logo

youtube-dl

[mirror] Download/Watch videos from video hostersgit clone https://hacktivis.me/git/mirror/youtube-dl.git

yandexmusic.py (18661B)

    

  1. # coding: utf-8
  2. from __future__ import unicode_literals
  4. import hashlib
  5. import itertools
  6. import re
  8. from .common import InfoExtractor
  9. from ..compat import compat_str
  10. from ..utils import (
  11.     ExtractorError,
  12.     int_or_none,
  13.     float_or_none,
  14.     try_get,
  15. )
  18. class YandexMusicBaseIE(InfoExtractor):
  19.     _VALID_URL_BASE = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by|com)'
  21.     @staticmethod
  22.     def _handle_error(response):
  23.         if isinstance(response, dict):
  24.             error = response.get('error')
  25.             if error:
  26.                 raise ExtractorError(error, expected=True)
  27.             if response.get('type') == 'captcha' or 'captcha' in response:
  28.                 YandexMusicBaseIE._raise_captcha()
  30.     @staticmethod
  31.     def _raise_captcha():
  32.         raise ExtractorError(
  33.             'YandexMusic has considered youtube-dl requests automated and '
  34.             'asks you to solve a CAPTCHA. You can either wait for some '
  35.             'time until unblocked and optionally use --sleep-interval '
  36.             'in future or alternatively you can go to https://music.yandex.ru/ '
  37.             'solve CAPTCHA, then export cookies and pass cookie file to '
  38.             'youtube-dl with --cookies',
  39.             expected=True)
  41.     def _download_webpage_handle(self, *args, **kwargs):
  42.         webpage = super(YandexMusicBaseIE, self)._download_webpage_handle(*args, **kwargs)
  43.         if 'Нам очень жаль, но&nbsp;запросы, поступившие с&nbsp;вашего IP-адреса, похожи на&nbsp;автоматические.' in webpage:
  44.             self._raise_captcha()
  45.         return webpage
  47.     def _download_json(self, *args, **kwargs):
  48.         response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
  49.         self._handle_error(response)
  50.         return response
  52.     def _call_api(self, ep, tld, url, item_id, note, query):
  53.         return self._download_json(
  54.             'https://music.yandex.%s/handlers/%s.jsx' % (tld, ep),
  55.             item_id, note,
  56.             fatal=False,
  57.             headers={
  58.                 'Referer': url,
  59.                 'X-Requested-With': 'XMLHttpRequest',
  60.                 'X-Retpath-Y': url,
  61.             },
  62.             query=query)
  65. class YandexMusicTrackIE(YandexMusicBaseIE):
  66.     IE_NAME = 'yandexmusic:track'
  67.     IE_DESC = 'Яндекс.Музыка - Трек'
  68.     _VALID_URL = r'%s/album/(?P<album_id>\d+)/track/(?P<id>\d+)' % YandexMusicBaseIE._VALID_URL_BASE
  70.     _TESTS = [{
  71.         'url': 'http://music.yandex.ru/album/540508/track/4878838',
  72.         'md5': 'dec8b661f12027ceaba33318787fff76',
  73.         'info_dict': {
  74.             'id': '4878838',
  75.             'ext': 'mp3',
  76.             'title': 'md5:c63e19341fdbe84e43425a30bc777856',
  77.             'filesize': int,
  78.             'duration': 193.04,
  79.             'track': 'md5:210508c6ffdfd67a493a6c378f22c3ff',
  80.             'album': 'md5:cd04fb13c4efeafdfa0a6a6aca36d01a',
  81.             'album_artist': 'md5:5f54c35462c07952df33d97cfb5fc200',
  82.             'artist': 'md5:e6fd86621825f14dc0b25db3acd68160',
  83.             'release_year': 2009,
  84.         },
  85.         # 'skip': 'Travis CI servers blocked by YandexMusic',
  86.     }, {
  87.         # multiple disks
  88.         'url': 'http://music.yandex.ru/album/3840501/track/705105',
  89.         'md5': '82a54e9e787301dd45aba093cf6e58c0',
  90.         'info_dict': {
  91.             'id': '705105',
  92.             'ext': 'mp3',
  93.             'title': 'md5:f86d4a9188279860a83000277024c1a6',
  94.             'filesize': int,
  95.             'duration': 239.27,
  96.             'track': 'md5:40f887f0666ba1aa10b835aca44807d1',
  97.             'album': 'md5:624f5224b14f5c88a8e812fd7fbf1873',
  98.             'album_artist': 'md5:dd35f2af4e8927100cbe6f5e62e1fb12',
