实现词云图

app/DataBase/__init__.py

@@ -13,9 +13,9 @@ from .micro_msg import MicroMsg
 # from . import output
 from .misc import Misc
 from .msg import Msg
-
+from .msg import MsgType
 misc_db = Misc()
 msg_db = Msg()
 micro_msg_db = MicroMsg()
 hard_link_db = HardLink()
-__all__ = ["data", 'output', 'misc_db', 'micro_msg_db', 'msg_db', 'hard_link_db']
+__all__ = ["data", 'output', 'misc_db', 'micro_msg_db', 'msg_db', 'hard_link_db','MsgType']

app/DataBase/msg.py

@@ -1,4 +1,5 @@
 import os.path
+import random
 import sqlite3
 import threading
 import traceback
@@ -27,7 +28,12 @@ def singleton(cls):
     return inner
 
 
-@singleton
+class MsgType:
+    TEXT = 1
+    IMAGE = 3
+    EMOJI = 47
+
+
 class Msg:
     def __init__(self):
         self.DB = None
@@ -35,8 +41,11 @@ class Msg:
         self.open_flag = False
         self.init_database()
 
-    def init_database(self):
+    def init_database(self, path=None):
+        global db_path
         if not self.open_flag:
+            if path:
+                db_path = path
             if os.path.exists(db_path):
                 self.DB = sqlite3.connect(db_path, check_same_thread=False)
                 # '''创建游标'''
@@ -102,6 +111,67 @@ class Msg:
         # result.sort(key=lambda x: x[5])
         return result
 
+    def get_messages_by_type(self, username_, type_):
+        if not self.open_flag:
+            return None
+        sql = '''
+            select localId,TalkerId,Type,SubType,IsSender,CreateTime,Status,StrContent,strftime('%Y-%m-%d %H:%M:%S',CreateTime,'unixepoch','localtime') as StrTime,MsgSvrID
+            from MSG
+            where StrTalker=? and Type=?
+            order by CreateTime
+        '''
+        try:
+            lock.acquire(True)
+            self.cursor.execute(sql, [username_, type_])
+            result = self.cursor.fetchall()
+        finally:
+            lock.release()
+        return result
+
+    def get_messages_by_keyword(self, username_, keyword, num=5):
+        if not self.open_flag:
+            return None
+        sql = '''
+            select localId,TalkerId,Type,SubType,IsSender,CreateTime,Status,StrContent,strftime('%Y-%m-%d %H:%M:%S',CreateTime,'unixepoch','localtime') as StrTime,MsgSvrID
+            from MSG
+            where StrTalker=? and Type=1 and StrContent like ?
+            order by CreateTime desc
+        '''
+        temp = []
+        try:
+            lock.acquire(True)
+            self.cursor.execute(sql, [username_, f'%{keyword}%'])
+            messages = self.cursor.fetchall()
+        finally:
+            lock.release()
+        if len(messages) > 5:
+            messages = random.sample(messages, num)
+        try:
+            lock.acquire(True)
+            for msg in messages:
+                local_id = msg[0]
+                is_send = msg[4]
+                sql = '''
+                    select localId,TalkerId,Type,SubType,IsSender,CreateTime,Status,StrContent,strftime('%Y-%m-%d %H:%M:%S',CreateTime,'unixepoch','localtime') as StrTime,MsgSvrID
+                    from MSG
+                    where localId > ? and StrTalker=? and Type=1 and IsSender=?
+                    limit 1
+                '''
+                self.cursor.execute(sql, [local_id, username_, 1 - is_send])
+                temp.append((msg, self.cursor.fetchone()))
+        finally:
+            lock.release()
+        res = []
+        for dialog in temp:
+            msg1 = dialog[0]
+            msg2 = dialog[1]
+            res.append((
+                (msg1[4], msg1[5], msg1[7].split(keyword), msg1[8]),
+                (msg2[4], msg2[5], msg2[7], msg2[8])
+            ))
+
+        return res
+
     def close(self):
         if self.open_flag:
             try:
@@ -123,4 +193,7 @@ if __name__ == '__main__':
     print(result)
     print(result[-1][0])
     local_id = result[-1][0]
+    wxid = 'wxid_0o18ef858vnu22'
     pprint(msg.get_message_by_num('wxid_0o18ef858vnu22', local_id))
+    print(msg.get_messages_by_keyword(wxid, '干嘛'))
+    pprint(msg.get_messages_by_keyword(wxid, '干嘛')[0])

