From 1a7dcca378e80a387923ee05c250d8ba122441c6 Mon Sep 17 00:00:00 2001 From: Jeroen Jacobs Date: Fri, 2 Jun 2023 20:29:00 +0200 Subject: [PATCH] [extractor/vrt] Overhaul extractors (#6244) * Fixes `VrtNU` extractor to work with the VRT MAX site change * Adapts `VRT`, `Ketnet` and `DagelijkseKost` extractors to the new VRT API * Removes `Canvas` and `CanvasEen` extractors; the sites and API no longer exist * Moves all remaining VRT-related extractors into the `vrt` module Closes #4908 Authored by: jeroenj, bergoid, bashonly Co-authored-by: bergoid Co-authored-by: bashonly <88596187+bashonly@users.noreply.github.com> --- yt_dlp/extractor/_extractors.py | 14 +- yt_dlp/extractor/canvas.py | 383 ----------------------------- yt_dlp/extractor/ketnet.py | 70 ------ yt_dlp/extractor/vrt.py | 413 +++++++++++++++++++++++++++++--- 4 files changed, 384 insertions(+), 496 deletions(-) delete mode 100644 yt_dlp/extractor/canvas.py delete mode 100644 yt_dlp/extractor/ketnet.py diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py index 808ede5ba..7120fd37d 100644 --- a/yt_dlp/extractor/_extractors.py +++ b/yt_dlp/extractor/_extractors.py @@ -295,12 +295,6 @@ from .canalalpha import CanalAlphaIE from .canalplus import CanalplusIE from .canalc2 import Canalc2IE -from .canvas import ( - CanvasIE, - CanvasEenIE, - VrtNUIE, - DagelijkseKostIE, -) from .carambatv import ( CarambaTVIE, CarambaTVPageIE, @@ -894,7 +888,6 @@ from .karrierevideos import KarriereVideosIE from .keezmovies import KeezMoviesIE from .kelbyone import KelbyOneIE -from .ketnet import KetnetIE from .khanacademy import ( KhanAcademyIE, KhanAcademyUnitIE, @@ -2285,7 +2278,12 @@ VoxMediaVolumeIE, VoxMediaIE, ) -from .vrt import VRTIE +from .vrt import ( + VRTIE, + VrtNUIE, + KetnetIE, + DagelijkseKostIE, +) from .vrak import VrakIE from .vrv import ( VRVIE, diff --git a/yt_dlp/extractor/canvas.py b/yt_dlp/extractor/canvas.py deleted file mode 100644 index ae6e03a4d..000000000 --- a/yt_dlp/extractor/canvas.py +++ /dev/null @@ -1,383 +0,0 @@ -import json - - -from .common import InfoExtractor -from .gigya import GigyaBaseIE -from ..compat import compat_HTTPError -from ..utils import ( - ExtractorError, - clean_html, - extract_attributes, - float_or_none, - get_element_by_class, - int_or_none, - merge_dicts, - str_or_none, - strip_or_none, - url_or_none, - urlencode_postdata -) - - -class CanvasIE(InfoExtractor): - _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?Pcanvas|een|ketnet|vrt(?:video|nieuws)|sporza|dako)/assets/(?P[^/?#&]+)' - _TESTS = [{ - 'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475', - 'md5': '37b2b7bb9b3dcaa05b67058dc3a714a9', - 'info_dict': { - 'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475', - 'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475', - 'ext': 'mp4', - 'title': 'Nachtwacht: De Greystook', - 'description': 'Nachtwacht: De Greystook', - 'thumbnail': r're:^https?://.*\.jpg$', - 'duration': 1468.02, - }, - 'expected_warnings': ['is not a supported codec'], - }, { - 'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e', - 'only_matching': True, - }] - _GEO_BYPASS = False - _HLS_ENTRY_PROTOCOLS_MAP = { - 'HLS': 'm3u8_native', - 'HLS_AES': 'm3u8_native', - } - _REST_API_BASE = 'https://media-services-public.vrt.be/vualto-video-aggregator-web/rest/external/v2' - - def _real_extract(self, url): - mobj = self._match_valid_url(url) - site_id, video_id = mobj.group('site_id'), mobj.group('id') - - data = None - if site_id != 'vrtvideo': - # Old API endpoint, serves more formats but may fail for some videos - data = self._download_json( - 'https://mediazone.vrt.be/api/v1/%s/assets/%s' - % (site_id, video_id), video_id, 'Downloading asset JSON', - 'Unable to download asset JSON', fatal=False) - - # New API endpoint - if not data: - vrtnutoken = self._download_json('https://token.vrt.be/refreshtoken', - video_id, note='refreshtoken: Retrieve vrtnutoken', - errnote='refreshtoken failed')['vrtnutoken'] - headers = self.geo_verification_headers() - headers.update({'Content-Type': 'application/json; charset=utf-8'}) - vrtPlayerToken = self._download_json( - '%s/tokens' % self._REST_API_BASE, video_id, - 'Downloading token', headers=headers, data=json.dumps({ - 'identityToken': vrtnutoken - }).encode('utf-8'))['vrtPlayerToken'] - data = self._download_json( - '%s/videos/%s' % (self._REST_API_BASE, video_id), - video_id, 'Downloading video JSON', query={ - 'vrtPlayerToken': vrtPlayerToken, - 'client': 'null', - }, expected_status=400) - if 'title' not in data: - code = data.get('code') - if code == 'AUTHENTICATION_REQUIRED': - self.raise_login_required() - elif code == 'INVALID_LOCATION': - self.raise_geo_restricted(countries=['BE']) - raise ExtractorError(data.get('message') or code, expected=True) - - # Note: The title may be an empty string - title = data['title'] or f'{site_id} {video_id}' - description = data.get('description') - - formats = [] - subtitles = {} - for target in data['targetUrls']: - format_url, format_type = url_or_none(target.get('url')), str_or_none(target.get('type')) - if not format_url or not format_type: - continue - format_type = format_type.upper() - if format_type in self._HLS_ENTRY_PROTOCOLS_MAP: - fmts, subs = self._extract_m3u8_formats_and_subtitles( - format_url, video_id, 'mp4', self._HLS_ENTRY_PROTOCOLS_MAP[format_type], - m3u8_id=format_type, fatal=False) - formats.extend(fmts) - subtitles = self._merge_subtitles(subtitles, subs) - elif format_type == 'HDS': - formats.extend(self._extract_f4m_formats( - format_url, video_id, f4m_id=format_type, fatal=False)) - elif format_type == 'MPEG_DASH': - fmts, subs = self._extract_mpd_formats_and_subtitles( - format_url, video_id, mpd_id=format_type, fatal=False) - formats.extend(fmts) - subtitles = self._merge_subtitles(subtitles, subs) - elif format_type == 'HSS': - fmts, subs = self._extract_ism_formats_and_subtitles( - format_url, video_id, ism_id='mss', fatal=False) - formats.extend(fmts) - subtitles = self._merge_subtitles(subtitles, subs) - else: - formats.append({ - 'format_id': format_type, - 'url': format_url, - }) - - subtitle_urls = data.get('subtitleUrls') - if isinstance(subtitle_urls, list): - for subtitle in subtitle_urls: - subtitle_url = subtitle.get('url') - if subtitle_url and subtitle.get('type') == 'CLOSED': - subtitles.setdefault('nl', []).append({'url': subtitle_url}) - - return { - 'id': video_id, - 'display_id': video_id, - 'title': title, - 'description': description, - 'formats': formats, - 'duration': float_or_none(data.get('duration'), 1000), - 'thumbnail': data.get('posterImageUrl'), - 'subtitles': subtitles, - } - - -class CanvasEenIE(InfoExtractor): - IE_DESC = 'canvas.be and een.be' - _VALID_URL = r'https?://(?:www\.)?(?Pcanvas|een)\.be/(?:[^/]+/)*(?P[^/?#&]+)' - _TESTS = [{ - 'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week', - 'md5': 'ed66976748d12350b118455979cca293', - 'info_dict': { - 'id': 'mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e', - 'display_id': 'de-afspraak-veilt-voor-de-warmste-week', - 'ext': 'flv', - 'title': 'De afspraak veilt voor de Warmste Week', - 'description': 'md5:24cb860c320dc2be7358e0e5aa317ba6', - 'thumbnail': r're:^https?://.*\.jpg$', - 'duration': 49.02, - }, - 'expected_warnings': ['is not a supported codec'], - }, { - # with subtitles - 'url': 'http://www.canvas.be/video/panorama/2016/pieter-0167', - 'info_dict': { - 'id': 'mz-ast-5240ff21-2d30-4101-bba6-92b5ec67c625', - 'display_id': 'pieter-0167', - 'ext': 'mp4', - 'title': 'Pieter 0167', - 'description': 'md5:943cd30f48a5d29ba02c3a104dc4ec4e', - 'thumbnail': r're:^https?://.