虾米音乐导出成kgl到网易音乐 含有artist title 和album的信息

xiami2kgl-2.py

#coding=utf-8
import requests
from pyquery import PyQuery as PQ
import urllib2
import sys
from bs4 import BeautifulSoup
from multiprocessing import Pool
from functools import partial
import codecs
import re


HEADERS = {'User-Agent': 'Mozilla/5.0'}

def upload_img42(img):
    url = 'http://img42.com'
    req = urllib2.Request(url, data=img)
    msg = urllib2.urlopen(req).read()
    return '%s/%s' % (url, json.loads(msg)['id'])

def captcha(jpeg):
    print '* captcha needed'
    print 'captcha url:', upload_img42(jpeg)
    foo = raw_input('captcha >')
    return foo

def captcha_handler(session, html):
    bs = BeautifulSoup(html)
    captcha_url = bs.find('img')['src']
    input_session_id = bs.find('input', {'name': 'sessionID'})['value']
    input_apply = bs.find('input', {'name': 'apply'})['value']
    input_referer = bs.find('input', {'name': 'referer'})['value']

    header = {'Referer': input_referer,
              'User-Agent': 'Mozilla/5.0'}

    captcha_data = session.get(captcha_url, headers=header).content

    code = captcha(captcha_data)

    url = 'http://www.xiami.com/alisec/captcha/tmdgetv3.php'
    data = {'code': code,
            'sessionID': input_session_id,
            'apply': input_apply,
            'referer': input_referer}

    ret = session.post(url, data=data, headers=header)

def get(session, *args, **kwargs):
    while True:
        try:
            kwargs['headers'] = HEADERS
            data = session.get(*args, **kwargs)

            if data.status_code == 403:
                sec = re.findall('<script>document.cookie="sec=(.*?);', data.content)
                if len(sec) != 0:
                    session.cookies['sec'] = sec[0]
                    continue
        except Exception as e:
            # 失败重试
            print 'Error', e
            continue

        if 'regcheckcode.taobao.com' in data.content:
            print 'Captcha needed.'
            # 目前只是把整个requests.Response和session对象传给验证码处理函数, 函数需自行处理完后重试.
            self.captcha_handler(session, data)
            continue
        return data

def get_stared_songs(uid, page):
    url = 'http://www.xiami.com/space/lib-song/u/{uid}/page/{page}'
    session = requests.session()
    def _bs_func(bsobj):
        foo = bsobj.findAll(attrs={'class': 'song_name'})
        if not foo:
            return None
        lst = []
        for i in foo:
            lst.append(i.find('a')['href'].replace('http://www.xiami.com/song/', ''))
        return lst

    ret = BeautifulSoup(get(session, url.format(uid=uid, page=page)).content)
    return _bs_func(ret)

def get_song_list(uid):
    pool = Pool(20)

    session = requests.session()
    foo = get(session, 'http://www.xiami.com/space/lib-song/u/%s/' % uid, headers=HEADERS).content

    song_count = int(BeautifulSoup(foo).find('span', attrs={'class': 'counts'}).text[:-1])
    page_amount = song_count/25.0
    page_amount = int(page_amount) + 1 if page_amount > int(page_amount) else int(page_amount)

    print 'Total pages:', page_amount
    ret = pool.map(partial(get_stared_songs, uid), range(1, page_amount + 1))

    song_list = []
    map(lambda x: song_list.extend(x), ret)
    return song_list

def get_song_info_task(song):
    suber = re.compile('[&><"\']')
    session = requests.session()
    while True:
        foo = session.get('http://www.xiami.com/song/%s' % song, headers={'User-Agent': 'Mozilla/5.0'}).content
        if 'document.cookie="sec' in foo:
            sec = re.findall('<script>document.cookie="sec=(.*?);', foo)
            if len(sec) != 0:
                session.cookies['sec'] = sec[0]
            else:
                raise Exception

            continue
        bar = PQ(foo)
        artist = bar('[property="og:music:artist"]').attr('content')
        title = bar('[property="og:title"]').attr('content')
        album_name = bar('[property="og:music:album"]').attr('content')
        artist = suber.sub('',  artist).strip()
        title = suber.sub('',  title).strip()
        album_name = suber.sub('',  album_name).strip()
        print '%s - %s - %s.mp3' % (artist, title, album_name)
        return '%s - %s - %s.mp3' % (artist, title, album_name)

def get_song_infos(songs):
    print 'Get song infos'

    pool = Pool(30)
    ret = pool.map(get_song_info_task, songs)
    return ret

def save(song_infos):
    ret = []
    templates = u'''<?xml version="1.0" encoding="utf-8"?>
    <List ListName="Music list">
    {file_nodes}
    </List>
    '''

    for i in song_infos:
        ret.append(u'''    <File>
        <FileName>%s</FileName>
    </File>''' % i.strip())

    f = codecs.open('xiami.kgl', mode='w', encoding='utf-8')
    f.write(templates.format(file_nodes=u'\n'.join(ret)))


def main():
    try:
        uid = int(sys.argv[1])
    except:
        print 'Arg 1 must be user id'
        sys.exit(1)

    song_list = get_song_list(uid)
    song_infos = get_song_infos(song_list)
    save(song_infos)



if __name__ == '__main__':
    main()