X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/47d80ec0b18245caeb97018d4c1af18d0b5b972b..233624c1db781ee7dabbaf88453cf18e248dd20d:/youtube_dl/extractor/nrk.py diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py index c89aac6..a231735 100644 --- a/youtube_dl/extractor/nrk.py +++ b/youtube_dl/extractor/nrk.py @@ -1,53 +1,56 @@ # coding: utf-8 from __future__ import unicode_literals -import random import re from .common import InfoExtractor -from ..compat import compat_urllib_parse_unquote +from ..compat import ( + compat_str, + compat_urllib_parse_unquote, +) from ..utils import ( ExtractorError, int_or_none, + JSON_LD_RE, + NO_DEFAULT, parse_age_limit, parse_duration, + try_get, ) class NRKBaseIE(InfoExtractor): - _faked_ip = None - - def _download_webpage_handle(self, *args, **kwargs): - # NRK checks X-Forwarded-For HTTP header in order to figure out the - # origin of the client behind proxy. This allows to bypass geo - # restriction by faking this header's value to some Norway IP. - # We will do so once we encounter any geo restriction error. - if self._faked_ip: - # NB: str is intentional - kwargs.setdefault(str('headers'), {})['X-Forwarded-For'] = self._faked_ip - return super(NRKBaseIE, self)._download_webpage_handle(*args, **kwargs) - - def _fake_ip(self): - # Use fake IP from 37.191.128.0/17 in order to workaround geo - # restriction - def octet(lb=0, ub=255): - return random.randint(lb, ub) - self._faked_ip = '37.191.%d.%d' % (octet(128), octet()) + _GEO_COUNTRIES = ['NO'] + + _api_host = None def _real_extract(self, url): video_id = self._match_id(url) - data = self._download_json( - 'http://%s/mediaelement/%s' % (self._API_HOST, video_id), - video_id, 'Downloading mediaelement JSON') + api_hosts = (self._api_host, ) if self._api_host else self._API_HOSTS + + for api_host in api_hosts: + data = self._download_json( + 'http://%s/mediaelement/%s' % (api_host, video_id), + video_id, 'Downloading mediaelement JSON', + fatal=api_host == api_hosts[-1]) + if not data: + continue + self._api_host = api_host + break title = data.get('fullTitle') or data.get('mainTitle') or data['title'] video_id = data.get('id') or video_id - http_headers = {'X-Forwarded-For': self._faked_ip} if self._faked_ip else {} - entries = [] + conviva = data.get('convivaStatistics') or {} + live = (data.get('mediaElementType') == 'Live' or + data.get('isLive') is True or conviva.get('isLive')) + + def make_title(t): + return self._live_title(t) if live else t + media_assets = data.get('mediaAssets') if media_assets and isinstance(media_assets, list): def video_id_and_title(idx): @@ -61,6 +64,13 @@ class NRKBaseIE(InfoExtractor): if not formats: continue self._sort_formats(formats) + + # Some f4m streams may not work with hdcore in fragments' URLs + for f in formats: + extra_param = f.get('extra_param_to_segment_url') + if extra_param and 'hdcore' in extra_param: + del f['extra_param_to_segment_url'] + entry_id, entry_title = video_id_and_title(num) duration = parse_duration(asset.get('duration')) subtitles = {} @@ -72,11 +82,10 @@ class NRKBaseIE(InfoExtractor): }) entries.append({ 'id': asset.get('carrierId') or entry_id, - 'title': entry_title, + 'title': make_title(entry_title), 'duration': duration, 'subtitles': subtitles, 'formats': formats, - 'http_headers': http_headers, }) if not entries: @@ -87,34 +96,47 @@ class NRKBaseIE(InfoExtractor): duration = parse_duration(data.get('duration')) entries = [{ 'id': video_id, - 'title': title, + 'title': make_title(title), 'duration': duration, 'formats': formats, }] if not entries: - message_type = data.get('messageType', '') - # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked* - if 'IsGeoBlocked' in message_type and not self._faked_ip: - self.report_warning( - 'Video is geo restricted, trying to fake IP') - self._fake_ip() - return self._real_extract(url) - MESSAGES = { 'ProgramRightsAreNotReady': 'Du kan dessverre ikke se eller høre programmet', 'ProgramRightsHasExpired': 'Programmet har gått ut', 'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge', } + message_type = data.get('messageType', '') + # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked* + if 'IsGeoBlocked' in message_type: + self.raise_geo_restricted( + msg=MESSAGES.get('ProgramIsGeoBlocked'), + countries=self._GEO_COUNTRIES) raise ExtractorError( '%s said: %s' % (self.IE_NAME, MESSAGES.get( message_type, message_type)), expected=True) - conviva = data.get('convivaStatistics') or {} series = conviva.get('seriesName') or data.get('seriesTitle') episode = conviva.get('episodeName') or data.get('episodeNumberOrDate') + season_number = None + episode_number = None + if data.get('mediaElementType') == 'Episode': + _season_episode = data.get('scoresStatistics', {}).get('springStreamStream') or \ + data.get('relativeOriginUrl', '') + EPISODENUM_RE = [ + r'/s(?P\d{,2})e(?P\d{,2})\.', + r'/sesong-(?P\d{,2})/episode-(?P\d{,2})', + ] + season_number = int_or_none(self._search_regex( + EPISODENUM_RE, _season_episode, 'season number', + default=None, group='season')) + episode_number = int_or_none(self._search_regex( + EPISODENUM_RE, _season_episode, 'episode number', + default=None, group='episode')) + thumbnails = None images = data.get('images') if images and isinstance(images, dict): @@ -127,24 +149,49 @@ class NRKBaseIE(InfoExtractor): } for image in web_images if image.get('imageUrl')] description = data.get('description') + category = data.get('mediaAnalytics', {}).get('category') common_info = { 'description': description, 'series': series, 'episode': episode, + 'season_number': season_number, + 'episode_number': episode_number, + 'categories': [category] if category else None, 'age_limit': parse_age_limit(data.get('legalAge')), 'thumbnails': thumbnails, } vcodec = 'none' if data.get('mediaType') == 'Audio' else None - # TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged - for entry in entries: entry.update(common_info) for f in entry['formats']: f['vcodec'] = vcodec + points = data.get('shortIndexPoints') + if isinstance(points, list): + chapters = [] + for next_num, point in enumerate(points, start=1): + if not isinstance(point, dict): + continue + start_time = parse_duration(point.get('startPoint')) + if start_time is None: + continue + end_time = parse_duration( + data.get('duration') + if next_num == len(points) + else points[next_num].get('startPoint')) + if end_time is None: + continue + chapters.append({ + 'start_time': start_time, + 'end_time': end_time, + 'title': point.get('title'), + }) + if chapters and len(entries) == 1: + entries[0]['chapters'] = chapters + return self.playlist_result(entries, video_id, title, description) @@ -155,12 +202,12 @@ class NRKIE(NRKBaseIE): https?:// (?: (?:www\.)?nrk\.no/video/PS\*| - v8-psapi\.nrk\.no/mediaelement/ + v8[-.]psapi\.nrk\.no/mediaelement/ ) ) - (?P[^/?#&]+) + (?P[^?#&]+) ''' - _API_HOST = 'v8.psapi.nrk.no' + _API_HOSTS = ('psapi.nrk.no', 'v8-psapi.nrk.no') _TESTS = [{ # video 'url': 'http://www.nrk.no/video/PS*150533', @@ -186,6 +233,9 @@ class NRKIE(NRKBaseIE): }, { 'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9', 'only_matching': True, + }, { + 'url': 'nrk:clip/7707d5a3-ebe7-434a-87d5-a3ebe7a34a70', + 'only_matching': True, }, { 'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9', 'only_matching': True, @@ -194,9 +244,16 @@ class NRKIE(NRKBaseIE): class NRKTVIE(NRKBaseIE): IE_DESC = 'NRK TV and NRK Radio' - _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:serie/[^/]+|program)/(?P[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P\d+))?' - _API_HOST = 'psapi-we.nrk.no' - + _EPISODE_RE = r'(?P[a-zA-Z]{4}\d{8})' + _VALID_URL = r'''(?x) + https?:// + (?:tv|radio)\.nrk(?:super)?