2013-12-09 11:08:58 -05:00
# encoding: utf-8
2014-01-16 21:29:41 -05:00
from __future__ import unicode_literals
2013-06-23 14:57:44 -04:00
import re
2013-09-14 15:41:49 -04:00
import itertools
2013-06-23 14:57:44 -04:00
2015-10-17 12:23:46 -04:00
from . common import (
InfoExtractor ,
SearchInfoExtractor
)
2014-12-13 06:24:42 -05:00
from . . compat import (
2013-06-23 14:57:44 -04:00
compat_str ,
2013-08-21 11:06:37 -04:00
compat_urlparse ,
2016-03-25 15:46:57 -04:00
compat_urllib_parse_urlencode ,
2014-12-13 06:24:42 -05:00
)
from . . utils import (
2013-06-23 14:57:44 -04:00
ExtractorError ,
2014-06-07 09:51:01 -04:00
int_or_none ,
2013-06-23 14:57:44 -04:00
unified_strdate ,
)
class SoundcloudIE ( InfoExtractor ) :
""" Information extractor for soundcloud.com
To access the media , the uid of the song and a stream token
must be extracted from the page source and the script must make
a request to media . soundcloud . com / crossdomain . xml . Then
the media can be grabbed by requesting from an url composed
of the stream token and uid
"""
2014-05-04 21:12:41 -04:00
_VALID_URL = r ''' (?x)^(?:https?://)?
2013-12-19 10:39:01 -05:00
( ? : ( ? : ( ? : www \. | m \. ) ? soundcloud \. com /
2013-12-09 13:57:00 -05:00
( ? P < uploader > [ \w \d - ] + ) /
2016-08-30 14:56:15 -04:00
( ? ! ( ? : tracks | sets ( ? : / . + ? ) ? | reposts | likes | spotlight ) / ? ( ? : $ | [ ? #]))
2014-08-27 18:58:24 -04:00
( ? P < title > [ \w \d - ] + ) / ?
2013-12-09 11:08:58 -05:00
( ? P < token > [ ^ ? ] + ? ) ? ( ? : [ ? ] . * ) ? $ )
2014-09-18 02:02:03 -04:00
| ( ? : api \. soundcloud \. com / tracks / ( ? P < track_id > \d + )
2014-12-03 05:49:53 -05:00
( ? : / ? \? secret_token = ( ? P < secret_token > [ ^ & ] + ) ) ? )
2014-01-02 10:18:51 -05:00
| ( ? P < player > ( ? : w | player | p . ) \. soundcloud \. com / player / ? . * ? url = . * )
2013-07-24 08:39:21 -04:00
)
'''
2014-01-16 21:29:41 -05:00
IE_NAME = ' soundcloud '
2013-11-09 12:06:09 -05:00
_TESTS = [
{
2014-01-16 21:29:41 -05:00
' url ' : ' http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy ' ,
' md5 ' : ' ebef0a451b909710ed1d7787dddbf0d7 ' ,
' info_dict ' : {
2014-10-25 16:32:01 -04:00
' id ' : ' 62986583 ' ,
' ext ' : ' mp3 ' ,
' upload_date ' : ' 20121011 ' ,
' 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 ' ,
' uploader ' : ' E.T. ExTerrestrial Music ' ,
' title ' : ' Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1 ' ,
' duration ' : 143 ,
2016-09-18 05:53:05 -04:00
' license ' : ' all-rights-reserved ' ,
2013-11-09 12:06:09 -05:00
}
} ,
# not streamable song
{
2014-01-16 21:29:41 -05:00
' url ' : ' https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep ' ,
' info_dict ' : {
' id ' : ' 47127627 ' ,
' ext ' : ' mp3 ' ,
' title ' : ' Goldrushed ' ,
2014-03-09 07:20:34 -04:00
' description ' : ' From Stockholm Sweden \r \n Povel / Magnus / Filip / David \r \n www.theroyalconcept.com ' ,
2014-01-16 21:29:41 -05:00
' uploader ' : ' The Royal Concept ' ,
' upload_date ' : ' 20120521 ' ,
2014-06-07 09:51:01 -04:00
' duration ' : 227 ,
