+# encoding: utf-8
+from __future__ import unicode_literals
+
import json
import re
import itertools
"""
_VALID_URL = r'''^(?:https?://)?
- (?:(?:(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)/?(?:[?].*)?$)
+ (?:(?:(?:www\.|m\.)?soundcloud\.com/
+ (?P<uploader>[\w\d-]+)/
+ (?!sets/)(?P<title>[\w\d-]+)/?
+ (?P<token>[^?]+?)?(?:[?].*)?$)
|(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+))
- |(?P<widget>w.soundcloud.com/player/?.*?url=.*)
+ |(?P<player>(?:w|player|p.)\.soundcloud\.com/player/?.*?url=.*)
)
'''
- IE_NAME = u'soundcloud'
+ IE_NAME = 'soundcloud'
_TESTS = [
{
- u'url': u'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
- u'file': u'62986583.mp3',
- u'md5': u'ebef0a451b909710ed1d7787dddbf0d7',
- u'info_dict': {
- u"upload_date": u"20121011",
- 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",
- u"uploader": u"E.T. ExTerrestrial Music",
- u"title": u"Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1"
+ 'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
+ 'file': '62986583.mp3',
+ 'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
+ 'info_dict': {
+ "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"
}
},
# not streamable song
{
- u'url': u'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
- u'info_dict': {
- u'id': u'47127627',
- u'ext': u'mp3',
- u'title': u'Goldrushed',
- u'uploader': u'The Royal Concept',
- u'upload_date': u'20120521',
+ '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',
+ 'uploader': 'The Royal Concept',
+ 'upload_date': '20120521',
},
- u'params': {
+ 'params': {
# rtmp
- u'skip_download': True,
+ 'skip_download': True,
+ },
+ },
+ # private link
+ {
+ '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',
+ },
+ },
+ # downloadable song
+ {
+ 'url': 'https://soundcloud.com/simgretina/just-your-problem-baby-1',
+ 'md5': '56a8b69568acaa967b4c49f9d1d52d19',
+ 'info_dict': {
+ 'id': '105614606',
+ 'ext': 'wav',
+ 'title': 'Just Your Problem Baby (Acapella)',
+ 'description': 'Vocals',
+ 'uploader': 'Sim Gretina',
+ 'upload_date': '20130815',
},
},
]
def _resolv_url(cls, url):
return 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID
- def _extract_info_dict(self, info, full_title=None, quiet=False):
+ def _extract_info_dict(self, info, full_title=None, quiet=False, secret_token=None):
track_id = compat_str(info['id'])
name = full_title or track_id
if quiet:
thumbnail = info['artwork_url']
if thumbnail is not None:
thumbnail = thumbnail.replace('-large', '-t500x500')
- ext = info.get('original_format', u'mp3')
+ ext = 'mp3'
result = {
'id': track_id,
'uploader': info['user']['username'],
if info.get('downloadable', False):
# We can build a direct link to the song
format_url = (
- u'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(
+ 'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(
track_id, self._CLIENT_ID))
result['formats'] = [{
'format_id': 'download',
- 'ext': ext,
+ 'ext': info.get('original_format', 'mp3'),
'url': format_url,
'vcodec': 'none',
}]
else:
# We have to retrieve the url
+ streams_url = ('http://api.soundcloud.com/i1/tracks/{0}/streams?'
+ 'client_id={1}&secret_token={2}'.format(track_id, self._IPHONE_CLIENT_ID, secret_token))
stream_json = self._download_webpage(
- 'http://api.soundcloud.com/i1/tracks/{0}/streams?client_id={1}'.format(track_id, self._IPHONE_CLIENT_ID),
- track_id, u'Downloading track url')
+ streams_url,
+ track_id, 'Downloading track url')
formats = []
format_dict = json.loads(stream_json)
# We fallback to the stream_url in the original info, this
# cannot be always used, sometimes it can give an HTTP 404 error
formats.append({
- 'format_id': u'fallback',
+ 'format_id': 'fallback',
'url': info['stream_url'] + '?client_id=' + self._CLIENT_ID,
'ext': ext,
'vcodec': 'none',
})
- def format_pref(f):
+ for f in formats:
if f['format_id'].startswith('http'):
- return 2
+ f['protocol'] = 'http'
if f['format_id'].startswith('rtmp'):
- return 1
- return 0
+ f['protocol'] = 'rtmp'
- formats.sort(key=format_pref)
+ self._sort_formats(formats)
result['formats'] = formats
return result
raise ExtractorError(u'Invalid URL: %s' % url)
track_id = mobj.group('track_id')
+ token = None
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
- elif mobj.group('widget'):
+ elif mobj.group('player'):
query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
return self.url_result(query['url'][0], ie='Soundcloud')
else:
# extract uploader (which is in the url)
- uploader = mobj.group(1)
+ uploader = mobj.group('uploader')
# extract simple title (uploader + slug of song title)
- slug_title = mobj.group(2)
- full_title = '%s/%s' % (uploader, slug_title)
+ slug_title = mobj.group('title')
+ token = mobj.group('token')
+ full_title = resolve_title = '%s/%s' % (uploader, slug_title)
+ if token:
+ resolve_title += '/%s' % token
self.report_resolve(full_title)
- url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title)
+ url = 'http://soundcloud.com/%s' % resolve_title
info_json_url = self._resolv_url(url)
- info_json = self._download_webpage(info_json_url, full_title, u'Downloading info JSON')
+ info_json = self._download_webpage(info_json_url, full_title, 'Downloading info JSON')
info = json.loads(info_json)
- return self._extract_info_dict(info, full_title)
+ return self._extract_info_dict(info, full_title, secret_token=token)
class SoundcloudSetIE(SoundcloudIE):
_VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$'
- IE_NAME = u'soundcloud:set'
+ IE_NAME = 'soundcloud:set'
# it's in tests/test_playlists.py
_TESTS = []
class SoundcloudUserIE(SoundcloudIE):
- _VALID_URL = r'https?://(www\.)?soundcloud.com/(?P<user>[^/]+)(/?(tracks/)?)?(\?.*)?$'
- IE_NAME = u'soundcloud:user'
+ _VALID_URL = r'https?://(www\.)?soundcloud\.com/(?P<user>[^/]+)(/?(tracks/)?)?(\?.*)?$'
+ IE_NAME = 'soundcloud:user'
# it's in tests/test_playlists.py
_TESTS = []
url = 'http://soundcloud.com/%s/' % uploader
resolv_url = self._resolv_url(url)
user_json = self._download_webpage(resolv_url, uploader,
- u'Downloading user info')
+ 'Downloading user info')
user = json.loads(user_json)
tracks = []
})
tracks_url = 'http://api.soundcloud.com/users/%s/tracks.json?' % user['id'] + data
response = self._download_webpage(tracks_url, uploader,
- u'Downloading tracks page %s' % (i+1))
+ 'Downloading tracks page %s' % (i+1))
new_tracks = json.loads(response)
tracks.extend(self._extract_info_dict(track, quiet=True) for track in new_tracks)
if len(new_tracks) < 50:
Rapsys Git / youtubedl / blobdiff