_ 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_urlparse,
   7     ExtractorError,
   8 )
   9
  10
  11 class BandcampIE(InfoExtractor):
  12     IE_NAME = u'Bandcamp'
  13     _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)'
  14     _TEST = {
  15         u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  16         u'file': u'1812978515.mp3',
  17         u'md5': u'cdeb30cdae1921719a3cbcab696ef53c',
  18         u'info_dict': {
  19             u"title": u"youtube-dl test song \"'/\\\u00e4\u21ad"
  20         },
  21         u'skip': u'There is a limit of 200 free downloads / month for the test song'
  22     }
  23
  24     def _real_extract(self, url):
  25         mobj = re.match(self._VALID_URL, url)
  26         title = mobj.group('title')
  27         webpage = self._download_webpage(url, title)
  28         # We get the link to the free download page
  29         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  30         if m_download is None:
  31             raise ExtractorError(u'No free songs found')
  32
  33         download_link = m_download.group(1)
  34         id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
  35                        webpage, re.MULTILINE|re.DOTALL).group('id')
  36
  37         download_webpage = self._download_webpage(download_link, id,
  38                                                   'Downloading free downloads page')
  39         # We get the dictionary of the track from some javascrip code
  40         info = re.search(r'items: (.*?),$',
  41                          download_webpage, re.MULTILINE).group(1)
  42         info = json.loads(info)[0]
  43         # We pick mp3-320 for now, until format selection can be easily implemented.
  44         mp3_info = info[u'downloads'][u'mp3-320']
  45         # If we try to use this url it says the link has expired
  46         initial_url = mp3_info[u'url']
  47         re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
  48         m_url = re.match(re_url, initial_url)
  49         #We build the url we will use to get the final track url
  50         # This url is build in Bandcamp in the script download_bunde_*.js
  51         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'))
  52         final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
  53         # If we could correctly generate the .rand field the url would be
  54         #in the "download_url" key
  55         final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
  56
  57         track_info = {'id':id,
  58                       'title' : info[u'title'],
  59                       'ext' :   'mp3',
  60                       'url' :   final_url,
  61                       'thumbnail' : info[u'thumb_url'],
  62                       'uploader' :  info[u'artist']
  63                       }
  64
  65         return [track_info]
  66
  67
  68 class BandcampAlbumIE(InfoExtractor):
  69     IE_NAME = u'Bandcamp:album'
  70     _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)'
  71
  72     def _real_extract(self, url):
  73         mobj = re.match(self._VALID_URL, url)
  74         title = mobj.group('title')
  75         webpage = self._download_webpage(url, title)
  76         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
  77         if not tracks_paths:
  78             raise ExtractorError(u'The page doesn\'t contain any track')
  79         entries = [
  80             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
  81             for t_path in tracks_paths]
  82         title = self._search_regex(r'album_title : "(.*?)"', webpage, u'title')
  83         return {
  84             '_type': 'playlist',
  85             'title': title,
  86             'entries': entries,
  87         }