diff options
Diffstat (limited to 'youtube_dl/extractor/common.py')
| -rw-r--r-- | youtube_dl/extractor/common.py | 45 |
1 files changed, 41 insertions, 4 deletions
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 0719c7bcd..2f574054d 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -34,6 +34,7 @@ from ..utils import ( fix_xml_ampersands, float_or_none, int_or_none, + parse_iso8601, RegexNotFoundError, sanitize_filename, sanitized_Request, @@ -313,9 +314,9 @@ class InfoExtractor(object): except ExtractorError: raise except compat_http_client.IncompleteRead as e: - raise ExtractorError('A network error has occured.', cause=e, expected=True) + raise ExtractorError('A network error has occurred.', cause=e, expected=True) except (KeyError, StopIteration) as e: - raise ExtractorError('An extractor error has occured.', cause=e) + raise ExtractorError('An extractor error has occurred.', cause=e) def set_downloader(self, downloader): """Sets the downloader for this IE.""" @@ -762,6 +763,42 @@ class InfoExtractor(object): return self._html_search_meta('twitter:player', html, 'twitter card player') + def _search_json_ld(self, html, video_id, **kwargs): + json_ld = self._search_regex( + r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>', + html, 'JSON-LD', group='json_ld', **kwargs) + if not json_ld: + return {} + return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True)) + + def _json_ld(self, json_ld, video_id, fatal=True): + if isinstance(json_ld, compat_str): + json_ld = self._parse_json(json_ld, video_id, fatal=fatal) + if not json_ld: + return {} + info = {} + if json_ld.get('@context') == 'http://schema.org': + item_type = json_ld.get('@type') + if item_type == 'TVEpisode': + info.update({ + 'episode': unescapeHTML(json_ld.get('name')), + 'episode_number': int_or_none(json_ld.get('episodeNumber')), + 'description': unescapeHTML(json_ld.get('description')), + }) + part_of_season = json_ld.get('partOfSeason') + if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason': + info['season_number'] = int_or_none(part_of_season.get('seasonNumber')) + part_of_series = json_ld.get('partOfSeries') + if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries': + info['series'] = unescapeHTML(part_of_series.get('name')) + elif item_type == 'Article': + info.update({ + 'timestamp': parse_iso8601(json_ld.get('datePublished')), + 'title': unescapeHTML(json_ld.get('headline')), + 'description': unescapeHTML(json_ld.get('articleBody')), + }) + return dict((k, v) for k, v in info.items() if v is not None) + @staticmethod def _hidden_inputs(html): html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html) @@ -1021,9 +1058,9 @@ class InfoExtractor(object): # TODO: looks like video codec is not always necessarily goes first va_codecs = codecs.split(',') if va_codecs[0]: - f['vcodec'] = va_codecs[0].partition('.')[0] + f['vcodec'] = va_codecs[0] if len(va_codecs) > 1 and va_codecs[1]: - f['acodec'] = va_codecs[1].partition('.')[0] + f['acodec'] = va_codecs[1] resolution = last_info.get('RESOLUTION') if resolution: width_str, height_str = resolution.split('x') |