href="weixin://revoke_edit_click">重新编辑</a>]]>', "")
+        doc.write(
+            f'''{str_time} {str_content}'''
+        )
+
+    def video(self, doc, message):
+        is_send = message[4]
+        doc.write(
+            f'''{self.title(message)}[视频]'''
+        )
+
+    def export(self):
+        # 实现导出为txt的逻辑
+        print(f"【开始导出 TXT {self.contact.remark}】")
+        origin_path = self.origin_path
+        os.makedirs(origin_path, exist_ok=True)
+        filename = os.path.join(origin_path, self.contact.remark + '_chat.txt')
+        messages = msg_db.get_messages_group_by_day(self.contact.wxid, time_range=self.time_range)
+        total_steps = len(messages)
+        with open(filename, mode='w', newline='', encoding='utf-8') as f:
+            for date, messages in messages.items():
+                f.write(f"\n\n{'*' * 20}{date}{'*' * 20}\n")
+                for index, message in enumerate(messages):
+                    type_ = message[2]
+                    sub_type = message[3]
+                    self.progressSignal.emit(int((index + 1) / total_steps * 100))
+                    if type_ == 1 and self.message_types.get(type_):
+                        self.text(f, message)
+        print(f"【完成导出 TXT {self.contact.remark}】")
+        self.okSignal.emit(1)

app/util/exporter/exporter_csv.py

@@ -0,0 +1,40 @@
+import csv
+import os
+
+from app.DataBase import msg_db
+from app.person import Me
+from app.util.exporter.exporter import ExporterBase
+from app.config import OUTPUT_DIR
+
+
+class CSVExporter(ExporterBase):
+    def to_csv(self):
+        print(f"【开始导出 CSV {self.contact.remark}】")
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        os.makedirs(origin_path, exist_ok=True)
+        filename = os.path.join(origin_path,f"{self.contact.remark}_utf8.csv")
+        columns = ['localId', 'TalkerId', 'Type', 'SubType',
+                   'IsSender', 'CreateTime', 'Status', 'StrContent',
+                   'StrTime', 'Remark', 'NickName', 'Sender']
+        messages = msg_db.get_messages(self.contact.wxid, time_range=self.time_range)
+        # 写入CSV文件
+        with open(filename, mode='w', newline='', encoding='utf-8-sig') as file:
+            writer = csv.writer(file)
+            writer.writerow(columns)
+            # 写入数据
+            # writer.writerows(messages)
+            for msg in messages:
+                if self.contact.is_chatroom:
+                    other_data = [msg[13].remark, msg[13].nickName, msg[13].wxid]
+                else:
+                    is_send = msg[4]
+                    Remark = Me().remark if is_send else self.contact.remark
+                    nickname = Me().nickName if is_send else self.contact.nickName
+                    wxid = Me().wxid if is_send else self.contact.wxid
+                    other_data = [Remark,nickname,wxid]
+                writer.writerow([*msg[:9], *other_data])
+        print(f"【完成导出 CSV {self.contact.remark}】")
+        self.okSignal.emit(1)
+
+    def run(self):
+        self.to_csv()

app/util/exporter/exporter_docx.py

@@ -0,0 +1,380 @@
+import os
+import shutil
+import time
+from re import findall
+
+import docx
+from docx import shared
+from docx.enum.table import WD_ALIGN_VERTICAL
+from docx.enum.text import WD_COLOR_INDEX, WD_PARAGRAPH_ALIGNMENT
+from docx.oxml.ns import qn
+from docxcompose.composer import Composer
+
+from app.DataBase import msg_db, hard_link_db
+from app.util.exporter.exporter import ExporterBase, escape_js_and_html
+from app.config import OUTPUT_DIR
+from app.log import logger
+from app.person import Me
+from app.util.compress_content import parser_reply, share_card, music_share
+from app.util.image import get_image_abs_path
+from app.util.music import get_music_path
+
+# 要删除的编码字符
+encoded_chars = b'\x00\x01\x02\x03\x04\x05\x06\x07\x08\x0b\x0c\x0e\x0f\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f'
+
+# 创建一个字典，将要删除的字符映射为 None
+char_mapping = {char: None for char in encoded_chars}
+
+
+def filter_control_characters(input_string):
+    """
+    过滤掉不可打印字符
+    @param input_string:
+    @return:
+    """
+
+    # 过滤掉非可打印字符
+    filtered_string = input_string.translate(char_mapping)
+
+    return filtered_string
+
+
+class DocxExporter(ExporterBase):
+    def text(self, doc, message):
+        type_ = message[2]
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+
+        display_name = self.get_display_name(is_send, message)
+        avatar = self.get_avatar_path(is_send, message, True)
+        content_cell = self.create_table(doc, is_send, avatar)
+        try:
+            content_cell.paragraphs[0].add_run(str_content)
+        except ValueError:
+            try:
+                str_content = filter_control_characters(str_content)
+                content_cell.paragraphs[0].add_run(str_content)
+            except ValueError:
+                logger.error(f'非法字符:{str_content}')
+                content_cell.paragraphs[0].add_run('非法字符')
+        content_cell.paragraphs[0].font_size = shared.Inches(0.5)
+        # doc.add_picture(avatar)
+        if is_send:
+            p = content_cell.paragraphs[0]
+            p.paragraph_format.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+        doc.add_paragraph()
+
+    def image(self, doc, message):
+        str_content = message[7]
+        is_send = message[4]
+        BytesExtra = message[10]
+        avatar = self.get_avatar_path(is_send, message, True)
+        content = self.create_table(doc, is_send, avatar)
+        run = content.paragraphs[0].add_run()
+        str_content = escape_js_and_html(str_content)
+        image_path = hard_link_db.get_image(str_content, BytesExtra, thumb=True)
+        base_path = os.path.join(OUTPUT_DIR, '聊天记录', self.contact.remark, 'image')
+        if not os.path.exists(os.path.join(Me().wx_dir, image_path)):
+            image_thumb_path = hard_link_db.get_image(str_content, BytesExtra, thumb=False)
+            if not os.path.exists(os.path.join(Me().wx_dir, image_thumb_path)):
+                return
+            image_path = image_thumb_path
+        image_path = get_image_abs_path(image_path, base_path=base_path)
+        try:
+            run.add_picture(image_path, height=shared.Inches(2))
+            doc.add_paragraph()
+        except Exception:
+            print("Error!image")
+
+    def audio(self, doc, message):
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        msgSvrId = message[9]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        avatar = self.get_avatar_path(is_send, message, True)
+        content_cell = self.create_table(doc, is_send, avatar)
+        content_cell.paragraphs[0].add_run('【语音】')
+        content_cell.paragraphs[0].font_size = shared.Inches(0.5)
+        if is_send:
+            p = content_cell.paragraphs[0]
+            p.paragraph_format.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+        doc.add_paragraph()
+
+    def emoji(self, doc, message):
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        avatar = self.get_avatar_path(is_send, message, True)
+        content_cell = self.create_table(doc, is_send, avatar)
+        content_cell.paragraphs[0].add_run('【表情包】')
+        content_cell.paragraphs[0].font_size = shared.Inches(0.5)
+        if is_send:
+            p = content_cell.paragraphs[0]
+            p.paragraph_format.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+        doc.add_paragraph()
+
+    def file(self, doc, message):
+        bytesExtra = message[10]
+        str_time = message[8]
+        is_send = message[4]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        avatar = self.get_avatar_path(is_send, message, True)
+        content_cell = self.create_table(doc, is_send, avatar)
+        content_cell.paragraphs[0].add_run('【文件】')
+        content_cell.paragraphs[0].font_size = shared.Inches(0.5)
+        if is_send:
+            p = content_cell.paragraphs[0]
+            p.paragraph_format.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+        doc.add_paragraph()
+
+    def refermsg(self, doc, message):
+        """
+        处理回复消息
+        @param doc:
+        @param message:
+        @return:
+        """
+        str_time = message[8]
+        is_send = message[4]
+        content = parser_reply(message[11])
+        refer_msg = content.get('refer')
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        avatar = self.get_avatar_path(is_send, message, True)
+        content_cell = self.create_table(doc, is_send, avatar)
+        content_cell.paragraphs[0].add_run(content.get('title'))
+        content_cell.paragraphs[0].font_size = shared.Inches(0.5)
