+ embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
+ embed_page = self._download_webpage(embed_url, video_id,
+ 'Downloading embed page')
+ info = self._search_regex(r'var info = ({.*?}),$', embed_page,
+ 'video info', flags=re.MULTILINE)
+ info = json.loads(info)
+ if info.get('error') is not None:
+ msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
+ raise ExtractorError(msg, expected=True)
+
+ formats = []
+ for (key, format_id) in self._FORMATS:
+ video_url = info.get(key)
+ if video_url is not None:
+ m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
+ if m_size is not None:
+ width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
+ else:
+ width, height = None, None
+ formats.append({
+ 'url': video_url,
+ 'ext': 'mp4',
+ 'format_id': format_id,
+ 'width': width,
+ 'height': height,
+ })
+ if not formats:
+ raise ExtractorError('Unable to extract video URL')
+
+ # subtitles
+ video_subtitles = self.extract_subtitles(video_id, webpage)
+
+ view_count = str_to_int(self._search_regex(
+ r'video_views_count[^>]+>\s+([\d\.,]+)',
+ webpage, 'view count', fatal=False))
+
+ title = self._og_search_title(webpage, default=None)
+ if title is None:
+ title = self._html_search_regex(
+ r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
+ 'title')
+
+ return {
+ 'id': video_id,
+ 'formats': formats,
+ 'uploader': info['owner.screenname'],
+ 'upload_date': video_upload_date,
+ 'title': title,
+ 'subtitles': video_subtitles,
+ 'thumbnail': info['thumbnail_url'],
+ 'age_limit': age_limit,
+ 'view_count': view_count,
+ }
+
+ def _get_subtitles(self, video_id, webpage):
+ try:
+ sub_list = self._download_webpage(
+ 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
+ video_id, note=False)
+ except ExtractorError as err:
+ self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
+ return {}
+ info = json.loads(sub_list)
+ if (info['total'] > 0):
+ sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
+ return sub_lang_list
+ self._downloader.report_warning('video doesn\'t have subtitles')
+ return {}
+
+
+class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
+ IE_NAME = 'dailymotion:playlist'
+ _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
+ _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
+ _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
+ _TESTS = [{
+ 'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
+ 'info_dict': {
+ 'title': 'SPORT',
+ 'id': 'xv4bw_nqtv_sport',
+ },
+ 'playlist_mincount': 20,
+ }]
+
+ def _extract_entries(self, id):
+ video_ids = []
+ for pagenum in itertools.count(1):
+ request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
+ webpage = self._download_webpage(request,
+ id, 'Downloading page %s' % pagenum)
+
+ video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
+
+ if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
+ break
+ return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
+ for video_id in orderedSet(video_ids)]
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ playlist_id = mobj.group('id')
+ webpage = self._download_webpage(url, playlist_id)
+
+ return {
+ '_type': 'playlist',
+ 'id': playlist_id,
+ 'title': self._og_search_title(webpage),
+ 'entries': self._extract_entries(playlist_id),
+ }
+
+
+class DailymotionUserIE(DailymotionPlaylistIE):
+ IE_NAME = 'dailymotion:user'
+ _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
+ _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
+ _TESTS = [{
+ 'url': 'https://www.dailymotion.com/user/nqtv',
+ 'info_dict': {
+ 'id': 'nqtv',
+ 'title': 'Rémi Gaillard',
+ },
+ 'playlist_mincount': 100,
+ }]
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ user = mobj.group('user')
+ webpage = self._download_webpage(url, user)
+ full_user = unescapeHTML(self._html_search_regex(
+ r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
+ webpage, 'user'))
+
+ return {
+ '_type': 'playlist',
+ 'id': user,
+ 'title': full_user,
+ 'entries': self._extract_entries(user),
+ }
Rapsys Git / youtubedl / blobdiff