_ 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     }]
  45
  46     def _real_extract(self, url):
  47         mobj = re.match(self._VALID_URL, url)
  48         title = mobj.group('title')
  49         webpage = self._download_webpage(url, title)
  50         # We get the link to the free download page
  51         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  52         if m_download is None:
  53             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
  54         if m_trackinfo:
  55             json_code = m_trackinfo.group(1)
  56             data = json.loads(json_code)
  57
  58             entries = []
  59             for d in data:
  60                 formats = [{
  61                     'format_id': 'format_id',
  62                     'url': format_url,
  63                     'ext': format_id.partition('-')[0]
  64                 } for format_id, format_url in sorted(d['file'].items())]
  65                 entries.append({
  66                     'id': compat_str(d['id']),
  67                     'title': d['title'],
  68                     'formats': formats,
  69                 })
  70
  71             return self.playlist_result(entries, title, title)
  72         else:
  73             raise ExtractorError(u'No free songs found')
  74
  75         download_link = m_download.group(1)
  76         id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
  77                        webpage, re.MULTILINE|re.DOTALL).group('id')
  78
  79         download_webpage = self._download_webpage(download_link, id,
  80                                                   'Downloading free downloads page')
  81         # We get the dictionary of the track from some javascrip code
  82         info = re.search(r'items: (.*?),$',
  83                          download_webpage, re.MULTILINE).group(1)
  84         info = json.loads(info)[0]
  85         # We pick mp3-320 for now, until format selection can be easily implemented.
  86         mp3_info = info[u'downloads'][u'mp3-320']
  87         # If we try to use this url it says the link has expired
  88         initial_url = mp3_info[u'url']
  89         re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
  90         m_url = re.match(re_url, initial_url)
  91         #We build the url we will use to get the final track url
  92         # This url is build in Bandcamp in the script download_bunde_*.js
  93         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'))
  94         final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
  95         # If we could correctly generate the .rand field the url would be
  96         #in the "download_url" key
  97         final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
  98
  99         track_info = {'id':id,
 100                       'title' : info[u'title'],
 101                       'ext' :   'mp3',
 102                       'url' :   final_url,
 103                       'thumbnail' : info[u'thumb_url'],
 104                       'uploader' :  info[u'artist']
 105                       }
 106
 107         return [track_info]
 108
 109
 110 class BandcampAlbumIE(InfoExtractor):
 111     IE_NAME = u'Bandcamp:album'
 112     _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)'
 113
 114     def _real_extract(self, url):
 115         mobj = re.match(self._VALID_URL, url)
 116         title = mobj.group('title')
 117         webpage = self._download_webpage(url, title)
 118         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
 119         if not tracks_paths:
 120             raise ExtractorError(u'The page doesn\'t contain any track')
 121         entries = [
 122             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
 123             for t_path in tracks_paths]
 124         title = self._search_regex(r'album_title : "(.*?)"', webpage, u'title')
 125         return {
 126             '_type': 'playlist',
 127             'title': title,
 128             'entries': entries,
 129         }