2 from __future__ import unicode_literals
8 from .common import InfoExtractor
23 class DailymotionBaseInfoExtractor(InfoExtractor):
25 def _build_request(url):
26 """Build a request with the family filter disabled"""
27 request = sanitized_Request(url)
28 request.add_header('Cookie', 'family_filter=off; ff=off')
31 def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
32 request = self._build_request(url)
33 return self._download_webpage_handle(request, *args, **kwargs)
35 def _download_webpage_no_ff(self, url, *args, **kwargs):
36 request = self._build_request(url)
37 return self._download_webpage(request, *args, **kwargs)
40 class DailymotionIE(DailymotionBaseInfoExtractor):
41 _VALID_URL = r'(?i)https?://(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:embed|swf|#)/)?video|swf)/(?P<id>[^/?_]+)'
42 IE_NAME = 'dailymotion'
45 ('stream_h264_ld_url', 'ld'),
46 ('stream_h264_url', 'standard'),
47 ('stream_h264_hq_url', 'hq'),
48 ('stream_h264_hd_url', 'hd'),
49 ('stream_h264_hd1080_url', 'hd180'),
53 'url': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news',
54 'md5': '074b95bdee76b9e3654137aee9c79dfe',
58 'title': 'Office Christmas Party Review – Jason Bateman, Olivia Munn, T.J. Miller',
59 'description': 'Office Christmas Party Review - Jason Bateman, Olivia Munn, T.J. Miller',
60 'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
62 'timestamp': 1493651285,
63 'upload_date': '20170501',
64 'uploader': 'Deadline',
65 'uploader_id': 'x1xm8ri',
70 'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
71 'md5': '2137c41a8e78554bb09225b8eb322406',
75 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
76 'description': 'Several come bundled with the Steam Controller.',
77 'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
79 'timestamp': 1425657362,
80 'upload_date': '20150306',
82 'uploader_id': 'xijv66',
89 'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
91 'title': 'Roar (Official)',
94 'uploader': 'Katy Perry',
95 'upload_date': '20130905',
98 'skip_download': True,
100 'skip': 'VEVO is only available in some countries',
102 # age-restricted video
103 'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
104 'md5': '0d667a7b9cebecc3c89ee93099c4159d',
108 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
109 'uploader': 'HotWaves1012',
112 'skip': 'video gone',
114 # geo-restricted, player v5
115 'url': 'http://www.dailymotion.com/video/xhza0o',
116 'only_matching': True,
119 'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
120 'only_matching': True,
122 'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
123 'only_matching': True,
125 'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun',
126 'only_matching': True,
130 def _extract_urls(webpage):
131 # Look for embedded Dailymotion player
132 matches = re.findall(
133 r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
134 return list(map(lambda m: unescapeHTML(m[1]), matches))
136 def _real_extract(self, url):
137 video_id = self._match_id(url)
139 webpage = self._download_webpage_no_ff(
140 'https://www.dailymotion.com/video/%s' % video_id, video_id)
142 age_limit = self._rta_search(webpage)
144 description = self._og_search_description(webpage) or self._html_search_meta(
145 'description', webpage, 'description')
147 view_count_str = self._search_regex(
148 (r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"',
149 r'video_views_count[^>]+>\s+([\s\d\,.]+)'),
150 webpage, 'view count', default=None)
152 view_count_str = re.sub(r'\s', '', view_count_str)
153 view_count = str_to_int(view_count_str)
154 comment_count = int_or_none(self._search_regex(
155 r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
156 webpage, 'comment count', default=None))
158 player_v5 = self._search_regex(
159 [r'buildPlayer\(({.+?})\);\n', # See https://github.com/rg3/youtube-dl/issues/7826
160 r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
161 r'buildPlayer\(({.+?})\);',
162 r'var\s+config\s*=\s*({.+?});',
163 # New layout regex (see https://github.com/rg3/youtube-dl/issues/13580)
164 r'__PLAYER_CONFIG__\s*=\s*({.+?});'],
165 webpage, 'player v5', default=None)
167 player = self._parse_json(player_v5, video_id)
168 metadata = player['metadata']
170 self._check_error(metadata)
173 for quality, media_list in metadata['qualities'].items():
174 for media in media_list:
175 media_url = media.get('url')
178 type_ = media.get('type')
179 if type_ == 'application/vnd.lumberjack.manifest':
181 ext = mimetype2ext(type_) or determine_ext(media_url)
183 m3u8_formats = self._extract_m3u8_formats(
184 media_url, video_id, 'mp4', preference=-1,
185 m3u8_id='hls', fatal=False)
186 for f in m3u8_formats:
187 f['url'] = f['url'].split('#')[0]
190 formats.extend(self._extract_f4m_formats(
191 media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
195 'format_id': 'http-%s' % quality,
198 m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
201 'width': int(m.group('width')),
202 'height': int(m.group('height')),
205 self._sort_formats(formats)
207 title = metadata['title']
208 duration = int_or_none(metadata.get('duration'))
209 timestamp = int_or_none(metadata.get('created_time'))
210 thumbnail = metadata.get('poster_url')
211 uploader = metadata.get('owner', {}).get('screenname')
212 uploader_id = metadata.get('owner', {}).get('id')
215 subtitles_data = metadata.get('subtitles', {}).get('data', {})
216 if subtitles_data and isinstance(subtitles_data, dict):
217 for subtitle_lang, subtitle in subtitles_data.items():
218 subtitles[subtitle_lang] = [{
219 'ext': determine_ext(subtitle_url),
221 } for subtitle_url in subtitle.get('urls', [])]
226 'description': description,
227 'thumbnail': thumbnail,
228 'duration': duration,
229 'timestamp': timestamp,
230 'uploader': uploader,
231 'uploader_id': uploader_id,
232 'age_limit': age_limit,
233 'view_count': view_count,
234 'comment_count': comment_count,
236 'subtitles': subtitles,
240 vevo_id = self._search_regex(
241 r'<link rel="video_src" href="[^"]*?vevo\.com[^"]*?video=(?P<id>[\w]*)',
242 webpage, 'vevo embed', default=None)
244 return self.url_result('vevo:%s' % vevo_id, 'Vevo')
246 # fallback old player
247 embed_page = self._download_webpage_no_ff(
248 'https://www.dailymotion.com/embed/video/%s' % video_id,
249 video_id, 'Downloading embed page')
251 timestamp = parse_iso8601(self._html_search_meta(
252 'video:release_date', webpage, 'upload date'))
254 info = self._parse_json(
256 r'var info = ({.*?}),$', embed_page,
257 'video info', flags=re.MULTILINE),
260 self._check_error(info)
263 for (key, format_id) in self._FORMATS:
264 video_url = info.get(key)
265 if video_url is not None:
266 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
267 if m_size is not None:
268 width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
270 width, height = None, None
274 'format_id': format_id,
278 self._sort_formats(formats)
281 video_subtitles = self.extract_subtitles(video_id, webpage)
283 title = self._og_search_title(webpage, default=None)
285 title = self._html_search_regex(
286 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
292 'uploader': info['owner.screenname'],
293 'timestamp': timestamp,
295 'description': description,
296 'subtitles': video_subtitles,
297 'thumbnail': info['thumbnail_url'],
298 'age_limit': age_limit,
299 'view_count': view_count,
300 'duration': info['duration']
303 def _check_error(self, info):
304 error = info.get('error')
305 if info.get('error') is not None:
306 title = error['title']
307 # See https://developer.dailymotion.com/api#access-error
308 if error.get('code') == 'DM007':
309 self.raise_geo_restricted(msg=title)
310 raise ExtractorError(
311 '%s said: %s' % (self.IE_NAME, title), expected=True)
313 def _get_subtitles(self, video_id, webpage):
315 sub_list = self._download_webpage(
316 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
317 video_id, note=False)
318 except ExtractorError as err:
319 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
321 info = json.loads(sub_list)
322 if (info['total'] > 0):
323 sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
325 self._downloader.report_warning('video doesn\'t have subtitles')
329 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
330 IE_NAME = 'dailymotion:playlist'
331 _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>[^/?#&]+)'
332 _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
333 _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
335 'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
338 'id': 'xv4bw_nqtv_sport',
340 'playlist_mincount': 20,
343 def _extract_entries(self, id):
345 processed_urls = set()
346 for pagenum in itertools.count(1):
347 page_url = self._PAGE_TEMPLATE % (id, pagenum)
348 webpage, urlh = self._download_webpage_handle_no_ff(
349 page_url, id, 'Downloading page %s' % pagenum)
350 if urlh.geturl() in processed_urls:
351 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
352 page_url, urlh.geturl()), id)
355 processed_urls.add(urlh.geturl())
357 for video_id in re.findall(r'data-xid="(.+?)"', webpage):
358 if video_id not in video_ids:
359 yield self.url_result(
360 'http://www.dailymotion.com/video/%s' % video_id,
361 DailymotionIE.ie_key(), video_id)
362 video_ids.add(video_id)
364 if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
367 def _real_extract(self, url):
368 mobj = re.match(self._VALID_URL, url)
369 playlist_id = mobj.group('id')
370 webpage = self._download_webpage(url, playlist_id)
375 'title': self._og_search_title(webpage),
376 'entries': self._extract_entries(playlist_id),
380 class DailymotionUserIE(DailymotionPlaylistIE):
381 IE_NAME = 'dailymotion:user'
382 _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
383 _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
385 'url': 'https://www.dailymotion.com/user/nqtv',
388 'title': 'Rémi Gaillard',
390 'playlist_mincount': 100,
392 'url': 'http://www.dailymotion.com/user/UnderProject',
394 'id': 'UnderProject',
395 'title': 'UnderProject',
397 'playlist_mincount': 1800,
398 'expected_warnings': [
399 'Stopped at duplicated page',
401 'skip': 'Takes too long time',
404 def _real_extract(self, url):
405 mobj = re.match(self._VALID_URL, url)
406 user = mobj.group('user')
407 webpage = self._download_webpage(
408 'https://www.dailymotion.com/user/%s' % user, user)
409 full_user = unescapeHTML(self._html_search_regex(
410 r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
417 'entries': self._extract_entries(user),