\.no/ + (?:serie/[^/]+|program)/ + (?![Ee]pisodes)%s + (?:/\d{2}-\d{2}-\d{4})? + (?:\#del=(?P\d+))? + ''' % _EPISODE_RE + _API_HOSTS = ('psapi-ne.nrk.no', 'psapi-we.nrk.no') _TESTS = [{ 'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014', 'md5': '4e9ca6629f09e588ed240fb11619922a', @@ -206,63 +263,321 @@ class NRKTVIE(NRKBaseIE): 'title': '20 spørsmål 23.05.2014', 'description': 'md5:bdea103bc35494c143c6a9acdd84887a', 'duration': 1741, + 'series': '20 spørsmål - TV', + 'episode': '23.05.2014', }, }, { 'url': 'https://tv.nrk.no/program/mdfp15000514', - 'md5': '43d0be26663d380603a9cf0c24366531', 'info_dict': { 'id': 'MDFP15000514CA', 'ext': 'mp4', 'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014', 'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db', 'duration': 4605, + 'series': 'Kunnskapskanalen', + 'episode': '24.05.2014', + }, + 'params': { + 'skip_download': True, }, }, { # single playlist video 'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2', - 'md5': 'adbd1dbd813edaf532b0a253780719c2', 'info_dict': { 'id': 'MSPO40010515-part2', 'ext': 'flv', 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)', 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', }, - 'skip': 'Only works from Norway', + 'params': { + 'skip_download': True, + }, + 'expected_warnings': ['Video is geo restricted'], + 'skip': 'particular part is not supported currently', }, { 'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015', 'playlist': [{ - 'md5': '9480285eff92d64f06e02a5367970a7a', 'info_dict': { - 'id': 'MSPO40010515-part1', - 'ext': 'flv', - 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)', - 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', + 'id': 'MSPO40010515AH', + 'ext': 'mp4', + 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 1)', + 'description': 'md5:c03aba1e917561eface5214020551b7a', + 'duration': 772, + 'series': 'Tour de Ski', + 'episode': '06.01.2015', + }, + 'params': { + 'skip_download': True, }, }, { - 'md5': 'adbd1dbd813edaf532b0a253780719c2', 'info_dict': { - 'id': 'MSPO40010515-part2', - 'ext': 'flv', - 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)', - 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', + 'id': 'MSPO40010515BH', + 'ext': 'mp4', + 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 2)', + 'description': 'md5:c03aba1e917561eface5214020551b7a', + 'duration': 6175, + 'series': 'Tour de Ski', + 'episode': '06.01.2015', + }, + 'params': { + 'skip_download': True, }, }], 'info_dict': { 'id': 'MSPO40010515', - 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn', - 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', - 'duration': 6947.52, + 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015', + 'description': 'md5:c03aba1e917561eface5214020551b7a', + }, + 'expected_warnings': ['Video is geo restricted'], + }, { + 'url': 'https://tv.nrk.no/serie/anno/KMTE50001317/sesong-3/episode-13', + 'info_dict': { + 'id': 'KMTE50001317AA', + 'ext': 'mp4', + 'title': 'Anno 13:30', + 'description': 'md5:11d9613661a8dbe6f9bef54e3a4cbbfa', + 'duration': 2340, + 'series': 'Anno', + 'episode': '13:30', + 'season_number': 3, + 'episode_number': 13, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'https://tv.nrk.no/serie/nytt-paa-nytt/MUHH46000317/27-01-2017', + 'info_dict': { + 'id': 'MUHH46000317AA', + 'ext': 'mp4', + 'title': 'Nytt på Nytt 27.01.2017', + 'description': 'md5:5358d6388fba0ea6f0b6d11c48b9eb4b', + 'duration': 1796, + 'series': 'Nytt på nytt', + 