Python爬虫之Beautiful Soup解析库的使用
Beautiful Soup-介绍
Python第三方库，用于从HTML或XML中提取数据
官方：http://www.crummv.com/software/BeautifulSoup/
安装：pip install beautifulsoup4
Beautiful Soup-语法
soup = BeautifulSoup(html_doc,'html.parser‘,from_encoding='utf-8' )
第一个参数：html文档字符串
第二个参数：html解析器
第三个参数：html文档的编码
Beautiful Soup-使用
标签选择器操作
注意：只会返回一个指定的标签，这也是标签选择器的特性
选择元素
from bs4 import BeautifulSoup html_doc=''' <div class="container"> <a href="/pc/home?sign=360_79aabe15" class="logo"></a> <nav id="nnav" data-mod="nnav"> <div class="nnav-wrap"> <ul class="nnav-items" id="nnav_main"> <li data-index="0"><a class="nnav-item" href="/pc/home?ch=youlike&sign=360_79aabe15" target="_blank" data-ch="youlike">推荐<span></span></a></li><li data-index="1"><a class="nnav-item" href="/pc/home?ch=good_safe2toera&sign=360_79aabe15" target="_blank" data-ch="good_safe2toera">新时代<span></span></a></li><li data-index="2"><a class="nnav-item" href="/pc/home?ch=fun&sign=360_79aabe15" target="_blank" data-ch="fun">娱乐<span></span></a></li><li data-index="3"><a class="nnav-item" href="/pc/home? data-index="4"><a class="nnav-item" href="/pc/home?ch=economy&sign=360_79aabe15" target="_blank" data-ch="economy">财经<span></span></a></li> ''' soup = BeautifulSoup(html_doc,'lxml')
#将html代码自动补全，并按html代码格式返回 print(soup.prettify())
#输出第一个a标签 print(soup.a)
#输出第一个span标签 print(soup.span)
运行结果如下：
<html>
<body>
<div class="container">
<a class="logo" href="/pc/home?sign=360_79aabe15">
</a>
<nav data-mod="nnav" id="nnav">
<div class="nnav-wrap">
<ul class="nnav-items" id="nnav_main">
<li data-index="0">
<a class="nnav-item" data-ch="youlike" href="/pc/home?ch=youlike&sign=360_79aabe15" target="_blank">
推荐
<span>
</span>
</a>
</li>
<li data-index="1">
<a class="nnav-item" data-ch="good_safe2toera" href="/pc/home?ch=good_safe2toera&sign=360_79aabe15" target="_blank">
新时代
<span>
</span>
</a>
</li>
<li data-index="2">
<a class="nnav-item" data-ch="fun" href="/pc/home?ch=fun&sign=360_79aabe15" target="_blank">
娱乐
<span>
</span>
</a>
</li>
<li data-index="3">
<a class="nnav-item" href="/pc/home?
data-index=">
</a>
<a class="nnav-item" data-ch="economy" href="/pc/home?ch=economy&sign=360_79aabe15" target="_blank">
财经
<span>
</span>
</a>
</li>
</ul>
</div>
</nav>
</div>
</body>
</html>
<a class="logo" href="/pc/home?sign=360_79aabe15"></a>
<span></span>
获取名称
获取属性
获取内容
from bs4 import BeautifulSoup html_doc=''' <div class="container"> <a href="/pc/home?sign=360_79aabe15" class="logo"></a> <nav id="nnav" data-mod="nnav"> <div class="nnav-wrap"> <ul class="nnav-items" id="nnav_main"> <li data-index="0"><a class="nnav-item" href="/pc/home?ch=youlike&sign=360_79aabe15" target="_blank" data-ch="youlike">推荐<span></span></a></li><li data-index="1"><a class="nnav-item" href="/pc/home?ch=good_safe2toera&sign=360_79aabe15" target="_blank" data-ch="good_safe2toera">新时代<span></span></a></li><li data-index="2"><a class="nnav-item" href="/pc/home?ch=fun&sign=360_79aabe15" target="_blank" data-ch="fun">娱乐<span></span></a></li><li data-index="3"><a class="nnav-item" href="/pc/home? data-index="4"><a class="nnav-item" href="/pc/home?ch=economy&sign=360_79aabe15" target="_blank" data-ch="economy">财经<span></span></a></li> ''' soup = BeautifulSoup(html_doc,'lxml') #输出第一个a标签的name print(soup.a.name) #输出第一个a标签的的class属性值，下面两种方法都可以 print(soup.a.attrs['class']) print(soup.a['class']) #输出第一个a标签的内容 print(soup.a.string)
运行结果如下：
a ['logo'] ['logo'] None
嵌套选择
from bs4 import BeautifulSoup html_doc=''' <a class="nnav-item" href="/pc/home?ch=youlike&sign=360_79aabe15" target="_blank" data-ch="youlike"><span>推荐</span></a> ''' soup = BeautifulSoup(html_doc,'lxml') print(soup.a.span.string)
