Python爬虫JD杜蕾斯源码

xpanxcom

于 2020-05-29 11:10:47 发布

阅读量563

点赞数

分类专栏： Python 文章标签： python 数据挖掘 selenium

本文链接：https://blog.csdn.net/HowardWood/article/details/106421229

版权

Python 专栏收录该内容

7 篇文章 1 订阅

订阅专栏

import requests
import json
import os
import time
import random
import jieba
from wordcloud import WordCloud
from imageio import imread

comments_file_path = 'jd_comments.txt'
def get_jd_comments(page = 0):
        #获取jd评论
        url ='https://club.jd.com/comment/productPageComments.action?callback=fetchJSON_comment98&productId=1340204&score=0&sortType=5&page=%s&pageSize=10&isShadowSku=0&fold=1'%page
        headers = {
            #从哪个页面发出的数据申请，每个网站都是不一样的
            'referer': 'https://item.jd.com/1340204.html',
            #'user-agent'指的是用户代理，也就是让网站知道你用的哪个浏览器登录的
            'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.149 Safari/537.36',
            #哪一类用户想要看数据，是游客还是会员,建议使用登录后的
            'cookie': '__jdu=1766075400; areaId=27; PCSYCityID=CN_610000_610100_610113; shshshfpa=a9dc241f-78b8-f3e1-edab-09485009987f-1585747224; shshshfpb=dwWV9IhxtSce3DU0STB1%20TQ%3D%3D; jwotest_product=99; unpl=V2_ZzNtbRAAFhJ3DUJTfhFcUGIAE1RKU0ZCdQoWU3kQXgcwBxJdclRCFnQUR1FnGF8UZAMZWEpcRhFFCEdkeBBVAWMDE1VGZxBFLV0CFSNGF1wjU00zQwBBQHcJFF0uSgwDYgcaDhFTQEJ2XBVQL0oMDDdRFAhyZ0AVRQhHZHsfWwJmBRZYQ1ZzJXI4dmR9EFoAYjMTbUNnAUEpDURSeRhbSGcFFVpDUUcQdAl2VUsa; __jdv=76161171|baidu-pinzhuan|t_288551095_baidupinzhuan|cpc|0f3d30c8dba7459bb52f2eb5eba8ac7d_0_cfd63456491d4208954f13a63833f511|1585835385193; __jda=122270672.1766075400.1585747219.1585829967.1585835353.3; __jdc=122270672; 3AB9D23F7A4B3C9B=AXAFRBHRKYDEJAQ4SPJBVU4J4TI6OQHDFRDGI7ISQFUQGA6OZOQN52T3QYSRWPSIHTFRYRN2QEG7AMEV2JG6NT2DFM; shshshfp=03ed62977bfa44b85be24ef65fbd9b87; ipLoc-djd=27-2376-4343-53952; JSESSIONID=51895EFB4EBD95BA3B3ADAC8C6C73CD8.s1; shshshsID=d2435956e0c158fa7db1980c3053033d_15_1585836826172; __jdb=122270672.16.1766075400|3.1585835353'
        }
        try:
            response = requests.get(url, headers = headers)
        except:
            print('something wrong!')

        #获取json格式数据集
        comments_json = response.text[20:-2]
        #将获取到的json数据集转换为json对象
        comments_json_obj = json.loads(comments_json)
        #获取comments里面全部的内容
        comments_all = comments_json_obj['comments']
        for comment in comments_all:
            with open(comments_file_path, 'a+', encoding = 'utf-8') as fin:
                fin.write(comment['content'] + '\n')
            print(comment['content'])

def batch_jd_comments():
    #每次写数据之前先清空
    if os.path.exists(comments_file_path):
        os.remove(comments_file_path)
    #我们指定page i的值时，它就可以获取固定页面的评论。
    for i in range(30):
        print('正在爬取'+str(i+1)+'页的数据....')
        get_jd_comments(i)
        #设置time用来模拟用户浏览，防止因为爬取太频繁导致ip被封。
        time.sleep(random.random()*5)

#对获取到的数据进行分词
def cut_comments():
    with open(comments_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

#引入图片byt.jpg来制作相同形状的词云图
def create_word_cloud():
    mask = imread('byt.jpg')
    wordcloud = WordCloud(font_path='msyh.ttc',mask = mask).generate(cut_comments())
    wordcloud.to_file('picture.png')

if __name__ == '__main__':
    create_word_cloud()

大家记得动手尝试啊！

https://www.xpanx.com/a/250.html