Python学习之网页抓取（二）

川长思鸟来 2021-12-22 04:11 404阅读 0赞

这一篇实现的功能是：对下载的网页内容进一步提取内容，如网页编码类型，标题等。

#!/usr/bin/env python
    # -*- coding: GBK -*-
    '''
    Created on Jul 17, 2013
    @author: belong
    '''
    
    import urllib
    import re
    
    class Tools:
        def write_log(self, level, info):
            print '[' + level + ']' + info
        def match_regex(self,pattern,src):
            result = ""
            com = re.compile(pattern,re.I)
            matchers = com.findall(src)
            print matchers
            for matcher in matchers:
                result += matcher +" "
            return result.strip()#strip函数返回去掉两端空格的字符串
               
    class Crawler:
        #获取一个url链接的所有urls
        def get_url(self, url):
            html = urllib.urlopen(url)
        #     pattern = re.compile("http://.*?\.com",re.I)
            pattern = re.compile(r'[a-zA-z]+://[^\s]*\.html?', re.I)
            while True:
                data = html.read()
                if data:
                    urls = pattern.findall(data)
                else:
                    break
            html.close()
            return urls
        
        #下载url对应网页
        def download_url(self, url, filename):
            Tools().write_log("info","开始下载")
            html = urllib.urlopen(url)
            f = open(filename, 'w')
            while True:
                page = html.read()
                if page:
                    f.write(page);
                else:
                    break
            html.close()
            f.close()
            Tools().write_log("info","网页下载成功")
            return page
        
        #广度优先遍历
        def broad_traverse(self, start_url, number):
            Tools().write_log("info","开始遍历")
            visited = []
            unvisited = []
            unvisited.append(start_url)
            while len(unvisited):
                if len(visited) < number:
                    url = unvisited.pop(0);
                    print url, len(visited)
                    i = len(visited)
    #                 self.download_url(url, str(i) + '.html')
                    visited.append(url)
                    url_list = self.get_url(url)
                    for eachlink in url_list:
                        if ((unvisited.count(eachlink) == 0) & (visited.count(eachlink) == 0)):
                            unvisited.append(eachlink)
                else:
                    break
            Tools().write_log("info","遍历成功")
            return visited 
        
        def main(self):
            start_url = "http://www.baidu.com"
            self.broad_traverse(start_url, 10)
    
    #数据提取类
    class DataExtractor:
        #提取标题
        def get_title(self,data):
            title = Tools().match_regex('<title>.*?<\/title>', data)
            if title == "":
                Tools().write_log( "Error", "标题匹配不成功")
            return title
        #提取内容格式
        def get_type(self,data):
            type = Tools().match_regex('<meta.*content=.*?\/>', data)
            return type
        #提取字符类型
        def get_charset(self,data):
            charset = Tools().match_regex('<meta.*?charset=.*?\/>', data)
            return charset
        
        def get_info(self,url):
            try:
                data = Crawler().download_url(url,str(11)+'.html')
            except:
                Tools().write_log("error", url+"抓取失败")
                raise
        
            Tools().write_log("info", "开始数据匹配")
            rst = {}#rst是一个字典
            rst['title'] = self.get_title(data)
            print "title:",rst['title'] 
            rst['type'] = self.get_type(data)
            print "type:",rst['type'] 
            rst['charset'] = self.get_charset(data)
            print "charset:",rst['charset'] 
            Tools().write_log("DEBUG",'title=%s,type=%s,type=%s'%(rst['title'],rst['type'],rst['charset']))
            return rst
    
    def main():
        start_url = "http://www.baidu.com"
        crawler = Crawler()
        url_list = crawler.broad_traverse(start_url, 10)   
        dataExtractor= DataExtractor()
        for url in url_list:
            dataExtractor.get_info(url)
    
    main()