_ 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         u'url': u'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
  25         u'playlist': [
  26             {
  27                 u'file': u'1353101989.mp3',
  28                 u'md5': u'39bc1eded3476e927c724321ddf116cf',
  29                 u'info_dict': {
  30                     u'title': u'Intro',
  31                 }
  32             },
  33             {
  34                 u'file': u'38097443.mp3',
  35                 u'md5': u'1a2c32e2691474643e912cc6cd4bffaa',
  36                 u'info_dict': {
  37                     u'title': u'Kero One - Keep It Alive (Blazo remix)',
  38                 }
  39             },
  40         ],
  41         u'params': {
  42             u'playlistend': 2
  43         },
  44         u'skip': u'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
  45     }]
  46
  47     def _real_extract(self, url):
  48         mobj = re.match(self._VALID_URL, url)
  49         title = mobj.group('title')
  50         webpage = self._download_webpage(url, title)
  51         # We get the link to the free download page
  52         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  53         if m_download is None:
  54             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
  55         if m_trackinfo:
  56             json_code = m_trackinfo.group(1)
  57             data = json.loads(json_code)
  58
  59             entries = []
  60             for d in data:
  61                 formats = [{
  62                     'format_id': 'format_id',
  63                     'url': format_url,
  64                     'ext': format_id.partition('-')[0]
  65                 } for format_id, format_url in sorted(d['file'].items())]
  66                 entries.append({
  67                     'id': compat_str(d['id']),
  68                     'title': d['title'],
  69                     'formats': formats,
  70                 })
  71
  72             return self.playlist_result(entries, title, title)
  73         else:
  74             raise ExtractorError(u'No free songs found')
  75
  76         download_link = m_download.group(1)
  77         id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
  78                        webpage, re.MULTILINE|re.DOTALL).group('id')
  79
  80         download_webpage = self._download_webpage(download_link, id,
  81                                                   'Downloading free downloads page')
  82         # We get the dictionary of the track from some javascrip code
  83         info = re.search(r'items: (.*?),$',
  84                          download_webpage, re.MULTILINE).group(1)
  85         info = json.loads(info)[0]
  86         # We pick mp3-320 for now, until format selection can be easily implemented.
  87         mp3_info = info[u'downloads'][u'mp3-320']
  88         # If we try to use this url it says the link has expired
  89         initial_url = mp3_info[u'url']
  90         re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
  91         m_url = re.match(re_url, initial_url)
  92         #We build the url we will use to get the final track url
  93         # This url is build in Bandcamp in the script download_bunde_*.js
  94         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'))
  95         final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
  96         # If we could correctly generate the .rand field the url would be
  97         #in the "download_url" key
  98         final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
  99
 100         track_info = {'id':id,
 101                       'title' : info[u'title'],
 102                       'ext' :   'mp3',
 103                       'url' :   final_url,
 104                       'thumbnail' : info[u'thumb_url'],
 105                       'uploader' :  info[u'artist']
 106                       }
 107
 108         return [track_info]
 109
 110
 111 class BandcampAlbumIE(InfoExtractor):
 112     IE_NAME = u'Bandcamp:album'
 113     _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)'
 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         }