- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % url)
- url_title = mobj.group('url_title')
- webpage = self._download_webpage(url, url_title)
-
- video_id = self._html_search_regex(r'<article class="video" data-id="(\d+?)"',
- webpage, u'video id')
-
- self.report_extraction(video_id)
-
- data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
- data = self._download_webpage(data_url, video_id, 'Downloading data webpage')
-
-
- qualities = [ '1080p', '720p', '1000k', '480p', '500k' ]
- best_quality_idx = len(qualities)+1 # First regex match may not be optimal
- for idx, quality in enumerate(qualities):
- regex = r'<file [^>]*type="(?:high|standard)".*?>(.*%s.*)</file>' % quality
- try:
- url = self._html_search_regex(regex, data, u'video URL')
- if idx < best_quality_idx:
- video_url = url
- best_quality_idx = idx
- except RegexNotFoundError:
- # Just catch fatal exc. Don't want the fatal=False warning
+
+ display_id = mobj.group('display_id')
+ webpage, urlh = self._download_webpage_handle(url, display_id)
+ if 'src=expired' in urlh.geturl():
+ raise ExtractorError('This video is expired.', expected=True)
+
+ video_id = mobj.group('video_id')
+ if not video_id:
+ video_id = self._html_search_regex(
+ self._VIDEO_ID_REGEXES, webpage, 'video id')
+
+ data = None
+
+ preload_codes = self._html_search_regex(
+ r'(function.+)setTimeout\(function\(\)\{playlist',
+ webpage, 'preload codes')
+ base64_fragments = re.findall(r'"([a-zA-Z0-9+/=]+)"', preload_codes)
+ base64_fragments.remove('init')
+
+ def _check_sequence(cur_fragments):
+ if not cur_fragments:
+ return
+ for i in range(len(cur_fragments)):
+ cur_sequence = (''.join(cur_fragments[i:] + cur_fragments[:i])).encode('ascii')
+ try:
+ raw_data = base64.b64decode(cur_sequence)
+ if compat_ord(raw_data[0]) == compat_ord('{'):
+ return json.loads(raw_data.decode('utf-8'))
+ except (TypeError, binascii.Error, UnicodeDecodeError, ValueError):
+ continue
+
+ def _check_data():
+ for i in range(len(base64_fragments) + 1):
+ for j in range(i, len(base64_fragments) + 1):
+ data = _check_sequence(base64_fragments[:i] + base64_fragments[j:])
+ if data:
+ return data
+
+ self.to_screen('Try to compute possible data sequence. This may take some time.')
+ data = _check_data()
+
+ if not data:
+ raise ExtractorError(
+ 'Preload information could not be extracted', expected=True)
+
+ formats = []
+ get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
+ for filed in data['files']:
+ if determine_ext(filed['url']) == 'm3u8':
+ # compat_urllib_parse.urljoin does not work here
+ if filed['url'].startswith('/'):
+ m3u8_url = 'http://ht.cdn.turner.com/tbs/big/teamcoco' + filed['url']
+ else:
+ m3u8_url = filed['url']
+ m3u8_formats = self._extract_m3u8_formats(
+ m3u8_url, video_id, ext='mp4')
+ for m3u8_format in m3u8_formats:
+ if m3u8_format not in formats:
+ formats.append(m3u8_format)
+ elif determine_ext(filed['url']) == 'f4m':
+ # TODO Correct f4m extraction