Prepare to upload

[youtubedl] / youtube_dl / extractor / yandexmusic.py

diff --git a/youtube_dl/extractor/yandexmusic.py b/youtube_dl/extractor/yandexmusic.py
index d3cc1a29fa473fee2f58e91323774633be00fc4b..eb1062142ecbc6a4702f0dc7763c414934fd3645 100644 (file)
--- a/youtube_dl/extractor/yandexmusic.py
+++ b/youtube_dl/extractor/yandexmusic.py
@@ -5,18 +5,48 @@ import re
 import hashlib
 
 from .common import InfoExtractor
 import hashlib
 
 from .common import InfoExtractor

-from ..compat import (
-    compat_str,
-    compat_urllib_parse,
-)
+from ..compat import compat_str

 from ..utils import (
 from ..utils import (

+    ExtractorError,

     int_or_none,
     float_or_none,
     int_or_none,
     float_or_none,

-    sanitized_Request,

 )
 
 
 )
 
 

-class YandexMusicTrackIE(InfoExtractor):
+class YandexMusicBaseIE(InfoExtractor):
+    @staticmethod
+    def _handle_error(response):
+        if isinstance(response, dict):
+            error = response.get('error')
+            if error:
+                raise ExtractorError(error, expected=True)
+            if response.get('type') == 'captcha' or 'captcha' in response:
+                YandexMusicBaseIE._raise_captcha()
+
+    @staticmethod
+    def _raise_captcha():
+        raise ExtractorError(
+            'YandexMusic has considered youtube-dl requests automated and '
+            'asks you to solve a CAPTCHA. You can either wait for some '
+            'time until unblocked and optionally use --sleep-interval '
+            'in future or alternatively you can go to https://music.yandex.ru/ '
+            'solve CAPTCHA, then export cookies and pass cookie file to '
+            'youtube-dl with --cookies',
+            expected=True)
+
+    def _download_webpage(self, *args, **kwargs):
+        webpage = super(YandexMusicBaseIE, self)._download_webpage(*args, **kwargs)
+        if 'Нам очень жаль, но запросы, поступившие с вашего IP-адреса, похожи на автоматические.' in webpage:
+            self._raise_captcha()
+        return webpage
+
+    def _download_json(self, *args, **kwargs):
+        response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
+        self._handle_error(response)
+        return response
+
+
+class YandexMusicTrackIE(YandexMusicBaseIE):

     IE_NAME = 'yandexmusic:track'
     IE_DESC = 'Яндекс.Музыка - Трек'
     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
     IE_NAME = 'yandexmusic:track'
     IE_DESC = 'Яндекс.Музыка - Трек'
     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'


@@ -27,10 +57,16 @@ class YandexMusicTrackIE(InfoExtractor):
         'info_dict': {
             'id': '4878838',
             'ext': 'mp3',
         'info_dict': {
             'id': '4878838',
             'ext': 'mp3',

-            'title': 'Carlo Ambrosio - Gypsy Eyes 1',
+            'title': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio - Gypsy Eyes 1',

             'filesize': 4628061,
             'duration': 193.04,
             'filesize': 4628061,
             'duration': 193.04,

-        }
+            'track': 'Gypsy Eyes 1',
+            'album': 'Gypsy Soul',
+            'album_artist': 'Carlo Ambrosio',
+            'artist': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio',
+            'release_year': '2009',
+        },
+        'skip': 'Travis CI servers blocked by YandexMusic',

     }
 
     def _get_track_url(self, storage_dir, track_id):
     }
 
     def _get_track_url(self, storage_dir, track_id):


@@ -39,6 +75,12 @@ class YandexMusicTrackIE(InfoExtractor):
             % storage_dir,
             track_id, 'Downloading track location JSON')
 
             % storage_dir,
             track_id, 'Downloading track location JSON')
 

+        # Each string is now wrapped in a list, this is probably only temporarily thus
+        # supporting both scenarios (see https://github.com/rg3/youtube-dl/issues/10193)
+        for k, v in data.items():
+            if v and isinstance(v, list):
+                data[k] = v[0]
+

         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
         storage = storage_dir.split('.')
 
