Imported Upstream version 2014.11.21

[youtubedl] / youtube_dl / extractor / soundcloud.py

diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py
index 393b5f17c53d5ed216b53ac0d1ff2941cb1d24f7..54256e1a29f02ad75dfdf828e7b4819455294a14 100644 (file)
--- a/youtube_dl/extractor/soundcloud.py
+++ b/youtube_dl/extractor/soundcloud.py
@@ -1,7 +1,6 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
 # encoding: utf-8
 from __future__ import unicode_literals
 

-import json

 import re
 import itertools
 
 import re
 import itertools
 


@@ -12,6 +11,7 @@ from ..utils import (
     compat_urllib_parse,
 
     ExtractorError,
     compat_urllib_parse,
 
     ExtractorError,

+    int_or_none,

     unified_strdate,
 )
 
     unified_strdate,
 )
 


@@ -25,12 +25,14 @@ class SoundcloudIE(InfoExtractor):
        of the stream token and uid
      """
 
        of the stream token and uid
      """
 

-    _VALID_URL = r'''^(?:https?://)?
+    _VALID_URL = r'''(?x)^(?:https?://)?

                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
                             (?P<uploader>[\w\d-]+)/
                     (?:(?:(?:www\.|m\.)?soundcloud\.com/
                             (?P<uploader>[\w\d-]+)/

-                            (?!sets/)(?P<title>[\w\d-]+)/?
+                            (?!sets/|likes/?(?:$|[?#]))
+                            (?P<title>[\w\d-]+)/?

                             (?P<token>[^?]+?)?(?:[?].*)?$)
                             (?P<token>[^?]+?)?(?:[?].*)?$)

-                       |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+))
+                       |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+)
+                          (?:/?\?secret_token=(?P<secret_token>[^&]+?))?$)

                        |(?P<player>(?:w|player|p.)\.soundcloud\.com/player/?.*?url=.*)
                     )
                     '''
                        |(?P<player>(?:w|player|p.)\.soundcloud\.com/player/?.*?url=.*)
                     )
                     '''


@@ -38,13 +40,15 @@ class SoundcloudIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
     _TESTS = [
         {
             'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',

-            'file': '62986583.mp3',

             'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
             'info_dict': {
             '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"
+                '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,

             }
         },
         # not streamable song
             }
         },
         # not streamable song


@@ -54,8 +58,10 @@ class SoundcloudIE(InfoExtractor):
                 'id': '47127627',
                 'ext': 'mp3',
                 'title': 'Goldrushed',
                 'id': '47127627',
                 'ext': 'mp3',
                 'title': 'Goldrushed',

+                'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',

                 'uploader': 'The Royal Concept',
                 'upload_date': '20120521',
                 'uploader': 'The Royal Concept',
                 'upload_date': '20120521',

+                'duration': 227,

             },
             'params': {
                 # rtmp
             },
             'params': {
                 # rtmp


@@ -73,19 +79,35 @@ class SoundcloudIE(InfoExtractor):
                 'uploader': 'jaimeMF',
                 'description': 'test chars:  \"\'/\\ä↭',
                 'upload_date': '20131209',
                 'uploader': 'jaimeMF',
                 'description': 'test chars:  \"\'/\\ä↭',
                 'upload_date': '20131209',

+                'duration': 9,
+            },
+        },
+        # 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,

             },
         },
         # downloadable song
         {
             },
         },
         # downloadable song
         {

-            'url': 'https://soundcloud.com/simgretina/just-your-problem-baby-1',
-            'md5': '56a8b69568acaa967b4c49f9d1d52d19',
+            'url': 'https://soundcloud.com/oddsamples/bus-brakes',
+            'md5': '7624f2351f8a3b2e7cd51522496e7631',

             'info_dict': {
             'info_dict': {

-                'id': '105614606',
-                'ext': 'wav',
-                'title': 'Just Your Problem Baby (Acapella)',
-                'description': 'Vocals',
-                'uploader': 'Sim Gretina',
-                'upload_date': '20130815',
+                'id': '128590877',
+                'ext': 'mp3',
+                'title': 'Bus Brakes',
+                'description': 'md5:0053ca6396e8d2fd7b7e1595ef12ab66',
+                'uploader': 'oddsamples',
+                'upload_date': '20140109',
+                'duration': 17,

             },
         },
     ]
             },
         },
     ]


@@ -93,13 +115,9 @@ class SoundcloudIE(InfoExtractor):
     _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28'
     _IPHONE_CLIENT_ID = '376f225bf427445fc4bfb6b99b72e0bf'
 
     _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28'
     _IPHONE_CLIENT_ID = '376f225bf427445fc4bfb6b99b72e0bf'
 

-    @classmethod
-    def suitable(cls, url):
-        return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None
-

     def report_resolve(self, video_id):
         """Report information extraction."""
     def report_resolve(self, video_id):
         """Report information extraction."""

