_ Git - youtube-dl/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     _TEST = {
  81         u"name": u"SoundcloudSet",
  82         u"url":"https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep",
  83         u"playlist": [
  84             {
  85                 u"file":"30510138.mp3",
  86                 u"md5":"f9136bf103901728f29e419d2c70f55d",
  87                 u"info_dict": {
  88                     u"upload_date": u"20111213",
  89                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
  90                     u"uploader": u"The Royal Concept",
  91                     u"title": u"D-D-Dance"
  92                 }
  93             },
  94             {
  95                 u"file":"47127625.mp3",
  96                 u"md5":"09b6758a018470570f8fd423c9453dd8",
  97                 u"info_dict": {
  98                     u"upload_date": u"20120521",
  99                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 100                     u"uploader": u"The Royal Concept",
 101                     u"title": u"The Royal Concept - Gimme Twice"
 102                 }
 103             },
 104             {
 105                 u"file":"47127627.mp3",
 106                 u"md5":"154abd4e418cea19c3b901f1e1306d9c",
 107                 u"info_dict": {
 108                     u"upload_date": u"20120521",
 109                     u"uploader": u"The Royal Concept",
 110                     u"title": u"Goldrushed"
 111                 }
 112             },
 113             {
 114                 u"file":"47127629.mp3",
 115                 u"md5":"2f5471edc79ad3f33a683153e96a79c1",
 116                 u"info_dict": {
 117                     u"upload_date": u"20120521",
 118                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 119                     u"uploader": u"The Royal Concept",
 120                     u"title": u"In the End"
 121                 }
 122             },
 123             {
 124                 u"file":"47127631.mp3",
 125                 u"md5":"f9ba87aa940af7213f98949254f1c6e2",
 126                 u"info_dict": {
 127                     u"upload_date": u"20120521",
 128                     u"description": u"The Royal Concept from Stockholm\r\nFilip / David / Povel / Magnus\r\nwww.theroyalconceptband.com",
 129                     u"uploader": u"The Royal Concept",
 130                     u"title": u"Knocked Up"
 131                 }
 132             },
 133             {
 134                 u"file":"75206121.mp3",
 135                 u"md5":"f9d1fe9406717e302980c30de4af9353",
 136                 u"info_dict": {
 137                     u"upload_date": u"20130116",
 138                     u"description": u"The unreleased track World on Fire premiered on the CW's hit show Arrow (8pm/7pm central).  \r\nAs a gift to our fans we would like to offer you a free download of the track!  ",
 139                     u"uploader": u"The Royal Concept",
 140                     u"title": u"World On Fire"
 141                 }
 142             }
 143         ]
 144     }
 145
 146     def report_resolve(self, video_id):
 147         """Report information extraction."""
 148         self.to_screen(u'%s: Resolving id' % video_id)
 149
 150     def _real_extract(self, url):
 151         mobj = re.match(self._VALID_URL, url)
 152         if mobj is None:
 153             raise ExtractorError(u'Invalid URL: %s' % url)
 154
 155         # extract uploader (which is in the url)
 156         uploader = mobj.group(1)
 157         # extract simple title (uploader + slug of song title)
 158         slug_title =  mobj.group(2)
 159         full_title = '%s/sets/%s' % (uploader, slug_title)
 160
 161         self.report_resolve(full_title)
 162
 163         url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
 164         resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
 165         info_json = self._download_webpage(resolv_url, full_title)
 166
 167         videos = []
 168         info = json.loads(info_json)
 169         if 'errors' in info:
 170             for err in info['errors']:
 171                 self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err['error_message']))
 172             return
 173
 174         self.report_extraction(full_title)
 175         for track in info['tracks']:
 176             video_id = track['id']
 177
 178             streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
 179             stream_json = self._download_webpage(streams_url, video_id, u'Downloading track info JSON')
 180
 181             self.report_extraction(video_id)
 182             streams = json.loads(stream_json)
 183             mediaURL = streams['http_mp3_128_url']
 184
 185             videos.append({
 186                 'id':       video_id,
 187                 'url':      mediaURL,
 188                 'uploader': track['user']['username'],
 189                 'upload_date':  unified_strdate(track['created_at']),
 190                 'title':    track['title'],
 191                 'ext':      u'mp3',
 192                 'description': track['description'],
 193             })
 194         return videos