Merge 35c656d3b9c97be5333e3f34c7926e76615987a1 into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

* thx seproDev, yt-dlp/yt-dlp/pull/12694/commits/ae69e3c
* also add documenting comment in `InfoExtractor`

youtube_dl/extractor/bokecc.py

@@ -32,7 +32,7 @@ class BokeCCBaseIE(InfoExtractor):
 
 
 class BokeCCIE(BokeCCBaseIE):
-    _IE_DESC = 'CC视频'
+    IE_DESC = 'CC视频'
     _VALID_URL = r'https?://union\.bokecc\.com/playvideo\.bo\?(?P<query>.*)'
 
     _TESTS = [{

youtube_dl/extractor/cloudy.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class CloudyIE(InfoExtractor):
-    _IE_DESC = 'cloudy.ec'
+    IE_DESC = 'cloudy.ec'
     _VALID_URL = r'https?://(?:www\.)?cloudy\.ec/(?:v/|embed\.php\?.*?\bid=)(?P<id>[A-Za-z0-9]+)'
     _TESTS = [{
         'url': 'https://www.cloudy.ec/v/af511e2527aac',

youtube_dl/extractor/common.py

@@ -422,6 +422,8 @@ class InfoExtractor(object):
     _GEO_COUNTRIES = None
     _GEO_IP_BLOCKS = None
     _WORKING = True
+    # supply this in public subclasses: used in supported sites list, etc
+    # IE_DESC = 'short description of IE'
 
     def __init__(self, downloader=None):
         """Constructor. Receives an optional downloader."""

youtube_dl/extractor/extractors.py

@@ -1625,6 +1625,7 @@ from .yahoo import (
 from .yandexdisk import YandexDiskIE
 from .yandexmusic import (
     YandexMusicTrackIE,
+    YandexMusicPrivateTrackIE,
     YandexMusicAlbumIE,
     YandexMusicPlaylistIE,
     YandexMusicArtistTracksIE,

youtube_dl/extractor/itv.py

@@ -35,15 +35,6 @@ from ..utils import (
 
 class ITVBaseIE(InfoExtractor):
 
-    def _search_nextjs_data(self, webpage, video_id, **kw):
-        transform_source = kw.pop('transform_source', None)
-        fatal = kw.pop('fatal', True)
-        return self._parse_json(
-            self._search_regex(
-                r'''<script\b[^>]+\bid=('|")__NEXT_DATA__\1[^>]*>(?P<js>[^<]+)</script>''',
-                webpage, 'next.js data', group='js', fatal=fatal, **kw),
-            video_id, transform_source=transform_source, fatal=fatal)
-
     def __handle_request_webpage_error(self, err, video_id=None, errnote=None, fatal=True):
         if errnote is False:
             return False
@@ -109,7 +100,9 @@ class ITVBaseIE(InfoExtractor):
 
 class ITVIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:(?P<w>watch)|hub)/[^/]+/(?(w)[\w-]+/)(?P<id>\w+)'
-    _IE_DESC = 'ITVX'
+    IE_DESC = 'ITVX'
+    _WORKING = False
+
     _TESTS = [{
         'note': 'Hub URLs redirect to ITVX',
         'url': 'https://www.itv.com/hub/liar/2a4547a0012',
@@ -270,7 +263,7 @@ class ITVIE(ITVBaseIE):
                 'ext': determine_ext(href, 'vtt'),
             })
 
-        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default='{}')
+        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default={})
         video_data.update(traverse_obj(next_data, ('props', 'pageProps', ('title', 'episode')), expected_type=dict)[0] or {})
         title = traverse_obj(video_data, 'headerTitle', 'episodeTitle')
         info = self._og_extract(webpage, require_title=not title)
@@ -323,7 +316,7 @@ class ITVIE(ITVBaseIE):
 
 class ITVBTCCIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?!(?:watch|hub)/)(?:[^/]+/)+(?P<id>[^/?#&]+)'
-    _IE_DESC = 'ITV articles: News, British Touring Car Championship'
+    IE_DESC = 'ITV articles: News, British Touring Car Championship'
     _TESTS = [{
         'note': 'British Touring Car Championship',
         'url': 'https://www.itv.com/btcc/articles/btcc-2018-all-the-action-from-brands-hatch',

