更新依赖项，将commentjson替换为pyjson5；优化README文档，修正项目文档链接；添加聊天平台适配表格以支持不同平台的消息类型。

README.md

@@ -21,7 +21,7 @@
   <a href="https://github.com/xming521/WeClone/blob/master/README_zh.md" target="_blank">简体中文</a>｜
   English</a>｜
   <a href="https://www.weclone.love/" target="_blank"> Project Homepage </a> ｜
-  <a href="https://www.weclone.love/what-is-weclone.html" target="_blank"> Documentation </a> 
+  <a href="https://docs.weclone.love/what-is-weclone.html" target="_blank"> Documentation </a> 
 </p>
 
 > [!IMPORTANT]
@@ -37,6 +37,15 @@
 > [!IMPORTANT]
 > ### WeClone is currently not partnered with any platform and has not issued any cryptocurrency. The only official website is: [weclone.love](https://www.weclone.love). Beware of imitations.
 
+### Chat Platform Support
+
+| Platform | Text | Images | Voice | Video | Animated Emojis | Links (Sharing) | Quote | Forward | Location | Files |
+|----------|------|--------|-------|-------|-----------------|-----------------|-------|---------|----------|-------|
+| WeChat | ✅ | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ |
+| Telegram | ✅ | ✅ | ❌ | ❌ | ⚠️Convert to Emoji | ❌ | ❌ | ✅ | ✅ | ❌ |
+
+ 
+
 > [!IMPORTANT]
 > - WeClone is still in rapid iteration phase, current performance does not represent final results.  
 > - LLM fine-tuning effectiveness largely depends on model size, quantity and quality of chat data. Theoretically, larger models with more data yield better results.

README_zh.md

@@ -21,7 +21,7 @@
 
 <p align="center">
   <a href="https://www.weclone.love/" target="_blank"> 项目主页 </a> ｜
-  <a href="https://www.weclone.love/what-is-weclone.html" target="_blank"> 项目文档 </a> ｜
+  <a href="https://docs.weclone.love/what-is-weclone.html" target="_blank"> 项目文档 </a> ｜
   <a href="https://blog.051088.xyz/2025/05/14/WeClone-%E7%94%A8%E5%BE%AE%E4%BF%A1%E8%81%8A%E5%A4%A9%E8%AE%B0%E5%BD%95%E6%89%93%E9%80%A0%E8%87%AA%E5%B7%B1%E7%9A%84AI%E6%95%B0%E5%AD%97%E5%88%86%E8%BA%AB/" target="_blank">Windows部署指南</a> ｜
   <a href="https://blog.051088.xyz/posts/weclone-linux-tutorial/" target="_blank"> Linux部署指南【保姆级】</a>
 </p>
@@ -39,6 +39,13 @@
 > [!IMPORTANT]
 > ### WeClone 目前未与任何平台合作，未发行任何数字货币。唯一官方网站：[weclone.love](https://www.weclone.love)，谨防仿冒。
 
+### 聊天平台适配
+
+| 平台 | 文字 | 图片 | 语音 | 视频 | 动画表情 | 链接(分享) | 引用 | 转发 | 位置 | 文件 |
+|------|------|------|------|------|----------|-----------|------|------|------|------|
+| 微信 | ✅ | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ |
+| Telegram | ✅ | ✅ | ❌ | ❌ | ⚠️转为Emjoy | ❌ | ❌ | ✅ | ✅ | ❌ |
+
 > [!IMPORTANT]
 > - WeClone仍在快速迭代期，当前效果不代表最终效果。  
 > - 微调LLM效果很大程度取决于模型大小、聊天数据的数量和质量，理论上模型越大，数据越多，效果越好。

pyproject.toml

@@ -8,7 +8,7 @@ requires-python = ">=3.10,<3.11"
 
 dependencies = [
   "pandas",
-  "commentjson",
+  "pyjson5",
   "omegaconf",
   "click",
   "tqdm",
@@ -40,6 +40,8 @@ main = [
   "transformers==4.52.1",
   "accelerate==1.7.0",
   "triton==3.3.0; platform_system == 'Linux'",
+  "presidio_analyzer[transformers]",
+  "presidio_anonymizer",
 ]
 sparktts = [
   "einops>=0.8.1",

weclone/cli.py

@@ -5,7 +5,7 @@ from pathlib import Path
 from typing import cast
 
 import click
-import commentjson
+import pyjson5
 
 from weclone.utils.config import load_config
 from weclone.utils.config_models import CliArgs
@@ -168,7 +168,7 @@ def _check_versions():
     if SETTINGS_PATH.exists():
         try:
             with open(SETTINGS_PATH, "r", encoding="utf-8") as f:
-                settings_data = commentjson.load(f)
+                settings_data = pyjson5.load(f)
                 settings_version = settings_data.get("version")
         except Exception as e:
             logger.error(f"错误：无法读取或解析 {SETTINGS_PATH}: {e}")

weclone/data/models.py

@@ -12,16 +12,16 @@ from weclone.utils.i18n import MultiLangList
 @dataclass
 class ChatMessage:
     id: int  # 顺序id
-    MsgSvrID: int
-    type_name: str
+    MsgSvrID: str  # 消息平台原始id
+    type_name: str  # 消息类型 参考cut_type_data和skip_type_data
     is_sender: int  # 0: 对方 1: 自己
-    talker: str
-    msg: str
-    src: str
-    CreateTime: Timestamp
-    room_name: Optional[str] = None
-    is_forward: bool = False
-    modality: Optional[DataModality] = None
+    talker: str  # 消息发送者
+    msg: str  # 消息内容
+    src: str  # 媒体文件路径、额外信息字段
+    CreateTime: Timestamp  # 消息发送时间
+    room_name: Optional[str] = None  # 聊天室名称
+    is_forward: bool = False  # 是否是转发消息
+    modality: Optional[DataModality] = None  # 消息模态  set in qa_generator.py
 
 
 @dataclass

weclone/utils/config.py

@@ -2,7 +2,7 @@ import os
 import sys
 from typing import Any, Dict, cast
 
-import commentjson
+import pyjson5
 from omegaconf import OmegaConf
 from pydantic import BaseModel
 
@@ -23,7 +23,7 @@ def load_base_config() -> WcConfig:
 
     try:
         with open(config_path, "r", encoding="utf-8") as f:
-            s_config_dict: Dict[str, Any] = commentjson.load(f)
+            s_config_dict: Dict[str, Any] = pyjson5.load(f)
     except FileNotFoundError:
         logger.error(f"Configuration file not found: {config_path}")
         sys.exit(1)