app/analysis/__init__.py

@@ -0,0 +1,4 @@
+
+from .analysis import Analysis
+
+__all__=['Analysis']

app/analysis/analysis.py

@@ -0,0 +1,66 @@
+from collections import Counter
+
+from app.DataBase import msg_db, MsgType
+from app.person_pc import ContactPC
+import jieba
+from pyecharts import options as opts
+from pyecharts.charts import Pie, WordCloud, Calendar, Bar, Line, Timeline, Grid
+
+charts_width = 800
+charts_height = 450
+wordcloud_width = 780
+wordcloud_height = 720
+
+
+def wordcloud(wxid):
+    import jieba
+    txt_messages = msg_db.get_messages_by_type(wxid, MsgType.TEXT)
+    text = ''.join(map(lambda x: x[7], txt_messages))
+    total_msg_len = len(text)
+    # 使用jieba进行分词，并加入停用词
+    words = jieba.cut(text)
+    # 统计词频
+    word_count = Counter(words)
+    # 过滤停用词
+    stopwords_file = '../data/stopwords.txt'
+    with open(stopwords_file, "r", encoding="utf-8") as stopword_file:
+        stopwords = set(stopword_file.read().splitlines())
+    filtered_word_count = {word: count for word, count in word_count.items() if len(word) > 1 and word not in stopwords}
+
+    # 转换为词云数据格式
+    data = [(word, count) for word, count in filtered_word_count.items()]
+    # text_data = data
+    data.sort(key=lambda x: x[1], reverse=True)
+
+    text_data = data[:100] if len(data) > 100 else data
+    # 创建词云图
+    keyword, max_num = text_data[0]
+    w = (
+        WordCloud(init_opts=opts.InitOpts(width=f"{wordcloud_width}px", height=f"{wordcloud_height}px"))
+        .add(series_name="聊天文字", data_pair=text_data, word_size_range=[20, 100])
+        .set_global_opts(
+            title_opts=opts.TitleOpts(
+                title=f"词云图", subtitle=f"总计{total_msg_len}字",
+                title_textstyle_opts=opts.TextStyleOpts(font_size=23)
+            ),
+            tooltip_opts=opts.TooltipOpts(is_show=True),
+            legend_opts=opts.LegendOpts(is_show=False)
+        )
+    )
+    # return w.render_embed()
+    return {
+        'chart_data': w.dump_options_with_quotes(),
+        'keyword': keyword,
+        'max_num': str(max_num),
+        'dialogs': msg_db.get_messages_by_keyword(wxid, keyword, num=5)
+    }
+
+
+class Analysis:
+    pass
+
+
+if __name__ == '__main__':
+    msg_db.init_database(path='../DataBase/Msg/MSG.db')
+    w = wordcloud('wxid_0o18ef858vnu22')
+    print(w)

app/data/stopwords.txt

@@ -1,4 +1,17 @@
 wxid
+就
+说
+啥
+好
+干
+哦
+好
+嗯
+恩
+噢
+喔
+行
+拿
 乡村
 炸弹
 腹肌
@@ -2518,3 +2531,11 @@ sup
 他
 她
 它
+听
+哪
+想
+打
+🙄
+奥
+真
+旺柴

