_ 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     _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28'
  37
  38     def report_resolve(self, video_id):
  39         """Report information extraction."""
  40         self.to_screen(u'%s: Resolving id' % video_id)
  41
  42     @classmethod
  43     def _resolv_url(cls, url):
  44         return 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID
  45
  46     def _extract_info_dict(self, info, full_title=None):
  47         video_id = info['id']
  48         name = full_title or video_id
  49         self.report_extraction(name)
  50
  51         thumbnail = info['artwork_url']
  52         if thumbnail is not None:
  53             thumbnail = thumbnail.replace('-large', '-t500x500')
  54         return {
  55             'id':       info['id'],
  56             'url':      info['stream_url'] + '?client_id=' + self._CLIENT_ID,
  57             'uploader': info['user']['username'],
  58             'upload_date': unified_strdate(info['created_at']),
  59             'title':    info['title'],
  60             'ext':      u'mp3',
  61             'description': info['description'],
  62             'thumbnail': thumbnail,
  63         }
  64
  65     def _real_extract(self, url):
  66         mobj = re.match(self._VALID_URL, url)
  67         if mobj is None:
  68             raise ExtractorError(u'Invalid URL: %s' % url)
  69
  70         # extract uploader (which is in the url)
  71         uploader = mobj.group(1)
  72         # extract simple title (uploader + slug of song title)
  73         slug_title =  mobj.group(2)
  74         full_title = '%s/%s' % (uploader, slug_title)
  75
  76         self.report_resolve(full_title)
  77
  78         url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title)
  79         resolv_url = self._resolv_url(url)
  80         info_json = self._download_webpage(resolv_url, full_title, u'Downloading info JSON')
  81
  82         info = json.loads(info_json)
  83         return self._extract_info_dict(info, full_title)
  84
  85 class SoundcloudSetIE(SoundcloudIE):
  86     _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$'
  87     IE_NAME = u'soundcloud:set'
  88     _TEST = {
  89         u"url":"https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep",
  90         u"playlist": [
  91             {
  92                 u"file":"30510138.mp3",
  93                 u"md5":"f9136bf103901728f29e419d2c70f55d",
  94                 u"info_dict": {
  95                     u"upload_date": u"20111213",
  96                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
  97                     u"uploader": u"The Royal Concept",
  98                     u"title": u"D-D-Dance"
  99                 }
 100             },
 101             {
 102                 u"file":"47127625.mp3",
 103                 u"md5":"09b6758a018470570f8fd423c9453dd8",
 104                 u"info_dict": {
 105                     u"upload_date": u"20120521",
 106                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 107                     u"uploader": u"The Royal Concept",
 108                     u"title": u"The Royal Concept - Gimme Twice"
 109                 }
 110             },
 111             {
 112                 u"file":"47127627.mp3",
 113                 u"md5":"154abd4e418cea19c3b901f1e1306d9c",
 114                 u"info_dict": {
 115                     u"upload_date": u"20120521",
 116                     u"uploader": u"The Royal Concept",
 117                     u"title": u"Goldrushed"
 118                 }
 119             },
 120             {
 121                 u"file":"47127629.mp3",
 122                 u"md5":"2f5471edc79ad3f33a683153e96a79c1",
 123                 u"info_dict": {
 124                     u"upload_date": u"20120521",
 125                     u"description": u"The Royal Concept from Stockholm\r\nFilip / Povel / David / Magnus\r\nwww.royalconceptband.com",
 126                     u"uploader": u"The Royal Concept",
 127                     u"title": u"In the End"
 128                 }
 129             },
 130             {
 131                 u"file":"47127631.mp3",
 132                 u"md5":"f9ba87aa940af7213f98949254f1c6e2",
 133                 u"info_dict": {
 134                     u"upload_date": u"20120521",
 135                     u"description": u"The Royal Concept from Stockholm\r\nFilip / David / Povel / Magnus\r\nwww.theroyalconceptband.com",
 136                     u"uploader": u"The Royal Concept",
 137                     u"title": u"Knocked Up"
 138                 }
 139             },
 140             {
 141                 u"file":"75206121.mp3",
 142                 u"md5":"f9d1fe9406717e302980c30de4af9353",
 143                 u"info_dict": {
 144                     u"upload_date": u"20130116",
 145                     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!  ",
 146                     u"uploader": u"The Royal Concept",
 147                     u"title": u"World On Fire"
 148                 }
 149             }
 150         ]
 151     }
 152
 153     def _real_extract(self, url):
 154         mobj = re.match(self._VALID_URL, url)
 155         if mobj is None:
 156             raise ExtractorError(u'Invalid URL: %s' % url)
 157
 158         # extract uploader (which is in the url)
 159         uploader = mobj.group(1)
 160         # extract simple title (uploader + slug of song title)
 161         slug_title =  mobj.group(2)
 162         full_title = '%s/sets/%s' % (uploader, slug_title)
 163
 164         self.report_resolve(full_title)
 165
 166         url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
 167         resolv_url = self._resolv_url(url)
 168         info_json = self._download_webpage(resolv_url, full_title)
 169
 170         videos = []
 171         info = json.loads(info_json)
 172         if 'errors' in info:
 173             for err in info['errors']:
 174                 self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err['error_message']))
 175             return
 176
 177         self.report_extraction(full_title)
 178         return {'_type': 'playlist',
 179                 'entries': [self._extract_info_dict(track) for track in info['tracks']],
 180                 'id': info['id'],
 181                 'title': info['title'],
 182                 }