youtube_dl/extractor/senateisvp.py

@@ -47,7 +47,7 @@ class SenateISVPIE(InfoExtractor):
         ['vetaff', '76462', 'http://vetaff-f.akamaihd.net'],
         ['arch', '', 'http://ussenate-f.akamaihd.net/']
     ]
-    _IE_NAME = 'senate.gov'
+    IE_NAME = 'senate.gov'
     _VALID_URL = r'https?://(?:www\.)?senate\.gov/isvp/?\?(?P<qs>.+)'
     _TESTS = [{
         'url': 'http://www.senate.gov/isvp/?comm=judiciary&type=live&stt=&filename=judiciary031715&auto_play=false&wmode=transparent&poster=http%3A%2F%2Fwww.judiciary.senate.gov%2Fthemes%2Fjudiciary%2Fimages%2Fvideo-poster-flash-fit.png',

youtube_dl/extractor/yandexmusic.py

@@ -219,42 +219,127 @@ class YandexMusicTrackIE(YandexMusicBaseIE):
         return track_info
 
 
+class YandexMusicPrivateTrackIE(YandexMusicBaseIE):
+    IE_NAME = 'yandexmusic:privatetrack'
+    IE_DESC = 'Яндекс.Музыка - Трек'
+    _VALID_URL = r'%s/track/(?P<id>[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12})' % YandexMusicBaseIE._VALID_URL_BASE
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        tld, track_id = mobj.group('tld'), mobj.group('id')
+
+        track = self._call_api(
+            'track', tld, url, track_id, 'Downloading track JSON',
+            {'track': '%s' % (track_id)})['track']
+        track_title = track['title']
+
+        download_data = self._download_json(
+            'https://music.yandex.ru/api/v2.1/handlers/track/%s/web-album_track-track-track-main/download/m' % (track_id),
+            track_id, 'Downloading track location url JSON',
+            headers={'X-Retpath-Y': url})
+
+        fd_data = self._download_json(
+            download_data['src'], track_id,
+            'Downloading track location JSON',
+            query={'format': 'json'})
+        key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + fd_data['path'][1:] + fd_data['s']).encode('utf-8')).hexdigest()
+        f_url = 'http://%s/get-mp3/%s/%s?track-id=%s ' % (fd_data['host'], key, fd_data['ts'] + fd_data['path'], track['id'])
+
+        thumbnail = None
+        cover_uri = track.get('coverUri')
+        if cover_uri:
+            thumbnail = cover_uri.replace('%%', 'orig')
+            if not thumbnail.startswith('http'):
+                thumbnail = 'http://' + thumbnail
+
+        track_info = {
+            'id': track_id,
+            'ext': 'mp3',
+            'url': f_url,
+            'filesize': int_or_none(track.get('fileSize')),
+            'duration': float_or_none(track.get('durationMs'), 1000),
+            'thumbnail': thumbnail,
+            'track': track_title,
+            'acodec': download_data.get('codec'),
+            'abr': int_or_none(download_data.get('bitrate')),
+        }
+
+        def extract_artist_name(artist):
+            decomposed = artist.get('decomposed')
+            if not isinstance(decomposed, list):
+                return artist['name']
+            parts = [artist['name']]
+            for element in decomposed:
+                if isinstance(element, dict) and element.get('name'):
+                    parts.append(element['name'])
+                elif isinstance(element, compat_str):
+                    parts.append(element)
+            return ''.join(parts)
+
+        def extract_artist(artist_list):
+            if artist_list and isinstance(artist_list, list):
+                artists_names = [extract_artist_name(a) 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')
+                disc_number = int_or_none(try_get(
+                    album, lambda x: x['trackPosition']['volume']))
+                track_number = int_or_none(try_get(
+                    album, lambda x: x['trackPosition']['index']))
+                track_info.update({
+                    'album': album.get('title'),
+                    'album_artist': extract_artist(album.get('artists')),
+                    'release_year': int_or_none(year),
+                    'genre': album.get('genre'),
+                    'disc_number': disc_number,
+                    'track_number': track_number,
+                })
+
+        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
+
+
 class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
