【脚本语言系列】关于PythonWeb服务自动化BeautifulSoup,你需要知道的事

如何使用BeautifulSoup

# -*- coding:utf-8 -*-
import requests
import sys
from bs4 import BeautifulSoup as soup
def get_links(url):
    result = requests.get(url)
    page = result.text
    doc = soup(page)
    links = [element.get("href") for element in doc.find_all("a")]
    return links

if __name__ == "__main__":
    url = "http://www.baidu.com"
    for num, link in enumerate(get_links(url), start=1):
            print num, link
    print

1 http://news.baidu.com
2 http://www.hao123.com
3 http://map.baidu.com
4 http://v.baidu.com
5 http://tieba.baidu.com
6 http://www.baidu.com/bdorz/login.gif?login&tpl=mn&u=http%3A%2F%2Fwww.baidu.com%2f%3fbdorz_come%3d1
7 //www.baidu.com/more/
8 http://home.baidu.com
9 http://ir.baidu.com
10 http://www.baidu.com/duty/
11 http://jianyi.baidu.com/