2016-09-18 05:53:05 -04:00
' license ' : ' all-rights-reserved ' ,
2013-11-09 12:06:09 -05:00
} ,
2014-01-16 21:29:41 -05:00
' params ' : {
2013-11-09 12:06:09 -05:00
# rtmp
2014-01-16 21:29:41 -05:00
' skip_download ' : True ,
2013-11-09 12:06:09 -05:00
} ,
} ,
2013-12-09 11:08:58 -05:00
# private link
{
2014-01-16 21:29:41 -05:00
' url ' : ' https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp ' ,
' md5 ' : ' aa0dd32bfea9b0c5ef4f02aacd080604 ' ,
' info_dict ' : {
' id ' : ' 123998367 ' ,
' ext ' : ' mp3 ' ,
' title ' : ' Youtube - Dl Test Video \' \' Ä↭ ' ,
' uploader ' : ' jaimeMF ' ,
' description ' : ' test chars: \" \' / \\ ä↭ ' ,
' upload_date ' : ' 20131209 ' ,
2014-06-07 09:51:01 -04:00
' duration ' : 9 ,
2016-09-18 05:53:05 -04:00
' license ' : ' all-rights-reserved ' ,
2013-12-09 11:08:58 -05:00
} ,
} ,
2014-09-18 02:02:03 -04:00
# private link (alt format)
{
' url ' : ' https://api.soundcloud.com/tracks/123998367?secret_token=s-8Pjrp ' ,
' md5 ' : ' aa0dd32bfea9b0c5ef4f02aacd080604 ' ,
' info_dict ' : {
' id ' : ' 123998367 ' ,
' ext ' : ' mp3 ' ,
' title ' : ' Youtube - Dl Test Video \' \' Ä↭ ' ,
' uploader ' : ' jaimeMF ' ,
' description ' : ' test chars: \" \' / \\ ä↭ ' ,
' upload_date ' : ' 20131209 ' ,
' duration ' : 9 ,
2016-09-18 05:53:05 -04:00
' license ' : ' all-rights-reserved ' ,
2014-09-18 02:02:03 -04:00
} ,
} ,
2013-12-10 07:04:21 -05:00
# downloadable song
{
2014-07-15 08:18:06 -04:00
' url ' : ' https://soundcloud.com/oddsamples/bus-brakes ' ,
2014-07-22 19:41:44 -04:00
' md5 ' : ' 7624f2351f8a3b2e7cd51522496e7631 ' ,
2014-01-16 21:29:41 -05:00
' info_dict ' : {
2014-07-15 08:18:06 -04:00
' id ' : ' 128590877 ' ,
2014-07-22 19:41:44 -04:00
' ext ' : ' mp3 ' ,
2014-07-15 08:18:06 -04:00
' title ' : ' Bus Brakes ' ,
2014-10-25 16:32:01 -04:00
' description ' : ' md5:0053ca6396e8d2fd7b7e1595ef12ab66 ' ,
2014-07-15 08:18:06 -04:00
' uploader ' : ' oddsamples ' ,
' upload_date ' : ' 20140109 ' ,
' duration ' : 17 ,
2016-09-18 05:53:05 -04:00
' license ' : ' cc-by-sa ' ,
2013-12-10 07:04:21 -05:00
} ,
} ,
2013-11-09 12:06:09 -05:00
]
2013-06-23 14:57:44 -04:00
2015-09-23 09:54:40 -04:00
_CLIENT_ID = ' 02gUJC0hH2ct1EGOcYXQIzRFU91c72Ea '
2013-11-21 07:16:19 -05:00
_IPHONE_CLIENT_ID = ' 376f225bf427445fc4bfb6b99b72e0bf '
2013-07-24 08:05:14 -04:00
2016-07-28 11:16:05 -04:00
@staticmethod
def _extract_urls ( webpage ) :
return [ m . group ( ' url ' ) for m in re . finditer (
r ' <iframe[^>]+src=([ " \' ])(?P<url>(?:https?://)?(?:w \ .)?soundcloud \ .com/player.+?) \ 1 ' ,
webpage ) ]
2013-06-23 14:57:44 -04:00
def report_resolve ( self , video_id ) :
""" Report information extraction. """
2014-03-23 21:15:31 -04:00
self . to_screen ( ' %s : Resolving id ' % video_id )
2013-06-23 14:57:44 -04:00
2013-07-24 08:05:14 -04:00
@classmethod
def _resolv_url ( cls , url ) :
return ' http://api.soundcloud.com/resolve.json?url= ' + url + ' &client_id= ' + cls . _CLIENT_ID
2013-12-09 11:08:58 -05:00
def _extract_info_dict ( self , info , full_title = None , quiet = False , secret_token = None ) :
