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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import hashlib
   6 import itertools
   7 import json
   8 import random
   9 import re
  10 import string
  11
  12 from .common import InfoExtractor
  13 from ..compat import compat_struct_pack
  14 from ..utils import (
  15     determine_ext,
  16     error_to_compat_str,
  17     ExtractorError,
  18     int_or_none,
  19     parse_iso8601,
  20     sanitized_Request,
  21     str_to_int,
  22     unescapeHTML,
  23     mimetype2ext,
  24 )
  25
  26
  27 class DailymotionBaseInfoExtractor(InfoExtractor):
  28     @staticmethod
  29     def _build_request(url):
  30         """Build a request with the family filter disabled"""
  31         request = sanitized_Request(url)
  32         request.add_header('Cookie', 'family_filter=off; ff=off')
  33         return request
  34
  35     def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
  36         request = self._build_request(url)
  37         return self._download_webpage_handle(request, *args, **kwargs)
  38
  39     def _download_webpage_no_ff(self, url, *args, **kwargs):
  40         request = self._build_request(url)
  41         return self._download_webpage(request, *args, **kwargs)
  42
  43
  44 class DailymotionIE(DailymotionBaseInfoExtractor):
  45     _VALID_URL = r'(?i)https?://(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:embed|swf|#)/)?video|swf)/(?P<id>[^/?_]+)'
  46     IE_NAME = 'dailymotion'
  47
  48     _FORMATS = [
  49         ('stream_h264_ld_url', 'ld'),
  50         ('stream_h264_url', 'standard'),
  51         ('stream_h264_hq_url', 'hq'),
  52         ('stream_h264_hd_url', 'hd'),
  53         ('stream_h264_hd1080_url', 'hd180'),
  54     ]
  55
  56     _TESTS = [{
  57         'url': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news',
  58         'md5': '074b95bdee76b9e3654137aee9c79dfe',
  59         'info_dict': {
  60             'id': 'x5kesuj',
  61             'ext': 'mp4',
  62             'title': 'Office Christmas Party Review –  Jason Bateman, Olivia Munn, T.J. Miller',
  63             'description': 'Office Christmas Party Review -  Jason Bateman, Olivia Munn, T.J. Miller',
  64             'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
  65             'duration': 187,
  66             'timestamp': 1493651285,
  67             'upload_date': '20170501',
  68             'uploader': 'Deadline',
  69             'uploader_id': 'x1xm8ri',
  70             'age_limit': 0,
  71         },
  72     }, {
  73         'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
  74         'md5': '2137c41a8e78554bb09225b8eb322406',
  75         'info_dict': {
  76             'id': 'x2iuewm',
  77             'ext': 'mp4',
  78             'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
  79             'description': 'Several come bundled with the Steam Controller.',
  80             'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
  81             'duration': 74,
  82             'timestamp': 1425657362,
  83             'upload_date': '20150306',
  84             'uploader': 'IGN',
  85             'uploader_id': 'xijv66',
  86             'age_limit': 0,
  87             'view_count': int,
  88         },
  89         'skip': 'video gone',
  90     }, {
  91         # Vevo video
  92         'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  93         'info_dict': {
  94             'title': 'Roar (Official)',
  95             'id': 'USUV71301934',
  96             'ext': 'mp4',
  97             'uploader': 'Katy Perry',
  98             'upload_date': '20130905',
  99         },
 100         'params': {
 101             'skip_download': True,
 102         },
 103         'skip': 'VEVO is only available in some countries',
 104     }, {
 105         # age-restricted video
 106         'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
 107         'md5': '0d667a7b9cebecc3c89ee93099c4159d',
 108         'info_dict': {
 109             'id': 'xyh2zz',
 110             'ext': 'mp4',
 111             'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
 112             'uploader': 'HotWaves1012',
 113             'age_limit': 18,
 114         },
 115         'skip': 'video gone',
 116     }, {
 117         # geo-restricted, player v5
 118         'url': 'http://www.dailymotion.com/video/xhza0o',
 119         'only_matching': True,
 120     }, {
 121         # with subtitles
 122         'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
 123         'only_matching': True,
 124     }, {
 125         'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
 126         'only_matching': True,
 127     }, {
 128         'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun',
 129         'only_matching': True,
 130     }]
 131
 132     @staticmethod
 133     def _extract_urls(webpage):
 134         # Look for embedded Dailymotion player
 135         matches = re.findall(
 136             r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
 137         return list(map(lambda m: unescapeHTML(m[1]), matches))
 138
 139     def _real_extract(self, url):
 140         video_id = self._match_id(url)
 141
 142         webpage = self._download_webpage_no_ff(
 143             'https://www.dailymotion.com/video/%s' % video_id, video_id)
 144
 145         age_limit = self._rta_search(webpage)
 146
 147         description = self._og_search_description(webpage) or self._html_search_meta(
 148             'description', webpage, 'description')
 149
 150         view_count_str = self._search_regex(
 151             (r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"',
 152              r'video_views_count[^>]+>\s+([\s\d\,.]+)'),
 153             webpage, 'view count', default=None)
 154         if view_count_str:
 155             view_count_str = re.sub(r'\s', '', view_count_str)
 156         view_count = str_to_int(view_count_str)
 157         comment_count = int_or_none(self._search_regex(
 158             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
