图片问题--未完全修复

app/DataBase/sns.py

@@ -49,7 +49,7 @@ class Sns:
             return None
         try:
             lock.acquire(True)
-            sql = '''select UserName, Content from FeedsV20  where CreateTime>=?
+            sql = '''select UserName, Content, FeedId from FeedsV20  where CreateTime>=?
                   and  CreateTime<=? order by CreateTime desc'''
             self.cursor.execute(sql, [start_time, end_time])
             res = self.cursor.fetchall()
@@ -58,6 +58,19 @@ class Sns:
 
         return res
 
+    def get_comment_by_feed_id(self, feed_id):
+        if not self.open_flag:
+            return None
+        try:
+            lock.acquire(True)
+            sql = '''select FromUserName, CommentType, Content from CommentV20 where FeedId=?
+                   order by CreateTime desc'''
+            self.cursor.execute(sql, [feed_id])
+            res = self.cursor.fetchall()
+        finally:
+            lock.release()
+        return res
+
     def get_cover_url(self) -> Optional[str]:
         if not self.open_flag:
             return None

decrypter/image_decrypt.py

@@ -0,0 +1,116 @@
+import hashlib
+import os
+import shutil
+import subprocess
+import sys
+import traceback
+from datetime import date
+from pathlib import Path
+import filetype
+
+import log
+
+
+class ImageDecrypter:
+
+    def __init__(self, gui: 'Gui', file_path):
+        self.file_path = file_path
+        self.gui = gui
+        self.sns_cache_path = file_path + "/FileStorage/Sns/Cache"
+
+    @staticmethod
+    def get_output_path(dir_name, md5, duration):
+        if getattr(sys, 'frozen', False) and hasattr(sys, '_MEIPASS'):
+            # 这是到_internal文件夹
+            resource_dir = getattr(sys, '_MEIPASS')
+            # 获取_internal上一级文件夹再拼接
+            return os.path.join(os.path.dirname(resource_dir), 'output', dir_name, 'videos', f'{md5}_{duration}.mp4')
+        else:
+            return os.path.join(os.getcwd(), 'output', dir_name, 'videos', f'{md5}_{duration}.mp4')
+
+    @staticmethod
+    def calculate_md5(file_path):
+        with open(file_path, "rb") as f:
+            file_content = f.read()
+        return hashlib.md5(file_content).hexdigest()
+
+    @staticmethod
+    def get_all_month_between_dates(start_date, end_date) -> list[str]:
+        result = []
+        current_date = start_date
+        while current_date <= end_date:
+            # 打印当前日期的年份和月份
+            result.append(current_date.strftime("%Y-%m"))
+            year = current_date.year + (current_date.month // 12)
+            month = current_date.month % 12 + 1
+            # 更新current_date到下个月的第一天
+            current_date = date(year, month, 1)
+        return result
+
+    @staticmethod
+    def decode(magic, buf):
+        return bytearray([b ^ magic for b in list(buf)])
+
+    @staticmethod
+    def guess_image_encoding_magic(buf):
+        header_code, check_code = 0xff, 0xd8
+        # 微信图片加密方法对字节逐一“异或”，即 源文件^magic(未知数)=加密后文件
+        # 已知jpg的头字节是0xff，将0xff与加密文件的头字节做异或运算求解magic码
+        magic = header_code ^ list(buf)[0] if buf else 0x00
+        # 尝试使用magic码解密，如果第二字节符合jpg特质，则图片解密成功
+        _, code = ImageDecrypter.decode(magic, buf[:2])
+        if check_code == code:
+            return magic
+
+    def decrypt_images(self, exporter, start_date, end_date, dir_name) -> None:
+        """将图片文件从缓存中复制出来，重命名为{主图字节数}_{缩略图字节数}.jpg
+        duration单位为秒
+        """
+        months = self.get_all_month_between_dates(start_date, end_date)
+
+        total_files = 0
+        processed_files = 0
+        for month in months:
+            source_dir = self.sns_cache_path + "/" + month
+            total_files = total_files + len(list(Path(source_dir).rglob('*')))
+
+        for month in months:
+            source_dir = self.sns_cache_path + "/" + month
+            for file in Path(source_dir).rglob('*'):
+                # 排除缩略图
+                if not exporter.stop_flag and file.is_file() and not file.name.endswith('_t'):
+                    try:
+                        with open(file, 'rb') as f:
+                            buff = bytearray(f.read())
+                        magic = self.guess_image_encoding_magic(buff)
+                        if magic:
+                            os.makedirs(f"output/{dir_name}/images/{month}/", exist_ok=True)
+                            os.makedirs(f"output/{dir_name}/thumbs/{month}/", exist_ok=True)
+                            main_file_size = file.stat().st_size
+                            thumb_file_size = 0
+                            # 找到对应缩略图
+                            thumb_file = Path(f'{source_dir}/{file.name}_t')
+                            if thumb_file.exists():
+                                thumb_file_size = thumb_file.stat().st_size
+                                # 读缩略图加密
+                                with open(thumb_file, 'rb') as f:
+                                    thumb_buff = bytearray(f.read())
+
+                                # 写缩略图
+                                thumb_destination = (f"output/{dir_name}/thumbs/{month}/"
+                                                     f"{main_file_size}_{thumb_file_size}.jpg")
+                                with open(thumb_destination, 'wb') as f:
+                                    new_thumb_buff = self.decode(magic, thumb_buff)
+                                    f.write(new_thumb_buff)
+
+                            destination = (f"output/{dir_name}/images/{month}/"
+                                           f"{main_file_size}_{thumb_file_size}.jpg")
+                            with open(destination, 'wb') as f:
+                                new_buf = self.decode(magic, buff)
+                                f.write(new_buf)
+                    except Exception:
+                        traceback.print_exc()
+                processed_files = processed_files + 1
+                # 15%的进度作为处理图片使用
+                progress = round(processed_files / total_files * 15)
+                self.gui.update_export_progressbar(progress)

