Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/soundcloud.py

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_str,
   7
   8     ExtractorError,
   9     unified_strdate,
  10 )
  11
  12
  13 class SoundcloudIE(InfoExtractor):
  14     """Information extractor for soundcloud.com
  15        To access the media, the uid of the song and a stream token
  16        must be extracted from the page source and the script must make
  17        a request to media.soundcloud.com/crossdomain.xml. Then
  18        the media can be grabbed by requesting from an url composed
  19        of the stream token and uid
  20      """
  21
  22     _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)'
  23     IE_NAME = u'soundcloud'
  24
  25     def report_resolve(self, video_id):
  26         """Report information extraction."""
  27         self.to_screen(u'%s: Resolving id' % video_id)
  28
  29     def _real_extract(self, url):
  30         mobj = re.match(self._VALID_URL, url)
  31         if mobj is None:
  32             raise ExtractorError(u'Invalid URL: %s' % url)
  33
  34         # extract uploader (which is in the url)
  35         uploader = mobj.group(1)
  36         # extract simple title (uploader + slug of song title)
  37         slug_title =  mobj.group(2)
  38         full_title = '%s/%s' % (uploader, slug_title)
  39
  40         self.report_resolve(full_title)
  41
  42         url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title)
  43         resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
  44         info_json = self._download_webpage(resolv_url, full_title, u'Downloading info JSON')
  45
  46         info = json.loads(info_json)
  47         video_id = info['id']
  48         self.report_extraction(full_title)
  49
  50         streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
  51         stream_json = self._download_webpage(streams_url, full_title,
  52                                              u'Downloading stream definitions',
  53                                              u'unable to download stream definitions')
  54
  55         streams = json.loads(stream_json)
  56         mediaURL = streams['http_mp3_128_url']
  57         upload_date = unified_strdate(info['created_at'])
  58
  59         return [{
  60             'id':       info['id'],
  61             'url':      mediaURL,
  62             'uploader': info['user']['username'],
  63             'upload_date': upload_date,
  64             'title':    info['title'],
  65             'ext':      u'mp3',
  66             'description': info['description'],
  67         }]
  68
  69 class SoundcloudSetIE(InfoExtractor):
  70     """Information extractor for soundcloud.com sets
  71        To access the media, the uid of the song and a stream token
  72        must be extracted from the page source and the script must make
  73        a request to media.soundcloud.com/crossdomain.xml. Then
  74        the media can be grabbed by requesting from an url composed
  75        of the stream token and uid
  76      """
  77
  78     _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)'
  79     IE_NAME = u'soundcloud:set'
  80
  81     def report_resolve(self, video_id):
  82         """Report information extraction."""
  83         self.to_screen(u'%s: Resolving id' % video_id)
  84
  85     def _real_extract(self, url):
  86         mobj = re.match(self._VALID_URL, url)
  87         if mobj is None:
  88             raise ExtractorError(u'Invalid URL: %s' % url)
  89
  90         # extract uploader (which is in the url)
  91         uploader = mobj.group(1)
  92         # extract simple title (uploader + slug of song title)
  93         slug_title =  mobj.group(2)
  94         full_title = '%s/sets/%s' % (uploader, slug_title)
  95
  96         self.report_resolve(full_title)
  97
  98         url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
  99         resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
 100         info_json = self._download_webpage(resolv_url, full_title)
 101
 102         videos = []
 103         info = json.loads(info_json)
 104         if 'errors' in info:
 105             for err in info['errors']:
 106                 self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err['error_message']))
 107             return
 108
 109         self.report_extraction(full_title)
 110         for track in info['tracks']:
 111             video_id = track['id']
 112
 113             streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
 114             stream_json = self._download_webpage(streams_url, video_id, u'Downloading track info JSON')
 115
 116             self.report_extraction(video_id)
 117             streams = json.loads(stream_json)
 118             mediaURL = streams['http_mp3_128_url']
 119
 120             videos.append({
 121                 'id':       video_id,
 122                 'url':      mediaURL,
 123                 'uploader': track['user']['username'],
 124                 'upload_date':  unified_strdate(track['created_at']),
 125                 'title':    track['title'],
 126                 'ext':      u'mp3',
 127                 'description': track['description'],
 128             })
 129         return videos