import requestsimport jsonimport osimport timeimport randomimport jiebafrom wordcloud import WordCloudfrom imageio import imread
comment_file_path = 'jd_comments.txt'
def get_spider_comments(page = 0):
    #爬取某東評論    url = 'https://sclub.jd.com/comment/productPageComments.action?callback=fetchJSON_comment98vv7990&productId=1070129528&score=0&sortType=5&page=%s&pageSize=10&isShadowSku=0&rid=0&fold=1'%page    headers = {
        'user-agent':'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36',        'referer':'https://item.jd.com/1070129528.html'    }    try:        response = requests.get(url, headers = headers)    except:        print("something wrong!")    #獲取json格式數(shù)據(jù)集    comments_json = response.text[26:-2]    #將json數(shù)據(jù)集轉為json對象    comments_json_obj = json.loads(comments_json)    #獲取comments里面的所有內容    comments_all = comments_json_obj['comments']    #獲取comments中評論content的內容    for comment in comments_all:        with open(comment_file_path,'a+' ,encoding='utf-8') as fin:            fin.write(comment['content']+'\n')        print(comment['content'])
def batch_spider_comments():    # 每次寫入數(shù)據(jù)之前先清空文件    if os.path.exists(comment_file_path):        os.remove(comment_file_path)    for i in range(100):        print('正在爬取'+str(i+1)+'頁數(shù)據(jù)。。。。')        get_spider_comments(i)        time.sleep(random.random()*5)
def cut_word():    with open(comment_file_path,encoding='utf-8')as file:        comment_text = file.read()        wordlist = jieba.lcut_for_search(comment_text)        new_wordlist = ' '.join(wordlist)        return new_wordlist
def create_word_cloud():    mask = imread('ball.jpg')    wordcloud = WordCloud(font_path='msyh.ttc',mask = mask).generate(cut_word())    wordcloud.to_file('picture.png')
if __name__ == '__main__':        create_word_cloud()