X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fsvt.py;h=1b5afb73ee473b23e78c27caba6708ce9ceb348f;hb=329e3dd5adf52520c87ba31395d090455114783b;hp=2ab30e45ff7c65ab7dd1d6cff7a1952764799cc0;hpb=d6712378e73951bede475569c887a1ac73f660a9;p=youtube-dl diff --git a/youtube_dl/extractor/svt.py b/youtube_dl/extractor/svt.py index 2ab30e45f..1b5afb73e 100644 --- a/youtube_dl/extractor/svt.py +++ b/youtube_dl/extractor/svt.py @@ -6,20 +6,19 @@ import re from .common import InfoExtractor from ..utils import ( determine_ext, + dict_get, + int_or_none, + try_get, ) class SVTBaseIE(InfoExtractor): - def _extract_video(self, url, video_id): - info = self._download_json(url, video_id) + _GEO_COUNTRIES = ['SE'] - title = info['context']['title'] - thumbnail = info['context'].get('thumbnailImage') - - video_info = info['video'] + def _extract_video(self, video_info, video_id): 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': @@ -40,27 +39,51 @@ class SVTBaseIE(InfoExtractor): 'format_id': player_type, 'url': vurl, }) + if not formats and video_info.get('rights', {}).get('geoBlockedSweden'): + self.raise_geo_restricted( + 'This video is only available in Sweden', + countries=self._GEO_COUNTRIES) self._sort_formats(formats) subtitles = {} - subtitle_references = video_info.get('subtitleReferences') + subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences')) if isinstance(subtitle_references, list): for sr in subtitle_references: subtitle_url = sr.get('url') + subtitle_lang = sr.get('language', 'sv') if subtitle_url: - subtitles.setdefault('sv', []).append({'url': subtitle_url}) + if determine_ext(subtitle_url) == 'm3u8': + # TODO(yan12125): handle WebVTT in m3u8 manifests + continue + + subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url}) + + title = video_info.get('title') - duration = video_info.get('materialLength') - age_limit = 18 if video_info.get('inappropriateForChildren') else 0 + series = video_info.get('programTitle') + season_number = int_or_none(video_info.get('season')) + episode = video_info.get('episodeTitle') + episode_number = int_or_none(video_info.get('episodeNumber')) + + duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration'))) + age_limit = None + adult = dict_get( + video_info, ('inappropriateForChildren', 'blockedForChildren'), + skip_false_values=False) + if adult is not None: + age_limit = 18 if adult else 0 return { 'id': video_id, 'title': title, 'formats': formats, 'subtitles': subtitles, - 'thumbnail': thumbnail, 'duration': duration, 'age_limit': age_limit, + 'series': series, + 'season_number': season_number, + 'episode': episode, + 'episode_number': episode_number, } @@ -68,11 +91,11 @@ class SVTIE(SVTBaseIE): _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P\d+)&.*?\barticleId=(?P\d+)' _TEST = { 'url': 'http://www.svt.se/wd?widgetId=23991§ionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false', - 'md5': '9648197555fc1b49e3dc22db4af51d46', + 'md5': '33e9a5d8f646523ce0868ecfb0eed77d', 'info_dict': { 'id': '2900353', - 'ext': 'flv', - 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)', + 'ext': 'mp4', + 'title': 'Stjärnorna skojar till det - under SVT-intervjun', 'duration': 27, 'age_limit': 0, }, @@ -89,15 +112,20 @@ class SVTIE(SVTBaseIE): mobj = re.match(self._VALID_URL, url) widget_id = mobj.group('widget_id') article_id = mobj.group('id') - return self._extract_video( + + info = self._download_json( 'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id), article_id) + info_dict = self._extract_video(info['video'], article_id) + info_dict['title'] = info['context']['title'] + return info_dict + class SVTPlayIE(SVTBaseIE): IE_DESC = 'SVT Play and Öppet arkiv' - _VALID_URL = r'https?://(?:www\.)?(?Psvtplay|oppetarkiv)\.se/video/(?P[0-9]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp)/(?P[0-9]+)' + _TESTS = [{ 'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2', 'md5': '2b6704fe4a28801e1a098bbf3c5ac611', 'info_dict': { @@ -105,7 +133,7 @@ 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': [{ @@ -113,12 +141,50 @@ 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, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - host = mobj.group('host') - return self._extract_video( - 'http://www.%s.se/video/%s?output=json' % (host, video_id), - video_id) + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + data = self._parse_json( + self._search_regex( + r'root\["__svtplay"\]\s*=\s*([^;]+);', + webpage, 'embedded data', default='{}'), + video_id, fatal=False) + + thumbnail = self._og_search_thumbnail(webpage) + + if data: + video_info = try_get( + data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'], + dict) + if video_info: + info_dict = self._extract_video(video_info, video_id) + info_dict.update({ + 'title': data['context']['dispatcher']['stores']['MetaStore']['title'], + 'thumbnail': thumbnail, + }) + return info_dict + + video_id = self._search_regex( + r']+data-video-id=["\']([\da-zA-Z-]+)', + webpage, 'video id', default=None) + + if video_id: + data = self._download_json( + 'http://www.svt.se/videoplayer-api/video/%s' % video_id, video_id) + 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)) + return info_dict