+        reply_p = content_cell.add_paragraph()
+        reply_content = f"{refer_msg.get('displayname')}:{refer_msg.get('content')}" if refer_msg else '未知引用'
+        run = content_cell.paragraphs[1].add_run(reply_content)
+        '''设置被回复内容格式'''
+        run.font.color.rgb = shared.RGBColor(121, 121, 121)
+        run.font_size = shared.Inches(0.3)
+        run.font.highlight_color = WD_COLOR_INDEX.GRAY_25
+
+        if is_send:
+            p = content_cell.paragraphs[0]
+            p.paragraph_format.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+            reply_p.paragraph_format.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+        doc.add_paragraph()
+
+    def system_msg(self, doc, message):
+        str_content = message[7]
+        is_send = message[4]
+        str_time = message[8]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+
+        str_content = str_content.replace('<![CDATA[', "").replace(
+            ' <a href="weixin://revoke_edit_click">重新编辑</a>]]>', "")
+        res = findall('(</{0,1}(img|revo|_wc_cus|a).*?>)', str_content)
+        for xmlstr, b in res:
+            str_content = str_content.replace(xmlstr, "")
+        doc.add_paragraph(str_content).alignment = WD_PARAGRAPH_ALIGNMENT.CENTER
+
+    def video(self, doc, message):
+        type_ = message[2]
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        BytesExtra = message[10]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        avatar = self.get_avatar_path(is_send, message, True)
+        content_cell = self.create_table(doc, is_send, avatar)
+        content_cell.paragraphs[0].add_run('【视频】')
+        content_cell.paragraphs[0].font_size = shared.Inches(0.5)
+        if is_send:
+            p = content_cell.paragraphs[0]
+            p.paragraph_format.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+        doc.add_paragraph()
+
+    def create_table(self, doc, is_send, avatar_path):
+        '''
+        #! 创建一个1*2表格
+        #! isSend = 1 (0,0)存聊天内容，(0,1)存头像
+        #! isSend = 0 (0,0)存头像，(0,1)存聊天内容
+        #! 返回聊天内容的坐标
+        '''
+        table = doc.add_table(rows=1, cols=2, style='Normal Table')
+        table.cell(0, 1).height = shared.Inches(0.5)
+        table.cell(0, 0).height = shared.Inches(0.5)
+        if is_send:
+            '''表格右对齐'''
+            table.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+            avatar = table.cell(0, 1).paragraphs[0].add_run()
+            '''插入头像，设置头像宽度'''
+            avatar.add_picture(avatar_path, width=shared.Inches(0.5))
+            '''设置单元格宽度跟头像一致'''
+            table.cell(0, 1).width = shared.Inches(0.5)
+            content_cell = table.cell(0, 0)
+            '''聊天内容右对齐'''
+            content_cell.paragraphs[0].paragraph_format.alignment = WD_PARAGRAPH_ALIGNMENT.RIGHT
+        else:
+            avatar = table.cell(0, 0).paragraphs[0].add_run()
+            avatar.add_picture(avatar_path, width=shared.Inches(0.5))
+            '''设置单元格宽度'''
+            table.cell(0, 0).width = shared.Inches(0.5)
+            content_cell = table.cell(0, 1)
+        '''聊天内容垂直居中对齐'''
+        content_cell.vertical_alignment = WD_ALIGN_VERTICAL.CENTER
+        return content_cell
+
+    def music_share(self, doc, message):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        is_send = message[4]
+        timestamp = message[5]
+        content = music_share(message[11])
+        music_path = ''
+        if content.get('audio_url') != '':
+            music_path = get_music_path(content.get('audio_url'), content.get('title'),
+                                        output_path=origin_path + '/music')
+            if music_path != '':
+                music_path = f'./music/{os.path.basename(music_path)}'
+                music_path = music_path.replace('\\', '/')
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+
+    def share_card(self, doc, message):
+        origin_path = f"{os.getcwd()}/data/聊天记录/{self.contact.remark}"
+        is_send = message[4]
+        timestamp = message[5]
+        bytesExtra = message[10]
+        compress_content_ = message[11]
+        card_data = share_card(bytesExtra, compress_content_)
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        thumbnail = ''
+        if card_data.get('thumbnail'):
+            thumbnail = os.path.join(Me().wx_dir, card_data.get('thumbnail'))
+            if os.path.exists(thumbnail):
+                shutil.copy(thumbnail, os.path.join(origin_path, 'image', os.path.basename(thumbnail)))
+                thumbnail = './image/' + os.path.basename(thumbnail)
+            else:
+                thumbnail = ''
+        app_logo = ''
+        if card_data.get('app_logo'):
+            app_logo = os.path.join(Me().wx_dir, card_data.get('app_logo'))
+            if os.path.exists(app_logo):
+                shutil.copy(app_logo, os.path.join(origin_path, 'image', os.path.basename(app_logo)))
+                app_logo = './image/' + os.path.basename(app_logo)
+            else:
+                app_logo = ''
+
+    def merge_docx(self, conRemark, n):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录')
+        all_file_path = []
+        for i in range(n):
+            file_name = f"{conRemark}{i}.docx"
+            all_file_path.append(origin_path + '/' + file_name)
+        filename = f"{conRemark}.docx"
+        # print(all_file_path)
+        doc = docx.Document()
+        doc.save(origin_path + '/' + filename)
+        master = docx.Document(origin_path + '/' + filename)
+        middle_new_docx = Composer(master)
+        num = 0
+        for word in all_file_path:
+            word_document = docx.Document(word)
+            word_document.add_page_break()
+            if num != 0:
+                middle_new_docx.append(word_document)
+            num = num + 1
+            os.remove(word)
+        middle_new_docx.save(origin_path + '/' + filename)
+
+    def export(self):
+        print(f"【开始导出 DOCX {self.contact.remark}】")
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        messages = msg_db.get_messages(self.contact.wxid, time_range=self.time_range)
+        Me().save_avatar(os.path.join(origin_path, 'avatar', f'{Me().wxid}.png'))
+        if self.contact.is_chatroom:
+            for message in messages:
+                if message[4]:  # is_send
+                    continue
+                try:
+                    chatroom_avatar_path =os.path.join(origin_path, 'avatar', f'{message[13].wxid}.png')
+                    message[13].save_avatar(chatroom_avatar_path)
+                except:
+                    print(message)
+                    pass
+        else:
+            self.contact.save_avatar(os.path.join(origin_path, 'avatar', f'{self.contact.wxid}.png'))
+        self.rangeSignal.emit(len(messages))
+
+        def newdoc():
+            nonlocal n, doc
+            doc = docx.Document()
+            doc.styles["Normal"].font.name = "Cambria"
+            doc.styles["Normal"]._element.rPr.rFonts.set(qn("w:eastAsia"), "宋体")
+            n += 1
+
+        doc = None
+        n = 0
+        index = 0
+        newdoc()
+        for index, message in enumerate(messages):
+            if index % 200 == 0 and index:
+                filename = os.path.join(origin_path, f"{self.contact.remark}_{n}.docx")
+                doc.save(filename)
+                self.okSignal.emit(n)
+                newdoc()
+
+            type_ = message[2]
+            sub_type = message[3]
+            timestamp = message[5]
+            self.progressSignal.emit(1)
+            if self.is_5_min(timestamp):
+                str_time = message[8]
+                doc.add_paragraph(str_time).alignment = WD_PARAGRAPH_ALIGNMENT.CENTER
+            if type_ == 1 and self.message_types.get(type_):
+                self.text(doc, message)
+            elif type_ == 3 and self.message_types.get(type_):
+                self.image(doc, message)
+            elif type_ == 34 and self.message_types.get(type_):
+                self.audio(doc, message)
+            elif type_ == 43 and self.message_types.get(type_):
+                self.video(doc, message)
+            elif type_ == 47 and self.message_types.get(type_):
+                self.emoji(doc, message)
+            elif type_ == 10000 and self.message_types.get(type_):
+                self.system_msg(doc, message)
+            elif type_ == 49 and sub_type == 57 and self.message_types.get(1):
+                self.refermsg(doc, message)
+            elif type_ == 49 and sub_type == 6 and self.message_types.get(4906):
+                self.file(doc, message)
+            if index % 25 == 0:
