debian/control: Drop really ancient Recommends to have ffprobe.

[youtubedl] / youtube_dl / extractor / soundcloud.py

diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py
index 54256e1a29f02ad75dfdf828e7b4819455294a14..118ca483265cfda0a4419d93e8ce2fd174148be4 100644 (file)
--- a/youtube_dl/extractor/soundcloud.py
+++ b/youtube_dl/extractor/soundcloud.py
@@ -5,11 +5,12 @@ import re
 import itertools
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_str,
     compat_urlparse,
     compat_urllib_parse,
-
+)
+from ..utils import (
     ExtractorError,
     int_or_none,
     unified_strdate,
@@ -28,11 +29,11 @@ class SoundcloudIE(InfoExtractor):
     _VALID_URL = r'''(?x)^(?:https?://)?
                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
                             (?P<uploader>[\w\d-]+)/
-                            (?!sets/|likes/?(?:$|[?#]))
+                            (?!sets/|(?:likes|tracks)/?(?:$|[?#]))
                             (?P<title>[\w\d-]+)/?
                             (?P<token>[^?]+?)?(?:[?].*)?$)
                        |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+)
-                          (?:/?\?secret_token=(?P<secret_token>[^&]+?))?$)
+                          (?:/?\?secret_token=(?P<secret_token>[^&]+))?)
                        |(?P<player>(?:w|player|p.)\.soundcloud\.com/player/?.*?url=.*)
                     )
                     '''
@@ -159,7 +160,7 @@ class SoundcloudIE(InfoExtractor):
 
         # 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))
+                       'client_id={1}&secret_token={2}'.format(track_id, self._IPHONE_CLIENT_ID, secret_token))
         format_dict = self._download_json(
             streams_url,
             track_id, 'Downloading track url')
@@ -179,7 +180,7 @@ class SoundcloudIE(InfoExtractor):
                     'format_id': key,
                     'url': url,
                     'play_path': 'mp3:' + path,
-                    'ext': ext,
+                    'ext': 'flv',
                     'vcodec': 'none',
                 })
 
@@ -199,8 +200,9 @@ class SoundcloudIE(InfoExtractor):
                 if f['format_id'].startswith('rtmp'):
                     f['protocol'] = 'rtmp'
 
-            self._sort_formats(formats)
-            result['formats'] = formats
+        self._check_formats(formats, track_id)
+        self._sort_formats(formats)
+        result['formats'] = formats
 
         return result
 
@@ -219,19 +221,24 @@ class SoundcloudIE(InfoExtractor):
                 info_json_url += "&secret_token=" + token
         elif mobj.group('player'):
             query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
-            return self.url_result(query['url'][0])
+            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)
         else:
             # extract uploader (which is in the url)
             uploader = mobj.group('uploader')
             # extract simple title (uploader + slug of song title)
-            slug_title =  mobj.group('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' % resolve_title
             info_json_url = self._resolv_url(url)
         info = self._download_json(info_json_url, full_title, 'Downloading info JSON')
@@ -240,11 +247,12 @@ class SoundcloudIE(InfoExtractor):
 
 
 class SoundcloudSetIE(SoundcloudIE):
-    _VALID_URL = r'https?://(?:www\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[\w\d-]+)(?:/(?P<token>[^?/]+))?'
+    _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[\w\d-]+)(?:/(?P<token>[^?/]+))?'
     IE_NAME = 'soundcloud:set'
     _TESTS = [{
         'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
         'info_dict': {
+            'id': '2284613',
             'title': 'The Royal Concept EP',
         },
         'playlist_mincount': 6,
@@ -271,20 +279,19 @@ class SoundcloudSetIE(SoundcloudIE):
         info = self._download_json(resolv_url, full_title)
 
         if 'errors' in info:
-            for err in info['errors']:
-                self._downloader.report_error('unable to download video webpage: %s' % compat_str(err['error_message']))
-            return
+            msgs = (compat_str(err['error_message']) for err in info['errors'])
+            raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs))
 
         return {
             '_type': 'playlist',
             'entries': [self._extract_info_dict(track, secret_token=token) for track in info['tracks']],
-            'id': info['id'],
+            'id': '%s' % info['id'],
             'title': info['title'],
         }
 
 
 class SoundcloudUserIE(SoundcloudIE):
-    _VALID_URL = r'https?://(www\.)?soundcloud\.com/(?P<user>[^/]+)/?((?P<rsrc>tracks|likes)/?)?(\?.*)?$'
+    _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<user>[^/]+)/?((?P<rsrc>tracks|likes)/?)?(\?.*)?$'
     IE_NAME = 'soundcloud:user'
     _TESTS = [{
         'url': 'https://soundcloud.com/the-concept-band',
@@ -300,6 +307,9 @@ class SoundcloudUserIE(SoundcloudIE):
             'title': 'The Royal Concept',
         },
         'playlist_mincount': 1,
+    }, {
+        'url': 'https://soundcloud.com/the-akashic-chronicler/tracks',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -329,7 +339,7 @@ class SoundcloudUserIE(SoundcloudIE):
             if len(new_entries) == 0:
                 self.to_screen('%s: End page received' % uploader)
                 break
-            entries.extend(self._extract_info_dict(e, quiet=True) for e in new_entries)
+            entries.extend(self.url_result(e['permalink_url'], 'Soundcloud') for e in new_entries)
 
         return {
             '_type': 'playlist',
@@ -371,7 +381,7 @@ class SoundcloudPlaylistIE(SoundcloudIE):
 
         entries = [
             self._extract_info_dict(t, quiet=True, secret_token=token)
-                for t in data['tracks']]
+            for t in data['tracks']]
 
         return {
             '_type': 'playlist',