比如要获取所有class=test的div元素，
1. 用Scrapy，示范代码：
def parse(self, response):
hxs = Selector(response)
items = []
divs = hxs.xpath('//div[@class="test"]')
2. 用lxml，示范代码：
from lxml import etree
#import mechanize
import lxml.html
#import cookielib
#br = mechanize.Browser()
#r = br.open('http://yourdomain.com')
#html = br.response().read()
#root = lxml.html.fromstring(html)
#divs = root.xpath("//div[@class='test']")
hparser = etree.HTMLParser(encoding='utf-8') #for avoiding unicode codec problems
htree = etree.parse('http://yourdomain.com',hparser)
htree.write('/tmp/bi.html')
divs= htree.xpath("//div[@class='test']")
要获取class包含test的所有div，比如<div class="test website"></div>
把上述xpath的参数修改为 "div[contains(@class,'test')]" 即可。
Scrapy内部解析引擎也是用的lxml。
参考链接：
http://lxml.de/dev/lxmlhtml.html#examples
http://doc.scrapy.org/en/latest/
by iefreer