decrypter/video_decrypt.py

@@ -127,6 +127,6 @@ class VideoDecrypter:
                     except Exception:
                         traceback.print_exc()
                 processed_files = processed_files + 1
-                # 前30%的进度作为 处理视频使用
+                # 15%的进度作为处理视频使用 + 15%(处理图像)
-                progress = round(processed_files / total_files * 30)
+                progress = round(processed_files / total_files * 15 + 15)
                 self.gui.update_export_progressbar(progress)

entity/comment.py

@@ -0,0 +1,8 @@
+from dataclasses import dataclass
+
+
+@dataclass
+class Comment:
+    from_user_name: str
+    comment_type: int
+    content: str

entity/moment_msg.py

@@ -22,12 +22,16 @@ class Url:
     type: str = field(metadata=config(field_name="@type"))
     text: str = field(metadata=config(field_name="#text"),  default="")
     md5: str = field(metadata=config(field_name="@md5"), default="")
+    token: str = field(metadata=config(field_name="@token"), default="")
+    enc_idx: str = field(metadata=config(field_name="@enc_idx"), default="")
 
 @dataclass_json
 @dataclass
 class Thumb:
     type: str = field(metadata=config(field_name="@type"))
     text: str = field(metadata=config(field_name="#text"))
+    token: str = field(metadata=config(field_name="@token"), default="")
+    enc_idx: str = field(metadata=config(field_name="@enc_idx"), default="")
 
 
 @dataclass_json
@@ -91,6 +95,13 @@ class TimelineObject:
         beijing_timezone = timezone(timedelta(hours=8))
         time_formatted = dt.astimezone(beijing_timezone).strftime('%Y-%m-%d %H:%M:%S')
         return time_formatted
+    @property
+    def create_year_month(self)->str:
+        dt = datetime.fromtimestamp(self.createTime, timezone.utc)
+        # 转换为北京时间（UTC+8）
+        beijing_timezone = timezone(timedelta(hours=8))
+        time_formatted = dt.astimezone(beijing_timezone).strftime('%Y-%m')
+        return time_formatted
 
 
 @dataclass_json

exporter/html_exporter.py

@@ -7,6 +7,7 @@ from typing import Tuple
 
 import xmltodict
 
+from entity.comment import Comment
 from entity.contact import Contact
 from exporter.avatar_exporter import AvatarExporter
 from exporter.emoji_exporter import EmojiExporter
@@ -67,7 +68,7 @@ def get_music_info(msg: MomentMsg) -> Tuple[str, str, str]:
 class HtmlExporter(threading.Thread):
 