  99.             'artist': 'md5:dd35f2af4e8927100cbe6f5e62e1fb12',
  100.             'release_year': 2016,
  101.             'genre': 'pop',
  102.             'disc_number': 2,
  103.             'track_number': 9,
  104.         },
  105.         # 'skip': 'Travis CI servers blocked by YandexMusic',
  106.     }, {
  107.         'url': 'http://music.yandex.com/album/540508/track/4878838',
  108.         'only_matching': True,
  109.     }, {
  110.         'url': 'https://music.yandex.ru/album/16302456/track/85430762',
  111.         'md5': '11b8d50ab03b57738deeaadf661a0a48',
  112.         'info_dict': {
  113.             'id': '85430762',
  114.             'ext': 'mp3',
  115.             'abr': 128,
  116.             'title': 'Haddadi Von Engst, Phonic Youth, Super Flu - Til The End (Super Flu Remix)',
  117.             'filesize': int,
  118.             'duration': 431.14,
  119.             'track': 'Til The End (Super Flu Remix)',
  120.             'album': 'Til The End',
  121.             'album_artist': 'Haddadi Von Engst, Phonic Youth',
  122.             'artist': 'Haddadi Von Engst, Phonic Youth, Super Flu',
  123.             'release_year': 2021,
  124.             'genre': 'house',
  125.             'disc_number': 1,
  126.             'track_number': 2,
  127.         }
  128.     }]
  130.     def _real_extract(self, url):
  131.         mobj = re.match(self._VALID_URL, url)
  132.         tld, album_id, track_id = mobj.group('tld'), mobj.group('album_id'), mobj.group('id')
  134.         track = self._call_api(
  135.             'track', tld, url, track_id, 'Downloading track JSON',
  136.             {'track': '%s:%s' % (track_id, album_id)})['track']
  137.         track_title = track['title']
  138.         track_version = track.get('version')
  139.         if track_version:
  140.             track_title = '%s (%s)' % (track_title, track_version)
  142.         download_data = self._download_json(
  143.             'https://music.yandex.ru/api/v2.1/handlers/track/%s:%s/web-album_track-track-track-main/download/m' % (track_id, album_id),
  144.             track_id, 'Downloading track location url JSON',
  145.             query={'hq': 1},
  146.             headers={'X-Retpath-Y': url})
  148.         fd_data = self._download_json(
  149.             download_data['src'], track_id,
  150.             'Downloading track location JSON',
  151.             query={'format': 'json'})
  152.         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + fd_data['path'][1:] + fd_data['s']).encode('utf-8')).hexdigest()
  153.         f_url = 'http://%s/get-mp3/%s/%s?track-id=%s ' % (fd_data['host'], key, fd_data['ts'] + fd_data['path'], track['id'])
  155.         thumbnail = None
  156.         cover_uri = track.get('albums', [{}])[0].get('coverUri')
  157.         if cover_uri:
  158.             thumbnail = cover_uri.replace('%%', 'orig')
  159.             if not thumbnail.startswith('http'):
  160.                 thumbnail = 'http://' + thumbnail
  162.         track_info = {
  163.             'id': track_id,
  164.             'ext': 'mp3',
  165.             'url': f_url,
  166.             'filesize': int_or_none(track.get('fileSize')),
  167.             'duration': float_or_none(track.get('durationMs'), 1000),
  168.             'thumbnail': thumbnail,
  169.             'track': track_title,
  170.             'acodec': download_data.get('codec'),
  171.             'abr': int_or_none(download_data.get('bitrate')),
  172.         }
  174.         def extract_artist_name(artist):
  175.             decomposed = artist.get('decomposed')
  176.             if not isinstance(decomposed, list):
  177.                 return artist['name']
  178.             parts = [artist['name']]
  179.             for element in decomposed:
  180.                 if isinstance(element, dict) and element.get('name'):
  181.                     parts.append(element['name'])
  182.                 elif isinstance(element, compat_str):
  183.                     parts.append(element)
  184.             return ''.join(parts)
  186.         def extract_artist(artist_list):
  187.             if artist_list and isinstance(artist_list, list):
  188.                 artists_names = [extract_artist_name(a) for a in artist_list if a.get('name')]