-        self.to_screen(u'%s: Resolving id' % video_id)
+        self.to_screen('%s: Resolving id' % video_id)

 
     @classmethod
     def _resolv_url(cls, url):
 
     @classmethod
     def _resolv_url(cls, url):


@@ -122,46 +140,48 @@ class SoundcloudIE(InfoExtractor):
             'title': info['title'],
             'description': info['description'],
             'thumbnail': thumbnail,
             'title': info['title'],
             'description': info['description'],
             'thumbnail': thumbnail,

+            'duration': int_or_none(info.get('duration'), 1000),
+            'webpage_url': info.get('permalink_url'),

         }
         }

+        formats = []

         if info.get('downloadable', False):
             # We can build a direct link to the song
             format_url = (
                 'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(
                     track_id, self._CLIENT_ID))
         if info.get('downloadable', False):
             # We can build a direct link to the song
             format_url = (
                 'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(
                     track_id, self._CLIENT_ID))

-            result['formats'] = [{
+            formats.append({

                 'format_id': 'download',
                 'ext': info.get('original_format', 'mp3'),
                 'url': format_url,
                 'vcodec': 'none',
                 'format_id': 'download',
                 '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(
-                streams_url,
-                track_id, 'Downloading track url')
-
-            formats = []
-            format_dict = json.loads(stream_json)
-            for key, stream_url in format_dict.items():
-                if key.startswith(u'http'):
-                    formats.append({
-                        'format_id': key,
-                        'ext': ext,
-                        'url': stream_url,
-                        'vcodec': 'none',
-                    })
-                elif key.startswith(u'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,
-                        'ext': ext,
-                        'vcodec': 'none',
-                    })
+                'preference': 10,
+            })
+
+        # 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))
+        format_dict = self._download_json(
+            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,
+                    'ext': ext,
+                    'vcodec': 'none',
+                })

 
             if not formats:
                 # We fallback to the stream_url in the original info, this
 
             if not formats:
                 # We fallback to the stream_url in the original info, this


@@ -187,16 +207,19 @@ class SoundcloudIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
         if mobj is None:
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
         if mobj is None:

-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError('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
 
         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

+            token = mobj.group('secret_token')
+            if token:
+                info_json_url += "&secret_token=" + token

         elif mobj.group('player'):
             query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
         elif mobj.group('player'):
             query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)

-            return self.url_result(query['url'][0], ie='Soundcloud')
+            return self.url_result(query['url'][0])

         else:
             # extract uploader (which is in the url)
             uploader = mobj.group('uploader')
         else:
             # extract uploader (which is in the url)
             uploader = mobj.group('uploader')


@@ -211,81 +234,149 @@ class SoundcloudIE(InfoExtractor):
     
             url = 'http://soundcloud.com/%s' % resolve_title
             info_json_url = self._resolv_url(url)
     
             url = 'http://soundcloud.com/%s' % resolve_title
             info_json_url = self._resolv_url(url)

-        info_json = self._download_webpage(info_json_url, full_title, 'Downloading info JSON')
+        info = self._download_json(info_json_url, full_title, 'Downloading info JSON')

 
 

-        info = json.loads(info_json)

         return self._extract_info_dict(info, full_title, secret_token=token)
 
         return self._extract_info_dict(info, full_title, secret_token=token)
 

+

 class SoundcloudSetIE(SoundcloudIE):
 class SoundcloudSetIE(SoundcloudIE):

-    _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$'
+    _VALID_URL = r'https?://(?:www\.)?soundcloud\.com/(?P<uploader>[\w\d-]+)/sets/(?P<slug_title>[\w\d-]+)(?:/(?P<token>[^?/]+))?'

     IE_NAME = 'soundcloud:set'
     IE_NAME = 'soundcloud:set'

-    # it's in tests/test_playlists.py
-    _TESTS = []
+    _TESTS = [{
+        'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep',
+        'info_dict': {
+            'title': 'The Royal Concept EP',
+        },
+        'playlist_mincount': 6,
+    }]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)

 
         # extract uploader (which is in the url)
 
         # extract uploader (which is in the url)

-        uploader = mobj.group(1)
+        uploader = mobj.group('uploader')

         # extract simple title (uploader + slug of song title)
         # extract simple title (uploader + slug of song title)

-        slug_title =  mobj.group(2)
+        slug_title = mobj.group('slug_title')

         full_title = '%s/sets/%s' % (uploader, slug_title)
         full_title = '%s/sets/%s' % (uploader, slug_title)

+        url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
+
+        token = mobj.group('token')
+        if token:
+            full_title += '/' + token
+            url += '/' + token

 
         self.report_resolve(full_title)
 
 
         self.report_resolve(full_title)
 

-        url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)

         resolv_url = self._resolv_url(url)
         resolv_url = self._resolv_url(url)

