]>
jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/soundcloud.py
2 from __future__
import unicode_literals
7 from . common
import InfoExtractor
19 class SoundcloudIE ( InfoExtractor
):
20 """Information extractor for soundcloud.com
21 To access the media, the uid of the song and a stream token
22 must be extracted from the page source and the script must make
23 a request to media.soundcloud.com/crossdomain.xml. Then
24 the media can be grabbed by requesting from an url composed
25 of the stream token and uid
28 _VALID_URL
= r
'''(?x)^(?:https?://)?
29 (?:(?:(?:www\.|m\.)?soundcloud\.com/
30 (?P<uploader>[\w\d-]+)/
31 (?!sets/|likes/?(?:$|[?#]))
33 (?P<token>[^?]+?)?(?:[?].*)?$)
34 |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+))
35 |(?P<player>(?:w|player|p.)\.soundcloud\.com/player/?.*?url=.*)
38 IE_NAME
= 'soundcloud'
41 'url' : 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy' ,
42 'file' : '62986583.mp3' ,
43 'md5' : 'ebef0a451b909710ed1d7787dddbf0d7' ,
45 "upload_date" : "20121011" ,
46 "description" : "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" ,
47 "uploader" : "E.T. ExTerrestrial Music" ,
48 "title" : "Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1" ,
54 'url' : 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep' ,
58 'title' : 'Goldrushed' ,
59 'description' : 'From Stockholm Sweden \r\n Povel / Magnus / Filip / David \r\n www.theroyalconcept.com' ,
60 'uploader' : 'The Royal Concept' ,
61 'upload_date' : '20120521' ,
66 'skip_download' : True ,
71 'url' : 'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp' ,
72 'md5' : 'aa0dd32bfea9b0c5ef4f02aacd080604' ,
76 'title' : 'Youtube - Dl Test Video \'\' Ä↭' ,
77 'uploader' : 'jaimeMF' ,
78 'description' : 'test chars: \"\' / \\ ä↭' ,
79 'upload_date' : '20131209' ,
85 'url' : 'https://soundcloud.com/oddsamples/bus-brakes' ,
86 'md5' : '7624f2351f8a3b2e7cd51522496e7631' ,
90 'title' : 'Bus Brakes' ,
91 'description' : 'md5:0170be75dd395c96025d210d261c784e' ,
92 'uploader' : 'oddsamples' ,
93 'upload_date' : '20140109' ,
99 _CLIENT_ID
= 'b45b1aa10f1ac2941910a7f0d10f8e28'
100 _IPHONE_CLIENT_ID
= '376f225bf427445fc4bfb6b99b72e0bf'
102 def report_resolve ( self
, video_id
):
103 """Report information extraction."""
104 self
. to_screen ( ' %s : Resolving id' % video_id
)
107 def _resolv_url ( cls
, url
):
108 return 'http://api.soundcloud.com/resolve.json?url=' + url
+ '&client_id=' + cls
._ CLIENT
_ ID
110 def _extract_info_dict ( self
, info
, full_title
= None , quiet
= False , secret_token
= None ):
111 track_id
= compat_str ( info
[ 'id' ])
112 name
= full_title
or track_id
114 self
. report_extraction ( name
)
116 thumbnail
= info
[ 'artwork_url' ]
117 if thumbnail
is not None :
118 thumbnail
= thumbnail
. replace ( '-large' , '-t500x500' )
122 'uploader' : info
[ 'user' ][ 'username' ],
123 'upload_date' : unified_strdate ( info
[ 'created_at' ]),
124 'title' : info
[ 'title' ],
125 'description' : info
[ 'description' ],
126 'thumbnail' : thumbnail
,
127 'duration' : int_or_none ( info
. get ( 'duration' ), 1000 ),
130 if info
. get ( 'downloadable' , False ):
131 # We can build a direct link to the song
133 'https://api.soundcloud.com/tracks/ {0} /download?client_id= {1} ' . format (
134 track_id
, self
._ CLIENT
_ ID
))
136 'format_id' : 'download' ,
137 'ext' : info
. get ( 'original_format' , 'mp3' ),
143 # We have to retrieve the url
144 streams_url
= ( 'http://api.soundcloud.com/i1/tracks/ {0} /streams?'