+                print(f"【导出 DOCX {self.contact.remark}】{index}/{len(messages)}")
+        if index % 25:
+            print(f"【导出 DOCX {self.contact.remark}】{index + 1}/{len(messages)}")
+        filename = os.path.join(origin_path, f"{self.contact.remark}_{n}.docx")
+        try:
+            # document.save(filename)
+            doc.save(filename)
+        except PermissionError:
+            filename = filename[:-5] + f'{time.time()}' + '.docx'
+            # document.save(filename)
+            doc.save(filename)
+        self.okSignal.emit(n)
+        print(f"【完成导出 DOCX {self.contact.remark}】")
+        self.okSignal.emit(10086)

app/util/exporter/exporter_html.py

@@ -0,0 +1,523 @@
+import os
+import shutil
+import sys
+import traceback
+from re import findall
+
+from PyQt5.QtCore import pyqtSignal, QThread
+
+from app.DataBase import msg_db, hard_link_db, media_msg_db
+from app.util.exporter.exporter import ExporterBase, escape_js_and_html
+from app.config import OUTPUT_DIR
+from app.log import logger
+from app.person import Me
+from app.util import path
+from app.util.compress_content import parser_reply, share_card, music_share, file, transfer_decompress, call_decompress
+from app.util.emoji import get_emoji_url
+from app.util.image import get_image_path, get_image
+from app.util.music import get_music_path
+
+icon_files = {
+    './icon/word.png': ['doc', 'docx'],
+    './icon/excel.png': ['xls', 'xlsx'],
+    './icon/csv.png': ['csv'],
+    './icon/txt.png': ['txt'],
+    './icon/zip.png': ['zip', '7z', 'rar'],
+    './icon/ppt.png': ['ppt', 'pptx'],
+    './icon/pdf.png': ['pdf'],
+}
+
+
+class HtmlExporter(ExporterBase):
+    def text(self, doc, message):
+        type_ = message[2]
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+
+        display_name = self.get_display_name(is_send, message)
+        avatar = self.get_avatar_path(is_send, message)
+        str_content = escape_js_and_html(str_content)
+        doc.write(
+            f'''{{ type:{1}, text: '{str_content}',is_send:{is_send},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}'}},'''
+        )
+
+    def image(self, doc, message):
+        base_path = os.path.join(OUTPUT_DIR, '聊天记录', self.contact.remark, 'image')
+        type_ = message[2]
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        BytesExtra = message[10]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        str_content = escape_js_and_html(str_content)
+        image_path = hard_link_db.get_image(str_content, BytesExtra, up_dir=Me().wx_dir, thumb=False)
+        image_path = get_image_path(image_path, base_path=base_path)
+        doc.write(
+            f'''{{ type:{type_}, text: '{image_path}',is_send:{is_send},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}'}},'''
+        )
+
+    def audio(self, doc, message):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        msgSvrId = message[9]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        try:
+            audio_path = media_msg_db.get_audio_path(msgSvrId, output_path=origin_path + "/voice")
+            audio_path = "./voice/" + os.path.basename(audio_path)
+        except:
+            logger.error(traceback.format_exc())
+            return
+        voice_to_text = media_msg_db.get_audio_text(str_content)
+        if voice_to_text and voice_to_text != "":
+            voice_to_text = escape_js_and_html(voice_to_text)
+        doc.write(
+            f'''{{ type:34, text:'{audio_path}',is_send:{is_send},avatar_path:'{avatar}',voice_to_text:'{voice_to_text}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}'}},'''
+        )
+
+    def emoji(self, doc, message):
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        emoji_path = get_emoji_url(str_content, thumb=True)
+        doc.write(
+            f'''{{ type:{3}, text: '{emoji_path}',is_send:{is_send},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}'}},'''
+        )
+
+    def file(self, doc, message):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        bytesExtra = message[10]
+        compress_content = message[11]
+        str_time = message[8]
+        is_send = message[4]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        file_info = file(bytesExtra, compress_content, output_path=origin_path + '/file')
+        if file_info.get('is_error') == False:
+            icon_path = None
+            for icon, extensions in icon_files.items():
+                if file_info.get('file_ext') in extensions:
+                    icon_path = icon
+                    break
+            # 如果没有与文件后缀匹配的图标，则使用默认图标
+            if icon_path is None:
+                default_icon = './icon/file.png'
+                icon_path = default_icon
+            file_path = file_info.get('file_path')
+            if file_path != "":
+                file_path = './file/' + file_info.get('file_name')
+            doc.write(
+                f'''{{ type:49, text: '{file_path}',is_send:{is_send},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}',icon_path: '{icon_path}',sub_type:6,file_name: '{file_info.get('file_name')}',file_size: '{file_info.get('file_len')}',app_name: '{file_info.get('app_name')}'}},'''
+            )
+
+    def refermsg(self, doc, message):
+        """
+        处理回复消息
+        @param doc:
+        @param message:
+        @return:
+        """
+        str_time = message[8]
+        is_send = message[4]
+        content = parser_reply(message[11])
+        refer_msg = content.get('refer')
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        contentText = escape_js_and_html(content.get('title'))
+        if refer_msg:
+            referText = f"{escape_js_and_html(refer_msg.get('displayname'))}：{escape_js_and_html(refer_msg.get('content'))}"
+            doc.write(
+                f'''{{ type:49, text: '{contentText}',is_send:{is_send},sub_type:{content.get('type')},refer_text: '{referText}',avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}'}},'''
+            )
+        else:
+            doc.write(
+                f'''{{ type:49, text: '{contentText}',is_send:{is_send},sub_type:{content.get('type')},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}'}},'''
+            )
+
+    def system_msg(self, doc, message):
+        str_content = message[7]
+        is_send = message[4]
+        str_time = message[8]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+
+        str_content = str_content.replace('<![CDATA[', "").replace(
+            ' <a href="weixin://revoke_edit_click">重新编辑</a>]]>', "")
+        res = findall('(</{0,1}(img|revo|_wc_cus|a).*?>)', str_content)
+        for xmlstr, b in res:
+            str_content = str_content.replace(xmlstr, "")
+        str_content = escape_js_and_html(str_content)
+        doc.write(
+            f'''{{ type:0, text: '{str_content}',is_send:{is_send},avatar_path:'',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:''}},'''
+        )
+
+    def video(self, doc, message):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        type_ = message[2]
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        BytesExtra = message[10]
+        timestamp = message[5]
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        video_path = hard_link_db.get_video(str_content, BytesExtra, thumb=False)
+        image_path = hard_link_db.get_video(str_content, BytesExtra, thumb=True)
+        if video_path is None and image_path is not None:
+            image_path = path.get_relative_path(image_path, base_path=f'/data/聊天记录/{self.contact.remark}/image')
+            try:
+                # todo 网络图片问题
+                print(origin_path + image_path[1:])
+                os.utime(origin_path + image_path[1:], (timestamp, timestamp))
+                doc.write(
+                    f'''{{ type:3, text: '{image_path}',is_send:{is_send},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}'}},'''
+                )
+            except:
+                doc.write(
+                    f'''{{ type:1, text: '视频丢失',is_send:{is_send},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}'}},'''
+                )
+            return
+        if video_path is None and image_path is None:
+            return
+        video_path = f'{Me().wx_dir}/{video_path}'
+        video_path = video_path.replace('\\', '/')
+        if os.path.exists(video_path):
+            new_path = origin_path + '/video/' + os.path.basename(video_path)
+            if not os.path.exists(new_path):
+                shutil.copy(video_path, os.path.join(origin_path, 'video'))
+            os.utime(new_path, (timestamp, timestamp))
+            video_path = f'./video/{os.path.basename(video_path)}'