运行结果如下：
推荐
子节点和子孙节点操作
获取所有的子节点
from bs4 import BeautifulSoup
html='''
<div class="bc">
<span class="fl" style="padding-top: 1px;"><a href="http://www.koolearn.com/" target="_blank" title="新东方在线网络课堂"><img src="http://images.koolearn.com/fe_upload/2015_9_2_1441179226504.jpg" width="105" height="48" alt="新东方在线网络课堂"></a></span>
<span class="fl" style="padding-top: 6px;">
<a href="http://cet4.koolearn.com/" target="_blank" rel="nofollow" class="ky">四级</a>
<a title="新东方在线网络课堂" href="http://www.koolearn.com/" target="_self">新东方在线</a> >
<a title="四级网络课堂" href="http://cet4.koolearn.com/" target="_self">四级</a> >
<a href="http://cet4.koolearn.com/cihui/" title="英语四级词汇">英语四级词汇</a> > 正文
</span>
<a href="http://www.xdf.cn/" target="_blank" rel="nofollow" class="fr logo_p2"><img src="http://images.koolearn.com/fe_upload/2015_9_2_1441179317774.jpg" width="208" height="24"></a>
</div>
'''
soup = BeautifulSoup(html,'lxml')
#第一种方法
print(soup.div.contents)
#第二种方法
print(soup.div.children)
for i,child in enumerate(soup.div.children):
print(i,child)
运行结果如下：
['
', <span class="fl" style="padding-top: 1px;"><a href="http://www.koolearn.com/" target="_blank" title="新东方在线网络课堂"><img alt="新东方在线网络课堂" height="48" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179226504.jpg" width="105"/></a></span>, '
', <span class="fl" style="padding-top: 6px;">
<a class="ky" href="http://cet4.koolearn.com/" rel="nofollow" target="_blank">四级</a>
<a href="http://www.koolearn.com/" target="_self" title="新东方在线网络课堂">新东方在线</a> >
<a href="http://cet4.koolearn.com/" target="_self" title="四级网络课堂">四级</a> >
<a href="http://cet4.koolearn.com/cihui/" title="英语四级词汇">英语四级词汇</a> > 正文
</span>, '
', <a class="fr logo_p2" href="http://www.xdf.cn/" rel="nofollow" target="_blank"><img height="24" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179317774.jpg" width="208"/></a>, '
']
<list_iterator object at 0x0000000002E498D0>
0
1 <span class="fl" style="padding-top: 1px;"><a href="http://www.koolearn.com/" target="_blank" title="新东方在线网络课堂"><img alt="新东方在线网络课堂" height="48" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179226504.jpg" width="105"/></a></span>
2
3 <span class="fl" style="padding-top: 6px;">
<a class="ky" href="http://cet4.koolearn.com/" rel="nofollow" target="_blank">四级</a>
<a href="http://www.koolearn.com/" target="_self" title="新东方在线网络课堂">新东方在线</a> >
<a href="http://cet4.koolearn.com/" target="_self" title="四级网络课堂">四级</a> >
<a href="http://cet4.koolearn.com/cihui/" title="英语四级词汇">英语四级词汇</a> > 正文
</span>
4
5 <a class="fr logo_p2" href="http://www.xdf.cn/" rel="nofollow" target="_blank"><img height="24" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179317774.jpg" width="208"/></a>
6
获取所有的子孙节点
from bs4 import BeautifulSoup
html='''
<div class="bc">
<span class="fl" style="padding-top: 1px;">
<a href="http://www.koolearn.com/" target="_blank" title="新东方在线网络课堂"><img src="http://images.koolearn.com/fe_upload/2015_9_2_1441179226504.jpg" width="105" height="48" alt="新东方在线网络课堂"></a></span>
<span class="fl" style="padding-top: 6px;">
<a href="http://cet4.koolearn.com/" target="_blank" rel="nofollow" class="ky">四级</a>
<a title="新东方在线网络课堂" href="http://www.koolearn.com/" target="_self">新东方在线</a> >
<a title="四级网络课堂" href="http://cet4.koolearn.com/" target="_self">四级</a> >
<a href="http://cet4.koolearn.com/cihui/" title="英语四级词汇">英语四级词汇</a> > 正文</span>
<a href="http://www.xdf.cn/" target="_blank" rel="nofollow" class="fr logo_p2"><img src="http://images.koolearn.com/fe_upload/2015_9_2_1441179317774.jpg" width="208" height="24"></a> </div>