     def __init__(self, gui: 'Gui', dir_name: str, contacts_map: dict[str, Contact], begin_date: datetime.date,
-                 end_date: datetime.date, download_pic: int, convert_video: int):
+                 end_date: datetime.date, convert_video: int):
         self.dir_name = dir_name
         if Path(f"output/{self.dir_name}").exists():
             shutil.rmtree(f"output/{self.dir_name}")
@@ -83,7 +84,6 @@ class HtmlExporter(threading.Thread):
         self.contacts_map = contacts_map
         self.begin_date = begin_date
         self.end_date = end_date
-        self.download_pic = download_pic
         self.convert_video = convert_video
         self.stop_flag = False
         super().__init__()
@@ -105,7 +105,7 @@ class HtmlExporter(threading.Thread):
         from app.DataBase import sns_db
         cover_url = sns_db.get_cover_url()
         if cover_url:
-            cover_path = self.image_exporter.save_image(cover_url, 'image')
+            cover_path = self.image_exporter.save_image((cover_url, "", ""), 'image')
             self.html_head = self.html_head.replace("{cover_path}", cover_path)
 
         self.file.write(self.html_head)
@@ -115,13 +115,20 @@ class HtmlExporter(threading.Thread):
             datetime.datetime(self.begin_date.year, self.begin_date.month, self.begin_date.day).timetuple())
         end_time = time.mktime(datetime.datetime(end_date.year, end_date.month, end_date.day).timetuple())
 
+        self.gui.image_decrypter.decrypt_images(self, self.begin_date, end_date, self.dir_name)
         self.gui.video_decrypter.decrypt_videos(self, self.begin_date, end_date, self.dir_name, self.convert_video)
 
+
         message_datas = sns_db.get_messages_in_time(begin_time, end_time)
         for index, message_data in enumerate(message_datas):
             if not self.stop_flag:
                 if message_data[0] in self.contacts_map:
-                    self.export_msg(message_data[1], self.contacts_map, self.download_pic)
+                    comments_datas = sns_db.get_comment_by_feed_id(message_data[2])
+                    comments: list[Comment] = []
+                    for c in comments_datas:
+                        contact = Comment(c[0], c[1], c[2])
+                        comments.append(contact)
+                    self.export_msg(message_data[1], comments, self.contacts_map)
                     # 更新进度条 前30%视频处理  后70%其他处理
                     progress = round(index / len(message_datas) * 70)
                     self.gui.update_export_progressbar(30 + progress)
@@ -132,7 +139,7 @@ class HtmlExporter(threading.Thread):
     def stop(self) -> None:
         self.stop_flag = True
 
-    def export_msg(self, message: str, contacts_map: dict[str, Contact], download_pic: int) -> None:
+    def export_msg(self, message: str, comments: list[Comment], contacts_map: dict[str, Contact]) -> None:
 
         LOG.info(message)
         # force_list: 强制要求转media为list
@@ -150,7 +157,7 @@ class HtmlExporter(threading.Thread):
         remark = contact.remark if contact.remark else contact.nickName
 
         # 朋友圈图片
-        images = self.image_exporter.get_images(msg, download_pic)
+        images = self.image_exporter.get_images(msg)
 
         # 朋友圈视频
         videos = self.video_exporter.get_videos(msg)
@@ -208,7 +215,8 @@ class HtmlExporter(threading.Thread):
             html += f'     <div style="width:10rem; overflow:hidden">\n'
             # 视频号图片
             thumb_path = self.image_exporter.get_finder_images(msg)
-            html += f'       <img src="{thumb_path}" onclick="openWarningOverlay(event)" style="width:10rem;height:10rem;object-fit:cover;cursor:pointer;"/>\n'
+            html += f"""       <img src=\"{thumb_path}\" onclick=\"openWarningOverlay(event)\" 
+                               style=\"width:10rem;height:10rem;object-fit:cover;cursor:pointer;\"/>\n"""
             html += '      </div>\n'
 
             # 视频号说明
@@ -221,7 +229,8 @@ class HtmlExporter(threading.Thread):
         else:
             html += f'     <div style="{get_img_div_css(len(images))}">\n'
             for thumb_path, image_path in images:
-                html += f'     <img src="{thumb_path}" full_img="{image_path}" onclick="openFullSize(event)" style="{get_img_css(len(images))}"/>\n'
+                html += f"""     <img src="{thumb_path}" full_img="{image_path}" onclick="openFullSize(event)" 
+                                 style="{get_img_css(len(images))}"/>\n"""
             html += '      </div>\n'
 
             html += '      <div>\n'

exporter/image_exporter.py

@@ -1,7 +1,8 @@
 import os
-import re
+from io import BytesIO
+from pathlib import Path
 from typing import Tuple, Optional
-
+from PIL import Image
 from entity.moment_msg import MomentMsg, Media
 import requests
 import uuid
@@ -15,36 +16,52 @@ class ImageExporter:
         if not os.path.exists(f'output/{self.dir_name}/images/'):
             os.mkdir(f'output/{self.dir_name}/images/')
 