*\.jpg$', - 'duration': 2553.08, - 'subtitles': { - 'nl': [{ - 'ext': 'vtt', - }], - }, - }, - 'params': { - 'skip_download': True, - }, - 'skip': 'Pagina niet gevonden', - }, { - 'url': 'https://www.een.be/thuis/emma-pakt-thilly-aan', - 'info_dict': { - 'id': 'md-ast-3a24ced2-64d7-44fb-b4ed-ed1aafbf90b8', - 'display_id': 'emma-pakt-thilly-aan', - 'ext': 'mp4', - 'title': 'Emma pakt Thilly aan', - 'description': 'md5:c5c9b572388a99b2690030afa3f3bad7', - 'thumbnail': r're:^https?://.*\.jpg$', - 'duration': 118.24, - }, - 'params': { - 'skip_download': True, - }, - 'expected_warnings': ['is not a supported codec'], - }, { - 'url': 'https://www.canvas.be/check-point/najaar-2016/de-politie-uw-vriend', - 'only_matching': True, - }] - - def _real_extract(self, url): - mobj = self._match_valid_url(url) - site_id, display_id = mobj.group('site_id'), mobj.group('id') - - webpage = self._download_webpage(url, display_id) - - title = strip_or_none(self._search_regex( - r']+class="video__body__header__title"[^>]*>(.+?)', - webpage, 'title', default=None) or self._og_search_title( - webpage, default=None)) - - video_id = self._html_search_regex( - r'data-video=(["\'])(?P(?:(?!\1).)+)\1', webpage, 'video id', - group='id') - - return { - '_type': 'url_transparent', - 'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (site_id, video_id), - 'ie_key': CanvasIE.ie_key(), - 'id': video_id, - 'display_id': display_id, - 'title': title, - 'description': self._og_search_description(webpage), - } - - -class VrtNUIE(GigyaBaseIE): - IE_DESC = 'VrtNU.be' - _VALID_URL = r'https?://(?:www\.)?vrt\.be/vrtnu/a-z/(?:[^/]+/){2}(?P[^/?#&]+)' - _TESTS = [{ - # Available via old API endpoint - 'url': 'https://www.vrt.be/vrtnu/a-z/postbus-x/1989/postbus-x-s1989a1/', - 'info_dict': { - 'id': 'pbs-pub-e8713dac-899e-41de-9313-81269f4c04ac$vid-90c932b1-e21d-4fb8-99b1-db7b49cf74de', - 'ext': 'mp4', - 'title': 'Postbus X - Aflevering 1 (Seizoen 1989)', - 'description': 'md5:b704f669eb9262da4c55b33d7c6ed4b7', - 'duration': 1457.04, - 'thumbnail': r're:^https?://.*\.jpg$', - 'series': 'Postbus X', - 'season': 'Seizoen 1989', - 'season_number': 1989, - 'episode': 'De zwarte weduwe', - 'episode_number': 1, - 'timestamp': 1595822400, - 'upload_date': '20200727', - }, - 'skip': 'This video is only available for registered users', - 'expected_warnings': ['is not a supported codec'], - }, { - # Only available via new API endpoint - 'url': 'https://www.vrt.be/vrtnu/a-z/kamp-waes/1/kamp-waes-s1a5/', - 'info_dict': { - 'id': 'pbs-pub-0763b56c-64fb-4d38-b95b-af60bf433c71$vid-ad36a73c-4735-4f1f-b2c0-a38e6e6aa7e1', - 'ext': 'mp4', - 'title': 'Aflevering 5', - 'description': 'Wie valt door de mand tijdens een missie?', - 'duration': 2967.06, - 'season': 'Season 1', - 'season_number': 1, - 'episode_number': 5, - }, - 'skip': 'This video is only available for registered users', - 'expected_warnings': ['Unable to download asset JSON', 'is not a supported codec', 'Unknown MIME type'], - }] - _NETRC_MACHINE = 'vrtnu' - _APIKEY = '3_0Z2HujMtiWq_pkAjgnS2Md2E11a1AwZjYiBETtwNE-EoEHDINgtnvcAOpNgmrVGy' - _CONTEXT_ID = 'R3595707040' - - def _perform_login(self, username, password): - auth_info = self._gigya_login({ - 'APIKey': self._APIKEY, - 'targetEnv': 'jssdk', - 'loginID': username, - 'password': password, - 'authMode': 'cookie', - }) - - if auth_info.get('errorDetails'): - raise ExtractorError('Unable to login: VrtNU said: ' + auth_info.get('errorDetails'), expected=True) - - # Sometimes authentication fails for no good reason, retry - login_attempt = 1 - while login_attempt <= 3: - try: - self._request_webpage('https://token.vrt.be/vrtnuinitlogin', - None, note='Requesting XSRF Token', errnote='Could not get XSRF Token', - query={'provider': 'site', 'destination': 