_ Git - youtube-dl/blob - youtube_dl/extractor/bandcamp.py

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_str,
   7     compat_urlparse,
   8     ExtractorError,
   9 )
  10
  11
  12 class BandcampIE(InfoExtractor):
  13     IE_NAME = u'Bandcamp'
  14     _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)'
  15     _TESTS = [{
  16         u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  17         u'file': u'1812978515.mp3',
  18         u'md5': u'cdeb30cdae1921719a3cbcab696ef53c',
  19         u'info_dict': {
  20             u"title": u"youtube-dl test song \"'/\\\u00e4\u21ad"
  21         },
  22         u'skip': u'There is a limit of 200 free downloads / month for the test song'
  23     }]
  24
  25     def _real_extract(self, url):
  26         mobj = re.match(self._VALID_URL, url)
  27         title = mobj.group('title')
  28         webpage = self._download_webpage(url, title)
  29         # We get the link to the free download page
  30         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  31         if m_download is None:
  32             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
  33         if m_trackinfo:
  34             json_code = m_trackinfo.group(1)
  35             data = json.loads(json_code)
  36
  37             entries = []
  38             for d in data:
  39                 formats = [{
  40                     'format_id': 'format_id',
  41                     'url': format_url,
  42                     'ext': format_id.partition('-')[0]
  43                 } for format_id, format_url in sorted(d['file'].items())]
  44                 return {
  45                     'id': compat_str(d['id']),
  46                     'title': d['title'],
  47                     'formats': formats,
  48                 }
  49         else:
  50             raise ExtractorError(u'No free songs found')
  51
  52         download_link = m_download.group(1)
  53         id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
  54                        webpage, re.MULTILINE|re.DOTALL).group('id')
  55
  56         download_webpage = self._download_webpage(download_link, id,
  57                                                   'Downloading free downloads page')
  58         # We get the dictionary of the track from some javascrip code
  59         info = re.search(r'items: (.*?),$',
  60                          download_webpage, re.MULTILINE).group(1)
  61         info = json.loads(info)[0]
  62         # We pick mp3-320 for now, until format selection can be easily implemented.
  63         mp3_info = info[u'downloads'][u'mp3-320']
  64         # If we try to use this url it says the link has expired
  65         initial_url = mp3_info[u'url']
  66         re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
  67         m_url = re.match(re_url, initial_url)
  68         #We build the url we will use to get the final track url
  69         # This url is build in Bandcamp in the script download_bunde_*.js
  70         request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), id, m_url.group('ts'))
  71         final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
  72         # If we could correctly generate the .rand field the url would be
  73         #in the "download_url" key
  74         final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
  75
  76         track_info = {'id':id,
  77                       'title' : info[u'title'],
  78                       'ext' :   'mp3',
  79                       'url' :   final_url,
  80                       'thumbnail' : info[u'thumb_url'],
  81                       'uploader' :  info[u'artist']
  82                       }
  83
  84         return [track_info]
  85
  86
  87 class BandcampAlbumIE(InfoExtractor):
  88     IE_NAME = u'Bandcamp:album'
  89     _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)'
  90
  91     _TEST = {
  92         u'url': u'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
  93         u'playlist': [
  94             {
  95                 u'file': u'1353101989.mp3',
  96                 u'md5': u'39bc1eded3476e927c724321ddf116cf',
  97                 u'info_dict': {
  98                     u'title': u'Intro',
  99                 }
 100             },
 101             {
 102                 u'file': u'38097443.mp3',
 103                 u'md5': u'1a2c32e2691474643e912cc6cd4bffaa',
 104                 u'info_dict': {
 105                     u'title': u'Kero One - Keep It Alive (Blazo remix)',
 106                 }
 107             },
 108         ],
 109         u'params': {
 110             u'playlistend': 2
 111         },
 112         u'skip': u'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
 113     }
 114
 115     def _real_extract(self, url):
 116         mobj = re.match(self._VALID_URL, url)
 117         title = mobj.group('title')
 118         webpage = self._download_webpage(url, title)
 119         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
 120         if not tracks_paths:
 121             raise ExtractorError(u'The page doesn\'t contain any track')
 122         entries = [
 123             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
 124             for t_path in tracks_paths]
 125         title = self._search_regex(r'album_title : "(.*?)"', webpage, u'title')
 126         return {
 127             '_type': 'playlist',
 128             'title': title,
 129             'entries': entries,
 130         }