-    def save_image(self, url: str, img_type: str) -> str:
+    @staticmethod
-        """ 下载图片
+    def get_image(link: tuple) -> bytes:
+        """ 向微信服务器请求图片
         """
-        if not (img_type == 'image' or img_type == 'thumb'):
+        url, idx, token = link
-            raise Exception("img_type 参数非法")
+        # 如果需要传递token
-        file_name = uuid.uuid4()
+        if idx and token:
+            url = f'{url}?idx={idx}&token={token}'
         response = requests.get(url)
         if response.ok:
+            return response.content
+
+    def save_image(self, link: tuple, img_type: str) -> str:
+        """ 下载图片
+        """
+        file_name = uuid.uuid4()
+        if not (img_type == 'image' or img_type == 'thumb'):
+            raise Exception("img_type 参数非法")
+        content = self.get_image(link)
+        if content:
             with open(f'output/{self.dir_name}/{img_type}s/{file_name}.jpg', 'wb') as file:
-                file.write(response.content)
+                file.write(content)
             return f'{img_type}s/{file_name}.jpg'
 
     @staticmethod
-    def get_image_thumb_and_url(media_item) -> Tuple[str, str]:
+    def get_image_thumb_and_url(media_item, content_style:int) -> Tuple[Tuple, Tuple]:
         """ 获取图片的缩略图与大图的链接
         """
         thumb = None
         url = None
         # 普通图片
         if media_item.type == "2":
-            thumb = media_item.thumb.text
+            thumb = (media_item.thumb.text, media_item.thumb.enc_idx, media_item.thumb.token)
-            url = media_item.url.text
+            url = (media_item.url.text, media_item.url.enc_idx, media_item.url.token)
         # 微信音乐
         if media_item.type == "5":
-            thumb = media_item.thumb.text
+            thumb = (media_item.thumb.text, "", "")
-            url = media_item.thumb.text
+            url = (media_item.thumb.text, "", "")
+        # 超链接类型
+        if content_style == 3:
+            thumb = (media_item.thumb.text, "", "")
+            url = (media_item.thumb.text, "", "")
 
         return thumb, url
 
-    def get_images(self, msg: MomentMsg, download_pic: int) -> list[Tuple]:
+    def get_images(self, msg: MomentMsg) -> list[Tuple]:
         """ 获取一条朋友圈的全部图像， 返回值是一个元组列表
             [(缩略图路径，原图路径)，(缩略图路径，原图路径)]
         """
@@ -54,14 +71,43 @@ class ImageExporter:
 
         media = msg.timelineObject.ContentObject.mediaList.media
         for media_item in media:
-            thumb, url = self.get_image_thumb_and_url(media_item)
+            thumb, url = self.get_image_thumb_and_url(media_item, msg.timelineObject.ContentObject.contentStyle)
             if thumb and url:
-                if download_pic:
+                thumb_path = None
-                    thumb_path = self.save_image(thumb, 'thumb')
+                image_path = None
-                    image_path = self.save_image(url, 'image')
+                # 主图内容
+                image_content = self.get_image(url)
+                # 如果拿不到主图数据
+                if not image_content:
+                    continue
+                # 如果在腾讯服务器获取到jpg图片
+                if image_content[:2] == b'\xff\xd8':
+                    file_name = uuid.uuid4()
+                    with open(f'output/{self.dir_name}/images/{file_name}.jpg', 'wb') as file:
+                        file.write(image_content)
+                    image_path = f'images/{file_name}.jpg'
+                    # 缩略图内容
+                    thumb_content = self.get_image(thumb)
+                    file_name = uuid.uuid4()
+                    with open(f'output/{self.dir_name}/thumbs/{file_name}.jpg', 'wb') as file:
+                        file.write(thumb_content)
+                    thumb_path = f'thumbs/{file_name}.jpg'
+                # 如果图片已加密，进入缓存图片中匹配
                 else:
-                    thumb_path = thumb
+                    # 获取2024-06格式的时间
-                    image_path = url
+                    month = msg.timelineObject.create_year_month
+                    image_content = self.get_image(url)
+                    thumb_content = self.get_image(thumb)
+                    # 从缓存里找文件
+                    image_file = Path((f"output/{self.dir_name}/images/{month}/"
+                                       f"{len(image_content)}_{len(thumb_content)}.jpg"))
+                    thumb_file = Path((f"output/{self.dir_name}/thumbs/{month}/"
+                                       f"{len(image_content)}_{len(thumb_content)}.jpg"))
+                    if image_file.exists():
+                        image_path = image_file.resolve()
+                    if thumb_file.exists():
+                        thumb_path = thumb_file.resolve()
+
                 if thumb_path and image_path:
                     results.append((thumb_path, image_path))
 