+        doc.write(
+            f'''{{ type:{type_}, text: '{video_path}',is_send:{is_send},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}'}},'''
+        )
+
+    def music_share(self, doc, message):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        is_send = message[4]
+        timestamp = message[5]
+        content = music_share(message[11])
+        music_path = ''
+        if content.get('is_error') == False:
+            if content.get('audio_url') != '':
+                music_path = get_music_path(content.get('audio_url'), content.get('title'),
+                                            output_path=origin_path + '/music')
+                if music_path != '':
+                    music_path = f'./music/{os.path.basename(music_path)}'
+                    music_path = music_path.replace('\\', '/')
+            is_chatroom = 1 if self.contact.is_chatroom else 0
+            avatar = self.get_avatar_path(is_send, message)
+            display_name = self.get_display_name(is_send, message)
+            music_path = escape_js_and_html(music_path)
+            doc.write(
+                f'''{{ type:49, text:'{music_path}',is_send:{is_send},avatar_path:'{avatar}',link_url:'{content.get('link_url')}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}',sub_type:3,title:'{content.get('title')}',artist:'{content.get('artist')}', website_name:'{content.get('website_name')}'}},'''
+            )
+
+    def share_card(self, doc, message):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        is_send = message[4]
+        timestamp = message[5]
+        bytesExtra = message[10]
+        compress_content_ = message[11]
+        card_data = share_card(bytesExtra, compress_content_)
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        thumbnail = ''
+        if card_data.get('thumbnail'):
+            thumbnail = os.path.join(Me().wx_dir, card_data.get('thumbnail'))
+            if os.path.exists(thumbnail):
+                shutil.copy(thumbnail, os.path.join(origin_path, 'image', os.path.basename(thumbnail)))
+                thumbnail = './image/' + os.path.basename(thumbnail)
+            else:
+                thumbnail = ''
+        app_logo = ''
+        if card_data.get('app_logo'):
+            app_logo = os.path.join(Me().wx_dir, card_data.get('app_logo'))
+            if os.path.exists(app_logo):
+                shutil.copy(app_logo, os.path.join(origin_path, 'image', os.path.basename(app_logo)))
+                app_logo = './image/' + os.path.basename(app_logo)
+            else:
+                app_logo = card_data.get('app_logo')
+        doc.write(
+            f'''{{ type:49,sub_type:5, text:'',is_send:{is_send},avatar_path:'{avatar}',url:'{card_data.get('url')}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}',title:'{card_data.get('title')}',description:'{card_data.get('description')}',thumbnail:'{thumbnail}',app_logo:'{app_logo}',app_name:'{card_data.get('app_name')}'}},\n'''
+        )
+
+    def transfer(self, doc, message):
+        is_send = message[4]
+        timestamp = message[5]
+        compress_content_ = message[11]
+        # open("test.bin", "wb").write(compress_content_)
+        transfer_detail = transfer_decompress(compress_content_)
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        try:
+            text_info_map = {
+                1: transfer_detail["pay_memo"] or "发起转账",
+                3: "已收款",
+                4: "已退还",
+                5: "非实时转账收款",
+                7: "发起非实时转账",
+                8: "未知",
+                9: "未知",
+            }
+            doc.write(
+                f"""{{ type:49,sub_type:2000,text:'{text_info_map[transfer_detail["paysubtype"]]}',is_send:{is_send},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}',paysubtype:{transfer_detail["paysubtype"]},pay_memo:'{transfer_detail["pay_memo"]}',feedesc:'{transfer_detail["feedesc"]}',}},\n""")
+        except Exception as e:
+            logger.error(f'转账解析错误：{transfer_detail}\n{traceback.format_exc()}')
+    def call(self, doc, message):
+        is_send = message[4]
+        timestamp = message[5]
+        str_content = message[7]
+        bytes_extra = message[10]
+        display_content = message[12]
+        call_detail = call_decompress(
+            is_send, bytes_extra, display_content, str_content
+        )
+        is_chatroom = 1 if self.contact.is_chatroom else 0
+        avatar = self.get_avatar_path(is_send, message)
+        display_name = self.get_display_name(is_send, message)
+        doc.write(
+            f"""{{ type:50, text:'{call_detail["display_content"]}',call_type:{call_detail["call_type"]},avatar_path:'{avatar}',timestamp:{timestamp},is_chatroom:{is_chatroom},displayname:'{display_name}',}},\n""")
+
+    def export(self):
+        print(f"【开始导出 HTML {self.contact.remark}】")
+        messages = msg_db.get_messages(self.contact.wxid, time_range=self.time_range)
+        filename = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark,
+                                f'{self.contact.remark}.html')
+        file_path = './app/resources/data/template.html'
+        if not os.path.exists(file_path):
+            resource_dir = getattr(sys, '_MEIPASS', os.path.abspath(os.path.dirname(__file__)))
+            file_path = os.path.join(resource_dir, 'app', 'resources', 'data', 'template.html')
+
+        with open(file_path, "r", encoding="utf-8") as f:
+            content = f.read()
+            html_head, html_end = content.split('/*注意看这是分割线*/')
+        f = open(filename, 'w', encoding='utf-8')
+        html_head = html_head.replace("<title>出错了</title>", f"<title>{self.contact.remark}</title>")
+        html_head = html_head.replace("<p id=\"title\">出错了</p>", f"<p id=\"title\">{self.contact.remark}</p>")
+        f.write(html_head)
+        self.rangeSignal.emit(len(messages))
+        for index, message in enumerate(messages):
+            type_ = message[2]
+            sub_type = message[3]
+            timestamp = message[5]
+            if (type_ == 3 and self.message_types.get(3)) or (type_ == 34 and self.message_types.get(34)) or (
+                    type_ == 47 and self.message_types.get(47)):
+                pass
+            else:
+                self.progressSignal.emit(1)
+
+            if type_ == 1 and self.message_types.get(type_):
+                self.text(f, message)
+            elif type_ == 3 and self.message_types.get(type_):
+                self.image(f, message)
+            elif type_ == 34 and self.message_types.get(type_):
+                self.audio(f, message)
+            elif type_ == 43 and self.message_types.get(type_):
+                self.video(f, message)
+            elif type_ == 47 and self.message_types.get(type_):
+                self.emoji(f, message)
+            elif type_ == 10000 and self.message_types.get(type_):
+                self.system_msg(f, message)
+            elif type_ == 49 and sub_type == 57 and self.message_types.get(1):
+                self.refermsg(f, message)
+            elif type_ == 49 and sub_type == 6 and self.message_types.get(4906):
+                self.file(f, message)
+            elif type_ == 49 and sub_type == 3 and self.message_types.get(4903):
+                self.music_share(f, message)
+            elif type_ == 49 and sub_type == 5 and self.message_types.get(4905):
+                self.share_card(f, message)
+            elif type_ == 49 and sub_type == 2000 and self.message_types.get(492000):
+                self.transfer(f, message)
+            elif type_ == 50 and self.message_types.get(50):
+                self.call(f, message)
+            if index % 2000 == 0:
+                print(f"【导出 HTML {self.contact.remark}】{index}/{len(messages)}")
+        f.write(html_end)
+        f.close()
+        print(f"【完成导出 HTML {self.contact.remark}】{len(messages)}")
+        self.count_finish_num(1)
+
+    def count_finish_num(self, num):
+        """
+        记录子线程完成个数
+        @param num:
+        @return:
+        """
+        self.num += 1
+        print("子线程完成", self.num, "/", self.total_num)
+        if self.num == self.total_num:
+            # 所有子线程都完成之后就发送完成信号
+            self.okSignal.emit(1)
+
+
+class OutputMedia(QThread):
+    """
+    导出语音消息
+    """
+
+    okSingal = pyqtSignal(int)
+    progressSignal = pyqtSignal(int)
+
+    def __init__(self, contact):
+        super().__init__()
+        self.contact = contact
+
+    def run(self):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        messages = msg_db.get_messages_by_type(self.contact.wxid, 34)
+        for message in messages:
+            is_send = message[4]
