python实现简单爬虫（二）---- 使用urllib等python模块

# -*- coding:utf-8 -*-
import urllib
import urllib2
import re
import json


class FilmSpider:
    def __init__(self, init_url_):
        self.start_url = init_url_
        self.contain = []
        self.curPage = 1

    def Parse(self):
        nexturl = self.start_url
        while nexturl != None:
            nexturl = self.parsePage(nexturl)
            self.curPage += 1

        f = open(‘movie2.json‘, ‘a‘)

        # JsonLinesItemExporter
        """
        for each_item in self.contain:
            tar = json.dumps(each_item, ensure_ascii=False)
            f.write(tar+‘\n‘)
        """
        # JsonItemExporter
        first_item = True
        f.write(‘[‘)
        for each_item in self.contain:
            if first_item:
                first_item = False
            else:
                f.write(‘,\n‘)
            tar = json.dumps(each_item, ensure_ascii=False)
            f.write(tar)
        f.write(‘]‘)
        f.close()

    def parsePage(self, cururl):
        print(‘sola is scrawling the %d page: ‘%self.curPage + cururl)
        response = urllib2.urlopen(cururl)
        body = response.read()
        #返回所有在<tr class="item"></tr>中的字符串的list
        myItems = re.findall(‘<tr class="item">(.*?)</tr>‘, body, re.S)
        for item in myItems:
            # 返回所有title 和link组成的tuple所组成的list, 其实list中只有一个tuple
            info = re.findall(‘<a.*?class="nbg".*?href="(.*?)".*?title="(.*?)">‘, item, re.S)
            # 返回<p>内的描述所构成的list, 同理只有一个元素
            desc = re.findall(‘<p.*?class="pl">(.*?)</p>‘, item, re.S)
            #newItem =  Item(info[0][0], info[0][1], desc[0])

            newItem = {}
            newItem[‘link‘] = info[0][0]
            newItem[‘title‘] = info[0][1]
            newItem[‘desc‘] = desc[0]
            self.contain.append(newItem)
            print newItem[‘title‘]
            print newItem[‘link‘]
            print newItem[‘desc‘]
            print (‘\n\n------------------------------------------------------------------------------------------------\n\n‘)

        Next = re.findall(‘<span.*?class="next">.*?<link.*?rel="next".*?href="(.*?)"/>‘, body, re.S)
        if Next:
            return Next[0]
        return None



#-------------------------Main--------------------
#program: spider for Love in Douban
#author : Patrick
#-------------------------------------------------

print(‘-----------------Start crawl----------------------‘)

initurl = ‘http://movie.douban.com/tag/%E7%88%B1%E6%83%85?start=0&type=T‘
solaSpider = FilmSpider(initurl)
solaSpider.Parse()

print(‘-----------------End crawl------------------------‘)