Download DoubanFM favourite songs

douban_album_download.py

# coding: utf-8

import os, urllib, urllib2, re, json

baseurl = 'http://douban.fm/j/mine/playlist?type=n&h=&channel=0&context=channel:0|subject_id:%s'
musicdir = os.path.abspath('./music')
num_pattern = re.compile(r'(\d+)')

def get_json_by_album(href):
    ''' get direct url of songs by album url '''
    subject_id = num_pattern.search(href).groups()[0]
    content = urllib2.urlopen(baseurl % subject_id).read()
    jsob = json.loads(content)
    return filter(lambda song: song['album'].endswith(subject_id+'/'), jsob['song'])

def download(song):
    ''' download by song json object then correct id3 infomation '''
    if not os.path.exists(musicdir):
        os.mkdir(musicdir)

    filename = '%s.mp3' % song['title']
    filepath = os.path.join(musicdir, filename)
    if os.path.exists(filepath):
        return
    urllib.urlretrieve(song['url'], filepath)

    picname = song['picture'][1+song['picture'].rindex('/'):]
    picpath = os.path.join(musicdir, picname)
    urllib.urlretrieve(song['picture'], picpath)

    import eyeD3
    tag = eyeD3.Tag()
    tag.link(filepath)
    tag.header.setVersion(eyeD3.ID3_V2_3)
    tag.encoding = '\x01'
    tag.setTitle(song['title'])
    tag.setAlbum(song['albumtitle'])
    tag.setArtist(song['artist'])
    tag.addImage(3, picpath, u'')
    os.remove(picpath)
    tag.update()

if __name__ == '__main__':
    from pprint import pprint
    songs = get_json_by_album('http://music.douban.com/subject/3649821/')

    #pprint(songs)

    print 'Found: %d' % len(songs)
    for song in songs:
        download(song)

from_douban.py

# coding: utf-8

# =======================================================

faillist = file('faillist.txt').read().split('\n')
faillist = map(lambda line: line.decode('utf-8'), faillist)
from fmdownload import *

# =======================================================

import sys, urllib2, cookielib
from BeautifulSoup import BeautifulSoup
from time import sleep
from pprint import pprint

import HTMLParser
hp = HTMLParser.HTMLParser()
ue = lambda s: hp.unescape(s)

# cookies
cookie = 'flag="ok";ck="XXXX";dbcl2="XXXXXXXXXXXXXXX";bid="XXXXXXXXX";'
# song per page
spp = 15
baseurl = 'http://douban.fm/mine?start=%d&type=liked'
pages = int(raw_input('number of pages: '))

opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookielib.CookieJar()))
urllib2.install_opener(opener)

for i in range(pages):
    sleep(1)
    req = urllib2.Request(baseurl % (i*spp,))
    req.add_header('User-Agent', 'Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)')
    req.add_header('Cookie', cookie)
    content = urllib2.urlopen(req).read()
    soup = BeautifulSoup(str(content))
    divs = soup.findAll('div', {'class': 'song_info'})
    for div in divs:
        p = div.findAll('p')
        name = u'%s %s' % (ue(p[0].string), ue(p[1].string))
        if name in faillist:
            songs = get_json_by_album(str(p[2].a))
            found = False
            for i in range(10):
                for song in songs:
                    if ue(p[0].string) == song['title']:
                        download(song)
                        #print 'DONE:', name.encode('utf-8')
                        found = True
                        break
                if not found:
                    if i == 9:
                        print 'FAIL:', name.encode('utf-8')
                else:
                    break

from_ting.py

# coding: utf-8
import os, urllib, urllib2, json, re

ref = 'https://github.com/alswl/tingdownload'

SEARCH_URL = u'http://openapi.baidu.com/public/2.0/mp3/info/suggestion?' \
            'format=json&word=%s&callback=window.baidu.sug'
DOWNLOAD_URL = u'http://ting.baidu.com/song/%s/download'
pattern = re.compile(r'<a href="([^"]*)"')
TARGET_URL = u'http://ting.baidu.com%s'
MUSICS_DIR = os.path.abspath('./musics')

def search(word):
    ''' search musics, return json infomation '''
    word = urllib2.quote(word)
    response = urllib2.urlopen(SEARCH_URL % word).read()
    json_text = response.strip()[17:-2]
    return json.loads(json_text)

def geturl(songid):
    ''' get download url by song id '''
    page = urllib2.urlopen(DOWNLOAD_URL % songid).read()
    link = pattern.search(page).groups()[0]
    return TARGET_URL % link

def download(url):
    ''' download song by url '''
    if not os.path.exists(MUSICS_DIR):
        os.mkdir(MUSICS_DIR)
    name = urllib2.unquote(url)[1+urllib2.unquote(url).rindex('/'):]
    #name = name.decode('utf-8')
    name = ''.join(map(chr,map(ord,list(name)))) # Orz
    print 'downloading:', name
    filepath = os.path.join(MUSICS_DIR, name)
    if not os.path.exists(filepath):
        urllib.urlretrieve(url, filepath)

if __name__ == '__main__':
    print 'Downloading DoubanFM favs...'
    songs = file('favs.txt').read().strip().split('\n')
    print len(songs), 'songs'

    faillist = []

    for line in songs:
        res = search(line)

        if not res['song']:
            print 'FAIL:', line
            faillist.append(line)
            continue
        
        download(geturl(res['song'][0]['songid']))
        print 'DONE:', line

    with file('faillist.txt','w') as f:
        f.write('\n'.join(faillist))

getlist.py

# coding: utf-8
import sys, urllib2, cookielib
from BeautifulSoup import BeautifulSoup
from time import sleep

import HTMLParser
hp = HTMLParser.HTMLParser()
ue = lambda s: hp.unescape(s)

# cookies
cookie = 'flag="ok";ck="XXXX";dbcl2="XXXXXXXXXXXXXXX";bid="XXXXXXXXX";'
# song per page
spp = 15
baseurl = 'http://douban.fm/mine?start=%d&type=liked'
pages = int(raw_input('number of pages: '))
result = 0

# output to file
reload(sys)
sys.setdefaultencoding(sys.getfilesystemencoding())
originout = sys.stdout
sys.stdout = open('favs.txt','w')

opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookielib.CookieJar()))
urllib2.install_opener(opener)

for i in range(pages):
    sleep(1)
    req = urllib2.Request(baseurl % (i*spp,))
    req.add_header('User-Agent', 'Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)')
    req.add_header('Cookie', cookie)
    content = urllib2.urlopen(req).read()
    soup = BeautifulSoup(str(content))
    divs = soup.findAll('div', {'class': 'song_info'})
    result += len(divs)
    for div in divs:
        p = div.findAll('p')
        print ue(p[0].string), ue(p[1].string)#, ue(p[2].a.string)

sys.stdout = originout
print 'Done with %d results.' % result