 159             webpage, 'comment count', default=None))
 160
 161         player_v5 = self._search_regex(
 162             [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
 163              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
 164              r'buildPlayer\(({.+?})\);',
 165              r'var\s+config\s*=\s*({.+?});',
 166              # New layout regex (see https://github.com/rg3/youtube-dl/issues/13580)
 167              r'__PLAYER_CONFIG__\s*=\s*({.+?});'],
 168             webpage, 'player v5', default=None)
 169         if player_v5:
 170             player = self._parse_json(player_v5, video_id)
 171             metadata = player['metadata']
 172
 173             if metadata.get('error', {}).get('type') == 'password_protected':
 174                 password = self._downloader.params.get('videopassword')
 175                 if password:
 176                     r = int(metadata['id'][1:], 36)
 177                     us64e = lambda x: base64.urlsafe_b64encode(x).decode().strip('=')
 178                     t = ''.join(random.choice(string.ascii_letters) for i in range(10))
 179                     n = us64e(compat_struct_pack('I', r))
 180                     i = us64e(hashlib.md5(('%s%d%s' % (password, r, t)).encode()).digest())
 181                     metadata = self._download_json(
 182                         'http://www.dailymotion.com/player/metadata/video/p' + i + t + n, video_id)
 183
 184             self._check_error(metadata)
 185
 186             formats = []
 187             for quality, media_list in metadata['qualities'].items():
 188                 for media in media_list:
 189                     media_url = media.get('url')
 190                     if not media_url:
 191                         continue
 192                     type_ = media.get('type')
 193                     if type_ == 'application/vnd.lumberjack.manifest':
 194                         continue
 195                     ext = mimetype2ext(type_) or determine_ext(media_url)
 196                     if ext == 'm3u8':
 197                         m3u8_formats = self._extract_m3u8_formats(
 198                             media_url, video_id, 'mp4', preference=-1,
 199                             m3u8_id='hls', fatal=False)
 200                         for f in m3u8_formats:
 201                             f['url'] = f['url'].split('#')[0]
 202                             formats.append(f)
 203                     elif ext == 'f4m':
 204                         formats.extend(self._extract_f4m_formats(
 205                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
 206                     else:
 207                         f = {
 208                             'url': media_url,
 209                             'format_id': 'http-%s' % quality,
 210                             'ext': ext,
 211                         }
 212                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 213                         if m:
 214                             f.update({
 215                                 'width': int(m.group('width')),
 216                                 'height': int(m.group('height')),
 217                             })
 218                         formats.append(f)
 219             self._sort_formats(formats)
 220
 221             title = metadata['title']
 222             duration = int_or_none(metadata.get('duration'))
 223             timestamp = int_or_none(metadata.get('created_time'))
 224             thumbnail = metadata.get('poster_url')
 225             uploader = metadata.get('owner', {}).get('screenname')
 226             uploader_id = metadata.get('owner', {}).get('id')
 227
 228             subtitles = {}
 229             subtitles_data = metadata.get('subtitles', {}).get('data', {})
 230             if subtitles_data and isinstance(subtitles_data, dict):
 231                 for subtitle_lang, subtitle in subtitles_data.items():
 232                     subtitles[subtitle_lang] = [{
 233                         'ext': determine_ext(subtitle_url),
 234                         'url': subtitle_url,
 235                     } for subtitle_url in subtitle.get('urls', [])]
 236
 237             return {
 238                 'id': video_id,
 239                 'title': title,
 240                 'description': description,
 241                 'thumbnail': thumbnail,
 242                 'duration': duration,
 243                 'timestamp': timestamp,
 244                 'uploader': uploader,
 245                 'uploader_id': uploader_id,
 246                 'age_limit': age_limit,
 247                 'view_count': view_count,
 248                 'comment_count': comment_count,
 249                 'formats': formats,
 250                 'subtitles': subtitles,
 251             }
 252
 253         # vevo embed
 254         vevo_id = self._search_regex(
 255             r'<link rel="video_src" href="[^"]*?vevo\.com[^"]*?video=(?P<id>[\w]*)',
 256             webpage, 'vevo embed', default=None)
 257         if vevo_id:
 258             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 259
 260         # fallback old player
 261         embed_page = self._download_webpage_no_ff(
 262             'https://www.dailymotion.com/embed/video/%s' % video_id,
 263             video_id, 'Downloading embed page')
 264
 265         timestamp = parse_iso8601(self._html_search_meta(
 266             'video:release_date', webpage, 'upload date'))
 267
 268         info = self._parse_json(
 269             self._search_regex(
 270                 r'var info = ({.*?}),$', embed_page,
 271                 'video info', flags=re.MULTILINE),
 272             video_id)
 273
 274         self._check_error(info)
 275
 276         formats = []
 277         for (key, format_id) in self._FORMATS:
 278             video_url = info.get(key)
 279             if video_url is not None:
 280                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 281                 if m_size is not None:
