X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Fcommon.py;h=daa10885fb27a583773e2f9b8421448309346ac7;hb=0d2f0b0357325823782884327a158aeccf4f9b49;hp=6d01f0800e5c6028a86915cc3e93878ab1257b5a;hpb=ff99fe529e52b2465f1d973e69df01a6391568d6;p=youtube-dl.git diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 6d01f0800..daa10885f 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -245,6 +245,10 @@ class InfoExtractor(object): specified in the URL. end_time: Time in seconds where the reproduction should end, as specified in the URL. + chapters: A list of dictionaries, with the following entries: + * "start_time" - The start time of the chapter in seconds + * "end_time" - The end time of the chapter in seconds + * "title" (optional, string) The following fields should only be used when the video belongs to some logical chapter or section: @@ -372,7 +376,7 @@ class InfoExtractor(object): cls._VALID_URL_RE = re.compile(cls._VALID_URL) m = cls._VALID_URL_RE.match(url) assert m - return m.group('id') + return compat_str(m.group('id')) @classmethod def working(cls): @@ -416,7 +420,7 @@ class InfoExtractor(object): if country_code: self._x_forwarded_for_ip = GeoUtils.random_ipv4(country_code) if self._downloader.params.get('verbose', False): - self._downloader.to_stdout( + self._downloader.to_screen( '[debug] Using fake IP %s (%s) as X-Forwarded-For.' % (self._x_forwarded_for_ip, country_code.upper())) @@ -990,6 +994,7 @@ class InfoExtractor(object): 'tbr': int_or_none(e.get('bitrate')), 'width': int_or_none(e.get('width')), 'height': int_or_none(e.get('height')), + 'view_count': int_or_none(e.get('interactionCount')), }) for e in json_ld: @@ -997,17 +1002,17 @@ class InfoExtractor(object): item_type = e.get('@type') if expected_type is not None and expected_type != item_type: return info - if item_type == 'TVEpisode': + if item_type in ('TVEpisode', 'Episode'): info.update({ 'episode': unescapeHTML(e.get('name')), 'episode_number': int_or_none(e.get('episodeNumber')), 'description': unescapeHTML(e.get('description')), }) part_of_season = e.get('partOfSeason') - if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason': + if isinstance(part_of_season, dict) and part_of_season.get('@type') in ('TVSeason', 'Season', 'CreativeWorkSeason'): info['season_number'] = int_or_none(part_of_season.get('seasonNumber')) part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries') - if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries': + if isinstance(part_of_series, dict) and part_of_series.get('@type') in ('TVSeries', 'Series', 'CreativeWorkSeries'): info['series'] = unescapeHTML(part_of_series.get('name')) elif item_type == 'Article': info.update({ @@ -1017,10 +1022,10 @@ class InfoExtractor(object): }) elif item_type == 'VideoObject': extract_video_object(e) - elif item_type == 'WebPage': - video = e.get('video') - if isinstance(video, dict) and video.get('@type') == 'VideoObject': - extract_video_object(video) + continue + video = e.get('video') + if isinstance(video, dict) and video.get('@type') == 'VideoObject': + extract_video_object(video) break return dict((k, v) for k, v in info.items() if v is not None) @@ -1386,6 +1391,7 @@ class InfoExtractor(object): f = { 'format_id': '-'.join(format_id), 'url': format_url(media_url), + 'manifest_url': m3u8_url, 'language': media.get('LANGUAGE'), 'ext': ext, 'protocol': entry_protocol, @@ -1995,6 +2001,12 @@ class InfoExtractor(object): compat_etree_fromstring(ism.encode('utf-8')), urlh.geturl(), ism_id) def _parse_ism_formats(self, ism_doc, ism_url, ism_id=None): + """ + Parse formats from ISM manifest. + References: + 1. [MS-SSTR]: Smooth Streaming Protocol, + https://msdn.microsoft.com/en-us/library/ff469518.aspx + """ if ism_doc.get('IsLive') == 'TRUE' or ism_doc.find('Protection') is not None: return [] @@ -2016,8 +2028,11 @@ class InfoExtractor(object): self.report_warning('%s is not a supported codec' % fourcc) continue tbr = int(track.attrib['Bitrate']) // 1000 - width = int_or_none(track.get('MaxWidth')) - height = int_or_none(track.get('MaxHeight')) + # [1] does not mention Width and Height attributes. However, + # they're often present while MaxWidth and MaxHeight are + # missing, so should be used as fallbacks + width = int_or_none(track.get('MaxWidth') or track.get('Width')) + height = int_or_none(track.get('MaxHeight') or track.get('Height')) sampling_rate = int_or_none(track.get('SamplingRate')) track_url_pattern = re.sub(r'{[Bb]itrate}', track.attrib['Bitrate'], url_pattern) @@ -2117,15 +2132,18 @@ class InfoExtractor(object): return is_plain_url, formats entries = [] + # amp-video and amp-audio are very similar to their HTML5 counterparts + # so we wll include them right here (see + # https://www.ampproject.org/docs/reference/components/amp-video) media_tags = [(media_tag, media_type, '') for media_tag, media_type - in re.findall(r'(?s)(<(video|audio)[^>]*/>)', webpage)] + in re.findall(r'(?s)(<(?:amp-)?(video|audio)[^>]*/>)', webpage)] media_tags.extend(re.findall( # We only allow video|audio followed by a whitespace or '>'. # Allowing more characters may end up in significant slow down (see # https://github.com/rg3/youtube-dl/issues/11979, example URL: # http://www.porntrex.com/maps/videositemap.xml). - r'(?s)(<(?Pvideo|audio)(?:\s+[^>]*)?>)(.*?)', webpage)) + r'(?s)(<(?P(?:amp-)?(?:video|audio))(?:\s+[^>]*)?>)(.*?)', webpage)) for media_tag, media_type, media_content in media_tags: media_info = { 'formats': [], @@ -2168,7 +2186,7 @@ class InfoExtractor(object): def _extract_akamai_formats(self, manifest_url, video_id, hosts={}): formats = [] hdcore_sign = 'hdcore=3.7.0' - f4m_url = re.sub(r'(https?://[^/+])/i/', r'\1/z/', manifest_url).replace('/master.m3u8', '/manifest.f4m') + f4m_url = re.sub(r'(https?://[^/]+)/i/', r'\1/z/', manifest_url).replace('/master.m3u8', '/manifest.f4m') hds_host = hosts.get('hds') if hds_host: f4m_url = re.sub(r'(https?://)[^/]+', r'\1' + hds_host, f4m_url) @@ -2190,8 +2208,9 @@ class InfoExtractor(object): def _extract_wowza_formats(self, url, video_id, m3u8_entry_protocol='m3u8_native', skip_protocols=[]): url = re.sub(r'/(?:manifest|playlist|jwplayer)\.(?:m3u8|f4m|mpd|smil)', '', url) - url_base = self._search_regex(r'(?:https?|rtmp|rtsp)(://[^?]+)', url, 'format url') - http_base_url = 'http' + url_base + url_base = self._search_regex( + r'(?:(?:https?|rtmp|rtsp):)?(//[^?]+)', url, 'format url') + http_base_url = '%s:%s' % ('http', url_base) formats = [] if 'm3u8' not in skip_protocols: formats.extend(self._extract_m3u8_formats( @@ -2225,7 +2244,7 @@ class InfoExtractor(object): for protocol in ('rtmp', 'rtsp'): if protocol not in skip_protocols: formats.append({ - 'url': protocol + url_base, + 'url': '%s:%s' % (protocol, url_base), 'format_id': protocol, 'protocol': protocol, }) @@ -2283,6 +2302,8 @@ class InfoExtractor(object): tracks = video_data.get('tracks') if tracks and isinstance(tracks, list): for track in tracks: + if not isinstance(track, dict): + continue if track.get('kind') != 'captions': continue track_url = urljoin(base_url, track.get('file')) @@ -2312,6 +2333,8 @@ class InfoExtractor(object): urls = [] formats = [] for source in jwplayer_sources_data: + if not isinstance(source, dict): + continue source_url = self._proto_relative_url(source.get('file')) if not source_url: continue