背景
在Python去写爬虫,网页解析等过程中,比如:
如何用Python,C#等语言去实现抓取静态网页+抓取动态网页+模拟登陆网站
常常需要涉及到HTML等网页的解析。
当然,对于简单的HTML中内容的提取,Python内置的正则表达式Re模块,就足够用了,
但是对于复杂的HTML的处理,尤其是一些非法的,有bug的html代码的处理,那么最好还是用专门的HTML的解析的库。
Python中的,专门用于HTML解析的库,比较好用的,就是BeautifulSoup。
BeautifulSoup简介
Python中,专门用于HTML/XML解析的库;
特点是:
即使是有bug,有问题的html代码,也可以解析。
功能很强大;
BeautifulSoup的主页是:
http://www.crummy.com/software/BeautifulSoup/
BeautifulSoup的版本
BeautifulSoup主要有两个版本:
BeautifulSoup 3
之前的,比较早的,是3.x的版本。
BeautifulSoup 3的在线文档
最新的,可用的,在线文档是:
http://www.crummy.com/software/BeautifulSoup/bs3/documentation.html
中文版的是:
http://www.crummy.com/software/BeautifulSoup/bs3/documentation.zh.html
下载BeautifulSoup 3
http://www.crummy.com/software/BeautifulSoup/bs3/download//3.x/
中可以下载到很多版本,比如我常用的3.0.6的版本:
http://www.crummy.com/software/BeautifulSoup/bs3/download//3.x/BeautifulSoup-3.0.6.py
BeautifulSoup 4:缩写为bs4
最新的v4版本的BeautifulSoup,改名为bs4了。
注意:
使用bs4时,导入BeautifulSoup的写法是:
from bs4 import BeautifulSoup;
然后就可以像之前3.x中一样,直接使用BeautifulSoup了。
详见:
【已解决】Python3中,已经安装了bs4(Beautifulsoup 4)了,但是却还是出错:ImportError: No module named BeautifulSoup
bs4的在线文档
http://www.crummy.com/software/BeautifulSoup/bs4/doc/
下载bs4
http://www.crummy.com/software/BeautifulSoup/bs4/download/
可以下载到对应的bs4的版本,比如:
此时最新的版本是:
http://www.crummy.com/software/BeautifulSoup/bs4/download/beautifulsoup4-4.1.3.tar.gz
BeautifulSoup的用法
如何安装BeautifulSoup
3.0.6之前:无需安装,放到和Python文件同目录下即可使用
3.0.6之前,都是不需要安装的,所以使用起来最简单,直接下载对应的版本,比如:
http://www.crummy.com/software/BeautifulSoup/bs3/download//3.x/BeautifulSoup-3.0.6.py
得到了BeautifulSoup-3.0.6.py,然后改名为:BeautifulSoup.py
然后,放到和你当前的python文件同目录下,比如我当前python文件是:
D:\tmp\tmp_dev_root\python\beautifulsoup_demo\beautifulsoup_demo.py
那就放到
D:\tmp\tmp_dev_root\python\beautifulsoup_demo\
下面,和beautifulsoup_demo.py同目录。
3.0.6之后:需要安装BeautifulSoup后才可使用
关于如何安装一个Python的第三方模块,简单说就是,进入对应目录,运行:
setup.py install
详细解释可参考:
如何使用BeautifulSoup
在你的Python文件,此处为beautifulsoup_demo.py,中直接import即可。
关于示例html代码,比如使用:
相关参考文档:
3.x版本的:
find(name, attrs, recursive, text, **kwargs)
使用BeautifulSoup提取html中的某个内容
关于最简单的,最基本的用法,提取html中的某个内容,具体用法,就死使用对应的find函数。
完整代码是:
#!/usr/bin/python # -*- coding: utf-8 -*- """ Function: 【教程】Python中第三方的用于解析HTML的库:BeautifulSoup https://www.crifan.com/python_third_party_lib_html_parser_beautifulsoup Author: Crifan Li Version: 2012-12-26 Contact: admin at crifan dot com """ from BeautifulSoup import BeautifulSoup; def beautifulsoupDemo(): demoHtml = """ <html> <body> <div class="icon_col"> <h1 class="h1user">crifan</h1> </div> </body> </html> """; soup = BeautifulSoup(demoHtml); print "type(soup)=",type(soup); #type(soup)= <type 'instance'> print "soup=",soup; # 1. extract content # method 1: no designate para name #h1userSoup = soup.find("h1", {"class":"h1user"}); # method 2: use para name h1userSoup = soup.find(name="h1", attrs={"class":"h1user"}); # more can found at: #http://www.crummy.com/software/BeautifulSoup/bs3/documentation.zh.html#find%28name,%20attrs,%20recursive,%20text,%20**kwargs%29 print "h1userSoup=",h1userSoup; #h1userSoup= <h1 class="h1user">crifan</h1> h1userUnicodeStr = h1userSoup.string; print "h1userUnicodeStr=",h1userUnicodeStr; #h1userUnicodeStr= crifan if __name__ == "__main__": beautifulsoupDemo();
输出为:
D:\tmp\tmp_dev_root\python\beautifulsoup_demo>beautifulsoup_demo.py type(soup)= <type 'instance'> soup= <html> <body> <div class="icon_col"> <h1 class="h1user">crifan</h1> </div> </body> </html> h1userSoup= <h1 class="h1user">crifan</h1> h1userUnicodeStr= crifan
使用BeautifulSoup修改/改变/替换原先html中的某个内容
如果需要改变原先html中的某个值,可以参考官网解释:
后来证实,只能改(Tag的)中的属性的值,不能改(Tag的)的值本身
完整示例代码为:
#!/usr/bin/python # -*- coding: utf-8 -*- """ Function: 【教程】Python中第三方的用于解析HTML的库:BeautifulSoup https://www.crifan.com/python_third_party_lib_html_parser_beautifulsoup Author: Crifan Li Version: 2013-02-01 Contact: admin at crifan dot com """ from BeautifulSoup import BeautifulSoup; def beautifulsoupDemo(): demoHtml = """ <html> <body> <div class="icon_col"> <h1 class="h1user">crifan</h1> </div> </body> </html> """; soup = BeautifulSoup(demoHtml); print "type(soup)=",type(soup); #type(soup)= <type 'instance'> print "soup=",soup; print '{0:=^80}'.format(" 1. extract content "); # method 1: no designate para name #h1userSoup = soup.find("h1", {"class":"h1user"}); # method 2: use para name h1userSoup = soup.find(name="h1", attrs={"class":"h1user"}); # more can found at: #http://www.crummy.com/software/BeautifulSoup/bs3/documentation.zh.html#find%28name,%20attrs,%20recursive,%20text,%20**kwargs%29 print "h1userSoup=",h1userSoup; #h1userSoup= <h1 class="h1user">crifan</h1> h1userUnicodeStr = h1userSoup.string; print "h1userUnicodeStr=",h1userUnicodeStr; #h1userUnicodeStr= crifan print '{0:=^80}'.format(" 2. demo change tag value and property "); print '{0:-^80}'.format(" 2.1 can NOT change tag value "); print "old tag value=",soup.body.div.h1.string; #old tag value= crifan changedToString = u"CrifanLi"; soup.body.div.h1.string = changedToString; print "changed tag value=",soup.body.div.h1.string; #changed tag value= CrifanLi print "After changed tag value, new h1=",soup.body.div.h1; #After changed tag value, new h1= <h1 class="h1user">crifan</h1> print '{0:-^80}'.format(" 2.2 can change tag property "); soup.body.div.h1['class'] = "newH1User"; print "changed tag property value=",soup.body.div.h1; #changed tag property value= <h1 class="newH1User">crifan</h1> if __name__ == "__main__": beautifulsoupDemo();
总结
更多的,用法和使用心得,部分内容,已整理到:
【总结】Python的第三方库BeautifulSoup的使用心得
【整理】关于Python中的html处理库函数BeautifulSoup使用注意事项
有空再统一整理到: