diff options
Diffstat (limited to 'youtube_dl/extractor/svt.py')
| -rw-r--r-- | youtube_dl/extractor/svt.py | 289 |
1 files changed, 262 insertions, 27 deletions
diff --git a/youtube_dl/extractor/svt.py b/youtube_dl/extractor/svt.py index 1b5afb73e..a5bb6daa7 100644 --- a/youtube_dl/extractor/svt.py +++ b/youtube_dl/extractor/svt.py @@ -4,10 +4,14 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_str from ..utils import ( determine_ext, dict_get, int_or_none, + unified_timestamp, + str_or_none, + strip_or_none, try_get, ) @@ -16,6 +20,8 @@ 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') or vr.get('format') @@ -24,7 +30,7 @@ class SVTBaseIE(InfoExtractor): 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( @@ -39,7 +45,8 @@ class SVTBaseIE(InfoExtractor): 'format_id': player_type, 'url': vurl, }) - if not formats and video_info.get('rights', {}).get('geoBlockedSweden'): + rights = try_get(video_info, lambda x: x['rights'], dict) or {} + if not formats and rights.get('geoBlockedSweden'): self.raise_geo_restricted( 'This video is only available in Sweden', countries=self._GEO_COUNTRIES) @@ -65,6 +72,7 @@ class SVTBaseIE(InfoExtractor): episode = video_info.get('episodeTitle') episode_number = int_or_none(video_info.get('episodeNumber')) + timestamp = unified_timestamp(rights.get('validFrom')) duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration'))) age_limit = None adult = dict_get( @@ -79,11 +87,13 @@ class SVTBaseIE(InfoExtractor): 'formats': formats, 'subtitles': subtitles, 'duration': duration, + 'timestamp': timestamp, 'age_limit': age_limit, 'series': series, 'season_number': season_number, 'episode': episode, 'episode_number': episode_number, + 'is_live': is_live, } @@ -122,25 +132,54 @@ class SVTIE(SVTBaseIE): return info_dict -class SVTPlayIE(SVTBaseIE): +class SVTPlayBaseIE(SVTBaseIE): + _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P<json>{.+?})\s*;\s*\n' + + +class SVTPlayIE(SVTPlayBaseIE): IE_DESC = 'SVT Play and Öppet arkiv' - _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp)/(?P<id>[0-9]+)' + _VALID_URL = r'''(?x) + (?: + (?: + svt:| + https?://(?:www\.)?svt\.se/barnkanalen/barnplay/[^/]+/ + ) + (?P<svt_id>[^/?#&]+)| + https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+) + (?:.*?(?:modalId|id)=(?P<modal_id>[\da-zA-Z-]+))? + ) + ''' _TESTS = [{ - 'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2', - 'md5': '2b6704fe4a28801e1a098bbf3c5ac611', + 'url': 'https://www.svtplay.se/video/30479064', + 'md5': '2382036fd6f8c994856c323fe51c426e', 'info_dict': { - 'id': '5996901', + 'id': '8zVbDPA', 'ext': 'mp4', - 'title': 'Flygplan till Haile Selassie', - 'duration': 3527, - 'thumbnail': r're:^https?://.*[\.-]jpg$', + 'title': 'Designdrömmar i Stenungsund', + 'timestamp': 1615770000, + 'upload_date': '20210315', + 'duration': 3519, + 'thumbnail': r're:^https?://(?:.*[\.-]jpg|www.svtstatic.se/image/.*)$', 'age_limit': 0, 'subtitles': { 'sv': [{ - 'ext': 'wsrt', + 'ext': 'vtt', }] }, }, + 'params': { + 'format': 'bestvideo', + # skip for now due to download test asserts that segment is > 10000 bytes and svt uses + # init segments that are smaller + # AssertionError: Expected test_SVTPlay_jNwpV9P.mp4 to be at least 9.77KiB, but it's only 864.00B + 'skip_download': True, + }, + }, { + 'url': 'https://www.svtplay.se/video/30479064/husdrommar/husdrommar-sasong-8-designdrommar-i-stenungsund?modalId=8zVbDPA', + 'only_matching': True, + }, { + 'url': 'https://www.svtplay.se/video/30684086/rapport/rapport-24-apr-18-00-7?id=e72gVpa', + 'only_matching': True, }, { # geo restricted to Sweden 'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten', @@ -148,17 +187,57 @@ class SVTPlayIE(SVTBaseIE): }, { 'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg', 'only_matching': True, + }, { + 'url': 'https://www.svtplay.se/kanaler/svt1', + 'only_matching': True, + }, { + 'url': 'svt:1376446-003A', + 'only_matching': True, + }, { + 'url': 'svt:14278044', + 'only_matching': True, + }, { + 'url': 'https://www.svt.se/barnkanalen/barnplay/kar/eWv5MLX/', + 'only_matching': True, + }, { + 'url': 'svt:eWv5MLX', + '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 = mobj.group('id') + svt_id = mobj.group('svt_id') or mobj.group('modal_id') + + if svt_id: + return self._extract_by_video_id(svt_id) webpage = self._download_webpage(url, video_id) 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) @@ -173,18 +252,174 @@ class SVTPlayIE(SVTBaseIE): 'title': data['context']['dispatcher']['stores']['MetaStore']['title'], 