]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/soundcloud.py
debian/control: Update list of extractors.
[youtubedl] / youtube_dl / extractor / soundcloud.py
index 80d7e1b543608a36170577c004d032a7c5e75d1b..3a19ab17222831d87ffde4992e5712b01359e6eb 100644 (file)
@@ -1,9 +1,12 @@
 import json
 import re
+import itertools
 
 from .common import InfoExtractor
 from ..utils import (
     compat_str,
+    compat_urlparse,
+    compat_urllib_parse,
 
     ExtractorError,
     unified_strdate,
@@ -19,68 +22,168 @@ class SoundcloudIE(InfoExtractor):
        of the stream token and uid
      """
 
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)'
+    _VALID_URL = r'''^(?:https?://)?
+                    (?:(?:(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)/?(?:[?].*)?$)
+                       |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+))
+                       |(?P<widget>w.soundcloud.com/player/?.*?url=.*)
+                    )
+                    '''
     IE_NAME = u'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"
+            }
+        },
+        # 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',
+            },
+            u'params': {
+                # rtmp
+                u'skip_download': True,
+            },
+        },
+    ]
+
+    _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."""
         self.to_screen(u'%s: Resolving id' % video_id)
 
+    @classmethod
+    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):
+        track_id = compat_str(info['id'])
+        name = full_title or track_id
+        if quiet:
+            self.report_extraction(name)
+
+        thumbnail = info['artwork_url']
+        if thumbnail is not None:
+            thumbnail = thumbnail.replace('-large', '-t500x500')
+        ext = info.get('original_format', u'mp3')
+        result = {
+            'id': track_id,
+            'uploader': info['user']['username'],
+            'upload_date': unified_strdate(info['created_at']),
+            'title': info['title'],
+            'description': info['description'],
+            'thumbnail': thumbnail,
+        }
+        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(
+                    track_id, self._CLIENT_ID))
+            result['formats'] = [{
+                'format_id': 'download',
+                'ext': ext,
+                'url': format_url,
+                'vcodec': 'none',
+            }]
+        else:
+            # We have to retrieve the url
+            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')
+
+            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',
+                    })
+
+            if not formats:
+                # 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',
+                    'url': info['stream_url'] + '?client_id=' + self._CLIENT_ID,
+                    'ext': ext,
+                    'vcodec': 'none',
+                })
+
+            def format_pref(f):
+                if f['format_id'].startswith('http'):
+                    return 2
+                if f['format_id'].startswith('rtmp'):
+                    return 1
+                return 0
+
+            formats.sort(key=format_pref)
+            result['formats'] = formats
+
+        return result
+
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
         if mobj is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
 
-        # extract uploader (which is in the url)
-        uploader = mobj.group(1)
-        # extract simple title (uploader + slug of song title)
-        slug_title =  mobj.group(2)
-        full_title = '%s/%s' % (uploader, slug_title)
-
-        self.report_resolve(full_title)
-
-        url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title)
-        resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
-        info_json = self._download_webpage(resolv_url, full_title, u'Downloading info JSON')
+        track_id = mobj.group('track_id')
+        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'):
+            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)
+            # extract simple title (uploader + slug of song title)
+            slug_title =  mobj.group(2)
+            full_title = '%s/%s' % (uploader, slug_title)
+    
+            self.report_resolve(full_title)
+    
+            url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title)
+            info_json_url = self._resolv_url(url)
+        info_json = self._download_webpage(info_json_url, full_title, u'Downloading info JSON')
 
         info = json.loads(info_json)
-        video_id = info['id']
-        self.report_extraction(full_title)
-
-        streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
-        stream_json = self._download_webpage(streams_url, full_title,
-                                             u'Downloading stream definitions',
-                                             u'unable to download stream definitions')
-
-        streams = json.loads(stream_json)
-        mediaURL = streams['http_mp3_128_url']
-        upload_date = unified_strdate(info['created_at'])
-
-        return [{
-            'id':       info['id'],
-            'url':      mediaURL,
-            'uploader': info['user']['username'],
-            'upload_date': upload_date,
-            'title':    info['title'],
-            'ext':      u'mp3',
-            'description': info['description'],
-        }]
+        return self._extract_info_dict(info, full_title)
 
-class SoundcloudSetIE(InfoExtractor):
-    """Information extractor for soundcloud.com sets
-       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
-     """
-
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)'
+class SoundcloudSetIE(SoundcloudIE):
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$'
     IE_NAME = u'soundcloud:set'
-
-    def report_resolve(self, video_id):
-        """Report information extraction."""
-        self.to_screen(u'%s: Resolving id' % video_id)
+    # it's in tests/test_playlists.py
+    _TESTS = []
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -96,10 +199,9 @@ class SoundcloudSetIE(InfoExtractor):
         self.report_resolve(full_title)
 
         url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
-        resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
+        resolv_url = self._resolv_url(url)
         info_json = self._download_webpage(resolv_url, full_title)
 
-        videos = []
         info = json.loads(info_json)
         if 'errors' in info:
             for err in info['errors']:
@@ -107,23 +209,46 @@ class SoundcloudSetIE(InfoExtractor):
             return
 
         self.report_extraction(full_title)
-        for track in info['tracks']:
-            video_id = track['id']
-
-            streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
-            stream_json = self._download_webpage(streams_url, video_id, u'Downloading track info JSON')
-
-            self.report_extraction(video_id)
-            streams = json.loads(stream_json)
-            mediaURL = streams['http_mp3_128_url']
-
-            videos.append({
-                'id':       video_id,
-                'url':      mediaURL,
-                'uploader': track['user']['username'],
-                'upload_date':  unified_strdate(track['created_at']),
-                'title':    track['title'],
-                'ext':      u'mp3',
-                'description': track['description'],
-            })
-        return videos
+        return {'_type': 'playlist',
+                'entries': [self._extract_info_dict(track) for track in info['tracks']],
+                'id': info['id'],
+                'title': info['title'],
+                }
+
+
+class SoundcloudUserIE(SoundcloudIE):
+    _VALID_URL = r'https?://(www\.)?soundcloud.com/(?P<user>[^/]+)(/?(tracks/)?)?(\?.*)?$'
+    IE_NAME = u'soundcloud:user'
+
+    # it's in tests/test_playlists.py
+    _TESTS = []
+
+    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)
+        user_json = self._download_webpage(resolv_url, uploader,
+            u'Downloading user info')
+        user = json.loads(user_json)
+
+        tracks = []
+        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, 
+                u'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:
+                break
+
+        return {
+            '_type': 'playlist',
+            'id': compat_str(user['id']),
+            'title': user['username'],
+            'entries': tracks,
+        }