yt_dlp/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     urljoin,
  16 )
  17
  18
  19 class MailRuIE(InfoExtractor):
  20     IE_NAME = 'mailru'
  21     IE_DESC = 'Видео@Mail.Ru'
  22     _VALID_URL = r'''(?x)
  23                     https?://
  24                         (?:(?:www|m|videoapi)\.)?my\.mail\.ru/+
  25                         (?:
  26                             video/.*\#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|
  27                             (?:videos/embed/)?(?:(?P<idv2prefix>(?:[^/]+/+){2})(?:video/(?:embed/)?)?(?P<idv2suffix>[^/]+/\d+))(?:\.html)?|
  28                             (?:video/embed|\+/video/meta)/(?P<metaid>\d+)
  29                         )
  30                     '''
  31     _TESTS = [
  32         {
  33             'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
  34             'md5': 'dea205f03120046894db4ebb6159879a',
  35             'info_dict': {
  36                 'id': '46301138_76',
  37                 'ext': 'mp4',
  38                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
  39                 'timestamp': 1393235077,
  40                 'upload_date': '20140224',
  41                 'uploader': 'sonypicturesrus',
  42                 'uploader_id': 'sonypicturesrus@mail.ru',
  43                 'duration': 184,
  44             },
  45             'skip': 'Not accessible from Travis CI server',
  46         },
  47         {
  48             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
  49             'md5': '00a91a58c3402204dcced523777b475f',
  50             'info_dict': {
  51                 'id': '46843144_1263',
  52                 'ext': 'mp4',
  53                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
  54                 'timestamp': 1397039888,
  55                 'upload_date': '20140409',
  56                 'uploader': 'hitech',
  57                 'uploader_id': 'hitech@corp.mail.ru',
  58                 'duration': 245,
  59             },
  60             'skip': 'Not accessible from Travis CI server',
  61         },
  62         {
  63             # only available via metaUrl API
  64             'url': 'http://my.mail.ru/mail/720pizle/video/_myvideo/502.html',
  65             'md5': '3b26d2491c6949d031a32b96bd97c096',
  66             'info_dict': {
  67                 'id': '56664382_502',
  68                 'ext': 'mp4',
  69                 'title': ':8336',
  70                 'timestamp': 1449094163,
  71                 'upload_date': '20151202',
  72                 'uploader': '720pizle@mail.ru',
  73                 'uploader_id': '720pizle@mail.ru',
  74                 'duration': 6001,
  75             },
  76             'skip': 'Not accessible from Travis CI server',
  77         },
  78         {
  79             'url': 'http://m.my.mail.ru/mail/3sktvtr/video/_myvideo/138.html',
  80             'only_matching': True,
  81         },
  82         {
  83             'url': 'https://my.mail.ru/video/embed/7949340477499637815',
  84             'only_matching': True,
  85         },
  86         {
  87             'url': 'http://my.mail.ru/+/video/meta/7949340477499637815',
  88             'only_matching': True,
  89         },
  90         {
  91             'url': 'https://my.mail.ru//list/sinyutin10/video/_myvideo/4.html',
  92             'only_matching': True,
  93         },
  94         {
  95             'url': 'https://my.mail.ru//list//sinyutin10/video/_myvideo/4.html',
  96             'only_matching': True,
  97         },
  98         {
  99             'url': 'https://my.mail.ru/mail/cloud-strife/video/embed/Games/2009',
 100             'only_matching': True,
 101         },
 102         {
 103             'url': 'https://videoapi.my.mail.ru/videos/embed/mail/cloud-strife/Games/2009.html',
 104             'only_matching': True,
 105         }
 106     ]
 107
 108     def _real_extract(self, url):
 109         mobj = re.match(self._VALID_URL, url)
 110         meta_id = mobj.group('metaid')
 111
 112         video_id = None
 113         if meta_id:
 114             meta_url = 'https://my.mail.ru/+/video/meta/%s' % meta_id
 115         else:
 116             video_id = mobj.group('idv1')
 117             if not video_id:
 118                 video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
 119             webpage = self._download_webpage(url, video_id)
 120             page_config = self._parse_json(self._search_regex([
 121                 r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>',
 122                 r'(?s)"video":\s*({.+?}),'],
 123                 webpage, 'page config', default='{}'), video_id, fatal=False)
 124             if page_config:
 125                 meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl') or page_config.get('metadataUrl')
 126             else:
 127                 meta_url = None
 128
 129         video_data = None
 130
 131         # fix meta_url if missing the host address
 132         if re.match(r'^\/\+\/', meta_url):
 133             meta_url = urljoin('https://my.mail.ru', meta_url)
 134
 135         if meta_url:
 136             video_data = self._download_json(
 137                 meta_url, video_id or meta_id, 'Downloading video meta JSON',
 138                 fatal=not video_id)
 139
 140         # Fallback old approach
 141         if not video_data:
 142             video_data = self._download_json(
 143                 'http://api.video.mail.ru/videos/%s.json?new=1' % video_id,
 144                 video_id, 'Downloading video JSON')
 145
 146         headers = {}
 147
 148         video_key = self._get_cookies('https://my.mail.ru').get('video_key')
 149         if video_key:
 150             headers['Cookie'] = 'video_key=%s' % video_key.value
 151
 152         formats = []
 153         for f in video_data['videos']:
 154             video_url = f.get('url')
 155             if not video_url:
 156                 continue
 157             format_id = f.get('key')
 158             height = int_or_none(self._search_regex(
 159                 r'^(\d+)[pP]$', format_id, 'height', default=None)) if format_id else None
 160             formats.append({
 161                 'url': video_url,
 162                 'format_id': format_id,
 163                 'height': height,
 164                 'http_headers': headers,
 165             })
 166         self._sort_formats(formats)
 167
 168         meta_data = video_data['meta']
 169         title = remove_end(meta_data['title'], '.mp4')
 170
 171         author = video_data.get('author')
 172         uploader = author.get('name')
 173         uploader_id = author.get('id') or author.get('email')
 174         view_count = int_or_none(video_data.get('viewsCount') or video_data.get('views_count'))
 175
 176         acc_id = meta_data.get('accId')
 177         item_id = meta_data.get('itemId')
 178         content_id = '%s_%s' % (acc_id, item_id) if acc_id and item_id else video_id
 179
 180         thumbnail = meta_data.get('poster')
 181         duration = int_or_none(meta_data.get('duration'))
 182         timestamp = int_or_none(meta_data.get('timestamp'))
 183
 184         return {
 185             'id': content_id,
 186             'title': title,
 187             'thumbnail': thumbnail,
 188             'timestamp': timestamp,
 189             'uploader': uploader,
 190             'uploader_id': uploader_id,
 191             'duration': duration,
 192             'view_count': view_count,
 193             'formats': formats,
 194         }
 195
 196
 197 class MailRuMusicSearchBaseIE(InfoExtractor):
 198     def _search(self, query, url, audio_id, limit=100, offset=0):
 199         search = self._download_json(
 200             'https://my.mail.ru/cgi-bin/my/ajax', audio_id,
 201             'Downloading songs JSON page %d' % (offset // limit + 1),
 202             headers={
 203                 'Referer': url,
 204                 'X-Requested-With': 'XMLHttpRequest',
 205             }, query={
 206                 'xemail': '',
 207                 'ajax_call': '1',
 208                 'func_name': 'music.search',
 209                 'mna': '',
 210                 'mnb': '',
 211                 'arg_query': query,
 212                 'arg_extended': '1',
 213                 'arg_search_params': json.dumps({
 214                     'music': {
 215                         'limit': limit,
 216                         'offset': offset,
 217                     },
 218                 }),
 219                 'arg_limit': limit,
 220                 'arg_offset': offset,
 221             })
 222         return next(e for e in search if isinstance(e, dict))
 223
 224     @staticmethod
 225     def _extract_track(t, fatal=True):
 226         audio_url = t['URL'] if fatal else t.get('URL')
 227         if not audio_url:
 228             return
 229
 230         audio_id = t['File'] if fatal else t.get('File')
 231         if not audio_id:
 232             return
 233
 234         thumbnail = t.get('AlbumCoverURL') or t.get('FiledAlbumCover')
 235         uploader = t.get('OwnerName') or t.get('OwnerName_Text_HTML')
 236         uploader_id = t.get('UploaderID')
 237         duration = int_or_none(t.get('DurationInSeconds')) or parse_duration(
 238             t.get('Duration') or t.get('DurationStr'))
 239         view_count = int_or_none(t.get('PlayCount') or t.get('PlayCount_hr'))
 240
 241         track = t.get('Name') or t.get('Name_Text_HTML')
 242         artist = t.get('Author') or t.get('Author_Text_HTML')
 243
 244         if track:
 245             title = '%s - %s' % (artist, track) if artist else track
 246         else:
 247             title = audio_id
 248
 249         return {
 250             'extractor_key': MailRuMusicIE.ie_key(),
 251             'id': audio_id,
 252             'title': title,
 253             'thumbnail': thumbnail,
 254             'uploader': uploader,
 255             'uploader_id': uploader_id,
 256             'duration': duration,
 257             'view_count': view_count,
 258             'vcodec': 'none',
 259             'abr': int_or_none(t.get('BitRate')),
 260             'track': track,
 261             'artist': artist,
 262             'album': t.get('Album'),
 263             'url': audio_url,
 264         }
 265
 266
 267 class MailRuMusicIE(MailRuMusicSearchBaseIE):
 268     IE_NAME = 'mailru:music'
 269     IE_DESC = 'Музыка@Mail.Ru'
 270     _VALID_URL = r'https?://my\.mail\.ru/+music/+songs/+[^/?#&]+-(?P<id>[\da-f]+)'
 271     _TESTS = [{
 272         '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',
 273         'md5': '0f8c22ef8c5d665b13ac709e63025610',
 274         'info_dict': {
 275             'id': '4e31f7125d0dfaef505d947642366893',
 276             'ext': 'mp3',
 277             'title': 'L.A.H. (Luciferian Aesthetics of Herrschaft) single, 2017 - М8Л8ТХ',
 278             'uploader': 'Игорь Мудрый',
 279             'uploader_id': '1459196328',
 280             'duration': 280,
 281             'view_count': int,
 282             'vcodec': 'none',
 283             'abr': 320,
 284             'track': 'L.A.H. (Luciferian Aesthetics of Herrschaft) single, 2017',
 285             'artist': 'М8Л8ТХ',
 286         },
 287     }]
 288
 289     def _real_extract(self, url):
 290         audio_id = self._match_id(url)
 291
 292         webpage = self._download_webpage(url, audio_id)
 293
 294         title = self._og_search_title(webpage)
 295         music_data = self._search(title, url, audio_id)['MusicData']
 296         t = next(t for t in music_data if t.get('File') == audio_id)
 297
 298         info = self._extract_track(t)
 299         info['title'] = title
 300         return info
 301
 302
 303 class MailRuMusicSearchIE(MailRuMusicSearchBaseIE):
 304     IE_NAME = 'mailru:music:search'
 305     IE_DESC = 'Музыка@Mail.Ru'
 306     _VALID_URL = r'https?://my\.mail\.ru/+music/+search/+(?P<id>[^/?#&]+)'
 307     _TESTS = [{
 308         'url': 'https://my.mail.ru/music/search/black%20shadow',
 309         'info_dict': {
 310             'id': 'black shadow',
 311         },
 312         'playlist_mincount': 532,
 313     }]
 314
 315     def _real_extract(self, url):
 316         query = compat_urllib_parse_unquote(self._match_id(url))
 317
 318         entries = []
 319
 320         LIMIT = 100
 321         offset = 0
 322
 323         for _ in itertools.count(1):
 324             search = self._search(query, url, query, LIMIT, offset)
 325
 326             music_data = search.get('MusicData')
 327             if not music_data or not isinstance(music_data, list):
 328                 break
 329
 330             for t in music_data:
 331                 track = self._extract_track(t, fatal=False)
 332                 if track:
 333                     entries.append(track)
 334
 335             total = try_get(
 336                 search, lambda x: x['Results']['music']['Total'], int)
 337
 338             if total is not None:
 339                 if offset > total:
 340                     break
 341
 342             offset += LIMIT
 343
 344         return self.playlist_result(entries, query)