+            msgSvrId = message[9]
+            try:
+                audio_path = media_msg_db.get_audio(
+                    msgSvrId, output_path=origin_path + "/voice"
+                )
+            except:
+                logger.error(traceback.format_exc())
+            finally:
+                self.progressSignal.emit(1)
+        self.okSingal.emit(34)
+
+
+class OutputEmoji(QThread):
+    """
+    导出表情包
+    """
+
+    okSingal = pyqtSignal(int)
+    progressSignal = pyqtSignal(int)
+
+    def __init__(self, contact):
+        super().__init__()
+        self.contact = contact
+
+    def run(self):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        messages = msg_db.get_messages_by_type(self.contact.wxid, 47)
+        for message in messages:
+            str_content = message[7]
+            try:
+                pass
+                # emoji_path = get_emoji(str_content, thumb=True, output_path=origin_path + '/emoji')
+            except:
+                logger.error(traceback.format_exc())
+            finally:
+                self.progressSignal.emit(1)
+        self.okSingal.emit(47)
+
+
+class OutputImage(QThread):
+    """
+    导出图片
+    """
+
+    okSingal = pyqtSignal(int)
+    progressSignal = pyqtSignal(int)
+
+    def __init__(self, contact):
+        super().__init__()
+        self.contact = contact
+        self.child_thread_num = 2
+        self.child_threads = [0] * (self.child_thread_num + 1)
+        self.num = 0
+
+    def count1(self, num):
+        self.num += 1
+        print("图片导出完成一个")
+        if self.num == self.child_thread_num:
+            self.okSingal.emit(47)
+            print("图片导出完成")
+
+    def run(self):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        messages = msg_db.get_messages_by_type(self.contact.wxid, 3)
+        base_path = os.path.join(OUTPUT_DIR, '聊天记录', self.contact.remark, 'image')
+        for message in messages:
+            str_content = message[7]
+            BytesExtra = message[10]
+            timestamp = message[5]
+            try:
+                image_path = hard_link_db.get_image(
+                    str_content, BytesExtra, thumb=False
+                )
+                if not os.path.exists(os.path.join(Me().wx_dir, image_path)):
+                    image_thumb_path = hard_link_db.get_image(
+                        str_content, BytesExtra, thumb=True
+                    )
+                    if not os.path.exists(os.path.join(Me().wx_dir, image_thumb_path)):
+                        continue
+                    image_path = image_thumb_path
+                image_path = get_image(
+                    image_path, base_path=base_path
+                )
+                try:
+                    os.utime(origin_path + image_path[1:], (timestamp, timestamp))
+                except:
+                    pass
+            except:
+                logger.error(traceback.format_exc())
+            finally:
+                self.progressSignal.emit(1)
+        self.okSingal.emit(47)
+
+
+class OutputImageChild(QThread):
+    okSingal = pyqtSignal(int)
+    progressSignal = pyqtSignal(int)
+
+    def __init__(self, contact, messages):
+        super().__init__()
+        self.contact = contact
+        self.messages = messages
+
+    def run(self):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        for message in self.messages:
+            str_content = message[7]
+            BytesExtra = message[10]
+            timestamp = message[5]
+            try:
+                image_path = hard_link_db.get_image(
+                    str_content, BytesExtra, thumb=False
+                )
+                if not os.path.exists(os.path.join(Me().wx_dir, image_path)):
+                    image_thumb_path = hard_link_db.get_image(
+                        str_content, BytesExtra, thumb=True
+                    )
+                    if not os.path.exists(os.path.join(Me().wx_dir, image_thumb_path)):
+                        continue
+                    image_path = image_thumb_path
+                image_path = get_image(
+                    image_path, base_path=f"/data/聊天记录/{self.contact.remark}/image"
+                )
+                try:
+                    os.utime(origin_path + image_path[1:], (timestamp, timestamp))
+                except:
+                    pass
+            except:
+                logger.error(traceback.format_exc())
+            finally:
+                self.progressSignal.emit(1)
+        self.okSingal.emit(47)
+        print("图片子线程完成")

app/util/exporter/exporter_json.py

@@ -0,0 +1,193 @@
+import json
+import random
+import os
+
+from app.DataBase import msg_db
+from app.person import Me
+from .exporter import ExporterBase
+
+
+def merge_content(conversions_list) -> list:
+    """
+    合并一组对话中连续发送的句子
+    @param conversions_list:
+    @return:
+    """
+    merged_data = []
+    current_role = None
+    current_content = ""
+    str_time = ''
+    for item in conversions_list:
+        if 'str_time' in item:
+            str_time = item['str_time']
+        else:
+            str_time = ''
+        if current_role is None:
+            current_role = item["role"]
+            current_content = item["content"]
+        elif current_role == item["role"]:
+            current_content += "\n" + item["content"]
+        else:
+            # merged_data.append({"role": current_role, "content": current_content, 'str_time': str_time})
+            merged_data.append({"role": current_role, "content": current_content})
+            current_role = item["role"]
+            current_content = item["content"]
+            str_time = item.get('str_time')
+
+    # 处理最后一组
+    if current_role is not None:
+        # merged_data.append({"role": current_role, "content": current_content,'str_time': str_time})
+        merged_data.append({"role": current_role, "content": current_content})
+    return merged_data
+
+
+def system_prompt():
+    system = {
+        "role": "system",
+        # "content": f"你是{Me().name}，一个聪明、热情、善良的男大学生，后面的对话来自{self.contact.remark}(！！！注意：对方的身份十分重要，你务必记住对方的身份，因为跟不同的人对话要用不同的态度、语气)，你要认真地回答他"
+        "content": f"你是{Me().name}，一个聪明、热情、善良的人，后面的对话来自你的朋友，你要认真地回答他"
+    }
+    return system
+
+
+def message_to_conversion(group):
+    conversions = [system_prompt()]
+    while len(group) and group[-1][4] == 0:
+        group.pop()
+    for message in group:
+        is_send = message[4]
+        if len(conversions) == 1 and is_send:
+            continue
+        if is_send:
+            json_msg = {
+                "role": "assistant",
+                "content": message[7]
+            }
+        else:
+            json_msg = {
+                "role": "user",
+                "content": message[7]
+            }
+        json_msg['str_time'] = message[8]
+        conversions.append(json_msg)
+    if len(conversions) == 1:
+        return []
+    return merge_content(conversions)
+
+
+class JsonExporter(ExporterBase):
+    def split_by_time(self, length=300):
+        messages = msg_db.get_messages_by_type(self.contact.wxid, type_=1, time_range=self.time_range)
+        start_time = 0
+        res = []
+        i = 0
+        while i < len(messages):
+            message = messages[i]
+            timestamp = message[5]
+            is_send = message[4]
+            group = [
+                system_prompt()
+            ]
+            while i < len(messages) and timestamp - start_time < length:
+                if is_send:
+                    json_msg = {
+                        "role": "assistant",
+                        "content": message[7]
+                    }
+                else:
+                    json_msg = {
+                        "role": "user",
+                        "content": message[7]
+                    }
+                group.append(json_msg)
+                i += 1
+                if i >= len(messages):
+                    break
+                message = messages[i]
+                timestamp = message[5]
+                is_send = message[4]
+            while is_send:
+                json_msg = {
+                    "role": "assistant",
+                    "content": message[7]
+                }
+                group.append(json_msg)
+                i += 1
+                if i >= len(messages):
+                    break
+                message = messages[i]
+                timestamp = message[5]
+                is_send = message[4]
+            start_time = timestamp
+            res.append(
+                {
+                    "conversations": group
+                }
+            )
+        res_ = []
