- 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 = compat_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)
+ def _real_extract(self, url):
+ display_id = self._match_id(url)
+
+ response = self._graphql_call('''{
+ %s(slug: "video/%s") {
+ ... on RecordSlug {
+ record {
+ id
+ title
+ teaser
+ publishOn
+ thumb {
+ preview
+ }
+ tags {
+ name
+ }
+ duration
+ }
+ }
+ ... on NotFoundSlug {
+ status
+ }
+ }
+}''', 'Slug', display_id)
+ if response.get('status'):
+ raise ExtractorError('This video is no longer available.', expected=True)
+
+ record = response['record']
+ video_id = record['id']
+
+ srcs = self._graphql_call('''{
+ %s(id: "%s") {
+ src
+ }
+}''', 'RecordVideoSource', video_id)['src']