145 'client_id= {1} &secret_token= {2} ' . format ( track_id
, self
._ IPHONE
_ CLIENT
_ ID
, secret_token
))
146 format_dict
= self
._ download
_ json
(
148 track_id
, 'Downloading track url' )
150 for key
, stream_url
in format_dict
. items ():
151 if key
. startswith ( 'http' ):
158 elif key
. startswith ( 'rtmp' ):
159 # The url doesn't have an rtmp app, we have to extract the playpath
160 url
, path
= stream_url
. split ( 'mp3:' , 1 )
164 'play_path' : 'mp3:' + path
,
170 # We fallback to the stream_url in the original info, this
171 # cannot be always used, sometimes it can give an HTTP 404 error
173 'format_id' : 'fallback' ,
174 'url' : info
[ 'stream_url' ] + '?client_id=' + self
._ CLIENT
_ ID
,
180 if f
[ 'format_id' ]. startswith ( 'http' ):
181 f
[ 'protocol' ] = 'http'
182 if f
[ 'format_id' ]. startswith ( 'rtmp' ):
183 f
[ 'protocol' ] = 'rtmp'
185 self
._ sort
_ formats
( formats
)
186 result
[ 'formats' ] = formats
190 def _real_extract ( self
, url
):
191 mobj
= re
. match ( self
._ VALID
_U RL
, url
, flags
= re
. VERBOSE
)
193 raise ExtractorError ( 'Invalid URL: %s ' % url
)
195 track_id
= mobj
. group ( 'track_id' )
197 if track_id
is not None :
198 info_json_url
= 'http://api.soundcloud.com/tracks/' + track_id
+ '.json?client_id=' + self
._ CLIENT
_ ID
199 full_title
= track_id
200 elif mobj
. group ( 'player' ):
201 query
= compat_urlparse
. parse_qs ( compat_urlparse
. urlparse ( url
). query
)
202 return self
. url_result ( query
[ 'url' ][ 0 ])
204 # extract uploader (which is in the url)
205 uploader
= mobj
. group ( 'uploader' )
206 # extract simple title (uploader + slug of song title)
207 slug_title
= mobj
. group ( 'title' )
208 token
= mobj
. group ( 'token' )
209 full_title
= resolve_title
= ' %s / %s ' % ( uploader
, slug_title
)
211 resolve_title
+= '/ %s ' % token
213 self
. report_resolve ( full_title
)
215 url
= 'http://soundcloud.com/ %s ' % resolve_title
216 info_json_url
= self
._ resolv
_u rl
( url
)
217 info
= self
._ download
_ json
( info_json_url
, full_title
, 'Downloading info JSON' )
219 return self
._ extract
_ info
_ dict
( info
, full_title
, secret_token
= token
)
222 class SoundcloudSetIE ( SoundcloudIE
):
223 _VALID_URL
= r
'https?://(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)'
224 IE_NAME
= 'soundcloud:set'
226 'url' : 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep' ,
228 'title' : 'The Royal Concept EP' ,
230 'playlist_mincount' : 6 ,
233 def _real_extract ( self
, url
):
234 mobj
= re
. match ( self
._ VALID
_U RL
, url
)
236 # extract uploader (which is in the url)
237 uploader
= mobj
. group ( 1 )
238 # extract simple title (uploader + slug of song title)
239 slug_title
= mobj
. group ( 2 )
240 full_title
= ' %s /sets/ %s ' % ( uploader
, slug_title
)
242 self
. report_resolve ( full_title
)
244 url
= 'http://soundcloud.com/ %s /sets/ %s ' % ( uploader
, slug_title
)
245 resolv_url
= self
._ resolv
_u rl
( url
)
246 info
= self
._ download
_ json
( resolv_url
, full_title
)
249 for err
in info
[ 'errors' ]:
250 self
._ downloader
. report_error ( 'unable to download video webpage: %s ' % compat_str ( err
[ 'error_message' ]))
255 'entries' : [ self
._ extract
_ info
_ dict
( track
) for track
in info
[ 'tracks' ]],
257 'title' : info
[ 'title' ],
261 class SoundcloudUserIE ( SoundcloudIE
):
262 _VALID_URL
= r
'https?://(www\.)?soundcloud\.com/(?P<user>[^/]+)/?((?P<rsrc>tracks|likes)/?)?(\?.*)?$'
263 IE_NAME
= 'soundcloud:user'
265 'url' : 'https://soundcloud.com/the-concept-band' ,
268 'title' : 'The Royal Concept' ,
270 'playlist_mincount' : 12
272 'url' : 'https://soundcloud.com/the-concept-band/likes' ,
275 'title' : 'The Royal Concept' ,
277 'playlist_mincount' : 1 ,
280 def _real_extract ( self
, url
):
281 mobj
= re
. match ( self
._ VALID
_U RL
, url
)
282 uploader
= mobj
. group ( 'user' )
283 resource
= mobj
. group ( 'rsrc' )
286 elif resource
== 'likes' :
287 resource
= 'favorites'
289 url
= 'http://soundcloud.com/ %s /' % uploader
290 resolv_url
= self
._ resolv
_u rl
( url
)
291 user
= self
._ download
_ json
(
292 resolv_url
, uploader
, 'Downloading user info' )
293 base_url
= 'http://api.soundcloud.com/users/ %s / %s .json?' % ( uploader
, resource
)
296 for i
in itertools
. count ():
297 data
= compat_urllib_parse
. urlencode ({
300 'client_id' : self
._ CLIENT
_ ID
,
302 new_entries
= self
._ download
_ json
(
303 base_url
+ data
, uploader
, 'Downloading track page %s ' % ( i
+ 1 ))
304 if len ( new_entries
) == 0 :
305 self
. to_screen ( ' %s : End page received' % uploader
)
307 entries
. extend ( self
._ extract
_ info
_ dict
( e
, quiet
= True ) for e
in new_entries
)
311 'id' : compat_str ( user
[ 'id' ]),
312 'title' : user
[ 'username' ],
317 class SoundcloudPlaylistIE ( SoundcloudIE
):
318 _VALID_URL
= r
'https?://api\.soundcloud\.com/playlists/(?P<id>[0-9]+)'
319 IE_NAME
= 'soundcloud:playlist'
323 'url' : 'http://api.soundcloud.com/playlists/4110309' ,
326 'title' : 'TILT Brass - Bowery Poetry Club, August \' 03 [Non-Site SCR 02]' ,
327 'description' : 're:.*?TILT Brass - Bowery Poetry Club' ,
333 def _real_extract ( self
, url
):
334 mobj
= re
. match ( self
._ VALID
_U RL
, url
)
335 playlist_id
= mobj
. group ( 'id' )
336 base_url
= ' %s //api.soundcloud.com/playlists/ %s .json?' % ( self
. http_scheme (), playlist_id
)
338 data
= compat_urllib_parse
. urlencode ({
339 'client_id' : self
._ CLIENT
_ ID
,
341 data
= self
._ download
_ json
(
342 base_url
+ data
, playlist_id
, 'Downloading playlist' )
345 self
._ extract
_ info
_ dict
( t
, quiet
= True ) for t
in data
[ 'tracks' ]]
350 'title' : data
. get ( 'title' ),
351 'description' : data
. get ( 'description' ),