     def _extract_tracks(self, source, item_id, url, tld):
-        tracks = source['tracks']
+        tracks = []
         track_ids = [compat_str(track_id) for track_id in source['trackIds']]
-
+        def create_chunks(lst, n):
+            """Yield successive n-sized chunks from lst."""
+            for i in range(0, len(lst), n):
+                yield lst[i:i + n]
         # tracks dictionary shipped with playlist.jsx API is limited to 150 tracks,
         # missing tracks should be retrieved manually.
-        if len(tracks) < len(track_ids):
+        _TRACKS_PER_CHUNK = 100
-            present_track_ids = set([
+        # Request missing tracks in chunks to avoid exceeding max HTTP header size,
-                compat_str(track['id'])
+        # see https://github.com/ytdl-org/youtube-dl/issues/27355
-                for track in tracks if track.get('id')])
+        chunks = create_chunks(track_ids, _TRACKS_PER_CHUNK)
-            missing_track_ids = [
+        for chunk_num, chunk in enumerate(chunks):
-                track_id for track_id in track_ids
+            missing_tracks = self._call_api(
-                if track_id not in present_track_ids]
+                'track-entries', tld, url, item_id,
-            # Request missing tracks in chunks to avoid exceeding max HTTP header size,
+                'Downloading tracks JSON chunk %d and size %d' % (chunk_num + 1, len(chunk)), {
-            # see https://github.com/ytdl-org/youtube-dl/issues/27355
+                    'entries': ','.join(chunk),
-            _TRACKS_PER_CHUNK = 250
+                    'lang': tld,
-            for chunk_num in itertools.count(0):
+                    'external-domain': 'music.yandex.%s' % tld,
-                start = chunk_num * _TRACKS_PER_CHUNK
+                    'experiments':'{"userFeed":"old","similarities":"default","genreRadio":"new-ichwill-matrixnet6","recommendedArtists":"ichwill_similar_artists","recommendedTracks":"recommended_tracks_by_artist_from_history","recommendedAlbumsOfFavoriteGenre":"recent","recommendedSimilarArtists":"default","recommendedArtistsWithArtistsFromHistory":"force_recent","adv":"a","loserArtistsWithArtists":"off","ny2015":"no"}',
-                end = start + _TRACKS_PER_CHUNK
+                    'overembed': 'false',
-                missing_track_ids_req = missing_track_ids[start:end]
+                    'strict': 'true',
-                assert missing_track_ids_req
+                })
-                missing_tracks = self._call_api(
+            if missing_tracks:
-                    'track-entries', tld, url, item_id,
+                tracks.extend(missing_tracks)
-                    'Downloading missing tracks JSON chunk %d' % (chunk_num + 1), {
+            
-                        'entries': ','.join(missing_track_ids_req),
-                        'lang': tld,
-                        'external-domain': 'music.yandex.%s' % tld,
-                        'overembed': 'false',
-                        'strict': 'true',
-                    })
-                if missing_tracks:
-                    tracks.extend(missing_tracks)
-                if end >= len(missing_track_ids):
-                    break
-
         return tracks
 
     def _build_playlist(self, tracks):
@@ -263,6 +348,11 @@ class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
             track_id = track.get('id') or track.get('realId')
             if not track_id:
                 continue
+            if re.match(r"[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}", track_id):
+                entries.append(self.url_result(
+                    'http://music.yandex.ru/track/%s' % (track_id),
+                    ie=YandexMusicTrackIE.ie_key(), video_id=track_id))
+                continue
             albums = track.get('albums')
             if not albums or not isinstance(albums, list):
                 continue

youtube_dl/jsinterp.py

@@ -686,6 +686,8 @@ class JSInterpreter(object):
             raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
 
     def _dump(self, obj, namespace):
+        if obj is JS_Undefined:
+            return 'undefined'
         try:
             return json.dumps(obj)
         except TypeError: