X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/fe979149c83b5a935f7d28baf75848a9137316fd..4d073ced87f1fff7fd6fdfd82462444769b9f90b:/youtube_dl/extractor/svt.py diff --git a/youtube_dl/extractor/svt.py b/youtube_dl/extractor/svt.py index 67f56fa..f71eab8 100644 --- a/youtube_dl/extractor/svt.py +++ b/youtube_dl/extractor/svt.py @@ -4,25 +4,35 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import ( + compat_parse_qs, + compat_urllib_parse_urlparse, +) from ..utils import ( determine_ext, dict_get, int_or_none, try_get, + urljoin, + compat_str, ) class SVTBaseIE(InfoExtractor): + _GEO_COUNTRIES = ['SE'] + def _extract_video(self, video_info, video_id): + is_live = dict_get(video_info, ('live', 'simulcast'), default=False) + m3u8_protocol = 'm3u8' if is_live else 'm3u8_native' formats = [] for vr in video_info['videoReferences']: - player_type = vr.get('playerType') + player_type = vr.get('playerType') or vr.get('format') vurl = vr['url'] ext = determine_ext(vurl) if ext == 'm3u8': formats.extend(self._extract_m3u8_formats( vurl, video_id, - ext='mp4', entry_protocol='m3u8_native', + ext='mp4', entry_protocol=m3u8_protocol, m3u8_id=player_type, fatal=False)) elif ext == 'f4m': formats.extend(self._extract_f4m_formats( @@ -38,7 +48,9 @@ class SVTBaseIE(InfoExtractor): 'url': vurl, }) if not formats and video_info.get('rights', {}).get('geoBlockedSweden'): - self.raise_geo_restricted('This video is only available in Sweden') + self.raise_geo_restricted( + 'This video is only available in Sweden', + countries=self._GEO_COUNTRIES) self._sort_formats(formats) subtitles = {} @@ -80,6 +92,7 @@ class SVTBaseIE(InfoExtractor): 'season_number': season_number, 'episode': episode, 'episode_number': episode_number, + 'is_live': is_live, } @@ -118,9 +131,13 @@ class SVTIE(SVTBaseIE): return info_dict -class SVTPlayIE(SVTBaseIE): +class SVTPlayBaseIE(SVTBaseIE): + _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P{.+?})\s*;\s*\n' + + +class SVTPlayIE(SVTPlayBaseIE): IE_DESC = 'SVT Play and Öppet arkiv' - _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/video/(?P[0-9]+)' + _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P[^/?#&]+)' _TESTS = [{ 'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2', 'md5': '2b6704fe4a28801e1a098bbf3c5ac611', @@ -129,7 +146,7 @@ class SVTPlayIE(SVTBaseIE): 'ext': 'mp4', 'title': 'Flygplan till Haile Selassie', 'duration': 3527, - 'thumbnail': 're:^https?://.*[\.-]jpg$', + 'thumbnail': r're:^https?://.*[\.-]jpg$', 'age_limit': 0, 'subtitles': { 'sv': [{ @@ -141,6 +158,12 @@ class SVTPlayIE(SVTBaseIE): # geo restricted to Sweden 'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten', 'only_matching': True, + }, { + 'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg', + 'only_matching': True, + }, { + 'url': 'https://www.svtplay.se/kanaler/svt1', + 'only_matching': True, }] def _real_extract(self, url): @@ -150,12 +173,16 @@ class SVTPlayIE(SVTBaseIE): data = self._parse_json( self._search_regex( - r'root\["__svtplay"\]\s*=\s*([^;]+);', - webpage, 'embedded data', default='{}'), + self._SVTPLAY_RE, webpage, 'embedded data', default='{}', + group='json'), video_id, fatal=False) thumbnail = self._og_search_thumbnail(webpage) + def adjust_title(info): + if info['is_live']: + info['title'] = self._live_title(info['title']) + if data: video_info = try_get( data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'], @@ -166,6 +193,7 @@ class SVTPlayIE(SVTBaseIE): 'title': data['context']['dispatcher']['stores']['MetaStore']['title'], 'thumbnail': thumbnail, }) + adjust_title(info_dict) return info_dict video_id = self._search_regex( @@ -174,10 +202,93 @@ class SVTPlayIE(SVTBaseIE): if video_id: data = self._download_json( - 'http://www.svt.se/videoplayer-api/video/%s' % video_id, video_id) + 'https://api.svt.se/videoplayer-api/video/%s' % video_id, + video_id, headers=self.geo_verification_headers()) info_dict = self._extract_video(data, video_id) if not info_dict.get('title'): info_dict['title'] = re.sub( r'\s*\|\s*.+?$', '', info_dict.get('episode') or self._og_search_title(webpage)) + adjust_title(info_dict) return info_dict + + +class SVTSeriesIE(SVTPlayBaseIE): + _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P[^/?&#]+)' + _TESTS = [{ + 'url': 'https://www.svtplay.se/rederiet', + 'info_dict': { + 'id': 'rederiet', + 'title': 'Rederiet', + 'description': 'md5:505d491a58f4fcf6eb418ecab947e69e', + }, + 'playlist_mincount': 318, + }, { + 'url': 'https://www.svtplay.se/rederiet?tab=sasong2', + 'info_dict': { + 'id': 'rederiet-sasong2', + 'title': 'Rederiet - Säsong 2', + 'description': 'md5:505d491a58f4fcf6eb418ecab947e69e', + }, + 'playlist_count': 12, + }] + + @classmethod + def suitable(cls, url): + return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url) + + def _real_extract(self, url): + series_id = self._match_id(url) + + qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query) + season_slug = qs.get('tab', [None])[0] + + if season_slug: + series_id += '-%s' % season_slug + + webpage = self._download_webpage( + url, series_id, 'Downloading series page') + + root = self._parse_json( + self._search_regex( + self._SVTPLAY_RE, webpage, 'content', group='json'), + series_id) + + season_name = None + + entries = [] + for season in root['relatedVideoContent']['relatedVideosAccordion']: + if not isinstance(season, dict): + continue + if season_slug: + if season.get('slug') != season_slug: + continue + season_name = season.get('name') + videos = season.get('videos') + if not isinstance(videos, list): + continue + for video in videos: + content_url = video.get('contentUrl') + if not content_url or not isinstance(content_url, compat_str): + continue + entries.append( + self.url_result( + urljoin(url, content_url), + ie=SVTPlayIE.ie_key(), + video_title=video.get('title') + )) + + metadata = root.get('metaData') + if not isinstance(metadata, dict): + metadata = {} + + title = metadata.get('title') + season_name = season_name or season_slug + + if title and season_name: + title = '%s - %s' % (title, season_name) + elif season_slug: + title = season_slug + + return self.playlist_result( + entries, series_id, title, metadata.get('description'))