app/util/dat2pic.py

@@ -29,9 +29,9 @@ def get_code(file_path):
         code = dat_read[0] ^ pic_head[head_index]
         idf_code = dat_read[1] ^ code
         head_index = head_index + 1
-        # if idf_code == pic_head[head_index]:
-        #     dat_file.close()
-        return head_index, code
+        if idf_code == pic_head[head_index]:
+            dat_file.close()
+            return head_index, code
         head_index = head_index + 1
     dat_file.close()
     print("not jpg, png, gif")
@@ -64,9 +64,8 @@ def decode_dat(file_path, out_path):
     with open(file_path, 'rb') as file_in:
         data = file_in.read()
     # 对数据进行异或加密/解密
-    encrypted_data = bytes([byte ^ decode_code for byte in data])
     with open(file_outpath, 'wb') as file_out:
-        file_out.write(encrypted_data)
+        file_out.write(bytes([byte ^ decode_code for byte in data]))
     print(file_path, '->', file_outpath)
     return file_outpath
 

app/util/emoji.py

@@ -1,3 +1,13 @@
+# -*- coding: utf-8 -*-
+"""
+emoji.py
+
+!!!声明：
+由于表情包并不属于个人，并且其可能具有版权风险，你只有浏览权没有拥有权
+另外访问腾讯API可能会给腾讯服务器造成压力
+所以禁止任何人以任何方式修改或间接修改该文件，违者后果自负
+"""
+
 import os
 import xml.etree.ElementTree as ET
 

app/web_ui/web.py

@@ -1,8 +1,13 @@
+import json
+
 from flask import Flask, render_template
 from pyecharts import options as opts
 from pyecharts.charts import Bar
 from pyecharts.globals import ThemeType
 
+from app.DataBase import msg_db
+from app.analysis import analysis
+
 app = Flask(__name__)
 
 
@@ -25,7 +30,7 @@ def index():
 
 @app.route("/index")
 def index0():
-    return render_template("index.html")
+    return render_template("index1.html")
 
 
 @app.route('/home')
@@ -41,7 +46,26 @@ def home():
 
 @app.route('/message_num')
 def one():
-    return "1hello world"
+    msg_db.init_database(path='../DataBase/Msg/MSG.db')
+    wxid = 'wxid_0o18ef858vnu22'
+    # wxid = 'wxid_8piw6sb4hvfm22'
+    wxid = 'wxid_lltzaezg38so22'
+    world_cloud_data = analysis.wordcloud(wxid)
+    # 创建一个简单的柱状图
+    with open('message_num_test.html','w',encoding='utf-8') as f:
+        f.write(render_template('message_num.html', **world_cloud_data))
+    return render_template('message_num.html', **world_cloud_data)
+
+
+@app.route('/test')
+def test():
+    bar = (
+        Bar(init_opts=opts.InitOpts(theme=ThemeType.LIGHT))
+        .add_xaxis(["A", "B", "C", "D", "E"])
+        .add_yaxis("Series", [5, 20, 36, 10, 75])
+        .set_global_opts(title_opts=opts.TitleOpts(title="Flask and Pyecharts Interaction"))
+    )
+    return bar.dump_options_with_quotes()
 
 
 if __name__ == "__main__":

readme.md

@@ -288,10 +288,11 @@ python main.py
 # 🏆致谢
 
 * PC微信解密工具:[https://github.com/xaoyaoo/PyWxDump](https://github.com/xaoyaoo/PyWxDump)
+* 我的得力助手:[ChatGPT](https://chat.openai.com/)
 
 ---
 
-> 说明：该项目仅可用于交流学习，禁止任何非法用途，创作者不承担任何责任🙄
+> 声明：该项目有且仅有一个目的：留痕——我的数据我做主，前提是“我的数据”其次才是“我做主”，禁止任何人以任何形式将其用于任何非法用途，对于使用该程序所造成的任何后果，创作者不承担任何责任🙄
 
 [![Star History Chart](https://api.star-history.com/svg?repos=LC044/WeChatMsg&type=Date)](https://star-history.com/?utm_source=bestxtools.com#LC044/WeChatMsg&Date)