'''
soup = BeautifulSoup(html,'lxml')
print(soup.div.descendants)
for i,child in enumerate(soup.div.descendants):
print(i,child)
运行结果如下：
<generator object descendants at 0x00000000028F5AF0>
0
1 <span class="fl" style="padding-top: 1px;">
<a href="http://www.koolearn.com/" target="_blank" title="新东方在线网络课堂"><img alt="新东方在线网络课堂" height="48" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179226504.jpg" width="105"/></a></span>
2
3 <a href="http://www.koolearn.com/" target="_blank" title="新东方在线网络课堂"><img alt="新东方在线网络课堂" height="48" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179226504.jpg" width="105"/></a>
4 <img alt="新东方在线网络课堂" height="48" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179226504.jpg" width="105"/>
5
6 <span class="fl" style="padding-top: 6px;">
<a class="ky" href="http://cet4.koolearn.com/" rel="nofollow" target="_blank">四级</a>
<a href="http://www.koolearn.com/" target="_self" title="新东方在线网络课堂">新东方在线</a> >
<a href="http://cet4.koolearn.com/" target="_self" title="四级网络课堂">四级</a> >
<a href="http://cet4.koolearn.com/cihui/" title="英语四级词汇">英语四级词汇</a> > 正文</span>
7
8 <a class="ky" href="http://cet4.koolearn.com/" rel="nofollow" target="_blank">四级</a>
9 四级
10
11 <a href="http://www.koolearn.com/" target="_self" title="新东方在线网络课堂">新东方在线</a>
12 新东方在线
13 >
14 <a href="http://cet4.koolearn.com/" target="_self" title="四级网络课堂">四级</a>
15 四级
16 >
17 <a href="http://cet4.koolearn.com/cihui/" title="英语四级词汇">英语四级词汇</a>
18 英语四级词汇
19 > 正文
20
21 <a class="fr logo_p2" href="http://www.xdf.cn/" rel="nofollow" target="_blank"><img height="24" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179317774.jpg" width="208"/></a>
22 <img height="24" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179317774.jpg" width="208"/>
23
父节点和祖先节点操作
获取父节点和祖先节点
from bs4 import BeautifulSoup
html='''
<div class="bc">
<span class="fl" style="padding-top: 1px;">
<a href="http://www.koolearn.com/" target="_blank" title="新东方在线网络课堂"><img src="http://images.koolearn.com/fe_upload/2015_9_2_1441179226504.jpg" width="105" height="48" alt="新东方在线网络课堂"></a></span>
<span class="fl" style="padding-top: 6px;">
<a href="http://cet4.koolearn.com/" target="_blank" rel="nofollow" class="ky">四级</a>
<a title="新东方在线网络课堂" href="http://www.koolearn.com/" target="_self">新东方在线</a> >
<a title="四级网络课堂" href="http://cet4.koolearn.com/" target="_self">四级</a> >
<a href="http://cet4.koolearn.com/cihui/" title="英语四级词汇">英语四级词汇</a> > 正文</span>
<a href="http://www.xdf.cn/" target="_blank" rel="nofollow" class="fr logo_p2"><img src="http://images.koolearn.com/fe_upload/2015_9_2_1441179317774.jpg" width="208" height="24"></a> </div>
'''
soup = BeautifulSoup(html,'lxml')
print(soup.a.parent) #获取父节点
print(soup.a.parents) #获取祖先节点
运行结果如下：
<span class="fl" style="padding-top: 1px;"> <a href="http://www.koolearn.com/" target="_blank" title="新东方在线网络课堂"><img alt="新东方在线网络课堂" height="48" src="http://images.koolearn.com/fe_upload/2015_9_2_1441179226504.jpg" width="105"/></a></span> <generator object parents at 0x00000000028C5B48>
兄弟节点操作
获取兄弟节点
from bs4 import BeautifulSoup
html='''
<div class="more_box" id="moreBox">
<h3>360识图</h3>
<a href="javascript:;" id="btnLoadMore" class="btn_loadmore">加载更多</a>
<p id="imgTotal" class="img_total">找到相关图片约 2637 张</p>
</div>
'''
soup = BeautifulSoup(html,'lxml')
print(soup.a.next_siblings) #获取前面的兄弟节点
print(soup.a.previous_siblings) #获取后面的兄弟节点
运行结果如下：
<generator object next_siblings at 0x0000000002885B48> <generator object previous_siblings at 0x0000000002885B48>
