# coding:utf-8 |
import requests |
import codecs |
import json |
# 爬取京东商品信息 |
def getCommentCounts(productId): |
url = 'https://sclub.jd.com/comment/productPageComments.action?callback=fetchJSON_comment98vv33&productId=' + productId + '&score=0&sortType=5&page=0&pageSize=10&isShadowSku=0&fold=1' |
# 模拟浏览器 |
header = { "User-Agent" : "Mozilla/5.0 (Windows NT 10.0; WOW64; Trident/7.0; rv:11.0) like Gecko" } |
# 请求 |
r = requests.get(url, headers = header) |
# 将json数据按照特殊的编码格式decode() |
html = r.content.decode() |
# 去除收尾 |
index = html.find( '{' ) |
html = html[index: - 2 ] |
# 将json字符串转为字典 |
data = json.loads(html) |
# 获取评论总数 |
commentCounts = data[ 'productCommentSummary' ][ 'commentCount' ] |
print ( '评论数:' , commentCounts) |
if __name__ = = '__main__' : |
productId = '7774757' |
getCommentCounts(productId) |