From 7b393f9cc5dc4790bcb623c768fa4a3046ef80bf Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Thu, 28 Jun 2018 04:29:11 +0700 Subject: [PATCH] [svt] Improve extraction and add support for pages (closes #16802) --- youtube_dl/extractor/extractors.py | 1 + youtube_dl/extractor/generic.py | 11 --- youtube_dl/extractor/svt.py | 117 ++++++++++++++++++++++++----- 3 files changed, 98 insertions(+), 31 deletions(-) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 27ece3b53..f2377521b 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -1040,6 +1040,7 @@ from .stretchinternet import StretchInternetIE from .sunporno import SunPornoIE from .svt import ( SVTIE, + SVTPageIE, SVTPlayIE, SVTSeriesIE, ) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index d71cb9050..aa04905ed 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -1395,17 +1395,6 @@ class GenericIE(InfoExtractor): 'skip_download': True, }, }, - # SVT embed - { - 'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun', - 'info_dict': { - 'id': '2900353', - 'ext': 'flv', - 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)', - 'duration': 27, - 'age_limit': 0, - }, - }, # Crooks and Liars embed { 'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists', diff --git a/youtube_dl/extractor/svt.py b/youtube_dl/extractor/svt.py index f71eab8b2..0901c3163 100644 --- a/youtube_dl/extractor/svt.py +++ b/youtube_dl/extractor/svt.py @@ -12,6 +12,8 @@ from ..utils import ( determine_ext, dict_get, int_or_none, + orderedSet, + strip_or_none, try_get, urljoin, compat_str, @@ -137,7 +139,12 @@ class SVTPlayBaseIE(SVTBaseIE): class SVTPlayIE(SVTPlayBaseIE): IE_DESC = 'SVT Play and Öppet arkiv' - _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P[^/?#&]+)' + _VALID_URL = r'''(?x) + (?: + svt:(?P[^/?#&]+)| + 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', @@ -164,10 +171,40 @@ class SVTPlayIE(SVTPlayBaseIE): }, { 'url': 'https://www.svtplay.se/kanaler/svt1', 'only_matching': True, + }, { + 'url': 'svt:1376446-003A', + 'only_matching': True, + }, { + 'url': 'svt:14278044', + 'only_matching': True, }] + def _adjust_title(self, info): + if info['is_live']: + info['title'] = self._live_title(info['title']) + + def _extract_by_video_id(self, video_id, webpage=None): + data = self._download_json( + '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'): + title = dict_get(info_dict, ('episode', 'series')) + if not title and webpage: + title = re.sub( + r'\s*\|\s*.+?$', '', self._og_search_title(webpage)) + if not title: + title = video_id + info_dict['title'] = title + self._adjust_title(info_dict) + return info_dict + def _real_extract(self, url): - video_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + video_id, svt_id = mobj.group('id', 'svt_id') + + if svt_id: + return self._extract_by_video_id(svt_id) webpage = self._download_webpage(url, video_id) @@ -179,10 +216,6 @@ class SVTPlayIE(SVTPlayBaseIE): thumbnail = self._og_search_thumbnail(webpage) - def adjust_title(info): - if info['is_live']: - info['title'] = self._live_title(info['title']) - if data: video_info = try_get( data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'], @@ -193,24 +226,14 @@ class SVTPlayIE(SVTPlayBaseIE): 'title': data['context']['dispatcher']['stores']['MetaStore']['title'], 'thumbnail': thumbnail, }) - adjust_title(info_dict) + self._adjust_title(info_dict) return info_dict - video_id = self._search_regex( + svt_id = self._search_regex( r']+data-video-id=["\']([\da-zA-Z-]+)', - webpage, 'video id', default=None) + webpage, 'video id') - if video_id: - data = self._download_json( - 'https://api.svt.se/videoplayer-api/video/%s' % video_id, - video_id, headers=self.geo_verification_headers()) - info_dict = self._extract_video(data, video_id) - if not info_dict.get('title'): - info_dict['title'] = re.sub( - r'\s*\|\s*.+?$', '', - info_dict.get('episode') or self._og_search_title(webpage)) - adjust_title(info_dict) - return info_dict + return self._extract_by_video_id(svt_id, webpage) class SVTSeriesIE(SVTPlayBaseIE): @@ -292,3 +315,57 @@ class SVTSeriesIE(SVTPlayBaseIE): return self.playlist_result( entries, series_id, title, metadata.get('description')) + + +class SVTPageIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?svt\.se/(?:[^/]+/)*(?P[^/?&#]+)' + _TESTS = [{ + 'url': 'https://www.svt.se/sport/oseedat/guide-sommartraningen-du-kan-gora-var-och-nar-du-vill', + 'info_dict': { + 'id': 'guide-sommartraningen-du-kan-gora-var-och-nar-du-vill', + 'title': 'GUIDE: Sommarträning du kan göra var och när du vill', + }, + 'playlist_count': 7, + }, { + 'url': 'https://www.svt.se/nyheter/inrikes/ebba-busch-thor-kd-har-delvis-ratt-om-no-go-zoner', + 'info_dict': { + 'id': 'ebba-busch-thor-kd-har-delvis-ratt-om-no-go-zoner', + 'title': 'Ebba Busch Thor har bara delvis rätt om ”no-go-zoner”', + }, + 'playlist_count': 1, + }, { + # only programTitle + 'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun', + 'info_dict': { + 'id': '2900353', + 'ext': 'mp4', + 'title': 'Stjärnorna skojar till det - under SVT-intervjun', + 'duration': 27, + 'age_limit': 0, + }, + }, { + 'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1', + 'only_matching': True, + }, { + 'url': 'https://www.svt.se/vader/manadskronikor/maj2018', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return False if SVTIE.suitable(url) else super(SVTPageIE, cls).suitable(url) + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + entries = [ + self.url_result( + 'svt:%s' % video_id, ie=SVTPlayIE.ie_key(), video_id=video_id) + for video_id in orderedSet(re.findall( + r'data-video-id=["\'](\d+)', webpage))] + + title = strip_or_none(self._og_search_title(webpage, default=None)) + + return self.playlist_result(entries, playlist_id, title) -- 2.22.2