Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/mailru.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import json
   6 import re
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_urllib_parse_unquote
  10 from ..utils import (
  11     int_or_none,
  12     parse_duration,
  13     remove_end,
  14     try_get,
  15 )
  16
  17
  18 class MailRuIE(InfoExtractor):
  19     IE_NAME = 'mailru'
  20     IE_DESC = 'Видео@Mail.Ru'
  21     _VALID_URL = r'''(?x)
  22                     https?://
  23                         (?:(?:www|m)\.)?my\.mail\.ru/
  24                         (?:
  25                             video/.*\#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|
  26                             (?:(?P<idv2prefix>(?:[^/]+/){2})video/(?P<idv2suffix>[^/]+/\d+))\.html|
  27                             (?:video/embed|\+/video/meta)/(?P<metaid>\d+)
  28                         )
  29                     '''
  30     _TESTS = [
  31         {
  32             'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
  33             'md5': 'dea205f03120046894db4ebb6159879a',
  34             'info_dict': {
  35                 'id': '46301138_76',
  36                 'ext': 'mp4',
  37                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
  38                 'timestamp': 1393235077,
  39                 'upload_date': '20140224',
  40                 'uploader': 'sonypicturesrus',
  41                 'uploader_id': 'sonypicturesrus@mail.ru',
  42                 'duration': 184,
  43             },
  44             'skip': 'Not accessible from Travis CI server',
  45         },
  46         {
  47             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
  48             'md5': '00a91a58c3402204dcced523777b475f',
  49             'info_dict': {
  50                 'id': '46843144_1263',
  51                 'ext': 'mp4',
  52                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
  53                 'timestamp': 1397039888,
  54                 'upload_date': '20140409',
  55                 'uploader': 'hitech',
  56                 'uploader_id': 'hitech@corp.mail.ru',
  57                 'duration': 245,
  58             },
  59             'skip': 'Not accessible from Travis CI server',
  60         },
  61         {
  62             # only available via metaUrl API
  63             'url': 'http://my.mail.ru/mail/720pizle/video/_myvideo/502.html',
  64             'md5': '3b26d2491c6949d031a32b96bd97c096',
  65             'info_dict': {
  66                 'id': '56664382_502',
  67                 'ext': 'mp4',
  68                 'title': ':8336',
  69                 'timestamp': 1449094163,
  70                 'upload_date': '20151202',
  71                 'uploader': '720pizle@mail.ru',
  72                 'uploader_id': '720pizle@mail.ru',
  73                 'duration': 6001,
  74             },
  75             'skip': 'Not accessible from Travis CI server',
  76         },
  77         {
  78             'url': 'http://m.my.mail.ru/mail/3sktvtr/video/_myvideo/138.html',
  79             'only_matching': True,
  80         },
  81         {
  82             'url': 'https://my.mail.ru/video/embed/7949340477499637815',
  83             'only_matching': True,
  84         },
  85         {
  86             'url': 'http://my.mail.ru/+/video/meta/7949340477499637815',
  87             'only_matching': True,
  88         }
  89     ]
  90
  91     def _real_extract(self, url):
  92         mobj = re.match(self._VALID_URL, url)
  93         meta_id = mobj.group('metaid')
  94
  95         video_id = None
  96         if meta_id:
  97             meta_url = 'https://my.mail.ru/+/video/meta/%s' % meta_id
  98         else:
  99             video_id = mobj.group('idv1')
 100             if not video_id:
 101                 video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
 102             webpage = self._download_webpage(url, video_id)
 103             page_config = self._parse_json(self._search_regex(
 104                 r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>',
 105                 webpage, 'page config', default='{}'), video_id, fatal=False)
 106             if page_config:
 107                 meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl')
 108             else:
 109                 meta_url = None
 110
 111         video_data = None
 112         if meta_url:
 113             video_data = self._download_json(
 114                 meta_url, video_id or meta_id, 'Downloading video meta JSON',
 115                 fatal=not video_id)
 116
 117         # Fallback old approach
 118         if not video_data:
 119             video_data = self._download_json(
 120                 'http://api.video.mail.ru/videos/%s.json?new=1' % video_id,
 121                 video_id, 'Downloading video JSON')
 122
 123         formats = []
 124         for f in video_data['videos']:
 125             video_url = f.get('url')
 126             if not video_url:
 127                 continue
 128             format_id = f.get('key')
 129             height = int_or_none(self._search_regex(
 130                 r'^(\d+)[pP]$', format_id, 'height', default=None)) if format_id else None
 131             formats.append({
 132                 'url': video_url,
 133                 'format_id': format_id,
 134                 'height': height,
 135             })
 136         self._sort_formats(formats)
 137
 138         meta_data = video_data['meta']
 139         title = remove_end(meta_data['title'], '.mp4')
 140
 141         author = video_data.get('author')
 142         uploader = author.get('name')
 143         uploader_id = author.get('id') or author.get('email')
 144         view_count = int_or_none(video_data.get('viewsCount') or video_data.get('views_count'))
 145
 146         acc_id = meta_data.get('accId')
 147         item_id = meta_data.get('itemId')
 148         content_id = '%s_%s' % (acc_id, item_id) if acc_id and item_id else video_id
 149
 150         thumbnail = meta_data.get('poster')
 151         duration = int_or_none(meta_data.get('duration'))
 152         timestamp = int_or_none(meta_data.get('timestamp'))
 153
 154         return {
 155             'id': content_id,
 156             'title': title,
 157             'thumbnail': thumbnail,
 158             'timestamp': timestamp,
 159             'uploader': uploader,
 160             'uploader_id': uploader_id,
 161             'duration': duration,
 162             'view_count': view_count,
 163             'formats': formats,
 164         }
 165
 166
 167 class MailRuMusicSearchBaseIE(InfoExtractor):
 168     def _search(self, query, url, audio_id, limit=100, offset=0):
 169         search = self._download_json(
 170             'https://my.mail.ru/cgi-bin/my/ajax', audio_id,
 171             'Downloading songs JSON page %d' % (offset // limit + 1),
 172             headers={
 173                 'Referer': url,
 174                 'X-Requested-With': 'XMLHttpRequest',
 175             }, query={
 176                 'xemail': '',
 177                 'ajax_call': '1',
 178                 'func_name': 'music.search',
 179                 'mna': '',
 180                 'mnb': '',
 181                 'arg_query': query,
 182                 'arg_extended': '1',
 183                 'arg_search_params': json.dumps({
 184                     'music': {
 185                         'limit': limit,
 186                         'offset': offset,
 187                     },
 188                 }),
 189                 'arg_limit': limit,
 190                 'arg_offset': offset,
 191             })
 192         return next(e for e in search if isinstance(e, dict))
 193
 194     @staticmethod
 195     def _extract_track(t, fatal=True):
 196         audio_url = t['URL'] if fatal else t.get('URL')
 197         if not audio_url:
 198             return
 199
 200         audio_id = t['File'] if fatal else t.get('File')
 201         if not audio_id:
 202             return
 203
 204         thumbnail = t.get('AlbumCoverURL') or t.get('FiledAlbumCover')
 205         uploader = t.get('OwnerName') or t.get('OwnerName_Text_HTML')
 206         uploader_id = t.get('UploaderID')
 207         duration = int_or_none(t.get('DurationInSeconds')) or parse_duration(
 208             t.get('Duration') or t.get('DurationStr'))
 209         view_count = int_or_none(t.get('PlayCount') or t.get('PlayCount_hr'))
 210
 211         track = t.get('Name') or t.get('Name_Text_HTML')
 212         artist = t.get('Author') or t.get('Author_Text_HTML')
 213
 214         if track:
 215             title = '%s - %s' % (artist, track) if artist else track
 216         else:
 217             title = audio_id
 218
 219         return {
 220             'extractor_key': MailRuMusicIE.ie_key(),
 221             'id': audio_id,
 222             'title': title,
 223             'thumbnail': thumbnail,
 224             'uploader': uploader,
 225             'uploader_id': uploader_id,
 226             'duration': duration,
 227             'view_count': view_count,
 228             'vcodec': 'none',
 229             'abr': int_or_none(t.get('BitRate')),
 230             'track': track,
 231             'artist': artist,
 232             'album': t.get('Album'),
 233             'url': audio_url,
 234         }
 235
 236
 237 class MailRuMusicIE(MailRuMusicSearchBaseIE):
 238     IE_NAME = 'mailru:music'
 239     IE_DESC = 'Музыка@Mail.Ru'
 240     _VALID_URL = r'https?://my\.mail\.ru/music/songs/[^/?#&]+-(?P<id>[\da-f]+)'
 241     _TESTS = [{
 242         'url': 'https://my.mail.ru/music/songs/%D0%BC8%D0%BB8%D1%82%D1%85-l-a-h-luciferian-aesthetics-of-herrschaft-single-2017-4e31f7125d0dfaef505d947642366893',
 243         'md5': '0f8c22ef8c5d665b13ac709e63025610',
 244         'info_dict': {
 245             'id': '4e31f7125d0dfaef505d947642366893',
 246             'ext': 'mp3',
 247             'title': 'L.A.H. (Luciferian Aesthetics of Herrschaft) single, 2017 - М8Л8ТХ',
 248             'uploader': 'Игорь Мудрый',
 249             'uploader_id': '1459196328',
 250             'duration': 280,
 251             'view_count': int,
 252             'vcodec': 'none',
 253             'abr': 320,
 254             'track': 'L.A.H. (Luciferian Aesthetics of Herrschaft) single, 2017',
 255             'artist': 'М8Л8ТХ',
 256         },
 257     }]
 258
 259     def _real_extract(self, url):
 260         audio_id = self._match_id(url)
 261
 262         webpage = self._download_webpage(url, audio_id)
 263
 264         title = self._og_search_title(webpage)
 265         music_data = self._search(title, url, audio_id)['MusicData']
 266         t = next(t for t in music_data if t.get('File') == audio_id)
 267
 268         info = self._extract_track(t)
 269         info['title'] = title
 270         return info
 271
 272
 273 class MailRuMusicSearchIE(MailRuMusicSearchBaseIE):
 274     IE_NAME = 'mailru:music:search'
 275     IE_DESC = 'Музыка@Mail.Ru'
 276     _VALID_URL = r'https?://my\.mail\.ru/music/search/(?P<id>[^/?#&]+)'
 277     _TESTS = [{
 278         'url': 'https://my.mail.ru/music/search/black%20shadow',
 279         'info_dict': {
 280             'id': 'black shadow',
 281         },
 282         'playlist_mincount': 532,
 283     }]
 284
 285     def _real_extract(self, url):
 286         query = compat_urllib_parse_unquote(self._match_id(url))
 287
 288         entries = []
 289
 290         LIMIT = 100
 291         offset = 0
 292
 293         for _ in itertools.count(1):
 294             search = self._search(query, url, query, LIMIT, offset)
 295
 296             music_data = search.get('MusicData')
 297             if not music_data or not isinstance(music_data, list):
 298                 break
 299
 300             for t in music_data:
 301                 track = self._extract_track(t, fatal=False)
 302                 if track:
 303                     entries.append(track)
 304
 305             total = try_get(
 306                 search, lambda x: x['Results']['music']['Total'], int)
 307
 308             if total is not None:
 309                 if offset > total:
 310                     break
 311
 312             offset += LIMIT
 313
 314         return self.playlist_result(entries, query)