+        for item in res:
+            conversations = item['conversations']
+            res_.append({
+                'conversations': merge_content(conversations)
+            })
+        return res_
+
+    def split_by_intervals(self, max_diff_seconds=300):
+        messages = msg_db.get_messages_by_type(self.contact.wxid, type_=1, time_range=self.time_range)
+        res = []
+        i = 0
+        current_group = []
+        while i < len(messages):
+            message = messages[i]
+            timestamp = message[5]
+            is_send = message[4]
+            while is_send and i + 1 < len(messages):
+                i += 1
+                message = messages[i]
+                is_send = message[4]
+            current_group = [messages[i]]
+            i += 1
+            while i < len(messages) and messages[i][5] - current_group[-1][5] <= max_diff_seconds:
+                current_group.append(messages[i])
+                i += 1
+            while i < len(messages) and messages[i][4]:
+                current_group.append(messages[i])
+                i += 1
+            res.append(current_group)
+        res_ = []
+        for group in res:
+            conversations = message_to_conversion(group)
+            if conversations:
+                res_.append({
+                    'conversations': conversations
+                })
+        return res_
+
+    def to_json(self):
+        print(f"【开始导出 json {self.contact.remark}】")
+        origin_path = self.origin_path
+        os.makedirs(origin_path, exist_ok=True)
+        filename = os.path.join(origin_path, f"{self.contact.remark}")
+
+        # res = self.split_by_time()
+        res = self.split_by_intervals(60)
+        # 打乱列表顺序
+        random.shuffle(res)
+
+        # 计算切分比例
+        split_ratio = 0.2  # 20% for the second list
+
+        # 计算切分点
+        split_point = int(len(res) * split_ratio)
+
+        # 分割列表
+        train_data = res[split_point:]
+        dev_data = res[:split_point]
+        with open(f'{filename}_train.json', "w", encoding="utf-8") as f:
+            json.dump(train_data, f, ensure_ascii=False, indent=4)
+        with open(f'{filename}_dev.json', "w", encoding="utf-8") as f:
+            json.dump(dev_data, f, ensure_ascii=False, indent=4)
+        self.okSignal.emit(1)
+
+    def run(self):
+        self.to_json()

app/util/exporter/exporter_txt.py

@@ -0,0 +1,146 @@
+import os
+
+from app.DataBase import msg_db
+from app.util.exporter.exporter import ExporterBase
+from app.config import OUTPUT_DIR
+from app.util.compress_content import parser_reply, share_card
+
+
+class TxtExporter(ExporterBase):
+    def text(self, doc, message):
+        str_content = message[7]
+        str_time = message[8]
+        is_send = message[4]
+        display_name = self.get_display_name(is_send, message)
+        name = display_name
+        doc.write(
+            f'''{str_time} {name}\n{str_content}\n\n'''
+        )
+
+    def image(self, doc, message):
+        str_time = message[8]
+        is_send = message[4]
+        display_name = self.get_display_name(is_send, message)
+        doc.write(
+            f'''{str_time} {display_name}\n[图片]\n\n'''
+        )
+
+    def audio(self, doc, message):
+        str_time = message[8]
+        is_send = message[4]
+        display_name = self.get_display_name(is_send, message)
+        doc.write(
+            f'''{str_time} {display_name}\n[语音]\n\n'''
+        )
+    def emoji(self, doc, message):
+        str_time = message[8]
+        is_send = message[4]
+        display_name = self.get_display_name(is_send, message)
+        doc.write(
+            f'''{str_time} {display_name}\n[表情包]\n\n'''
+        )
+
+    def file(self, doc, message):
+        str_time = message[8]
+        is_send = message[4]
+        display_name = self.get_display_name(is_send, message)
+        doc.write(
+            f'''{str_time} {display_name}\n[文件]\n\n'''
+        )
+
+    def refermsg(self, doc, message):
+        """
+        处理回复消息
+        @param doc:
+        @param message:
+        @return:
+        """
+        str_time = message[8]
+        is_send = message[4]
+        content = parser_reply(message[11])
+        refer_msg = content.get('refer')
+        display_name = self.get_display_name(is_send, message)
+        if refer_msg:
+            doc.write(
+                f'''{str_time} {display_name}\n{content.get('title')}\n引用:{refer_msg.get('displayname')}:{refer_msg.get('content')}\n\n'''
+            )
+        else:
+            doc.write(
+                f'''{str_time} {display_name}\n{content.get('title')}\n引用:未知\n\n'''
+            )
+
+    def system_msg(self, doc, message):
+        str_content = message[7]
+        str_time = message[8]
+        str_content = str_content.replace('<![CDATA[', "").replace(
+            ' <a href="weixin://revoke_edit_click">重新编辑</a>]]>', "")
+        doc.write(
+            f'''{str_time} {str_content}\n\n'''
+        )
+
+    def video(self, doc, message):
+        str_time = message[8]
+        is_send = message[4]
+        display_name = self.get_display_name(is_send, message)
+        doc.write(
+            f'''{str_time} {display_name}\n[视频]\n\n'''
+        )
+    def music_share(self, doc, message):
+        is_send = message[4]
+        str_time = message[8]
+        display_name = self.get_display_name(is_send, message)
+        doc.write(
+            f'''{str_time} {display_name}\n[音乐分享]\n\n'''
+        )
+
+    def share_card(self, doc, message):
+        is_send = message[4]
+        bytesExtra = message[10]
+        compress_content_ = message[11]
+        str_time = message[8]
+        card_data = share_card(bytesExtra, compress_content_)
+        display_name = self.get_display_name(is_send, message)
+        doc.write(
+            f'''{str_time} {display_name}
+            [链接]:title:{card_data.get('title')}
+            description:{card_data.get('description')}
+            url:{card_data.get('url')}
+            name:{card_data.get('app_name')}
+            \n\n'''
+        )
+
+    def export(self):
+        # 实现导出为txt的逻辑
+        print(f"【开始导出 TXT {self.contact.remark}】")
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        os.makedirs(origin_path, exist_ok=True)
+        filename = os.path.join(origin_path, self.contact.remark+'.txt')
+        messages = msg_db.get_messages(self.contact.wxid, time_range=self.time_range)
+        total_steps = len(messages)
+        with open(filename, mode='w', newline='', encoding='utf-8') as f:
+            for index, message in enumerate(messages):
+                type_ = message[2]
+                sub_type = message[3]
+                self.progressSignal.emit(int((index + 1) / total_steps * 100))
+                if type_ == 1 and self.message_types.get(type_):
+                    self.text(f, message)
+                elif type_ == 3 and self.message_types.get(type_):
+                    self.image(f, message)
+                elif type_ == 34 and self.message_types.get(type_):
+                    self.audio(f, message)
+                elif type_ == 43 and self.message_types.get(type_):
+                    self.video(f, message)
+                elif type_ == 47 and self.message_types.get(type_):
+                    self.emoji(f, message)
+                elif type_ == 10000 and self.message_types.get(type_):
+                    self.system_msg(f, message)
+                elif type_ == 49 and sub_type == 57 and self.message_types.get(1):
+                    self.refermsg(f, message)
+                elif type_ == 49 and sub_type == 6 and self.message_types.get(4906):
+                    self.file(f, message)
+                elif type_ == 49 and sub_type == 3 and self.message_types.get(4903):
+                    self.music_share(f, message)
+                elif type_ == 49 and sub_type == 5 and self.message_types.get(4905):
+                    self.share_card(f, message)
+        print(f"【完成导出 TXT {self.contact.remark}】")
+        self.okSignal.emit(1)

app/util/exporter/output.py

@@ -0,0 +1,466 @@
+import csv
+import os
+import time
+import traceback
+from typing import List
+
+import docx
+from PyQt5.QtCore import pyqtSignal, QThread
+from PyQt5.QtWidgets import QFileDialog
+from docx.oxml.ns import qn
+from docxcompose.composer import Composer
+
+from app.util.exporter.exporter_ai_txt import AiTxtExporter
+from app.util.exporter.exporter_csv import CSVExporter
+from app.util.exporter.exporter_docx import DocxExporter
+from app.util.exporter.exporter_html import HtmlExporter
+from app.util.exporter.exporter_json import JsonExporter
+from app.util.exporter.exporter_txt import TxtExporter
