diff --git a/app/DataBase/msg.py b/app/DataBase/msg.py index 70f28cf..56e402f 100644 --- a/app/DataBase/msg.py +++ b/app/DataBase/msg.py @@ -329,14 +329,16 @@ class Msg: result = self.cursor.fetchall() return result - def get_messages_by_keyword(self, username_, keyword, num=5, max_len=10, year_='all'): + def get_messages_by_keyword(self, username_, keyword, num=5, max_len=10,time_range=None, year_='all'): if not self.open_flag: return None + if time_range: + start_time, end_time = convert_to_timestamp(time_range) sql = f''' select localId,TalkerId,Type,SubType,IsSender,CreateTime,Status,StrContent,strftime('%Y-%m-%d %H:%M:%S',CreateTime,'unixepoch','localtime') as StrTime,MsgSvrID,BytesExtra from MSG where StrTalker=? and Type=1 and LENGTH(StrContent)' + str(start_time) + ' AND CreateTime<' + str(end_time) if time_range else ''} order by CreateTime desc ''' temp = [] @@ -489,46 +491,30 @@ class Msg: lock.release() return result - def get_messages_by_hour(self, username_, year_='all'): + def get_messages_by_hour(self, username_, time_range=None,year_='all'): result = [] if not self.open_flag: return result - if year_ == 'all': - sql = ''' - SELECT strftime('%H:00',CreateTime,'unixepoch','localtime') as hours,count(MsgSvrID) - from ( - SELECT MsgSvrID, CreateTime - FROM MSG - where StrTalker = ? - ) - group by hours - ''' - try: - lock.acquire(True) - self.cursor.execute(sql, [username_]) - except sqlite3.DatabaseError: - logger.error(f'{traceback.format_exc()}\n数据库损坏请删除msg文件夹重试') - finally: - lock.release() - result = self.cursor.fetchall() - else: - sql = ''' - SELECT strftime('%H:00',CreateTime,'unixepoch','localtime') as hours,count(MsgSvrID) - from ( - SELECT MsgSvrID, CreateTime - FROM MSG - where StrTalker = ? and strftime('%Y', CreateTime, 'unixepoch', 'localtime') = ? - ) - group by hours - ''' - try: - lock.acquire(True) - self.cursor.execute(sql, [username_, year_]) - except sqlite3.DatabaseError: - logger.error(f'{traceback.format_exc()}\n数据库损坏请删除msg文件夹重试') - finally: - lock.release() - result = self.cursor.fetchall() + if time_range: + start_time, end_time = convert_to_timestamp(time_range) + sql = f''' + SELECT strftime('%H:00',CreateTime,'unixepoch','localtime') as hours,count(MsgSvrID) + from ( + SELECT MsgSvrID, CreateTime + FROM MSG + where StrTalker = ? + {'AND CreateTime>' + str(start_time) + ' AND CreateTime<' + str(end_time) if time_range else ''} + ) + group by hours + ''' + try: + lock.acquire(True) + self.cursor.execute(sql, [username_]) + except sqlite3.DatabaseError: + logger.error(f'{traceback.format_exc()}\n数据库损坏请删除msg文件夹重试') + finally: + lock.release() + result = self.cursor.fetchall() return result def get_first_time_of_message(self, username_=''): @@ -549,9 +535,11 @@ class Msg: lock.release() return result - def get_latest_time_of_message(self, username_='', year_='all'): + def get_latest_time_of_message(self, username_='', time_range=None,year_='all'): if not self.open_flag: return None + if time_range: + start_time, end_time = convert_to_timestamp(time_range) sql = f''' SELECT isSender,StrContent,strftime('%Y-%m-%d %H:%M:%S',CreateTime,'unixepoch','localtime') as StrTime, strftime('%H:%M:%S', CreateTime,'unixepoch','localtime') as hour @@ -559,7 +547,7 @@ class Msg: WHERE Type=1 AND {'StrTalker = ? AND ' if username_ else