#-*- coding: UTF- -*-

from lxml import etree

source = u'''

<div><p class="p1" data-a="">测试数据1</p>

<p class="p1" data-a="">测试数据2</p>

<p class="p1" data-a="" style="height:100px;">

<strong class="s">测试数据3</strong></p>

<p class="p1" data-a="" width=""><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