+from app.DataBase.hard_link import decodeExtraBuf
+from app.config import OUTPUT_DIR
+from app.DataBase.package_msg import PackageMsg
+from app.DataBase import media_msg_db, hard_link_db, micro_msg_db, msg_db
+from app.log import logger
+from app.person import Me
+from app.util.image import get_image
+
+os.makedirs(os.path.join(OUTPUT_DIR, '聊天记录'), exist_ok=True)
+
+
+class Output(QThread):
+    """
+    发送信息线程
+    """
+    startSignal = pyqtSignal(int)
+    progressSignal = pyqtSignal(int)
+    rangeSignal = pyqtSignal(int)
+    okSignal = pyqtSignal(int)
+    batchOkSignal = pyqtSignal(int)
+    nowContact = pyqtSignal(str)
+    i = 1
+    CSV = 0
+    DOCX = 1
+    HTML = 2
+    CSV_ALL = 3
+    CONTACT_CSV = 4
+    TXT = 5
+    JSON = 6
+    AI_TXT = 7
+    Batch = 10086
+
+    def __init__(self, contact, type_=DOCX, message_types={}, sub_type=[], time_range=None, parent=None):
+        super().__init__(parent)
+        self.children = []
+        self.last_timestamp = 0
+        self.sub_type = sub_type
+        self.time_range = time_range
+        self.message_types = message_types
+        self.sec = 2  # 默认1000秒
+        self.contact = contact
+        self.msg_id = 0
+        self.output_type: int | List[int] = type_
+        self.total_num = 1
+        self.num = 0
+
+    def progress(self, value):
+        self.progressSignal.emit(value)
+
+    def output_image(self):
+        """
+        导出全部图片
+        @return:
+        """
+        return
+
+    def output_emoji(self):
+        """
+        导出全部表情包
+        @return:
+        """
+        return
+
+    def to_csv_all(self):
+        """
+        导出全部聊天记录到CSV
+        @return:
+        """
+
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录')
+        os.makedirs(origin_path, exist_ok=True)
+        filename = QFileDialog.getSaveFileName(None, "save file", os.path.join(os.getcwd(), 'messages.csv'),
+                                               "csv files (*.csv);;all files(*.*)")
+        if not filename[0]:
+            return
+        self.startSignal.emit(1)
+        filename = filename[0]
+        # columns = ["用户名", "消息内容", "发送时间", "发送状态", "消息类型", "isSend", "msgId"]
+        columns = ['localId', 'TalkerId', 'Type', 'SubType',
+                   'IsSender', 'CreateTime', 'Status', 'StrContent',
+                   'StrTime', 'Remark', 'NickName', 'Sender']
+
+        packagemsg = PackageMsg()
+        messages = packagemsg.get_package_message_all()
+        # 写入CSV文件
+        with open(filename, mode='w', newline='', encoding='utf-8-sig') as file:
+            writer = csv.writer(file)
+            writer.writerow(columns)
+            # 写入数据
+            writer.writerows(messages)
+        self.okSignal.emit(1)
+
+    def contact_to_csv(self):
+        """
+        导出联系人到CSV
+        @return:
+        """
+        filename = QFileDialog.getSaveFileName(None, "save file", os.path.join(os.getcwd(), 'contacts.csv'),
+                                               "csv files (*.csv);;all files(*.*)")
+        if not filename[0]:
+            return
+        self.startSignal.emit(1)
+        filename = filename[0]
+        # columns = ["用户名", "消息内容", "发送时间", "发送状态", "消息类型", "isSend", "msgId"]
+        columns = ['UserName', 'Alias', 'Type', 'Remark', 'NickName', 'PYInitial', 'RemarkPYInitial', 'smallHeadImgUrl',
+                   'bigHeadImgUrl', 'label', 'gender', 'telephone', 'signature', 'country/region', 'province', 'city']
+        contacts = micro_msg_db.get_contact()
+        # 写入CSV文件
+        with open(filename, mode='w', newline='', encoding='utf-8-sig') as file:
+            writer = csv.writer(file)
+            writer.writerow(columns)
+            # 写入数据
+            # writer.writerows(contacts)
+            for contact in contacts:
+                detail = decodeExtraBuf(contact[9])
+                gender_code = detail.get('gender')
+                if gender_code == 0:
+                    gender = '未知'
+                elif gender_code == 1:
+                    gender = '男'
+                else:
+                    gender = '女'
+                writer.writerow([*contact[:9], contact[10], gender, detail.get('telephone'), detail.get('signature'),
+                                 *detail.get('region')])
+
+        self.okSignal.emit(1)
+
+    def batch_export(self):
+        print('开始批量导出')
+        print(self.sub_type, self.message_types)
+        print(len(self.contact))
+        print([contact.remark for contact in self.contact])
+        self.batch_num_total = len(self.contact) * len(self.sub_type)
+        self.batch_num = 0
+        self.rangeSignal.emit(self.batch_num_total)
+        for contact in self.contact:
+            # print('联系人', contact.remark)
+            for type_ in self.sub_type:
+                # print('导出类型', type_)
+                if type_ == self.DOCX:
+                    self.to_docx(contact, self.message_types, True)
+                elif type_ == self.TXT:
+                    # print('批量导出txt')
+                    self.to_txt(contact, self.message_types, True)
+                elif type_ == self.AI_TXT:
+                    # print('批量导出txt')
+                    self.to_ai_txt(contact, self.message_types, True)
+                elif type_ == self.CSV:
+                    self.to_csv(contact, self.message_types, True)
+                elif type_ == self.HTML:
+                    self.to_html(contact, self.message_types, True)
+                elif type_ == self.JSON:
+                    self.to_json(contact,self.message_types,True)
+
+    def batch_finish_one(self, num):
+        self.nowContact.emit(self.contact[self.batch_num // len(self.sub_type)].remark)
+        self.batch_num += 1
+        if self.batch_num == self.batch_num_total:
+            self.okSignal.emit(1)
+
+    def merge_docx(self, n):
+        conRemark = self.contact.remark
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', conRemark)
+        filename = f"{origin_path}/{conRemark}_{n}.docx"
+        if n == 10086:
+            # self.document.append(self.document)
+            file = os.path.join(origin_path, f'{conRemark}.docx')
+            try:
+                self.document.save(file)
+            except PermissionError:
+                file = file[:-5] + f'{time.time()}' + '.docx'
+                self.document.save(file)
+            self.okSignal.emit(1)
+            return
+        doc = docx.Document(filename)
+        self.document.append(doc)
+        os.remove(filename)
+        if n % 50 == 0:
+            # self.document.append(self.document)
+            file = os.path.join(origin_path, f'{conRemark}-{n // 50}.docx')
+            try:
+                self.document.save(file)
+            except PermissionError:
+                file = file[:-5] + f'{time.time()}' + '.docx'
+                self.document.save(file)
+            doc = docx.Document()
+            doc.styles["Normal"].font.name = "Cambria"
+            doc.styles["Normal"]._element.rPr.rFonts.set(qn("w:eastAsia"), "宋体")
+            self.document = Composer(doc)
+
+    def to_docx(self, contact, message_types, is_batch=False):
+        doc = docx.Document()
+        doc.styles["Normal"].font.name = "Cambria"
+        doc.styles["Normal"]._element.rPr.rFonts.set(qn("w:eastAsia"), "宋体")
+        self.document = Composer(doc)
+        Child = DocxExporter(contact, type_=self.DOCX, message_types=message_types, time_range=self.time_range)
+        self.children.append(Child)
+        Child.progressSignal.connect(self.progress)
+        if not is_batch:
+            Child.rangeSignal.connect(self.rangeSignal)
+        Child.okSignal.connect(self.merge_docx if not is_batch else self.batch_finish_one)
+        Child.start()
+
+    def to_json(self, contact, message_types, is_batch=False):
+        Child = JsonExporter(contact, type_=self.JSON, message_types=message_types, time_range=self.time_range)
+        self.children.append(Child)
+        Child.progressSignal.connect(self.progress)
+        if not is_batch:
+            Child.rangeSignal.connect(self.rangeSignal)
+        Child.okSignal.connect(self.okSignal if not is_batch else self.batch_finish_one)
+        Child.start()
+
+    def to_txt(self, contact, message_types, is_batch=False):
+        Child = TxtExporter(contact, type_=self.TXT, message_types=message_types, time_range=self.time_range)
+        self.children.append(Child)
+        Child.progressSignal.connect(self.progress)
+        if not is_batch:
+            Child.rangeSignal.connect(self.rangeSignal)
+        Child.okSignal.connect(self.okSignal if not is_batch else self.batch_finish_one)