@@ -79,5 +125,5 @@ class ImageExporter:
 
         media = msg.timelineObject.ContentObject.finderFeed.mediaList.media
         for media_item in media:
-            thumb_path = self.save_image(media_item.thumbUrl, 'thumb')
+            thumb_path = self.save_image((media_item.thumbUrl, "", ""), 'thumb')
             return thumb_path

gui/gui.py

@@ -7,6 +7,7 @@ from pathlib import Path
 from typing import Optional
 import tkcalendar
 from decrypter.db_decrypt import DatabaseDecrypter
+from decrypter.image_decrypt import ImageDecrypter
 from decrypter.video_decrypt import VideoDecrypter
 from gui.auto_scroll_guide import AutoScrollGuide
 from gui.auto_scrolls_single_guide import AutoScrollSingleGuide
@@ -35,8 +36,6 @@ class Gui:
         self.confirm_button_text = None
         self.succeed_label_2 = None
         self.succeed_label = None
-        self.download_pic_var = Optional[tkinter.IntVar]
-        self.download_pic = None
         self.auto_scroll_button_text = None
         self.warning_label = None
         self.root = None
@@ -57,6 +56,7 @@ class Gui:
         self.decrypt_note_text = None
         self.account_info = None
         self.video_decrypter = None
+        self.image_decrypter = None
         self.export_dir_name = None
         self.exporting = False
         # 1: 自动滚动数据 2: 解密数据库 3: 导出
@@ -167,6 +167,9 @@ class Gui:
             self.next_step_button.place_forget()
             # 初始化视频导出器
             self.video_decrypter = VideoDecrypter(self, self.account_info.get("filePath"))
+            # 初始化图片导出器
+            self.image_decrypter = ImageDecrypter(self, self.account_info.get("filePath"))
+
 
         self.page_stage = self.page_stage + 1
 
@@ -216,10 +219,6 @@ class Gui:
         self.end_calendar = tkcalendar.DateEntry(master=self.root, locale="zh_CN", maxdate=datetime.now())
         self.end_calendar.place(relx=0.65, rely=0.3)
 
-        self.download_pic_var = tkinter.IntVar(value=0)
-        self.download_pic = tkinter.ttk.Checkbutton(self.root, text='下载图片', variable=self.download_pic_var)
-        self.download_pic.place(relx=0.65, rely=0.4)
-        ToolTip(self.download_pic, "将图片下载到电脑上，网页\n可离线查看，导出速度变慢")
 
         self.convert_video_var = tkinter.IntVar(value=0)
         self.convert_video = tkinter.ttk.Checkbutton(self.root, text='视频转码', variable=self.convert_video_var)
@@ -278,7 +277,7 @@ class Gui:
             # 导出线程
             self.html_exporter_thread = HtmlExporter(self, self.export_dir_name, contact_map,
                                                      self.begin_calendar.get_date(), self.end_calendar.get_date(),
-                                                     self.download_pic_var.get(), self.convert_video_var.get())
+                                                     self.convert_video_var.get())
             self.html_exporter_thread.start()
 
     def update_decrypt_progressbar(self, progress):

test.py

@@ -1,4 +1,6 @@
 import datetime
+
+from decrypter.image_decrypt import ImageDecrypter
 from decrypter.video_decrypt import VideoDecrypter
 import threading
 from time import sleep
@@ -12,8 +14,8 @@ def stage_3():
     gui_thread.start()
     gui.init_export_page()
 
-    gui.begin_calendar.set_date(datetime.date(2024, 3, 6))
+    gui.begin_calendar.set_date(datetime.date(2024, 5, 6))
-    gui.end_calendar.set_date(datetime.date(2024, 3, 6))
+    gui.end_calendar.set_date(datetime.date(2024, 5, 6))
 
     # 后台读取微信信息
     # 请等待完全接入微信再进行UI操作
@@ -31,6 +33,7 @@ def stage_3():
             gui.waiting_label.config(text="微信已登录")
             # 初始化视频导出器
             gui.video_decrypter = VideoDecrypter(gui, gui.account_info.get("filePath"))
+            gui.image_decrypter = ImageDecrypter(gui, gui.account_info.get("filePath"))
             gui.waiting_label.place_forget()
             break