<html>
<head>
<title>The Dormouse's story </title>
</head>
<body>
<p id="p1">p1p1p1
<b id='b1'>b1b1b1</b>
</p>
<p id="p2">p2p2p2
<ul id='u1'>u1u1u1</ul>
<a id="a1">a1a1a1</a>
<p id='d1'>
<a id="a2">a2a2a2 </a>
<b id='b2'>b2b2b2</b>
<p id='p3'>p3p3p3</p>
</p>
<a id="a3">a3a3a3 </a>
</p>
<p id="p4">p4p4p4</p>
</body>
</html>比如第一个a元素:a#a1,要获取这个元素以上的所有网页源码:
<html>
<head>
<title>The Dormouse's story </title>
</head>
<body>
<p id="p1">p1p1p1
<b id='b1'>b1b1b1</b>
</p>
<p id="p2">p2p2p2
<ul id='u1'>u1u1u1</ul>
<a id="a1">a1a1a1</a>
</p>
</body>
</html>
阿神2017-04-18 09:49:08
由於你原來的html不合規範，我改了點。
下面是用 lxml 做的。
doc = '''
<html>
<head>
<title>The Dormouse's story </title>
</head>
<body>
<p id="p1">p1p1p1
<b id='b1'>b1b1b1</b>
</p>
<p id="p2">p2p2p2</p>
<p id='d1'>
<ul id='u1'>u1u1u1</ul>
<a id="a1">a1a1a1</a>
<p id='d2'>
<a id="a2">a2a2a2 </a>
<b id='b2'>b2b2b2</b>
<p id='p3'>p3p3p3</p>
</p>
<a id="a3">a3a3a3 </a>
</p>
<p id="p4">p4p4p4</p>
</body>
</html>
'''
from lxml import html
tree = html.fromstring(doc)
a = tree.get_element_by_id("a1")
print(html.tostring(a))
print(html.tostring(tree).decode())
def dropnode(e=None):
if e is None: return
if e.tag == 'body': return
nd = e.getnext()
while nd is not None:
nd.drop_tree()
nd = e.getnext()
dropnode(e.getparent())
dropnode(a)
print(html.tostring(tree).decode()) 
阿神2017-04-18 09:49:08
新手,我只學了re模組，所以只用re模組+普通方式來提取
>>> html = '''
<html>
<head>
<title>The Dormouse's story </title>
</head>
<body>
<p id="p1">p1p1p1
<b id='b1'>b1b1b1</b>
</p>
<p id="p2">p2p2p2
<ul id='u1'>u1u1u1</ul>
<a id="a1">a1a1a1</a>
<p id='d1'>
<a id="a2">a2a2a2 </a>
<b id='b2'>b2b2b2</b>
<p id='p3'>p3p3p3</p>
</p>
<a id="a3">a3a3a3 </a>
</p>
<p id="p4">p4p4p4</p>
</body>
</html>
'''
>>> html_1 = re.search(r'<a',html)
>>> html_1 = html_1.span()
>>> print(html[:html_1[0]])
<html>
<head>
<title>The Dormouse's story </title>
</head>
<body>
<p id="p1">p1p1p1
<b id='b1'>b1b1b1</b>
</p>
<p id="p2">p2p2p2
<ul id='u1'>u1u1u1</ul>
>>> 