【python】一个简单的贪婪爬虫

这个爬虫的作用是，对于一个给定的url，查找页面里面所有的url连接并依次贪婪爬取

主要需要注意的地方：

1.lxml.html.iterlinks()  可以实现对页面所有url的查找

2.获取页面要设置超时，否则遇到没有响应的页面，代码容易卡死

3.对所有异常的捕获

4.广度优先搜索实现

具体代码如下：

#!/usr/bin/env python
#encoding:utf8

#这是一个爬虫代码，从初始url页面中查找所有的url并根据广度优先的顺序依次贪婪爬取
#爬取的页面以1.html,2.html...的方式命名
#author：匡子语

import re
import time
import lxml.html
import urllib2
from collections import deque

class ScrawURL:
    def __init__(self):
        self.total = 50000   #要获取的页面数
        self.urls = ["http://www.cnblogs.com/"]  #初始url
        self.html_num = 13047   #当前存储的是第几个页面
        self.url_num = 0   #当前要爬取的url在urls中的位置

    def getHtml(self, url):
        print "url:%s" % url
        user_agent = 'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'
        headers = { 'User-Agent' : user_agent }
        try:
            request = urllib2.Request(url, headers = headers)
            response = urllib2.urlopen(request, timeout=5)   #要设置timeout,有时页面长时间无法响应不加timeout会导致程序卡死
            content = response.read()
            self.html_num += 1
            name = "htmls/%d.html" % self.html_num
            with open(name, "w") as f:
                f.write(content)
            print self.html_num
            return content
        except urllib2.HTTPError, e:
            return ''
        except urllib2.URLError, e:
            return ''
        except Exception, ex:   #对于所有异常形式，都返回空内容
            return ''

    def getUrl(self, content):
        try:
            print "getUrl"
            html = lxml.html.fromstring(content)
            links = lxml.html.iterlinks(html)   #这个函数可以自动获取页面中的所有url，非常好用
            urls = []
            for link in links:
                if "http" in link[2]:
                    urls.append(link[2])
            return urls
        except Exception, ex:
            return []

    def scrawl(self):   #广度优先爬取
        print "scrawl"
        while len(self.urls) < self.total and len(self.urls) > self.url_num:   #获取足够的url
            url = self.urls[self.url_num]
            self.url_num += 1
            content = self.getHtml(url)
            if content:
                urls = self.getUrl(content)
                if urls:
                    for url in urls:
                        if url not in self.urls:   #对于已经爬取过的url跳过
                            self.urls.append(url)

        while self.html_num < self.total and len(self.urls) > 0:   #获取足够的页面
            url = self.urls[self.url_num]
            self.url_num += 1
            self.getHtml(url)

if __name__ == "__main__":
    surl = ScrawURL()
    surl.scrawl()