_ 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     ExtractorError,
   7 )
   8
   9
  10 class BandcampIE(InfoExtractor):
  11     _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)'
  12
  13     def _real_extract(self, url):
  14         mobj = re.match(self._VALID_URL, url)
  15         title = mobj.group('title')
  16         webpage = self._download_webpage(url, title)
  17         # We get the link to the free download page
  18         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  19         if m_download is None:
  20             raise ExtractorError(u'No free songs found')
  21
  22         download_link = m_download.group(1)
  23         id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
  24                        webpage, re.MULTILINE|re.DOTALL).group('id')
  25
  26         download_webpage = self._download_webpage(download_link, id,
  27                                                   'Downloading free downloads page')
  28         # We get the dictionary of the track from some javascrip code
  29         info = re.search(r'items: (.*?),$',
  30                          download_webpage, re.MULTILINE).group(1)
  31         info = json.loads(info)[0]
  32         # We pick mp3-320 for now, until format selection can be easily implemented.
  33         mp3_info = info[u'downloads'][u'mp3-320']
  34         # If we try to use this url it says the link has expired
  35         initial_url = mp3_info[u'url']
  36         re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
  37         m_url = re.match(re_url, initial_url)
  38         #We build the url we will use to get the final track url
  39         # This url is build in Bandcamp in the script download_bunde_*.js
  40         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'))
  41         final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
  42         # If we could correctly generate the .rand field the url would be
  43         #in the "download_url" key
  44         final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
  45
  46         track_info = {'id':id,
  47                       'title' : info[u'title'],
  48                       'ext' :   'mp3',
  49                       'url' :   final_url,
  50                       'thumbnail' : info[u'thumb_url'],
  51                       'uploader' :  info[u'artist']
  52                       }
  53
  54         return [track_info]