2013-11-09 12:06:09 -05:00
track_id = compat_str ( info [ ' id ' ] )
name = full_title or track_id
2013-11-25 14:30:41 -05:00
if quiet :
2013-09-14 15:41:49 -04:00
self . report_extraction ( name )
2016-09-20 10:55:57 -04:00
thumbnail = info . get ( ' artwork_url ' )
if isinstance ( thumbnail , compat_str ) :
2013-07-24 08:05:14 -04:00
thumbnail = thumbnail . replace ( ' -large ' , ' -t500x500 ' )
2014-01-16 21:29:41 -05:00
ext = ' mp3 '
2013-11-09 12:06:09 -05:00
result = {
2013-11-25 14:30:41 -05:00
' id ' : track_id ,
2016-09-20 10:55:57 -04:00
' uploader ' : info . get ( ' user ' , { } ) . get ( ' username ' ) ,
' upload_date ' : unified_strdate ( info . get ( ' created_at ' ) ) ,
2013-11-25 14:30:41 -05:00
' title ' : info [ ' title ' ] ,
2016-09-20 10:55:57 -04:00
' description ' : info . get ( ' description ' ) ,
2013-07-24 08:05:14 -04:00
' thumbnail ' : thumbnail ,
2014-06-07 09:51:01 -04:00
' duration ' : int_or_none ( info . get ( ' duration ' ) , 1000 ) ,
2014-10-26 14:08:36 -04:00
' webpage_url ' : info . get ( ' permalink_url ' ) ,
2016-09-20 10:55:57 -04:00
' license ' : info . get ( ' license ' ) ,
2013-07-24 08:05:14 -04:00
}
2014-03-23 21:21:17 -04:00
formats = [ ]
2013-11-09 12:06:09 -05:00
if info . get ( ' downloadable ' , False ) :
2013-11-21 07:16:19 -05:00
# We can build a direct link to the song
2013-11-25 14:30:41 -05:00
format_url = (
2014-01-16 21:29:41 -05:00
' https://api.soundcloud.com/tracks/ {0} /download?client_id= {1} ' . format (
2013-11-25 14:30:41 -05:00
track_id , self . _CLIENT_ID ) )
2014-03-23 21:21:17 -04:00
formats . append ( {
2013-11-25 14:30:41 -05:00
' format_id ' : ' download ' ,
2014-01-16 21:29:41 -05:00
' ext ' : info . get ( ' original_format ' , ' mp3 ' ) ,
2013-11-25 14:30:41 -05:00
' url ' : format_url ,
2013-11-25 16:34:56 -05:00
' vcodec ' : ' none ' ,
2014-03-23 21:21:17 -04:00
' preference ' : 10 ,
} )
# We have to retrieve the url
streams_url = ( ' http://api.soundcloud.com/i1/tracks/ {0} /streams? '
2014-11-23 15:39:15 -05:00
' client_id= {1} &secret_token= {2} ' . format ( track_id , self . _IPHONE_CLIENT_ID , secret_token ) )
2014-05-04 21:12:41 -04:00
format_dict = self . _download_json (
2014-03-23 21:21:17 -04:00
streams_url ,
track_id , ' Downloading track url ' )
for key , stream_url in format_dict . items ( ) :
if key . startswith ( ' http ' ) :
formats . append ( {
' format_id ' : key ,
' ext ' : ext ,
' url ' : stream_url ,
' vcodec ' : ' none ' ,
} )
elif key . startswith ( ' rtmp ' ) :
# The url doesn't have an rtmp app, we have to extract the playpath
url , path = stream_url . split ( ' mp3: ' , 1 )
formats . append ( {
' format_id ' : key ,
' url ' : url ,
' play_path ' : ' mp3: ' + path ,
2015-03-02 11:47:07 -05:00
' ext ' : ' flv ' ,
2014-03-23 21:21:17 -04:00
' vcodec ' : ' none ' ,
} )
2013-11-25 14:30:41 -05:00
if not formats :
2013-11-21 07:16:19 -05:00
# We fallback to the stream_url in the original info, this
# cannot be always used, sometimes it can give an HTTP 404 error
2013-11-25 14:30:41 -05:00
