网页编程网

# -*- coding: utf-8 -*-
import requests
from pyquery import PyQuery as pq

headers={
        'User-Agent':'Mozilla/5.0 (Windows NT 6.1; ) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/82.0.4068.4 Safari/537.36'
        }
#获取首页，以求内容页url
def Index():
    url='http://mil.news.sina.com.cn/roll/index.d.html?cid=57918'
    #decode()解码
    response=requests.get(url,headers=headers).content.decode()
    #print(response)
    #数据初始化
    listURL=pq(response)
    #id选择器，
    linkNews=listURL('.linkNews li a').items()
    #print(linkNews)
    for i in linkNews:
        #print(i)
        #通过属性取数据attr()
        #print(i.attr('href'))
        details=i.attr('href')
        Detail(details)

def Detail(urlDetail):
    #urlDetail='https://mil.news.sina.com.cn/china/2020-03-19/doc-iimxyqwa1556368.shtml'
    response=requests.get(urlDetail,headers=headers).content.decode()
    contentAll=pq(response)
    content=contentAll('.article p').text()
    print(content)

if __name__=='__main__':
    Index()