  189.                 if artists_names:
  190.                     return ', '.join(artists_names)
  192.         albums = track.get('albums')
  193.         if albums and isinstance(albums, list):
  194.             album = albums[0]
  195.             if isinstance(album, dict):
  196.                 year = album.get('year')
  197.                 disc_number = int_or_none(try_get(
  198.                     album, lambda x: x['trackPosition']['volume']))
  199.                 track_number = int_or_none(try_get(
  200.                     album, lambda x: x['trackPosition']['index']))
  201.                 track_info.update({
  202.                     'album': album.get('title'),
  203.                     'album_artist': extract_artist(album.get('artists')),
  204.                     'release_year': int_or_none(year),
  205.                     'genre': album.get('genre'),
  206.                     'disc_number': disc_number,
  207.                     'track_number': track_number,
  208.                 })
  210.         track_artist = extract_artist(track.get('artists'))
  211.         if track_artist:
  212.             track_info.update({
  213.                 'artist': track_artist,
  214.                 'title': '%s - %s' % (track_artist, track_title),
  215.             })
  216.         else:
  217.             track_info['title'] = track_title
  219.         return track_info
  222. class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
  223.     def _extract_tracks(self, source, item_id, url, tld):
  224.         tracks = source['tracks']
  225.         track_ids = [compat_str(track_id) for track_id in source['trackIds']]
  227.         # tracks dictionary shipped with playlist.jsx API is limited to 150 tracks,
  228.         # missing tracks should be retrieved manually.
  229.         if len(tracks) < len(track_ids):
  230.             present_track_ids = set([
  231.                 compat_str(track['id'])
  232.                 for track in tracks if track.get('id')])
  233.             missing_track_ids = [
  234.                 track_id for track_id in track_ids
  235.                 if track_id not in present_track_ids]
  236.             # Request missing tracks in chunks to avoid exceeding max HTTP header size,
  237.             # see https://github.com/ytdl-org/youtube-dl/issues/27355
  238.             _TRACKS_PER_CHUNK = 250
  239.             for chunk_num in itertools.count(0):
  240.                 start = chunk_num * _TRACKS_PER_CHUNK
  241.                 end = start + _TRACKS_PER_CHUNK
  242.                 missing_track_ids_req = missing_track_ids[start:end]
  243.                 assert missing_track_ids_req
  244.                 missing_tracks = self._call_api(
  245.                     'track-entries', tld, url, item_id,
  246.                     'Downloading missing tracks JSON chunk %d' % (chunk_num + 1), {
  247.                         'entries': ','.join(missing_track_ids_req),
  248.                         'lang': tld,
  249.                         'external-domain': 'music.yandex.%s' % tld,
  250.                         'overembed': 'false',
  251.                         'strict': 'true',
  252.                     })
  253.                 if missing_tracks:
  254.                     tracks.extend(missing_tracks)
  255.                 if end >= len(missing_track_ids):
  256.                     break
  258.         return tracks
  260.     def _build_playlist(self, tracks):
  261.         entries = []
  262.         for track in tracks:
  263.             track_id = track.get('id') or track.get('realId')
  264.             if not track_id:
  265.                 continue
  266.             albums = track.get('albums')
  267.             if not albums or not isinstance(albums, list):
  268.                 continue
  269.             album = albums[0]
  270.             if not isinstance(album, dict):
  271.                 continue
  272.             album_id = album.get('id')
  273.             if not album_id:
  274.                 continue
  275.             entries.append(self.url_result(
  276.                 'http://music.yandex.ru/album/%s/track/%s' % (album_id, track_id),