formats . append ( {
2014-01-16 21:29:41 -05:00
' format_id ' : ' fallback ' ,
2013-11-25 14:30:41 -05:00
' url ' : info [ ' stream_url ' ] + ' ?client_id= ' + self . _CLIENT_ID ,
' ext ' : ext ,
2013-11-25 16:34:56 -05:00
' vcodec ' : ' none ' ,
2013-11-25 14:30:41 -05:00
} )
2014-01-16 21:29:41 -05:00
for f in formats :
2013-11-25 14:30:41 -05:00
if f [ ' format_id ' ] . startswith ( ' http ' ) :
2014-01-16 21:29:41 -05:00
f [ ' protocol ' ] = ' http '
2013-11-25 14:30:41 -05:00
if f [ ' format_id ' ] . startswith ( ' rtmp ' ) :
2014-01-16 21:29:41 -05:00
f [ ' protocol ' ] = ' rtmp '
2013-11-25 14:30:41 -05:00
2015-03-02 11:39:32 -05:00
self . _check_formats ( formats , track_id )
self . _sort_formats ( formats )
result [ ' formats ' ] = formats
2013-11-21 07:16:19 -05:00
2013-11-09 12:06:09 -05:00
return result
2013-07-24 08:05:14 -04:00
2013-06-23 14:57:44 -04:00
def _real_extract ( self , url ) :
2013-07-24 08:39:21 -04:00
mobj = re . match ( self . _VALID_URL , url , flags = re . VERBOSE )
2013-06-23 14:57:44 -04:00
if mobj is None :
2014-03-23 21:15:31 -04:00
raise ExtractorError ( ' Invalid URL: %s ' % url )
2013-06-23 14:57:44 -04:00
2013-07-24 08:39:21 -04:00
track_id = mobj . group ( ' track_id ' )
2016-09-18 05:53:05 -04:00
2013-07-24 08:39:21 -04:00
if track_id is not None :
info_json_url = ' http://api.soundcloud.com/tracks/ ' + track_id + ' .json?client_id= ' + self . _CLIENT_ID
full_title = track_id
2014-09-18 02:02:03 -04:00
token = mobj . group ( ' secret_token ' )
if token :
2016-02-14 04:37:17 -05:00
info_json_url + = ' &secret_token= ' + token
2014-01-02 10:18:51 -05:00
elif mobj . group ( ' player ' ) :
2013-08-21 11:06:37 -04:00
query = compat_urlparse . parse_qs ( compat_urlparse . urlparse ( url ) . query )
2015-04-17 04:46:25 -04:00
real_url = query [ ' url ' ] [ 0 ]
# If the token is in the query of the original url we have to
# manually add it
if ' secret_token ' in query :
real_url + = ' ?secret_token= ' + query [ ' secret_token ' ] [ 0 ]
return self . url_result ( real_url )
2013-07-24 08:39:21 -04:00
else :
# extract uploader (which is in the url)
2013-12-09 11:08:58 -05:00
uploader = mobj . group ( ' uploader ' )
2013-07-24 08:39:21 -04:00
# extract simple title (uploader + slug of song title)
2014-11-23 15:20:46 -05:00
slug_title = mobj . group ( ' title ' )
2013-12-09 11:08:58 -05:00
token = mobj . group ( ' token ' )
full_title = resolve_title = ' %s / %s ' % ( uploader , slug_title )
if token :
resolve_title + = ' / %s ' % token
2014-11-23 14:41:03 -05:00
2013-07-24 08:39:21 -04:00
self . report_resolve ( full_title )
2014-11-23 14:41:03 -05:00
2013-12-09 11:08:58 -05:00
url = ' http://soundcloud.com/ %s ' % resolve_title
2013-07-24 08:39:21 -04:00
info_json_url = self . _resolv_url ( url )
2014-05-04 21:12:41 -04:00
info = self . _download_json ( info_json_url , full_title , ' Downloading info JSON ' )
2013-06-23 14:57:44 -04:00
2013-12-09 11:08:58 -05:00
return self . _extract_info_dict ( info , full_title , secret_token = token )
2013-06-23 14:57:44 -04:00
2014-05-04 21:12:41 -04:00
2016-09-24 08:29:49 -04:00
class SoundcloudPlaylistBaseIE ( SoundcloudIE ) :
2016-09-24 08:18:01 -04:00
@staticmethod
def _extract_id ( e ) :
return compat_str ( e [ ' id ' ] ) if e . get ( ' id ' ) else None
def _extract_track_entries ( self , tracks ) :
return [
self . url_result (
track [ ' permalink_url ' ] , SoundcloudIE . ie_key ( ) ,
video_id = self . _extract_id ( track ) )
for track in tracks if track . get ( ' permalink_url ' ) ]
2016-09-24 08:29:49 -04:00
class SoundcloudSetIE ( SoundcloudPlaylistBaseIE ) :
2015-03-30 11:03:38 -04:00
_VALID_URL = r ' https?://(?:(?:www|m) \ .)?soundcloud \ .com/(?P<uploader>[ \ w \ d-]+)/sets/(?P<slug_title>[ \ w \ d-]+)(?:/(?P<token>[^?/]+))? '
2014-01-16 21:29:41 -05:00
IE_NAME = ' soundcloud:set '
2014-08-27 18:58:24 -04:00
_TESTS = [ {
' url ' : ' https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep ' ,
' info_dict ' : {
2015-02-01 09:24:38 -05:00
' id ' : ' 2284613 ' ,
2014-08-27 18:58:24 -04:00
' title ' : ' The Royal Concept EP ' ,
} ,
' playlist_mincount ' : 6 ,
2016-08-30 14:56:15 -04:00
} , {
' url ' : ' https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token ' ,
' only_matching ' : True ,
2014-08-27 18:58:24 -04:00
} ]
2013-06-23 14:57:44 -04:00
def _real_extract ( self , url ) :
mobj = re . match ( self . _VALID_URL , url )
# extract uploader (which is in the url)
2014-09-18 05:35:11 -04:00
uploader = mobj . group ( ' uploader ' )
2013-06-23 14:57:44 -04:00
# extract simple title (uploader + slug of song title)
2014-09-18 05:35:11 -04:00
slug_title = mobj . group ( ' slug_title ' )
2013-06-23 14:57:44 -04:00
full_title = ' %s /sets/ %s ' % ( uploader , slug_title )
2014-09-18 05:35:11 -04:00
url = ' http://soundcloud.com/ %s /sets/ %s ' % ( uploader , slug_title )
token = mobj . group ( ' token ' )
if token :
full_title + = ' / ' + token
url + = ' / ' + token
2013-06-23 14:57:44 -04:00
self . report_resolve ( full_title )
2013-07-24 08:05:14 -04:00
resolv_url = self . _resolv_url ( url )
2014-05-04 21:12:41 -04:00
info = self . _download_json ( resolv_url , full_title )
2013-06-23 14:57:44 -04:00
if ' errors ' in info :
2015-04-17 13:24:30 -04:00
msgs = ( compat_str ( err [ ' error_message ' ] ) for err in info [ ' errors ' ] )
raise ExtractorError ( ' unable to download video webpage: %s ' % ' , ' . join ( msgs ) )
2013-06-23 14:57:44 -04:00
2016-09-24 08:18:01 -04:00
entries = self . _extract_track_entries ( info [ ' tracks ' ] )
2015-07-25 05:47:53 -04:00
2014-08-27 18:58:24 -04:00
return {
' _type ' : ' playlist ' ,
2015-07-25 05:47:53 -04:00
' entries ' : entries ,
2015-02-01 09:24:38 -05:00
' id ' : ' %s ' % info [ ' id ' ] ,
2014-08-27 18:58:24 -04:00
' title ' : info [ ' title ' ] ,
}
2013-09-14 15:41:49 -04:00
2016-09-24 08:29:49 -04:00
class SoundcloudUserIE ( SoundcloudPlaylistBaseIE ) :
2015-07-31 17:50:03 -04:00
_VALID_URL = r ''' (?x)
https ? : / /
( ? : ( ? : www | m ) \. ) ? soundcloud \. com /
( ? P < user > [ ^ / ] + )
( ? : /
( ? P < rsrc > tracks | sets | reposts | likes | spotlight )
) ?
/ ? ( ? : [ ? #].*)?$
'''
2014-01-16 21:29:41 -05:00
IE_NAME = ' soundcloud:user '
2014-08-27 18:58:24 -04:00
_TESTS = [ {
2015-07-30 14:54:26 -04:00
' url ' : ' https://soundcloud.com/the-akashic-chronicler ' ,
2014-08-27 18:58:24 -04:00
' info_dict ' : {
2015-07-30 14:54:26 -04:00
' id ' : ' 114582580 ' ,
' title ' : ' The Akashic Chronicler (All) ' ,
2014-08-27 18:58:24 -04:00
} ,
2016-09-24 08:18:01 -04:00
' playlist_mincount ' : 74 ,
2014-08-27 18:58:24 -04:00
} , {
2015-07-30 14:54:26 -04:00
' url ' : ' https://soundcloud.com/the-akashic-chronicler/tracks ' ,
2014-08-27 18:58:24 -04:00
' info_dict ' : {
2015-07-30 14:54:26 -04:00
' id ' : ' 114582580 ' ,
' title ' : ' The Akashic Chronicler (Tracks) ' ,
2014-08-27 18:58:24 -04:00
} ,
2016-09-24 08:18:01 -04:00
' playlist_mincount ' : 37 ,
2015-06-24 10:49:22 -04:00
} , {
2015-07-30 14:54:26 -04:00
' url ' : ' https://soundcloud.com/the-akashic-chronicler/sets ' ,
' info_dict ' : {
' id ' : ' 114582580 ' ,
' title ' : ' The Akashic Chronicler (Playlists) ' ,
} ,
2016-09-24 08:18:01 -04:00
' playlist_mincount ' : 2 ,
2015-07-30 14:54:26 -04:00
} , {
' url ' : ' https://soundcloud.com/the-akashic-chronicler/reposts ' ,
' info_dict ' : {
' id ' : ' 114582580 ' ,
' title ' : ' The Akashic Chronicler (Reposts) ' ,
} ,
2015-08-21 20:30:00 -04:00
' playlist_mincount ' : 7 ,
2015-07-30 14:54:26 -04:00
} , {
' url ' : ' https://soundcloud.com/the-akashic-chronicler/likes ' ,
' info_dict ' : {
' id ' : ' 114582580 ' ,
' title ' : ' The Akashic Chronicler (Likes) ' ,
} ,
2015-08-21 20:30:00 -04:00
' playlist_mincount ' : 321 ,
2015-07-30 14:54:26 -04:00
} , {
' url ' : ' https://soundcloud.com/grynpyret/spotlight ' ,
' info_dict ' : {
' id ' : ' 7098329 ' ,
2016-09-24 08:18:01 -04:00
' title ' : ' GRYNPYRET (Spotlight) ' ,
2015-07-30 14:54:26 -04:00
} ,
' playlist_mincount ' : 1 ,
2014-08-27 18:58:24 -04:00
} ]
2013-09-14 15:41:49 -04:00
2015-07-30 14:54:26 -04:00
_API_BASE = ' https://api.soundcloud.com '
_API_V2_BASE = ' https://api-v2.soundcloud.com '
_BASE_URL_MAP = {
' all ' : ' %s /profile/soundcloud:users: %% s ' % _API_V2_BASE ,
' tracks ' : ' %s /users/ %% s/tracks ' % _API_BASE ,
' sets ' : ' %s /users/ %% s/playlists ' % _API_V2_BASE ,
' reposts ' : ' %s /profile/soundcloud:users: %% s/reposts ' % _API_V2_BASE ,
' likes ' : ' %s /users/ %% s/likes ' % _API_V2_BASE ,
' spotlight ' : ' %s /users/ %% s/spotlight ' % _API_V2_BASE ,
}
_TITLE_MAP = {
' all ' : ' All ' ,
' tracks ' : ' Tracks ' ,
' sets ' : ' Playlists ' ,
' reposts ' : ' Reposts ' ,
' likes ' : ' Likes ' ,
' spotlight ' : ' Spotlight ' ,
}
2013-09-14 15:41:49 -04:00
def _real_extract ( self , url ) :
mobj = re . match ( self . _VALID_URL , url )
uploader = mobj . group ( ' user ' )
url = ' http://soundcloud.com/ %s / ' % uploader
resolv_url = self . _resolv_url ( url )
2014-05-04 21:12:41 -04:00
user = self . _download_json (
resolv_url , uploader , ' Downloading user info ' )
2015-07-30 14:54:26 -04:00
resource = mobj . group ( ' rsrc ' ) or ' all '
base_url = self . _BASE_URL_MAP [ resource ] % user [ ' id ' ]
2016-01-07 14:54:31 -05:00
COMMON_QUERY = {
' limit ' : 50 ,
' client_id ' : self . _CLIENT_ID ,
' linked_partitioning ' : ' 1 ' ,
}
query = COMMON_QUERY . copy ( )
query [ ' offset ' ] = 0
2016-03-25 15:46:57 -04:00
next_href = base_url + ' ? ' + compat_urllib_parse_urlencode ( query )
2013-09-14 15:41:49 -04:00
2014-05-04 21:12:41 -04:00
entries = [ ]
2013-09-14 15:41:49 -04:00
for i in itertools . count ( ) :
2015-07-30 14:54:26 -04:00
response = self . _download_json (
next_href , uploader , ' Downloading track page %s ' % ( i + 1 ) )
collection = response [ ' collection ' ]
if not collection :
2013-09-14 15:41:49 -04:00
break
2015-07-30 14:54:26 -04:00
def resolve_permalink_url ( candidates ) :
for cand in candidates :
if isinstance ( cand , dict ) :
permalink_url = cand . get ( ' permalink_url ' )
2016-09-24 08:18:01 -04:00
entry_id = self . _extract_id ( cand )
2015-07-30 14:54:26 -04:00
if permalink_url and permalink_url . startswith ( ' http ' ) :
2016-09-24 08:18:01 -04:00
return permalink_url , entry_id
2015-07-30 14:54:26 -04:00
for e in collection :
2016-09-24 08:18:01 -04:00
permalink_url , entry_id = resolve_permalink_url ( ( e , e . get ( ' track ' ) , e . get ( ' playlist ' ) ) )
2015-07-30 14:54:26 -04:00
if permalink_url :
2016-09-24 08:18:01 -04:00
entries . append ( self . url_result ( permalink_url , video_id = entry_id ) )
2015-07-30 14:54:26 -04:00
2016-01-07 14:54:31 -05:00
next_href = response . get ( ' next_href ' )
if not next_href :
break
parsed_next_href = compat_urlparse . urlparse ( response [ ' next_href ' ] )
qs = compat_urlparse . parse_qs ( parsed_next_href . query )
qs . update ( COMMON_QUERY )
next_href = compat_urlparse . urlunparse (
2016-03-25 15:46:57 -04:00
parsed_next_href . _replace ( query = compat_urllib_parse_urlencode ( qs , True ) ) )
2013-09-14 15:41:49 -04:00
return {
' _type ' : ' playlist ' ,
' id ' : compat_str ( user [ ' id ' ] ) ,
2015-07-30 14:54:26 -04:00
' title ' : ' %s ( %s ) ' % ( user [ ' username ' ] , self . _TITLE_MAP [ resource ] ) ,
2014-05-04 21:12:41 -04:00
' entries ' : entries ,
}
2016-09-24 08:29:49 -04:00
class SoundcloudPlaylistIE ( SoundcloudPlaylistBaseIE ) :
2014-09-18 12:57:04 -04:00
_VALID_URL = r ' https?://api \ .soundcloud \ .com/playlists/(?P<id>[0-9]+)(?:/? \ ?secret_token=(?P<token>[^&]+?))?$ '
2014-05-04 21:12:41 -04:00
IE_NAME = ' soundcloud:playlist '
2014-09-18 12:57:04 -04:00
_TESTS = [ {
' url ' : ' http://api.soundcloud.com/playlists/4110309 ' ,
' info_dict ' : {
' id ' : ' 4110309 ' ,
' title ' : ' TILT Brass - Bowery Poetry Club, August \' 03 [Non-Site SCR 02] ' ,
' description ' : ' re:.*?TILT Brass - Bowery Poetry Club ' ,
} ,
' playlist_count ' : 6 ,
} ]
2014-05-04 21:12:41 -04:00
def _real_extract ( self , url ) :
mobj = re . match ( self . _VALID_URL , url )
playlist_id = mobj . group ( ' id ' )
base_url = ' %s //api.soundcloud.com/playlists/ %s .json? ' % ( self . http_scheme ( ) , playlist_id )
2014-09-18 05:35:11 -04:00
data_dict = {
2014-05-04 21:12:41 -04:00
' client_id ' : self . _CLIENT_ID ,
2014-09-18 05:35:11 -04:00
}
token = mobj . group ( ' token ' )