-        info_json = self._download_webpage(resolv_url, full_title)
+        info = self._download_json(resolv_url, full_title)

 
 

-        info = json.loads(info_json)

         if 'errors' in info:
             for err in info['errors']:
         if 'errors' in info:
             for err in info['errors']:

-                self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err['error_message']))
+                self._downloader.report_error('unable to download video webpage: %s' % compat_str(err['error_message']))

             return
 
             return
 

-        self.report_extraction(full_title)
-        return {'_type': 'playlist',
-                'entries': [self._extract_info_dict(track) for track in info['tracks']],
-                'id': info['id'],
-                'title': info['title'],
-                }
+        return {
+            '_type': 'playlist',
+            'entries': [self._extract_info_dict(track, secret_token=token) for track in info['tracks']],
+            'id': info['id'],
+            'title': info['title'],
+        }

 
 
 class SoundcloudUserIE(SoundcloudIE):
 
 
 class SoundcloudUserIE(SoundcloudIE):

-    _VALID_URL = r'https?://(www\.)?soundcloud\.com/(?P<user>[^/]+)(/?(tracks/)?)?(\?.*)?$'
+    _VALID_URL = r'https?://(www\.)?soundcloud\.com/(?P<user>[^/]+)/?((?P<rsrc>tracks|likes)/?)?(\?.*)?$'

     IE_NAME = 'soundcloud:user'
     IE_NAME = 'soundcloud:user'

-
-    # it's in tests/test_playlists.py
-    _TESTS = []
+    _TESTS = [{
+        'url': 'https://soundcloud.com/the-concept-band',
+        'info_dict': {
+            'id': '9615865',
+            'title': 'The Royal Concept',
+        },
+        'playlist_mincount': 12
+    }, {
+        'url': 'https://soundcloud.com/the-concept-band/likes',
+        'info_dict': {
+            'id': '9615865',
+            'title': 'The Royal Concept',
+        },
+        'playlist_mincount': 1,
+    }]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         uploader = mobj.group('user')
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         uploader = mobj.group('user')

+        resource = mobj.group('rsrc')
+        if resource is None:
+            resource = 'tracks'
+        elif resource == 'likes':
+            resource = 'favorites'

 
         url = 'http://soundcloud.com/%s/' % uploader
         resolv_url = self._resolv_url(url)
 
         url = 'http://soundcloud.com/%s/' % uploader
         resolv_url = self._resolv_url(url)

-        user_json = self._download_webpage(resolv_url, uploader,
-            'Downloading user info')
-        user = json.loads(user_json)
+        user = self._download_json(
+            resolv_url, uploader, 'Downloading user info')
+        base_url = 'http://api.soundcloud.com/users/%s/%s.json?' % (uploader, resource)

 
 

-        tracks = []
+        entries = []

         for i in itertools.count():
         for i in itertools.count():

-            data = compat_urllib_parse.urlencode({'offset': i*50,
-                                                  'client_id': self._CLIENT_ID,
-                                                  })
-            tracks_url = 'http://api.soundcloud.com/users/%s/tracks.json?' % user['id'] + data
-            response = self._download_webpage(tracks_url, uploader, 
-                '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:
+            data = compat_urllib_parse.urlencode({
+                'offset': i * 50,
+                'limit': 50,
+                'client_id': self._CLIENT_ID,
+            })
+            new_entries = self._download_json(
+                base_url + data, uploader, 'Downloading track page %s' % (i + 1))
+            if len(new_entries) == 0:
+                self.to_screen('%s: End page received' % uploader)

                 break
                 break

+            entries.extend(self._extract_info_dict(e, quiet=True) for e in new_entries)

 
         return {
             '_type': 'playlist',
             'id': compat_str(user['id']),
             'title': user['username'],
 
         return {
             '_type': 'playlist',
             'id': compat_str(user['id']),
             'title': user['username'],

-            'entries': tracks,
+            'entries': entries,
+        }
+
+
+class SoundcloudPlaylistIE(SoundcloudIE):
+    _VALID_URL = r'https?://api\.soundcloud\.com/playlists/(?P<id>[0-9]+)(?:/?\?secret_token=(?P<token>[^&]+?))?$'
+    IE_NAME = 'soundcloud:playlist'
+    _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,
+    }]
+
+    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)
+
+        data_dict = {
+            'client_id': self._CLIENT_ID,
+        }
+        token = mobj.group('token')
+
+        if token:
+            data_dict['secret_token'] = token
+
+        data = compat_urllib_parse.urlencode(data_dict)
+        data = self._download_json(
+            base_url + data, playlist_id, 'Downloading playlist')
+
+        entries = [
+            self._extract_info_dict(t, quiet=True, secret_token=token)
+                for t in data['tracks']]
+
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': data.get('title'),
+            'description': data.get('description'),
+            'entries': entries,

         }
         }