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?://)?
  23                     (?:(?:(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)/?(?:[?].*)?$)
  24                        |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+))
  25                     )
  26                     '''
  27     IE_NAME = u'soundcloud'
  28     _TEST = {
  29         u'url': u'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
  30         u'file': u'62986583.mp3',
  31         u'md5': u'ebef0a451b909710ed1d7787dddbf0d7',
  32         u'info_dict': {
  33             u"upload_date": u"20121011",
  34             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",
  35             u"uploader": u"E.T. ExTerrestrial Music",
  36             u"title": u"Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1"
  37         }
  38     }
  39
  40     _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28'
  41
  42     @classmethod
  43     def suitable(cls, url):
  44         return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None
  45
  46     def report_resolve(self, video_id):
  47         """Report information extraction."""
  48         self.to_screen(u'%s: Resolving id' % video_id)
  49
  50     @classmethod
  51     def _resolv_url(cls, url):
  52         return 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID
  53
  54     def _extract_info_dict(self, info, full_title=None):
  55         video_id = info['id']
  56         name = full_title or video_id
  57         self.report_extraction(name)
  58
  59         thumbnail = info['artwork_url']
  60         if thumbnail is not None:
  61             thumbnail = thumbnail.replace('-large', '-t500x500')
  62         return {
  63             'id':       info['id'],
  64             'url':      info['stream_url'] + '?client_id=' + self._CLIENT_ID,
  65             'uploader': info['user']['username'],
  66             'upload_date': unified_strdate(info['created_at']),
  67             'title':    info['title'],
  68             'ext':      u'mp3',
  69             'description': info['description'],
  70             'thumbnail': thumbnail,
  71         }
  72
  73     def _real_extract(self, url):
  74         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
  75         if mobj is None:
  76             raise ExtractorError(u'Invalid URL: %s' % url)
  77
  78         track_id = mobj.group('track_id')
  79         if track_id is not None:
  80             info_json_url = 'http://api.soundcloud.com/tracks/' + track_id + '.json?client_id=' + self._CLIENT_ID
  81             full_title = track_id
  82         else:
  83             # extract uploader (which is in the url)
  84             uploader = mobj.group(1)
  85             # extract simple title (uploader + slug of song title)
  86             slug_title =  mobj.group(2)
  87             full_title = '%s/%s' % (uploader, slug_title)
  88
  89             self.report_resolve(full_title)
  90
  91             url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title)
  92             info_json_url = self._resolv_url(url)
  93         info_json = self._download_webpage(info_json_url, full_title, u'Downloading info JSON')
  94
  95         info = json.loads(info_json)
  96         return self._extract_info_dict(info, full_title)
  97
  98 class SoundcloudSetIE(SoundcloudIE):
  99     _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$'
 100     IE_NAME = u'soundcloud:set'
 101     _TEST = {
 102         u"url":"https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep",
 103         u"playlist": [
 104             {
 105                 u"file":"30510138.mp3",
 106                 u"md5":"f9136bf103901728f29e419d2c70f55d",
 107                 u"info_dict": {
 108                     u"upload_date": u"20111213",
 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"D-D-Dance"
 112                 }
 113             },
 114             {
 115                 u"file":"47127625.mp3",
 116                 u"md5":"09b6758a018470570f8fd423c9453dd8",
 117                 u"info_dict": {
 118                     u"upload_date": u"20120521",
 119                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 120                     u"uploader": u"The Royal Concept",
 121                     u"title": u"The Royal Concept - Gimme Twice"
 122                 }
 123             },
 124             {
 125                 u"file":"47127627.mp3",
 126                 u"md5":"154abd4e418cea19c3b901f1e1306d9c",
 127                 u"info_dict": {
 128                     u"upload_date": u"20120521",
 129                     u"uploader": u"The Royal Concept",
 130                     u"title": u"Goldrushed"
 131                 }
 132             },
 133             {
 134                 u"file":"47127629.mp3",
 135                 u"md5":"2f5471edc79ad3f33a683153e96a79c1",
 136                 u"info_dict": {
 137                     u"upload_date": u"20120521",
 138                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 139                     u"uploader": u"The Royal Concept",
 140                     u"title": u"In the End"
 141                 }
 142             },
 143             {
 144                 u"file":"47127631.mp3",
 145                 u"md5":"f9ba87aa940af7213f98949254f1c6e2",
 146                 u"info_dict": {
 147                     u"upload_date": u"20120521",
 148                     u"description": u"The Royal Concept from Stockholm\r\nFilip / David / Povel / Magnus\r\nwww.theroyalconceptband.com",
 149                     u"uploader": u"The Royal Concept",
 150                     u"title": u"Knocked Up"
 151                 }
 152             },
 153             {
 154                 u"file":"75206121.mp3",
 155                 u"md5":"f9d1fe9406717e302980c30de4af9353",
 156                 u"info_dict": {
 157                     u"upload_date": u"20130116",
 158                     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!  ",
 159                     u"uploader": u"The Royal Concept",
 160                     u"title": u"World On Fire"
 161                 }
 162             }
 163         ]
 164     }
 165
 166     def _real_extract(self, url):
 167         mobj = re.match(self._VALID_URL, url)
 168         if mobj is None:
 169             raise ExtractorError(u'Invalid URL: %s' % url)
 170
 171         # extract uploader (which is in the url)
 172         uploader = mobj.group(1)
 173         # extract simple title (uploader + slug of song title)
 174         slug_title =  mobj.group(2)
 175         full_title = '%s/sets/%s' % (uploader, slug_title)
 176
 177         self.report_resolve(full_title)
 178
 179         url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
 180         resolv_url = self._resolv_url(url)
 181         info_json = self._download_webpage(resolv_url, full_title)
 182
 183         videos = []
 184         info = json.loads(info_json)
 185         if 'errors' in info:
 186             for err in info['errors']:
 187                 self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err['error_message']))
 188             return
 189
 190         self.report_extraction(full_title)
 191         return {'_type': 'playlist',
 192                 'entries': [self._extract_info_dict(track) for track in info['tracks']],
 193                 'id': info['id'],
 194                 'title': info['title'],
 195                 }