X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/b4a0c9f9de9d715538a1718922d6ab01a40f7ce3..16a91194b542b099c141633fd95e77db4ff075ed:/youtube_dl/extractor/itv.py diff --git a/youtube_dl/extractor/itv.py b/youtube_dl/extractor/itv.py index 18a7d7f..de65b6b 100644 --- a/youtube_dl/extractor/itv.py +++ b/youtube_dl/extractor/itv.py @@ -7,19 +7,23 @@ import json import re from .common import InfoExtractor +from .brightcove import BrightcoveNewIE from ..compat import ( compat_str, compat_etree_register_namespace, ) from ..utils import ( + determine_ext, + ExtractorError, extract_attributes, + int_or_none, + merge_dicts, + parse_duration, + smuggle_url, + url_or_none, xpath_with_ns, xpath_element, xpath_text, - int_or_none, - parse_duration, - ExtractorError, - determine_ext, ) @@ -41,6 +45,14 @@ class ITVIE(InfoExtractor): # unavailable via data-playlist-url 'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033', 'only_matching': True, + }, { + # InvalidVodcrid + 'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034', + 'only_matching': True, + }, { + # ContentUnavailable + 'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024', + 'only_matching': True, }] def _real_extract(self, url): @@ -119,63 +131,65 @@ class ITVIE(InfoExtractor): resp_env = self._download_xml( params['data-playlist-url'], video_id, - headers=headers, data=etree.tostring(req_env)) - playlist = xpath_element(resp_env, './/Playlist') - if playlist is None: - fault_code = xpath_text(resp_env, './/faultcode') - fault_string = xpath_text(resp_env, './/faultstring') - if fault_code == 'InvalidGeoRegion': - self.raise_geo_restricted( - msg=fault_string, countries=self._GEO_COUNTRIES) - elif fault_code != 'InvalidEntity': - raise ExtractorError( - '%s said: %s' % (self.IE_NAME, fault_string), expected=True) - info.update({ - 'title': self._og_search_title(webpage), - 'episode_title': params.get('data-video-episode'), - 'series': params.get('data-video-title'), - }) - else: - title = xpath_text(playlist, 'EpisodeTitle', default=None) - info.update({ - 'title': title, - 'episode_title': title, - 'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')), - 'series': xpath_text(playlist, 'ProgrammeTitle'), - 'duration': parse_duration(xpath_text(playlist, 'Duration')), - }) - video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True) - media_files = xpath_element(video_element, 'MediaFiles', fatal=True) - rtmp_url = media_files.attrib['base'] - - for media_file in media_files.findall('MediaFile'): - play_path = xpath_text(media_file, 'URL') - if not play_path: - continue - tbr = int_or_none(media_file.get('bitrate'), 1000) - f = { - 'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''), - 'play_path': play_path, - # Providing this swfVfy allows to avoid truncated downloads - 'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf', - 'page_url': url, - 'tbr': tbr, - 'ext': 'flv', - } - app = self._search_regex( - 'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None) - if app: - f.update({ - 'url': rtmp_url.split('?', 1)[0], - 'app': app, - }) - else: - f['url'] = rtmp_url - formats.append(f) - - for caption_url in video_element.findall('ClosedCaptioningURIs/URL'): - if caption_url.text: - extract_subtitle(caption_url.text) + headers=headers, data=etree.tostring(req_env), fatal=False) + if resp_env: + playlist = xpath_element(resp_env, './/Playlist') + if playlist is None: + fault_code = xpath_text(resp_env, './/faultcode') + fault_string = xpath_text(resp_env, './/faultstring') + if fault_code == 'InvalidGeoRegion': + self.raise_geo_restricted( + msg=fault_string, countries=self._GEO_COUNTRIES) + elif fault_code not in ( + 'InvalidEntity', 'InvalidVodcrid', 'ContentUnavailable'): + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, fault_string), expected=True) + info.update({ + 'title': self._og_search_title(webpage), + 'episode_title': params.get('data-video-episode'), + 'series': params.get('data-video-title'), + }) + else: + title = xpath_text(playlist, 'EpisodeTitle', default=None) + info.update({ + 'title': title, + 'episode_title': title, + 'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')), + 'series': xpath_text(playlist, 'ProgrammeTitle'), + 'duration': parse_duration(xpath_text(playlist, 'Duration')), + }) + video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True) + media_files = xpath_element(video_element, 'MediaFiles', fatal=True) + rtmp_url = media_files.attrib['base'] + + for media_file in media_files.findall('MediaFile'): + play_path = xpath_text(media_file, 'URL') + if not play_path: + continue + tbr = int_or_none(media_file.get('bitrate'), 1000) + f = { + 'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''), + 'play_path': play_path, + # Providing this swfVfy allows to avoid truncated downloads + 'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf', + 'page_url': url, + 'tbr': tbr, + 'ext': 'flv', + } + app = self._search_regex( + 'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None) + if app: + f.update({ + 'url': rtmp_url.split('?', 1)[0], + 'app': app, + }) + else: + f['url'] = rtmp_url + formats.append(f) + + for caption_url in video_element.findall('ClosedCaptioningURIs/URL'): + if caption_url.text: + extract_subtitle(caption_url.text) ios_playlist_url = params.get('data-video-playlist') or params.get('data-video-id') hmac = params.get('data-video-hmac') @@ -237,8 +251,8 @@ class ITVIE(InfoExtractor): for sub in subs: if not isinstance(sub, dict): continue - href = sub.get('Href') - if isinstance(href, compat_str): + href = url_or_none(sub.get('Href')) + if href: extract_subtitle(href) if not info.get('duration'): info['duration'] = parse_duration(video_data.get('Duration')) @@ -250,4 +264,49 @@ class ITVIE(InfoExtractor): 'formats': formats, 'subtitles': subtitles, }) - return info + + webpage_info = self._search_json_ld(webpage, video_id, default={}) + if not webpage_info.get('title'): + webpage_info['title'] = self._html_search_regex( + r'(?s)]+\bclass=["\'][^>]*episode-title["\'][^>]*>([^<]+)<', + webpage, 'title', default=None) or self._og_search_title( + webpage, default=None) or self._html_search_meta( + 'twitter:title', webpage, 'title', + default=None) or webpage_info['episode'] + + return merge_dicts(info, webpage_info) + + +class ITVBTCCIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?itv\.com/btcc/(?:[^/]+/)*(?P[^/?#&]+)' + _TEST = { + 'url': 'http://www.itv.com/btcc/races/btcc-2018-all-the-action-from-brands-hatch', + 'info_dict': { + 'id': 'btcc-2018-all-the-action-from-brands-hatch', + 'title': 'BTCC 2018: All the action from Brands Hatch', + }, + 'playlist_mincount': 9, + } + BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1582188683001/HkiHLnNRx_default/index.html?videoId=%s' + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + entries = [ + self.url_result( + smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % video_id, { + # ITV does not like some GB IP ranges, so here are some + # IP blocks it accepts + 'geo_ip_blocks': [ + '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21' + ], + 'referrer': url, + }), + ie=BrightcoveNewIE.ie_key(), video_id=video_id) + for video_id in re.findall(r'data-video-id=["\'](\d+)', webpage)] + + title = self._og_search_title(webpage, fatal=False) + + return self.playlist_result(entries, playlist_id, title)