#-*- coding: UTF-8 -*-

from lxml import etree

source = u'''
<div><p class="p1" data-a="1">测试数据1</p>
<p class="p1" data-a="2">测试数据2</p>
<p class="p1" data-a="3" style="height:100px;">
<strong class="s">测试数据3</strong></p>
<p class="p1" data-a="4" width="200"><img src="1.jpg" class="img"/><br/>
图片</p>
'''

# 从字符串解析
page = etree.HTML(source)

# 元素列表
ps = page.xpath("//p")
for p in ps:
  print u"属性：%s" % p.attrib
  print u"文本：%s" % p.text

# 文本列表
ts = page.xpath("//p/text()")
for t in ts:
  print t

# xpath定位 
ls = page.xpath('//p[@class="p1"][last()]/img')
for l in ls:
  print l.attrib