f"'{username_}'=? AND "} hour BETWEEN '00:00:00' AND '05:00:00' - {"and strftime('%Y', CreateTime, 'unixepoch', 'localtime') = ?" if year_ != "all" else ""} + {'AND CreateTime>' + str(start_time) + ' AND CreateTime<' + str(end_time) if time_range else ''} ORDER BY hour DESC LIMIT 20; ''' diff --git a/app/analysis/analysis.py b/app/analysis/analysis.py index 76707a7..2690776 100644 --- a/app/analysis/analysis.py +++ b/app/analysis/analysis.py @@ -114,9 +114,10 @@ def get_wordcloud(text): } -def wordcloud_christmas(wxid, year='2023'): +def wordcloud_christmas(wxid,time_range=None, year='2023'): import jieba - txt_messages = msg_db.get_messages_by_type(wxid, MsgType.TEXT, year) + + txt_messages = msg_db.get_messages_by_type(wxid, MsgType.TEXT, time_range=time_range) if not txt_messages: return { 'wordcloud_chart_data': None, @@ -127,42 +128,14 @@ def wordcloud_christmas(wxid, year='2023'): } 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 = './app/data/stopwords.txt' - with open(stopwords_file, "r", encoding="utf-8") as stopword_file: - stopwords1 = set(stopword_file.read().splitlines()) - # 构建 FFmpeg 可执行文件的路径 - stopwords = set() - stopwords_file = './app/resources/data/stopwords.txt' - if not os.path.exists(stopwords_file): - resource_dir = getattr(sys, '_MEIPASS', os.path.abspath(os.path.dirname(__file__))) - stopwords_file = os.path.join(resource_dir, 'app', 'resources', 'data', 'stopwords.txt') - with open(stopwords_file, "r", encoding="utf-8") as stopword_file: - stopwords = set(stopword_file.read().splitlines()) - stopwords = stopwords.union(stopwords1) - - 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() - .add(series_name="聊天文字", data_pair=text_data, word_size_range=[5, 40]) - ) + wordcloud_data = get_wordcloud(text) # return w.render_embed() - dialogs = msg_db.get_messages_by_keyword(wxid, keyword, num=3, max_len=12, year_=year) + keyword = wordcloud_data.get('keyword') + max_num = wordcloud_data.get('keyword_max_num') + dialogs = msg_db.get_messages_by_keyword(wxid, keyword, num=3, max_len=12, time_range=time_range) return { - 'wordcloud_chart_data': w.dump_options_with_quotes(), + 'wordcloud_chart_data': wordcloud_data.get('chart_data_wordcloud'), 'keyword': keyword, 'keyword_max_num': str(max_num), 'dialogs': dialogs, diff --git a/app/web_ui/web.py b/app/web_ui/web.py index 5b53f87..ff84f4c 100644 --- a/app/web_ui/web.py +++ b/app/web_ui/web.py @@ -94,8 +94,8 @@ def christmas(wxid): 'my_nickname': Me().name, 'first_time': first_time, } - wordcloud_cloud_data = analysis.wordcloud_christmas(contact.wxid) - msg_data = msg_db.get_messages_by_hour(contact.wxid, year_="2023") + wordcloud_cloud_data = analysis.wordcloud_christmas(contact.wxid,time_range=time_range) + msg_data = msg_db.get_messages_by_hour(contact.wxid, time_range=time_range) msg_data.sort(key=lambda x: x[1], reverse=True) desc = { '夜猫子': {'22:00', '23:00', '00:00', '01:00', '02:00', '03:00', '04:00', '05:00'}, @@ -108,7 +108,7 @@ def christmas(wxid): for key, item in desc.items(): if time_ in item: label = key - latest_dialog = msg_db.get_latest_time_of_message(contact.wxid, year_='2023') + latest_dialog = msg_db.get_latest_time_of_message(contact.wxid, time_range=time_range) latest_time = latest_dialog[0][2] if latest_dialog else '' time_data = { 'latest_time': latest_time,