看kindle网站电子书用Python爬取下载

#!/usr/bin/env python

# coding=utf-8

from bs4 import BeautifulSoup

import urllib2

import socket

import re

import unicodedata

import os

from urwid.text_layout import trim_line

def download(url):

    print 'starting download %s' % url

    response=urllib2.urlopen(url,timeout=30)

    html_data=response.read()

    soup=BeautifulSoup(html_data)

    print 'start to analayse---------------'

    title_soup=soup.find_all(class_='yanshi_xiazai')

    name_soup = soup.find_all('h1')

    tag_a = title_soup[0].a.attrs['href']

    tag_name= title_soup[0].a.contents

    link_name = name_soup[0]

    link_name = str(link_name).replace("<h1>","").replace("</h1>","")

    #print tag_name[0]

    #print link_name

    filename = link_name+".mobi"

    filename = "ebook/"+filename

    print 'filename is :%s' % filename

    print "downloading with urllib2 %s" % tag_a

    if os.path.exists(filename):

        print 'already donwload ,ignore'

    else:

        try:

            f = urllib2.urlopen(tag_a,timeout=60)

            data = f.read()

            #print 'the data is %s'% data

            with open(filename, "wb") as code:

                code.write(data)

        except Exception,e:

            print e

def get_all_link(url):

    print 'Starting get all the list'

    response=urllib2.urlopen(url,timeout=30)

    html_data=response.read()

    #print html_data

    soup=BeautifulSoup(html_data)

    link_soup = soup.find_all('a')

    #print link_soup

    for each_link in link_soup:

        if re.search('view',str(each_link)):

            #print each_link

            print each_link

            print each_link.attrs['href']

            download(each_link.attrs['href'])

if __name__ == '__main__':

    for page in range(1,13):

        url = "http://kankindle.com/simple/page/3"+str(page)

        url = url.strip()

        print url

        get_all_link(url)