'episode': '27.01.2017', + }, + 'params': { + 'skip_download': True, }, - 'skip': 'Only works from Norway', }, { 'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#', 'only_matching': True, }] -class NRKPlaylistIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P[^/]+)' +class NRKTVEpisodeIE(InfoExtractor): + _VALID_URL = r'https?://tv\.nrk\.no/serie/(?P[^/]+/sesong/\d+/episode/\d+)' + _TEST = { + 'url': 'https://tv.nrk.no/serie/backstage/sesong/1/episode/8', + 'info_dict': { + 'id': 'MSUI14000816AA', + 'ext': 'mp4', + 'title': 'Backstage 8:30', + 'description': 'md5:de6ca5d5a2d56849e4021f2bf2850df4', + 'duration': 1320, + 'series': 'Backstage', + 'season_number': 1, + 'episode_number': 8, + 'episode': '8:30', + }, + 'params': { + 'skip_download': True, + }, + } + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + nrk_id = self._parse_json( + self._search_regex(JSON_LD_RE, webpage, 'JSON-LD', group='json_ld'), + display_id)['@id'] + + assert re.match(NRKTVIE._EPISODE_RE, nrk_id) + return self.url_result( + 'nrk:%s' % nrk_id, ie=NRKIE.ie_key(), video_id=nrk_id) + + +class NRKTVSerieBaseIE(InfoExtractor): + def _extract_series(self, webpage, display_id, fatal=True): + config = self._parse_json( + self._search_regex( + r'({.+?})\s*,\s*"[^"]+"\s*\)\s*', webpage, 'config', + default='{}' if not fatal else NO_DEFAULT), + display_id, fatal=False) + if not config: + return + return try_get(config, lambda x: x['series'], dict) + + def _extract_episodes(self, season): + entries = [] + if not isinstance(season, dict): + return entries + episodes = season.get('episodes') + if not isinstance(episodes, list): + return entries + for episode in episodes: + nrk_id = episode.get('prfId') + if not nrk_id or not isinstance(nrk_id, compat_str): + continue + entries.append(self.url_result( + 'nrk:%s' % nrk_id, ie=NRKIE.ie_key(), video_id=nrk_id)) + return entries + + +class NRKTVSeasonIE(NRKTVSerieBaseIE): + _VALID_URL = r'https?://tv\.nrk\.no/serie/[^/]+/sesong/(?P\d+)' + _TEST = { + 'url': 'https://tv.nrk.no/serie/backstage/sesong/1', + 'info_dict': { + 'id': '1', + 'title': 'Sesong 1', + }, + 'playlist_mincount': 30, + } + + @classmethod + def suitable(cls, url): + return (False if NRKTVIE.suitable(url) or NRKTVEpisodeIE.suitable(url) + else super(NRKTVSeasonIE, cls).suitable(url)) + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + series = self._extract_series(webpage, display_id) + + season = next( + s for s in series['seasons'] + if int(display_id) == s.get('seasonNumber')) + + title = try_get(season, lambda x: x['titles']['title'], compat_str) + return self.playlist_result( + self._extract_episodes(season), display_id, title) + + +class NRKTVSeriesIE(NRKTVSerieBaseIE): + _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/serie/(?P[^/]+)' + _ITEM_RE = r'(?:data-season=["\']|id=["\']season-)(?P\d+)' + _TESTS = [{ + # new layout + 'url': 'https://tv.nrk.no/serie/backstage', + 'info_dict': { + 'id': 'backstage', + 'title': 'Backstage', + 'description': 'md5:c3ec3a35736fca0f9e1207b5511143d3', + }, + 'playlist_mincount': 60, + }, { + # old layout + 'url': 'https://tv.nrk.no/serie/groenn-glede', + 'info_dict': { + 'id': 'groenn-glede', + 'title': 'Grønn glede', + 'description': 'md5:7576e92ae7f65da6993cf90ee29e4608', + }, + 'playlist_mincount': 9, + }, { + 'url': 'http://tv.nrksuper.no/serie/labyrint', + 'info_dict': { + 'id': 'labyrint', + 'title': 'Labyrint', + 'description': 'md5:58afd450974c89e27d5a19212eee7115', + }, + 'playlist_mincount': 3, + }, { + 'url': 'https://tv.nrk.no/serie/broedrene-dal-og-spektralsteinene', + 'only_matching': True, + }, { + 'url': 'https://tv.nrk.no/serie/saving-the-human-race', + 'only_matching': True, + }, { + 'url': 