import requests
import json
def get_comment():
  for i in range(1,7000):
    # 打印頁碼
    print(i)
    # headers頭部
    headers = {'User-Agent': 'Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:59.0) Gecko/20100101 Firefox/59.0',
 'Referer': "https://y.qq.com/n/yqq/song/0031TAKo0095np.html"}
    # 請求的url
    url = 'https://c.y.qq.com/base/fcgi-bin/fcg_global_comment_h5.fcg?g_tk=2058499274&jsonpCallback=jsoncallback06927647062927766&loginUin=2230661779&hostUin=0&format=jsonp&inCharset=utf8&outCharset=GB2312&notice=0&platform=yqq&needNewCode=0&cid=205360772&reqtype=2&biztype=1&topid=212877900&cmd=8&needmusiccrit=0&pagenum=%s&pagesize=25&lasthotcommentid=song_212877900_3035803620_1526783365&callback=jsoncallback06927647062927766&domain=qq.com&ct=24&cv=101010' %i
    # 打印當(dāng)前訪問的url地址
    print (url)
    # 將請求得到的頁面賦值為req
    req = requests.get(url,headers=headers,verify=False)
    # 對獲取到的內(nèi)容進(jìn)行utf-8編碼
    html = str(req.content,'UTF-8')
    # 對非正規(guī)的json進(jìn)行處理,去掉頭部跟尾部多余的部分
    html= html.strip("jsoncallback06927647062927766(")
    html = html.replace(")","")
    # 去掉兩邊的空格
    html = html.strip()
    # 將處理后的json轉(zhuǎn)為python的json
    data = json.loads(html)
    # 獲取json中評論的部分
    list = data['comment']['commentlist']
    # 每次都重新定義一個列表來存儲每一頁的評論
    content = []
    # 遍歷當(dāng)前頁的評論并通過調(diào)用write()函數(shù)來保存
    for i in list:
      # 偶爾也會有一頁的評論獲取不到，這時候如果報錯了可以直接忽略那一頁，繼續(xù)運(yùn)行
      try:
        content.append(i['rootcommentcontent'].replace("[em]","").replace("[/em]","").replace("e400",""))
      except KeyError:
        content = []
        break
    write(content)
# 將當(dāng)前頁面的評論傳遞過來
def write(content):
  # 打開一個文件,將列表的內(nèi)容一行一行的存儲下來
  with open('comments.txt', 'a', encoding = 'UTF-8') as f:
    for i in range(len(content)):
      # 因?yàn)檗D(zhuǎn)為json后\n不胡自動換行，所以我們這里將\n給手換行
      string = content[i].split("\\n")
      for i in string:
        # 因?yàn)槌霈F(xiàn)了很多評論被刪除的情況,所有我們把這句給過濾掉
        i = i.replace("該評論已經(jīng)被刪除", "")
        # 打印每條評論
        print (i)
        # 將評論寫入文本
        f.writelines(i)
        # 給評論換行
        f.write("\n")
if __name__ == "__main__":
  get_comment()
          

            
# -*- coding: utf-8 -*-
import jieba
from wordcloud import WordCloud, STOPWORDS
from os import path
from scipy.misc import imread
# 讀取mask/color圖片
d = path.dirname(__file__)
color_mask = imread("cyx.png")
#將爬到的評論放在string中
with open('nbzd.txt', 'r', encoding = 'UTF-8') as f:
  string = f.read()
  word = " ".join(jieba.cut(string))
  wordcloud = WordCloud(background_color='white',
             mask=color_mask,
             max_words=100,
             stopwords=STOPWORDS,
             font_path='/home/azhao/桌面/素材/simsun.ttc',
             max_font_size=100,
             random_state=30,
             margin=2).generate_from_text(word)
import matplotlib.pyplot as plt
plt.imshow(wordcloud, interpolation='bilinear')
plt.axis("off")
plt.show()