import csv
import re

import numpy as np
import pandas as pd
from wordcloud import WordCloud
import matplotlib.pyplot as plt

def purification(self,new_self):
    with open(self,"r",newline='',encoding='utf-8') as self:
        reader = csv.reader(self)
        data_list = []
        for row in reader:
            new_row_data = str([row[13]])
            pattern = re.compile("[\u4e00-\u9fa5]+")
            new_row_data = pattern.findall(new_row_data)
            data_list.append(new_row_data)

    with open(new_self,'w',newline='',encoding='utf-8') as  new_self:
        writer = csv.writer(new_self)
        for row in data_list:
            writer.writerow(row)

def wordcloud():
    # 读取 csv 文件
    df = pd.read_csv('analyse.csv',sep = '$')
    # 提取第四列数据并去重
    column_data = df.iloc[:,0]  # 通过 iloc 方法选取第一列数据
    unique_data = column_data.drop_duplicates()
    # 将去重后的数据转换为字符串类型
    text = ' '.join(unique_data.astype(str).tolist())
    # 生成词云
    my_stopwords = []
    wordcloud = WordCloud(font_path='simhei.ttf',width=1000, height=600, background_color='white', stopwords=my_stopwords).generate(text)
    # 显示词云
    plt.figure(figsize=(10, 6))
    plt.imshow(wordcloud, interpolation='bilinear')
    plt.axis("off")
    plt.show()

self = 'congtent.csv'
new_self = 'analyse.csv'
purification(self,new_self)
wordcloud()