'thumbnail': thumbnail, }) + self._adjust_title(info_dict) return info_dict - video_id = self._search_regex( - r'<video[^>]+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 + svt_id = try_get( + data, lambda x: x['statistics']['dataLake']['content']['id'], + compat_str) + + if not svt_id: + svt_id = self._search_regex( + (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)', + r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/%s/[^"\']*\b(?:modalId|id)=([\da-zA-Z-]+)' % re.escape(video_id), + r'["\']videoSvtId["\']\s*:\s*["\']([\da-zA-Z-]+)', + r'["\']videoSvtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)', + r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"', + r'["\']svtId["\']\s*:\s*["\']([\da-zA-Z-]+)', + r'["\']svtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)'), + webpage, 'video id') + + info_dict = self._extract_by_video_id(svt_id, webpage) + info_dict['thumbnail'] = thumbnail + + return info_dict + + +class SVTSeriesIE(SVTPlayBaseIE): + _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)(?:.+?\btab=(?P<season_slug>[^&#]+))?' + _TESTS = [{ + 'url': 'https://www.svtplay.se/rederiet', + 'info_dict': { + 'id': '14445680', + 'title': 'Rederiet', + 'description': 'md5:d9fdfff17f5d8f73468176ecd2836039', + }, + 'playlist_mincount': 318, + }, { + 'url': 'https://www.svtplay.se/rederiet?tab=season-2-14445680', + 'info_dict': { + 'id': 'season-2-14445680', + 'title': 'Rederiet - Säsong 2', + 'description': 'md5:d9fdfff17f5d8f73468176ecd2836039', + }, + 'playlist_mincount': 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_slug, season_id = re.match(self._VALID_URL, url).groups() + + series = self._download_json( + 'https://api.svt.se/contento/graphql', series_slug, + 'Downloading series page', query={ + 'query': '''{ + listablesBySlug(slugs: ["%s"]) { + associatedContent(include: [productionPeriod, season]) { + items { + item { + ... on Episode { + videoSvtId + } + } + } + id + name + } + id + longDescription + name + shortDescription + } +}''' % series_slug, + })['data']['listablesBySlug'][0] + + season_name = None + + entries = [] + for season in series['associatedContent']: + if not isinstance(season, dict): + continue + if season_id: + if season.get('id') != season_id: + continue + season_name = season.get('name') + items = season.get('items') + if not isinstance(items, list): + continue + for item in items: + video = item.get('item') or {} + content_id = video.get('videoSvtId') + if not content_id or not isinstance(content_id, compat_str): + continue + entries.append(self.url_result( + 'svt:' + content_id, SVTPlayIE.ie_key(), content_id)) + + title = series.get('name') + season_name = season_name or season_id + + if title and season_name: + title = '%s - %s' % (title, season_name) + elif season_id: + title = season_id + + return self.playlist_result( + entries, season_id or series.get('id'), title, + dict_get(series, ('longDescription', 'shortDescription'))) + + +class SVTPageIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?svt\.se/(?P<path>(?:[^/]+/)*(?P<id>[^/?&#]+))' + _TESTS = [{ + 'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa', + 'info_dict': { + 'id': '25298267', + 'title': 'Bakom masken – Lehners kamp mot mental ohälsa', + }, + 'playlist_count': 4, + }, { + 'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien', + 'info_dict': { + 'id': '24243746', + 'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien', + }, + 'playlist_count': 2, + }, { + # only programTitle + 'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun', + 'info_dict': { + 'id': '8439V2K', + '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) or SVTPlayIE.suitable(url) else super(SVTPageIE, cls).suitable(url) + + def _real_extract(self, url): + path, display_id = re.match(self._VALID_URL, url).groups() + + article = self._download_json( + 'https://api.svt.se/nss-api/page/' + path, display_id, + query={'q': 'articles'})['articles']['content'][0] + + entries = [] + + def _process_content(content): + if content.get('_type') in ('VIDEOCLIP', 'VIDEOEPISODE'): + video_id = compat_str(content['image']['svtId']) + entries.append(self.url_result( + 'svt:' + video_id, SVTPlayIE.ie_key(), video_id)) + + for media in article.get('media', []): + _process_content(media) + + for obj in article.get('structuredBody', []): + _process_content(obj.get('content') or {}) + + return self.playlist_result( + entries, str_or_none(article.get('id')), + strip_or_none(article.get('title'))) |