'https://tv.nrk.no/serie/postmann-pat', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return ( + False if any(ie.suitable(url) + for ie in (NRKTVIE, NRKTVEpisodeIE, NRKTVSeasonIE)) + else super(NRKTVSeriesIE, cls).suitable(url)) + + def _real_extract(self, url): + series_id = self._match_id(url) + + webpage = self._download_webpage(url, series_id) + + # New layout (e.g. https://tv.nrk.no/serie/backstage) + series = self._extract_series(webpage, series_id, fatal=False) + if series: + title = try_get(series, lambda x: x['titles']['title'], compat_str) + description = try_get( + series, lambda x: x['titles']['subtitle'], compat_str) + entries = [] + for season in series['seasons']: + entries.extend(self._extract_episodes(season)) + return self.playlist_result(entries, series_id, title, description) + + # Old layout (e.g. https://tv.nrk.no/serie/groenn-glede) + entries = [ + self.url_result( + 'https://tv.nrk.no/program/Episodes/{series}/{season}'.format( + series=series_id, season=season_id)) + for season_id in re.findall(self._ITEM_RE, webpage) + ] + + title = self._html_search_meta( + 'seriestitle', webpage, + 'title', default=None) or self._og_search_title( + webpage, fatal=False) + + description = self._html_search_meta( + 'series_description', webpage, + 'description', default=None) or self._og_search_description(webpage) + + return self.playlist_result(entries, series_id, title, description) + + +class NRKTVDirekteIE(NRKTVIE): + IE_DESC = 'NRK TV Direkte and NRK Radio Direkte' + _VALID_URL = r'https?://(?:tv|radio)\.nrk\.no/direkte/(?P[^/?#&]+)' + _TESTS = [{ + 'url': 'https://tv.nrk.no/direkte/nrk1', + 'only_matching': True, + }, { + 'url': 'https://radio.nrk.no/direkte/p1_oslo_akershus', + 'only_matching': True, + }] + + +class NRKPlaylistBaseIE(InfoExtractor): + def _extract_description(self, webpage): + pass + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + entries = [ + self.url_result('nrk:%s' % video_id, NRKIE.ie_key()) + for video_id in re.findall(self._ITEM_RE, webpage) + ] + + playlist_title = self. _extract_title(webpage) + playlist_description = self._extract_description(webpage) + + return self.playlist_result( + entries, playlist_id, playlist_title, playlist_description) + + +class NRKPlaylistIE(NRKPlaylistBaseIE): + _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P[^/]+)' + _ITEM_RE = r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"' _TESTS = [{ 'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763', 'info_dict': { @@ -281,23 +596,28 @@ class NRKPlaylistIE(InfoExtractor): 'playlist_count': 5, }] - def _real_extract(self, url): - playlist_id = self._match_id(url) + def _extract_title(self, webpage): + return self._og_search_title(webpage, fatal=False) - webpage = self._download_webpage(url, playlist_id) + def _extract_description(self, webpage): + return self._og_search_description(webpage) - entries = [ - self.url_result('nrk:%s' % video_id, 'NRK') - for video_id in re.findall( - r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"', - webpage) - ] - playlist_title = self._og_search_title(webpage) - playlist_description = self._og_search_description(webpage) +class NRKTVEpisodesIE(NRKPlaylistBaseIE): + _VALID_URL = r'https?://tv\.nrk\.no/program/[Ee]pisodes/[^/]+/(?P\d+)' + _ITEM_RE = r'data-episode=["\']%s' % NRKTVIE._EPISODE_RE + _TESTS = [{ + 'url': 'https://tv.nrk.no/program/episodes/nytt-paa-nytt/69031', + 'info_dict': { + 'id': '69031', + 'title': 'Nytt på nytt, sesong: 201210', + }, + 'playlist_count': 4, + }] - return self.playlist_result( - entries, playlist_id, playlist_title, playlist_description) + def _extract_title(self, webpage): + return self._html_search_regex( + r'

([^<]+)

', webpage, 'title', fatal=False) class NRKSkoleIE(InfoExtractor):