python生成器generator
l = [x * x for x in range(10)] g = (x * x for x in range(10)) print(l) print(g)
运行结果如下：
[0, 1, 4, 9, 16, 25, 36, 49, 64, 81] <generator object <genexpr> at 0x000000000251C468>
L 是一个list， 而 G 是一个generator：它们在创建时候最基本的不同就list是 [ ] ,而generator是 ( )
如果要一个个打印出来，可以通过next（）函数来获得generator的下一个返回值
g = (x * x for x in range(10)) for i in range(10): print(next(g))
运行结果如下
0 1 4 9 16 25 36 49 64 81
标准选择器操作
#可根据标签名、属性、内容查找文档，返回所有匹配结果
find_all(name,attrs,recusive,text,**kwargs)
#查找所有标签为a的节点
soup.find_all('a')
#查找所有标签为a，链接符合/view/123/htm形式的节点
soup.find_all('a',href='/view/123.htm')
soup.find_all('a',href=re.compile(r'/view/d+.htm'))
#查找所有标签为div,class为abc，文字为python的节点
soup.find_all('div',class_='abc',string='python')
属性：
#获取查到的节点的标签名称
node.name
#获取查找到的a节点的href属性
node['href']
#获取查找到的a节点的链接文字
node.get_text()
find(name,attrs,recusive,text,**kwargs)
可根据标签名、属性、内容查找文档，和find_all使用方法差不多，只不过返回第一个符合匹配的结果
find_parents() find_parent()
find_parents()返回所有祖先节点 ，find_parent()返回直接父节点
find_next_siblings() find_next_sibling()
find_next_siblings()返回前面所有兄弟节点，find_next_sibling()返回后面第一个兄弟节点
find_previous_siblings() find_previous_sibling()
find_previous_siblings()返回前面所有兄弟节点 ， find_previous_sibling()返回前面第一个兄弟节点
find_all_next() find_next()
find_all_next()返回节点后所有符合条件的节点 ， find_next()返回第一个符合条件的节点
find_all_previous() find_previous()
find_all_previous()返回节点后所有符合条件的节点 ，find_previous()返回第一个符合条件的节点
测试实例：
import bs4
html_doc='''
<div class="container"> <a href="/pc/home?sign=360_79aabe15" class="logo"></a> <nav id="nnav" data-mod="nnav"> <div class="nnav-wrap"> <ul class="nnav-items" id="nnav_main"> <li data-index="0"><a class="nnav-item" href="/pc/home?ch=youlike&sign=360_79aabe15" target="_blank" data-ch="youlike">推荐<span></span></a></li><li data-index="1"><a class="nnav-item" href="/pc/home?ch=good_safe2toera&sign=360_79aabe15" target="_blank" data-ch="good_safe2toera">新时代<span></span></a></li><li data-index="2"><a class="nnav-item" href="/pc/home?ch=fun&sign=360_79aabe15" target="_blank" data-ch="fun">娱乐<span></span></a></li><li data-index="3"><a class="nnav-item" href="/pc/home?
data-index="4"><a class="nnav-item" href="/pc/home?ch=economy&sign=360_79aabe15" target="_blank" data-ch="economy">财经<span></span></a></li><li data-index="5"><a class="nnav-item" href="/pc/home?ch=estate&sign=360_79aabe15" target="_blank" data-ch="estate">房产<span></span></a></li><li data-index="6"><a class="nnav-item" href="/pc/home?ch=car&sign=360_79aabe15" target="_blank" data-ch="car">汽车<span></span></a></li><li data-index="7"><a class="nnav-item" href="/pc/home?ch=sport&sign=360_79aabe15" target="_blank" data-ch="sport">体育<span></span></a></li><li data-index="8"><a class="nnav-item" href="/pc/home?ch=domestic&sign=360_79aabe15" target="_blank" data-ch="domestic">国内
'''
#创建BeautifulSoup对象
soup = bs4.BeautifulSoup(html_doc,'html.parser')
#获取所有的链接
links = soup.find_all('a')
for link in links:
print(link.name,link['href'],link.get_text())
#获取/pc/home?sign=360_79aabe15的链接
link_node = soup.find('a',href='/pc/home?sign=360_79aabe15')
print(link_node.name,link_node['href'],link_node.get_text())
运行结果如下：
a /pc/home?sign=360_79aabe15 a /pc/home?ch=youlike&sign=360_79aabe15 推荐 a /pc/home?ch=good_safe2toera&sign=360_79aabe15 新时代 a /pc/home?ch=fun&sign=360_79aabe15 娱乐 a /pc/home? data-index= 财经 a /pc/home?ch=economy&sign=360_79aabe15 财经 a /pc/home?ch=estate&sign=360_79aabe15 房产 a /pc/home?ch=car&sign=360_79aabe15 汽车 a /pc/home?ch=sport&sign=360_79aabe15 体育 a /pc/home?ch=domestic&sign=360_79aabe15 国内 a /pc/home?sign=360_79aabe15