_ 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     _TEST = {
  25         u'url': u'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
  26         u'file': u'62986583.mp3',
  27         u'md5': u'ebef0a451b909710ed1d7787dddbf0d7',
  28         u'info_dict': {
  29             u"upload_date": u"20121011",
  30             u"description": u"No Downloads untill we record the finished version this weekend, i was too pumped n i had to post it , earl is prolly gonna b hella p.o'd",
  31             u"uploader": u"E.T. ExTerrestrial Music",
  32             u"title": u"Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1"
  33         }
  34     }
  35
  36     def report_resolve(self, video_id):
  37         """Report information extraction."""
  38         self.to_screen(u'%s: Resolving id' % video_id)
  39
  40     def _real_extract(self, url):
  41         mobj = re.match(self._VALID_URL, url)
  42         if mobj is None:
  43             raise ExtractorError(u'Invalid URL: %s' % url)
  44
  45         # extract uploader (which is in the url)
  46         uploader = mobj.group(1)
  47         # extract simple title (uploader + slug of song title)
  48         slug_title =  mobj.group(2)
  49         full_title = '%s/%s' % (uploader, slug_title)
  50
  51         self.report_resolve(full_title)
  52
  53         url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title)
  54         resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
  55         info_json = self._download_webpage(resolv_url, full_title, u'Downloading info JSON')
  56
  57         info = json.loads(info_json)
  58         video_id = info['id']
  59         self.report_extraction(full_title)
  60
  61         streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
  62         stream_json = self._download_webpage(streams_url, full_title,
  63                                              u'Downloading stream definitions',
  64                                              u'unable to download stream definitions')
  65
  66         streams = json.loads(stream_json)
  67         mediaURL = streams['http_mp3_128_url']
  68         upload_date = unified_strdate(info['created_at'])
  69
  70         return [{
  71             'id':       info['id'],
  72             'url':      mediaURL,
  73             'uploader': info['user']['username'],
  74             'upload_date': upload_date,
  75             'title':    info['title'],
  76             'ext':      u'mp3',
  77             'description': info['description'],
  78         }]
  79
  80 class SoundcloudSetIE(InfoExtractor):
  81     """Information extractor for soundcloud.com sets
  82        To access the media, the uid of the song and a stream token
  83        must be extracted from the page source and the script must make
  84        a request to media.soundcloud.com/crossdomain.xml. Then
  85        the media can be grabbed by requesting from an url composed
  86        of the stream token and uid
  87      """
  88
  89     _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$'
  90     IE_NAME = u'soundcloud:set'
  91     _TEST = {
  92         u"url":"https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep",
  93         u"playlist": [
  94             {
  95                 u"file":"30510138.mp3",
  96                 u"md5":"f9136bf103901728f29e419d2c70f55d",
  97                 u"info_dict": {
  98                     u"upload_date": u"20111213",
  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"D-D-Dance"
 102                 }
 103             },
 104             {
 105                 u"file":"47127625.mp3",
 106                 u"md5":"09b6758a018470570f8fd423c9453dd8",
 107                 u"info_dict": {
 108                     u"upload_date": u"20120521",
 109                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 110                     u"uploader": u"The Royal Concept",
 111                     u"title": u"The Royal Concept - Gimme Twice"
 112                 }
 113             },
 114             {
 115                 u"file":"47127627.mp3",
 116                 u"md5":"154abd4e418cea19c3b901f1e1306d9c",
 117                 u"info_dict": {
 118                     u"upload_date": u"20120521",
 119                     u"uploader": u"The Royal Concept",
 120                     u"title": u"Goldrushed"
 121                 }
 122             },
 123             {
 124                 u"file":"47127629.mp3",
 125                 u"md5":"2f5471edc79ad3f33a683153e96a79c1",
 126                 u"info_dict": {
 127                     u"upload_date": u"20120521",
 128                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 129                     u"uploader": u"The Royal Concept",
 130                     u"title": u"In the End"
 131                 }
 132             },
 133             {
 134                 u"file":"47127631.mp3",
 135                 u"md5":"f9ba87aa940af7213f98949254f1c6e2",
 136                 u"info_dict": {
 137                     u"upload_date": u"20120521",
 138                     u"description": u"The Royal Concept from Stockholm\r\nFilip / David / Povel / Magnus\r\nwww.theroyalconceptband.com",
 139                     u"uploader": u"The Royal Concept",
 140                     u"title": u"Knocked Up"
 141                 }
 142             },
 143             {
 144                 u"file":"75206121.mp3",
 145                 u"md5":"f9d1fe9406717e302980c30de4af9353",
 146                 u"info_dict": {
 147                     u"upload_date": u"20130116",
 148                     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!  ",
 149                     u"uploader": u"The Royal Concept",
 150                     u"title": u"World On Fire"
 151                 }
 152             }
 153         ]
 154     }
 155
 156     def report_resolve(self, video_id):
 157         """Report information extraction."""
 158         self.to_screen(u'%s: Resolving id' % video_id)
 159
 160     def _real_extract(self, url):
 161         mobj = re.match(self._VALID_URL, url)
 162         if mobj is None:
 163             raise ExtractorError(u'Invalid URL: %s' % url)
 164
 165         # extract uploader (which is in the url)
 166         uploader = mobj.group(1)
 167         # extract simple title (uploader + slug of song title)
 168         slug_title =  mobj.group(2)
 169         full_title = '%s/sets/%s' % (uploader, slug_title)
 170
 171         self.report_resolve(full_title)
 172
 173         url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
 174         resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
 175         info_json = self._download_webpage(resolv_url, full_title)
 176
 177         videos = []
 178         info = json.loads(info_json)
 179         if 'errors' in info:
 180             for err in info['errors']:
 181                 self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err['error_message']))
 182             return
 183
 184         self.report_extraction(full_title)
 185         for track in info['tracks']:
 186             video_id = track['id']
 187
 188             streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
 189             stream_json = self._download_webpage(streams_url, video_id, u'Downloading track info JSON')
 190
 191             self.report_extraction(video_id)
 192             streams = json.loads(stream_json)
 193             mediaURL = streams['http_mp3_128_url']
 194
 195             videos.append({
 196                 'id':       video_id,
 197                 'url':      mediaURL,
 198                 'uploader': track['user']['username'],
 199                 'upload_date':  unified_strdate(track['created_at']),
 200                 'title':    track['title'],
 201                 'ext':      u'mp3',
 202                 'description': track['description'],
 203             })
 204         return videos