+        Child.start()
+
+    def to_ai_txt(self, contact, message_types, is_batch=False):
+        Child = AiTxtExporter(contact, type_=self.TXT, message_types=message_types, time_range=self.time_range)
+        self.children.append(Child)
+        Child.progressSignal.connect(self.progress)
+        if not is_batch:
+            Child.rangeSignal.connect(self.rangeSignal)
+        Child.okSignal.connect(self.okSignal if not is_batch else self.batch_finish_one)
+        Child.start()
+
+    def to_html(self, contact, message_types, is_batch=False):
+        Child = HtmlExporter(contact, type_=self.output_type, message_types=message_types, time_range=self.time_range)
+        self.children.append(Child)
+        Child.progressSignal.connect(self.progress)
+        if not is_batch:
+            Child.rangeSignal.connect(self.rangeSignal)
+        Child.okSignal.connect(self.count_finish_num)
+        Child.start()
+        self.total_num = 1
+        if message_types.get(34):
+            # 语音消息单独的线程
+            self.total_num += 1
+            output_media = OutputMedia(contact, time_range=self.time_range)
+            self.children.append(output_media)
+            output_media.okSingal.connect(self.count_finish_num)
+            output_media.progressSignal.connect(self.progressSignal)
+            output_media.start()
+        if message_types.get(47):
+            # emoji消息单独的线程
+            self.total_num += 1
+            output_emoji = OutputEmoji(contact, time_range=self.time_range)
+            self.children.append(output_emoji)
+            output_emoji.okSingal.connect(self.count_finish_num)
+            output_emoji.progressSignal.connect(self.progressSignal)
+            output_emoji.start()
+        if message_types.get(3):
+            # 图片消息单独的线程
+            self.total_num += 1
+            output_image = OutputImage(contact, time_range=self.time_range)
+            self.children.append(output_image)
+            output_image.okSingal.connect(self.count_finish_num)
+            output_image.progressSignal.connect(self.progressSignal)
+            output_image.start()
+
+    def to_csv(self, contact, message_types, is_batch=False):
+        Child = CSVExporter(contact, type_=self.CSV, message_types=message_types, time_range=self.time_range)
+        self.children.append(Child)
+        Child.progressSignal.connect(self.progress)
+        if not is_batch:
+            Child.rangeSignal.connect(self.rangeSignal)
+        Child.okSignal.connect(self.okSignal if not is_batch else self.batch_finish_one)
+        Child.start()
+
+    def run(self):
+        if self.output_type == self.DOCX:
+            self.to_docx(self.contact, self.message_types)
+        elif self.output_type == self.CSV_ALL:
+            self.to_csv_all()
+        elif self.output_type == self.CONTACT_CSV:
+            self.contact_to_csv()
+        elif self.output_type == self.TXT:
+            self.to_txt(self.contact, self.message_types)
+        elif self.output_type == self.AI_TXT:
+            self.to_ai_txt(self.contact, self.message_types)
+        elif self.output_type == self.CSV:
+            self.to_csv(self.contact, self.message_types)
+        elif self.output_type == self.HTML:
+            self.to_html(self.contact, self.message_types)
+        elif self.output_type == self.JSON:
+            self.to_json(self.contact, self.message_types)
+        elif self.output_type == self.Batch:
+            self.batch_export()
+
+    def count_finish_num(self, num):
+        """
+        记录子线程完成个数
+        @param num:
+        @return:
+        """
+        self.num += 1
+        if self.num == self.total_num:
+            # 所有子线程都完成之后就发送完成信号
+            if self.output_type == self.Batch:
+                self.batch_finish_one(1)
+            else:
+                self.okSignal.emit(1)
+            self.num = 0
+
+    def cancel(self):
+        self.requestInterruption()
+
+
+class OutputMedia(QThread):
+    """
+    导出语音消息
+    """
+    okSingal = pyqtSignal(int)
+    progressSignal = pyqtSignal(int)
+
+    def __init__(self, contact, time_range=None):
+        super().__init__()
+        self.contact = contact
+        self.time_range = time_range
+
+    def run(self):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        messages = msg_db.get_messages_by_type(self.contact.wxid, 34, time_range=self.time_range)
+        for message in messages:
+            is_send = message[4]
+            msgSvrId = message[9]
+            try:
+                audio_path = media_msg_db.get_audio(msgSvrId, output_path=origin_path + "/voice")
+            except:
+                logger.error(traceback.format_exc())
+            finally:
+                self.progressSignal.emit(1)
+        self.okSingal.emit(34)
+
+
+class OutputEmoji(QThread):
+    """
+    导出表情包
+    """
+    okSingal = pyqtSignal(int)
+    progressSignal = pyqtSignal(int)
+
+    def __init__(self, contact, time_range=None):
+        super().__init__()
+        self.contact = contact
+        self.time_range = time_range
+
+    def run(self):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        messages = msg_db.get_messages_by_type(self.contact.wxid, 47, time_range=self.time_range)
+        for message in messages:
+            str_content = message[7]
+            try:
+                pass
+                # emoji_path = get_emoji(str_content, thumb=True, output_path=origin_path + '/emoji')
+            except:
+                logger.error(traceback.format_exc())
+            finally:
+                self.progressSignal.emit(1)
+        self.okSingal.emit(47)
+
+
+class OutputImage(QThread):
+    """
+    导出图片
+    """
+    okSingal = pyqtSignal(int)
+    progressSignal = pyqtSignal(int)
+
+    def __init__(self, contact, time_range):
+        super().__init__()
+        self.contact = contact
+        self.child_thread_num = 2
+        self.time_range = time_range
+        self.child_threads = [0] * (self.child_thread_num + 1)
+        self.num = 0
+
+    def count1(self, num):
+        self.num += 1
+        print('图片导出完成一个')
+        if self.num == self.child_thread_num:
+            self.okSingal.emit(47)
+            print('图片导出完成')
+
+    def run(self):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        messages = msg_db.get_messages_by_type(self.contact.wxid, 3, time_range=self.time_range)
+        base_path = os.path.join(OUTPUT_DIR, '聊天记录', self.contact.remark, 'image')
+        for message in messages:
+            str_content = message[7]
+            BytesExtra = message[10]
+            timestamp = message[5]
+            try:
+                image_path = hard_link_db.get_image(str_content, BytesExtra, up_dir=Me().wx_dir, thumb=False)
+                image_path = get_image(image_path, base_path=base_path)
+                try:
+                    os.utime(origin_path + image_path[1:], (timestamp, timestamp))
+                except:
+                    pass
+            except:
+                logger.error(traceback.format_exc())
+            finally:
+                self.progressSignal.emit(1)
+        self.okSingal.emit(47)
+
+
+class OutputImageChild(QThread):
+    okSingal = pyqtSignal(int)
+    progressSignal = pyqtSignal(int)
+
+    def __init__(self, contact, messages, time_range):
+        super().__init__()
+        self.contact = contact
+        self.messages = messages
+        self.time_range = time_range
+
+    def run(self):
+        origin_path = os.path.join(os.getcwd(), OUTPUT_DIR, '聊天记录', self.contact.remark)
+        for message in self.messages:
+            str_content = message[7]
+            BytesExtra = message[10]
+            timestamp = message[5]
+            try:
+                image_path = hard_link_db.get_image(str_content, BytesExtra, thumb=False)
+                if not os.path.exists(os.path.join(Me().wx_dir, image_path)):
+                    image_thumb_path = hard_link_db.get_image(str_content, BytesExtra, thumb=True)
+                    if not os.path.exists(os.path.join(Me().wx_dir, image_thumb_path)):
+                        continue
+                    image_path = image_thumb_path
+                image_path = get_image(image_path, base_path=f'/data/聊天记录/{self.contact.remark}/image')
+                try:
+                    os.utime(origin_path + image_path[1:], (timestamp, timestamp))
+                except:
+                    pass
+            except:
+                logger.error(traceback.format_exc())
+            finally:
+                self.progressSignal.emit(1)
+        self.okSingal.emit(47)
+        print('图片子线程完成')
+
+
+if __name__ == "__main__":
+    pass