  277.                 ie=YandexMusicTrackIE.ie_key(), video_id=track_id))
  278.         return entries
  281. class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
  282.     IE_NAME = 'yandexmusic:album'
  283.     IE_DESC = 'Яндекс.Музыка - Альбом'
  284.     _VALID_URL = r'%s/album/(?P<id>\d+)' % YandexMusicBaseIE._VALID_URL_BASE
  286.     _TESTS = [{
  287.         'url': 'http://music.yandex.ru/album/540508',
  288.         'info_dict': {
  289.             'id': '540508',
  290.             'title': 'md5:7ed1c3567f28d14be9f61179116f5571',
  291.         },
  292.         'playlist_count': 50,
  293.         # 'skip': 'Travis CI servers blocked by YandexMusic',
  294.     }, {
  295.         'url': 'https://music.yandex.ru/album/3840501',
  296.         'info_dict': {
  297.             'id': '3840501',
  298.             'title': 'md5:36733472cdaa7dcb1fd9473f7da8e50f',
  299.         },
  300.         'playlist_count': 33,
  301.         # 'skip': 'Travis CI servers blocked by YandexMusic',
  302.     }, {
  303.         # empty artists
  304.         'url': 'https://music.yandex.ru/album/9091882',
  305.         'info_dict': {
  306.             'id': '9091882',
  307.             'title': 'ТЕД на русском',
  308.         },
  309.         'playlist_count': 187,
  310.     }]
  312.     @classmethod
  313.     def suitable(cls, url):
  314.         return False if YandexMusicTrackIE.suitable(url) else super(YandexMusicAlbumIE, cls).suitable(url)
  316.     def _real_extract(self, url):
  317.         mobj = re.match(self._VALID_URL, url)
  318.         tld = mobj.group('tld')
  319.         album_id = mobj.group('id')
  321.         album = self._call_api(
  322.             'album', tld, url, album_id, 'Downloading album JSON',
  323.             {'album': album_id})
  325.         entries = self._build_playlist([track for volume in album['volumes'] for track in volume])
  327.         title = album['title']
  328.         artist = try_get(album, lambda x: x['artists'][0]['name'], compat_str)
  329.         if artist:
  330.             title = '%s - %s' % (artist, title)
  331.         year = album.get('year')
  332.         if year:
  333.             title += ' (%s)' % year
  335.         return self.playlist_result(entries, compat_str(album['id']), title)
  338. class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
  339.     IE_NAME = 'yandexmusic:playlist'
  340.     IE_DESC = 'Яндекс.Музыка - Плейлист'
  341.     _VALID_URL = r'%s/users/(?P<user>[^/]+)/playlists/(?P<id>\d+)' % YandexMusicBaseIE._VALID_URL_BASE
  343.     _TESTS = [{
  344.         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
  345.         'info_dict': {
  346.             'id': '1245',
  347.             'title': 'md5:841559b3fe2b998eca88d0d2e22a3097',
  348.             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
  349.         },
  350.         'playlist_count': 5,
  351.         # 'skip': 'Travis CI servers blocked by YandexMusic',
  352.     }, {
  353.         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
  354.         'only_matching': True,
  355.     }, {
  356.         # playlist exceeding the limit of 150 tracks (see
  357.         # https://github.com/ytdl-org/youtube-dl/issues/6666)
  358.         'url': 'https://music.yandex.ru/users/mesiaz/playlists/1364',
  359.         'info_dict': {
  360.             'id': '1364',
  361.             'title': 'md5:b3b400f997d3f878a13ae0699653f7db',
  362.         },
  363.         'playlist_mincount': 437,
  364.         # 'skip': 'Travis CI servers blocked by YandexMusic',
  365.     }]
  367.     def _real_extract(self, url):
  368.         mobj = re.match(self._VALID_URL, url)
  369.         tld = mobj.group('tld')
  370.         user = mobj.group('user')
  371.         playlist_id = mobj.group('id')
  373.         playlist = self._call_api(
  374.             'playlist', tld, url, playlist_id, 'Downloading playlist JSON', {
  375.                 'owner': user,
  376.                 'kinds': playlist_id,
  377.                 'light': 'true',
  378.                 'lang': tld,
  379.                 'external-domain': 'music.yandex.%s' % tld,
  380.                 'overembed': 'false',
  381.             })['playlist']
  383.         tracks = self._extract_tracks(playlist, playlist_id, url, tld)
