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     compat_urlparse,
   8
   9     ExtractorError,
  10     unified_strdate,
  11 )
  12
  13
  14 class SoundcloudIE(InfoExtractor):
  15     """Information extractor for soundcloud.com
  16        To access the media, the uid of the song and a stream token
  17        must be extracted from the page source and the script must make
  18        a request to media.soundcloud.com/crossdomain.xml. Then
  19        the media can be grabbed by requesting from an url composed
  20        of the stream token and uid
  21      """
  22
  23     _VALID_URL = r'''^(?:https?://)?
  24                     (?:(?:(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)/?(?:[?].*)?$)
  25                        |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+))
  26                        |(?P<widget>w.soundcloud.com/player/?.*?url=.*)
  27                     )
  28                     '''
  29     IE_NAME = u'soundcloud'
  30     _TEST = {
  31         u'url': u'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
  32         u'file': u'62986583.mp3',
  33         u'md5': u'ebef0a451b909710ed1d7787dddbf0d7',
  34         u'info_dict': {
  35             u"upload_date": u"20121011",
  36             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",
  37             u"uploader": u"E.T. ExTerrestrial Music",
  38             u"title": u"Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1"
  39         }
  40     }
  41
  42     _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28'
  43
  44     @classmethod
  45     def suitable(cls, url):
  46         return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None
  47
  48     def report_resolve(self, video_id):
  49         """Report information extraction."""
  50         self.to_screen(u'%s: Resolving id' % video_id)
  51
  52     @classmethod
  53     def _resolv_url(cls, url):
  54         return 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID
  55
  56     def _extract_info_dict(self, info, full_title=None):
  57         video_id = info['id']
  58         name = full_title or video_id
  59         self.report_extraction(name)
  60
  61         thumbnail = info['artwork_url']
  62         if thumbnail is not None:
  63             thumbnail = thumbnail.replace('-large', '-t500x500')
  64         return {
  65             'id':       info['id'],
  66             'url':      info['stream_url'] + '?client_id=' + self._CLIENT_ID,
  67             'uploader': info['user']['username'],
  68             'upload_date': unified_strdate(info['created_at']),
  69             'title':    info['title'],
  70             'ext':      u'mp3',
  71             'description': info['description'],
  72             'thumbnail': thumbnail,
  73         }
  74
  75     def _real_extract(self, url):
  76         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
  77         if mobj is None:
  78             raise ExtractorError(u'Invalid URL: %s' % url)
  79
  80         track_id = mobj.group('track_id')
  81         if track_id is not None:
  82             info_json_url = 'http://api.soundcloud.com/tracks/' + track_id + '.json?client_id=' + self._CLIENT_ID
  83             full_title = track_id
  84         elif mobj.group('widget'):
  85             query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
  86             return self.url_result(query['url'][0], ie='Soundcloud')
  87         else:
  88             # extract uploader (which is in the url)
  89             uploader = mobj.group(1)
  90             # extract simple title (uploader + slug of song title)
  91             slug_title =  mobj.group(2)
  92             full_title = '%s/%s' % (uploader, slug_title)
  93
  94             self.report_resolve(full_title)
  95
  96             url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title)
  97             info_json_url = self._resolv_url(url)
  98         info_json = self._download_webpage(info_json_url, full_title, u'Downloading info JSON')
  99
 100         info = json.loads(info_json)
 101         return self._extract_info_dict(info, full_title)
 102
 103 class SoundcloudSetIE(SoundcloudIE):
 104     _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$'
 105     IE_NAME = u'soundcloud:set'
 106     _TEST = {
 107         u"url":"https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep",
 108         u"playlist": [
 109             {
 110                 u"file":"30510138.mp3",
 111                 u"md5":"f9136bf103901728f29e419d2c70f55d",
 112                 u"info_dict": {
 113                     u"upload_date": u"20111213",
 114                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 115                     u"uploader": u"The Royal Concept",
 116                     u"title": u"D-D-Dance"
 117                 }
 118             },
 119             {
 120                 u"file":"47127625.mp3",
 121                 u"md5":"09b6758a018470570f8fd423c9453dd8",
 122                 u"info_dict": {
 123                     u"upload_date": u"20120521",
 124                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 125                     u"uploader": u"The Royal Concept",
 126                     u"title": u"The Royal Concept - Gimme Twice"
 127                 }
 128             },
 129             {
 130                 u"file":"47127627.mp3",
 131                 u"md5":"154abd4e418cea19c3b901f1e1306d9c",
 132                 u"info_dict": {
 133                     u"upload_date": u"20120521",
 134                     u"uploader": u"The Royal Concept",
 135                     u"title": u"Goldrushed"
 136                 }
 137             },
 138             {
 139                 u"file":"47127629.mp3",
 140                 u"md5":"2f5471edc79ad3f33a683153e96a79c1",
 141                 u"info_dict": {
 142                     u"upload_date": u"20120521",
 143                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 144                     u"uploader": u"The Royal Concept",
 145                     u"title": u"In the End"
 146                 }
 147             },
 148             {
 149                 u"file":"47127631.mp3",
 150                 u"md5":"f9ba87aa940af7213f98949254f1c6e2",
 151                 u"info_dict": {
 152                     u"upload_date": u"20120521",
 153                     u"description": u"The Royal Concept from Stockholm\r\nFilip / David / Povel / Magnus\r\nwww.theroyalconceptband.com",
 154                     u"uploader": u"The Royal Concept",
 155                     u"title": u"Knocked Up"
 156                 }
 157             },
 158             {
 159                 u"file":"75206121.mp3",
 160                 u"md5":"f9d1fe9406717e302980c30de4af9353",
 161                 u"info_dict": {
 162                     u"upload_date": u"20130116",
 163                     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!  ",
 164                     u"uploader": u"The Royal Concept",
 165                     u"title": u"World On Fire"
 166                 }
 167             }
 168         ]
 169     }
 170
 171     def _real_extract(self, url):
 172         mobj = re.match(self._VALID_URL, url)
 173         if mobj is None:
 174             raise ExtractorError(u'Invalid URL: %s' % url)
 175
 176         # extract uploader (which is in the url)
 177         uploader = mobj.group(1)
 178         # extract simple title (uploader + slug of song title)
 179         slug_title =  mobj.group(2)
 180         full_title = '%s/sets/%s' % (uploader, slug_title)
 181
 182         self.report_resolve(full_title)
 183
 184         url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
 185         resolv_url = self._resolv_url(url)
 186         info_json = self._download_webpage(resolv_url, full_title)
 187
 188         videos = []
 189         info = json.loads(info_json)
 190         if 'errors' in info:
 191             for err in info['errors']:
 192                 self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err['error_message']))
 193             return
 194
 195         self.report_extraction(full_title)
 196         return {'_type': 'playlist',
 197                 'entries': [self._extract_info_dict(track) for track in info['tracks']],
 198                 'id': info['id'],
 199                 'title': info['title'],
 200                 }