if token :
data_dict [ ' secret_token ' ] = token
2016-03-25 15:46:57 -04:00
data = compat_urllib_parse_urlencode ( data_dict )
2014-05-04 21:12:41 -04:00
data = self . _download_json (
base_url + data , playlist_id , ' Downloading playlist ' )
2016-09-24 08:18:01 -04:00
entries = self . _extract_track_entries ( data [ ' tracks ' ] )
2014-05-04 21:12:41 -04:00
return {
' _type ' : ' playlist ' ,
' id ' : playlist_id ,
' title ' : data . get ( ' title ' ) ,
' description ' : data . get ( ' description ' ) ,
' entries ' : entries ,
2013-09-14 15:41:49 -04:00
}
2015-10-17 12:23:46 -04:00
class SoundcloudSearchIE ( SearchInfoExtractor , SoundcloudIE ) :
IE_NAME = ' soundcloud:search '
IE_DESC = ' Soundcloud search '
2015-10-30 18:56:07 -04:00
_MAX_RESULTS = float ( ' inf ' )
2015-10-17 12:23:46 -04:00
_TESTS = [ {
' url ' : ' scsearch15:post-avant jazzcore ' ,
' info_dict ' : {
' title ' : ' post-avant jazzcore ' ,
} ,
' playlist_count ' : 15 ,
} ]
_SEARCH_KEY = ' scsearch '
2015-10-30 18:56:07 -04:00
_MAX_RESULTS_PER_PAGE = 200
_DEFAULT_RESULTS_PER_PAGE = 50
2015-10-17 15:18:42 -04:00
_API_V2_BASE = ' https://api-v2.soundcloud.com '
2015-10-17 12:23:46 -04:00
def _get_collection ( self , endpoint , collection_id , * * query ) :
2015-11-21 13:49:58 -05:00
limit = min (
2015-10-30 18:56:07 -04:00
query . get ( ' limit ' , self . _DEFAULT_RESULTS_PER_PAGE ) ,
self . _MAX_RESULTS_PER_PAGE )
2015-11-21 10:21:21 -05:00
query [ ' limit ' ] = limit
2015-10-17 12:23:46 -04:00
query [ ' client_id ' ] = self . _CLIENT_ID
query [ ' linked_partitioning ' ] = ' 1 '
2015-11-21 10:21:21 -05:00
query [ ' offset ' ] = 0
2016-03-25 15:46:57 -04:00
data = compat_urllib_parse_urlencode ( query )
2015-11-21 10:21:21 -05:00
next_url = ' {0} {1} ? {2} ' . format ( self . _API_V2_BASE , endpoint , data )
2015-10-17 12:23:46 -04:00
collected_results = 0
2015-11-21 10:21:21 -05:00
for i in itertools . count ( 1 ) :
2015-11-21 09:04:35 -05:00
response = self . _download_json (
2015-11-21 10:21:21 -05:00
next_url , collection_id , ' Downloading page {0} ' . format ( i ) ,
2015-11-21 09:04:35 -05:00
' Unable to download API page ' )
2015-10-17 12:23:46 -04:00
2015-11-21 10:21:21 -05:00
collection = response . get ( ' collection ' , [ ] )
if not collection :
break
2015-10-17 12:23:46 -04:00
2015-11-21 10:21:21 -05:00
collection = list ( filter ( bool , collection ) )
2015-10-17 12:23:46 -04:00
collected_results + = len ( collection )
2015-11-21 10:21:21 -05:00
for item in collection :
yield self . url_result ( item [ ' uri ' ] , SoundcloudIE . ie_key ( ) )
2015-10-17 12:23:46 -04:00
2015-11-21 10:21:21 -05:00
if not collection or collected_results > = limit :
2015-10-17 12:23:46 -04:00
break
2015-11-21 09:04:35 -05:00
next_url = response . get ( ' next_href ' )
2015-11-21 10:21:21 -05:00
if not next_url :
break
2015-10-17 12:23:46 -04:00
def _get_n_results ( self , query , n ) :
2015-11-21 10:21:21 -05:00
tracks = self . _get_collection ( ' /search/tracks ' , query , limit = n , q = query )
return self . playlist_result ( tracks , playlist_title = query )