 282                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 283                 else:
 284                     width, height = None, None
 285                 formats.append({
 286                     'url': video_url,
 287                     'ext': 'mp4',
 288                     'format_id': format_id,
 289                     'width': width,
 290                     'height': height,
 291                 })
 292         self._sort_formats(formats)
 293
 294         # subtitles
 295         video_subtitles = self.extract_subtitles(video_id, webpage)
 296
 297         title = self._og_search_title(webpage, default=None)
 298         if title is None:
 299             title = self._html_search_regex(
 300                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 301                 'title')
 302
 303         return {
 304             'id': video_id,
 305             'formats': formats,
 306             'uploader': info['owner.screenname'],
 307             'timestamp': timestamp,
 308             'title': title,
 309             'description': description,
 310             'subtitles': video_subtitles,
 311             'thumbnail': info['thumbnail_url'],
 312             'age_limit': age_limit,
 313             'view_count': view_count,
 314             'duration': info['duration']
 315         }
 316
 317     def _check_error(self, info):
 318         error = info.get('error')
 319         if error:
 320             title = error.get('title') or error['message']
 321             # See https://developer.dailymotion.com/api#access-error
 322             if error.get('code') == 'DM007':
 323                 self.raise_geo_restricted(msg=title)
 324             raise ExtractorError(
 325                 '%s said: %s' % (self.IE_NAME, title), expected=True)
 326
 327     def _get_subtitles(self, video_id, webpage):
 328         try:
 329             sub_list = self._download_webpage(
 330                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 331                 video_id, note=False)
 332         except ExtractorError as err:
 333             self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
 334             return {}
 335         info = json.loads(sub_list)
 336         if (info['total'] > 0):
 337             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 338             return sub_lang_list
 339         self._downloader.report_warning('video doesn\'t have subtitles')
 340         return {}
 341
 342
 343 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 344     IE_NAME = 'dailymotion:playlist'
 345     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>[^/?#&]+)'
 346     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 347     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 348     _TESTS = [{
 349         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 350         'info_dict': {
 351             'title': 'SPORT',
 352             'id': 'xv4bw_nqtv_sport',
 353         },
 354         'playlist_mincount': 20,
 355     }]
 356
 357     def _extract_entries(self, id):
 358         video_ids = set()
 359         processed_urls = set()
 360         for pagenum in itertools.count(1):
 361             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 362             webpage, urlh = self._download_webpage_handle_no_ff(
 363                 page_url, id, 'Downloading page %s' % pagenum)
 364             if urlh.geturl() in processed_urls:
 365                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 366                     page_url, urlh.geturl()), id)
 367                 break
 368
 369             processed_urls.add(urlh.geturl())
 370
 371             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
 372                 if video_id not in video_ids:
 373                     yield self.url_result(
 374                         'http://www.dailymotion.com/video/%s' % video_id,
 375                         DailymotionIE.ie_key(), video_id)
 376                     video_ids.add(video_id)
 377
 378             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 379                 break
 380
 381     def _real_extract(self, url):
 382         mobj = re.match(self._VALID_URL, url)
 383         playlist_id = mobj.group('id')
 384         webpage = self._download_webpage(url, playlist_id)
 385
 386         return {
 387             '_type': 'playlist',
 388             'id': playlist_id,
 389             'title': self._og_search_title(webpage),
 390             'entries': self._extract_entries(playlist_id),
 391         }
 392
 393
 394 class DailymotionUserIE(DailymotionPlaylistIE):
 395     IE_NAME = 'dailymotion:user'
 396     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
 397     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 398     _TESTS = [{
 399         'url': 'https://www.dailymotion.com/user/nqtv',
 400         'info_dict': {
 401             'id': 'nqtv',
 402             'title': 'Rémi Gaillard',
 403         },
 404         'playlist_mincount': 100,
 405     }, {
 406         'url': 'http://www.dailymotion.com/user/UnderProject',
 407         'info_dict': {
 408             'id': 'UnderProject',
 409             'title': 'UnderProject',
 410         },
 411         'playlist_mincount': 1800,
 412         'expected_warnings': [
 413             'Stopped at duplicated page',
 414         ],
 415         'skip': 'Takes too long time',
 416     }]
 417
 418     def _real_extract(self, url):
 419         mobj = re.match(self._VALID_URL, url)
 420         user = mobj.group('user')
 421         webpage = self._download_webpage(
 422             'https://www.dailymotion.com/user/%s' % user, user)
 423         full_user = unescapeHTML(self._html_search_regex(
 424             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 425             webpage, 'user'))
 426
 427         return {
 428             '_type': 'playlist',
 429             'id': user,
 430             'title': full_user,
 431             'entries': self._extract_entries(user),
 432         }