  385.         return self.playlist_result(
  386.             self._build_playlist(tracks),
  387.             compat_str(playlist_id),
  388.             playlist.get('title'), playlist.get('description'))
  391. class YandexMusicArtistBaseIE(YandexMusicPlaylistBaseIE):
  392.     def _call_artist(self, tld, url, artist_id):
  393.         return self._call_api(
  394.             'artist', tld, url, artist_id,
  395.             'Downloading artist %s JSON' % self._ARTIST_WHAT, {
  396.                 'artist': artist_id,
  397.                 'what': self._ARTIST_WHAT,
  398.                 'sort': self._ARTIST_SORT or '',
  399.                 'dir': '',
  400.                 'period': '',
  401.                 'lang': tld,
  402.                 'external-domain': 'music.yandex.%s' % tld,
  403.                 'overembed': 'false',
  404.             })
  406.     def _real_extract(self, url):
  407.         mobj = re.match(self._VALID_URL, url)
  408.         tld = mobj.group('tld')
  409.         artist_id = mobj.group('id')
  410.         data = self._call_artist(tld, url, artist_id)
  411.         tracks = self._extract_tracks(data, artist_id, url, tld)
  412.         title = try_get(data, lambda x: x['artist']['name'], compat_str)
  413.         return self.playlist_result(
  414.             self._build_playlist(tracks), artist_id, title)
  417. class YandexMusicArtistTracksIE(YandexMusicArtistBaseIE):
  418.     IE_NAME = 'yandexmusic:artist:tracks'
  419.     IE_DESC = 'Яндекс.Музыка - Артист - Треки'
  420.     _VALID_URL = r'%s/artist/(?P<id>\d+)/tracks' % YandexMusicBaseIE._VALID_URL_BASE
  422.     _TESTS = [{
  423.         'url': 'https://music.yandex.ru/artist/617526/tracks',
  424.         'info_dict': {
  425.             'id': '617526',
  426.             'title': 'md5:131aef29d45fd5a965ca613e708c040b',
  427.         },
  428.         'playlist_count': 507,
  429.         # 'skip': 'Travis CI servers blocked by YandexMusic',
  430.     }]
  432.     _ARTIST_SORT = ''
  433.     _ARTIST_WHAT = 'tracks'
  435.     def _real_extract(self, url):
  436.         mobj = re.match(self._VALID_URL, url)
  437.         tld = mobj.group('tld')
  438.         artist_id = mobj.group('id')
  439.         data = self._call_artist(tld, url, artist_id)
  440.         tracks = self._extract_tracks(data, artist_id, url, tld)
  441.         artist = try_get(data, lambda x: x['artist']['name'], compat_str)
  442.         title = '%s - %s' % (artist or artist_id, 'Треки')
  443.         return self.playlist_result(
  444.             self._build_playlist(tracks), artist_id, title)
  447. class YandexMusicArtistAlbumsIE(YandexMusicArtistBaseIE):
  448.     IE_NAME = 'yandexmusic:artist:albums'
  449.     IE_DESC = 'Яндекс.Музыка - Артист - Альбомы'
  450.     _VALID_URL = r'%s/artist/(?P<id>\d+)/albums' % YandexMusicBaseIE._VALID_URL_BASE
  452.     _TESTS = [{
  453.         'url': 'https://music.yandex.ru/artist/617526/albums',
  454.         'info_dict': {
  455.             'id': '617526',
  456.             'title': 'md5:55dc58d5c85699b7fb41ee926700236c',
  457.         },
  458.         'playlist_count': 8,
  459.         # 'skip': 'Travis CI servers blocked by YandexMusic',
  460.     }]
  462.     _ARTIST_SORT = 'year'
  463.     _ARTIST_WHAT = 'albums'
  465.     def _real_extract(self, url):
  466.         mobj = re.match(self._VALID_URL, url)
  467.         tld = mobj.group('tld')
  468.         artist_id = mobj.group('id')
  469.         data = self._call_artist(tld, url, artist_id)
  470.         entries = []
  471.         for album in data['albums']:
  472.             if not isinstance(album, dict):
  473.                 continue
  474.             album_id = album.get('id')
  475.             if not album_id:
  476.                 continue
  477.             entries.append(self.url_result(
  478.                 'http://music.yandex.ru/album/%s' % album_id,
  479.                 ie=YandexMusicAlbumIE.ie_key(), video_id=album_id))
  480.         artist = try_get(data, lambda x: x['artist']['name'], compat_str)
  481.         title = '%s - %s' % (artist or artist_id, 'Альбомы')
  482.         return self.playlist_result(entries, artist_id, title)