         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
         storage = storage_dir.split('.')
 


@@ -52,16 +94,45 @@ class YandexMusicTrackIE(InfoExtractor):
             thumbnail = cover_uri.replace('%%', 'orig')
             if not thumbnail.startswith('http'):
                 thumbnail = 'http://' + thumbnail
             thumbnail = cover_uri.replace('%%', 'orig')
             if not thumbnail.startswith('http'):
                 thumbnail = 'http://' + thumbnail

-        return {
+
+        track_title = track['title']
+        track_info = {

             'id': track['id'],
             'ext': 'mp3',
             'url': self._get_track_url(track['storageDir'], track['id']),
             'id': track['id'],
             'ext': 'mp3',
             'url': self._get_track_url(track['storageDir'], track['id']),

-            'title': '%s - %s' % (track['artists'][0]['name'], track['title']),

             'filesize': int_or_none(track.get('fileSize')),
             'duration': float_or_none(track.get('durationMs'), 1000),
             'thumbnail': thumbnail,
             'filesize': int_or_none(track.get('fileSize')),
             'duration': float_or_none(track.get('durationMs'), 1000),
             'thumbnail': thumbnail,

+            'track': track_title,

         }
 
         }
 

+        def extract_artist(artist_list):
+            if artist_list and isinstance(artist_list, list):
+                artists_names = [a['name'] for a in artist_list if a.get('name')]
+                if artists_names:
+                    return ', '.join(artists_names)
+
+        albums = track.get('albums')
+        if albums and isinstance(albums, list):
+            album = albums[0]
+            if isinstance(album, dict):
+                year = album.get('year')
+                track_info.update({
+                    'album': album.get('title'),
+                    'album_artist': extract_artist(album.get('artists')),
+                    'release_year': compat_str(year) if year else None,
+                })
+
+        track_artist = extract_artist(track.get('artists'))
+        if track_artist:
+            track_info.update({
+                'artist': track_artist,
+                'title': '%s - %s' % (track_artist, track_title),
+            })
+        else:
+            track_info['title'] = track_title
+        return track_info
+

     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         album_id, track_id = mobj.group('album_id'), mobj.group('id')
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         album_id, track_id = mobj.group('album_id'), mobj.group('id')


@@ -73,7 +144,7 @@ class YandexMusicTrackIE(InfoExtractor):
         return self._get_track_info(track)
 
 
         return self._get_track_info(track)
 
 

-class YandexMusicPlaylistBaseIE(InfoExtractor):
+class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):

     def _build_playlist(self, tracks):
         return [
             self.url_result(
     def _build_playlist(self, tracks):
         return [
             self.url_result(


@@ -93,6 +164,7 @@ class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
         },
         'playlist_count': 50,
             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
         },
         'playlist_count': 50,

+        'skip': 'Travis CI servers blocked by YandexMusic',

     }
 
     def _real_extract(self, url):
     }
 
     def _real_extract(self, url):


@@ -115,7 +187,7 @@ class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
     IE_NAME = 'yandexmusic:playlist'
     IE_DESC = 'Яндекс.Музыка - Плейлист'
 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
     IE_NAME = 'yandexmusic:playlist'
     IE_DESC = 'Яндекс.Музыка - Плейлист'

-    _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/users/[^/]+/playlists/(?P<id>\d+)'
+    _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/users/(?P<user>[^/]+)/playlists/(?P<id>\d+)'

 
     _TESTS = [{
         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 
     _TESTS = [{
         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',


@@ -125,6 +197,7 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
         },
         'playlist_count': 6,
             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
         },
         'playlist_count': 6,

+        'skip': 'Travis CI servers blocked by YandexMusic',

     }, {
         # playlist exceeding the limit of 150 tracks shipped with webpage (see
         # https://github.com/rg3/youtube-dl/issues/6666)
     }, {
         # playlist exceeding the limit of 150 tracks shipped with webpage (see
         # https://github.com/rg3/youtube-dl/issues/6666)


@@ -133,46 +206,65 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
             'id': '1036',
             'title': 'Музыка 90-х',
         },
             'id': '1036',
             'title': 'Музыка 90-х',
         },

-        'playlist_count': 310,
+        'playlist_mincount': 300,
+        'skip': 'Travis CI servers blocked by YandexMusic',

     }]
 
     def _real_extract(self, url):
     }]
 
     def _real_extract(self, url):

-        playlist_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, playlist_id)
-
-        mu = self._parse_json(
-            self._search_regex(
-                r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'),
-            playlist_id)
-
-        playlist = mu['pageData']['playlist']
-        tracks, track_ids = playlist['tracks'], playlist['trackIds']
-
-        # tracks dictionary shipped with webpage is limited to 150 tracks,
+        mobj = re.match(self._VALID_URL, url)
+        tld = mobj.group('tld')
+        user = mobj.group('user')
+        playlist_id = mobj.group('id')
+
+        playlist = self._download_json(
+            'https://music.yandex.%s/handlers/playlist.jsx' % tld,
+            playlist_id, 'Downloading missing tracks JSON',
+            fatal=False,
+            headers={
+                'Referer': url,
+                'X-Requested-With': 'XMLHttpRequest',
+                'X-Retpath-Y': url,
+            },
+            query={
+                'owner': user,
+                'kinds': playlist_id,
+                'light': 'true',
+                'lang': tld,
+                'external-domain': 'music.yandex.%s' % tld,
+                'overembed': 'false',
+            })['playlist']
+
+        tracks = playlist['tracks']
+        track_ids = [compat_str(track_id) for track_id in playlist['trackIds']]
+
+        # tracks dictionary shipped with playlist.jsx API is limited to 150 tracks,

         # missing tracks should be retrieved manually.
         if len(tracks) < len(track_ids):
         # missing tracks should be retrieved manually.
         if len(tracks) < len(track_ids):

-            present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')])
-            missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids)
-            request = sanitized_Request(
-                'https://music.yandex.ru/handlers/track-entries.jsx',
-                compat_urllib_parse.urlencode({
+            present_track_ids = set([
+                compat_str(track['id'])
+                for track in tracks if track.get('id')])
+            missing_track_ids = [
+                track_id for track_id in track_ids
+                if track_id not in present_track_ids]
+            missing_tracks = self._download_json(
+                'https://music.yandex.%s/handlers/track-entries.jsx' % tld,
+                playlist_id, 'Downloading missing tracks JSON',
+                fatal=False,
+                headers={
+                    'Referer': url,
+                    'X-Requested-With': 'XMLHttpRequest',
+                },
+                query={

                     'entries': ','.join(missing_track_ids),
                     'entries': ','.join(missing_track_ids),

-                    'lang': mu.get('settings', {}).get('lang', 'en'),
-                    'external-domain': 'music.yandex.ru',
+                    'lang': tld,
+                    'external-domain': 'music.yandex.%s' % tld,

                     'overembed': 'false',
                     'overembed': 'false',

-                    'sign': mu.get('authData', {}).get('user', {}).get('sign'),

                     'strict': 'true',
                     'strict': 'true',

-                }).encode('utf-8'))
-            request.add_header('Referer', url)
-            request.add_header('X-Requested-With', 'XMLHttpRequest')
-
-            missing_tracks = self._download_json(
-                request, playlist_id, 'Downloading missing tracks JSON', fatal=False)
+                })

             if missing_tracks:
                 tracks.extend(missing_tracks)
 
         return self.playlist_result(
             self._build_playlist(tracks),
             compat_str(playlist_id),
             if missing_tracks:
                 tracks.extend(missing_tracks)
 
         return self.playlist_result(
             self._build_playlist(tracks),
             compat_str(playlist_id),

-            playlist['title'], playlist.get('description'))
+            playlist.get('title'), playlist.get('description'))