'https://www.vrt.be/vrtnu/'}) - - post_data = { - 'UID': auth_info['UID'], - 'UIDSignature': auth_info['UIDSignature'], - 'signatureTimestamp': auth_info['signatureTimestamp'], - '_csrf': self._get_cookies('https://login.vrt.be').get('OIDCXSRF').value, - } - - self._request_webpage( - 'https://login.vrt.be/perform_login', - None, note='Performing login', errnote='perform login failed', - headers={}, query={ - 'client_id': 'vrtnu-site' - }, data=urlencode_postdata(post_data)) - - except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401: - login_attempt += 1 - self.report_warning('Authentication failed') - self._sleep(1, None, msg_template='Waiting for %(timeout)s seconds before trying again') - else: - raise e - else: - break - - def _real_extract(self, url): - display_id = self._match_id(url) - - webpage = self._download_webpage(url, display_id) - - attrs = extract_attributes(self._search_regex( - r'(]+>)', webpage, 'media element')) - video_id = attrs['videoid'] - publication_id = attrs.get('publicationid') - if publication_id: - video_id = publication_id + '$' + video_id - - page = (self._parse_json(self._search_regex( - r'digitalData\s*=\s*({.+?});', webpage, 'digial data', - default='{}'), video_id, fatal=False) or {}).get('page') or {} - - info = self._search_json_ld(webpage, display_id, default={}) - return merge_dicts(info, { - '_type': 'url_transparent', - 'url': 'https://mediazone.vrt.be/api/v1/vrtvideo/assets/%s' % video_id, - 'ie_key': CanvasIE.ie_key(), - 'id': video_id, - 'display_id': display_id, - 'season_number': int_or_none(page.get('episode_season')), - }) - - -class DagelijkseKostIE(InfoExtractor): - IE_DESC = 'dagelijksekost.een.be' - _VALID_URL = r'https?://dagelijksekost\.een\.be/gerechten/(?P[^/?#&]+)' - _TEST = { - 'url': 'https://dagelijksekost.een.be/gerechten/hachis-parmentier-met-witloof', - 'md5': '30bfffc323009a3e5f689bef6efa2365', - 'info_dict': { - 'id': 'md-ast-27a4d1ff-7d7b-425e-b84f-a4d227f592fa', - 'display_id': 'hachis-parmentier-met-witloof', - 'ext': 'mp4', - 'title': 'Hachis parmentier met witloof', - 'description': 'md5:9960478392d87f63567b5b117688cdc5', - 'thumbnail': r're:^https?://.*\.jpg$', - 'duration': 283.02, - }, - 'expected_warnings': ['is not a supported codec'], - } - - def _real_extract(self, url): - display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) - - title = strip_or_none(get_element_by_class( - 'dish-metadata__title', webpage - ) or self._html_search_meta( - 'twitter:title', webpage)) - - description = clean_html(get_element_by_class( - 'dish-description', webpage) - ) or self._html_search_meta( - ('description', 'twitter:description', 'og:description'), - webpage) - - video_id = self._html_search_regex( - r'data-url=(["\'])(?P(?:(?!\1).)+)\1', webpage, 'video id', - group='id') - - return { - '_type': 'url_transparent', - 'url': 'https://mediazone.vrt.be/api/v1/dako/assets/%s' % video_id, - 'ie_key': CanvasIE.ie_key(), - 'id': video_id, - 'display_id': display_id, - 'title': title, - 'description': description, - } diff --git a/yt_dlp/extractor/ketnet.py b/yt_dlp/extractor/ketnet.py deleted file mode 100644 index ab6276727..000000000 --- a/yt_dlp/extractor/ketnet.py +++ /dev/null @@ -1,70 +0,0 @@ -from .canvas import CanvasIE -from .common import InfoExtractor -from ..compat import compat_urllib_parse_unquote -from ..utils import ( - int_or_none, - parse_iso8601, -) - - -class KetnetIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?ketnet\.be/(?P(?:[^/]+/)*[^/?#&]+)' - _TESTS = [{ - 'url': 'https://www.ketnet.be/kijken/n/nachtwacht/3/nachtwacht-s3a1-de-greystook', - 'md5': '37b2b7bb9b3dcaa05b67058dc3a714a9', - 'info_dict': { - 'id': 'pbs-pub-aef8b526-115e-4006-aa24-e59ff6c6ef6f$vid-ddb815bf-c8e7-467b-8879-6bad7a32cebd', - 'ext': 'mp4', - 'title': 'Nachtwacht - Reeks 3: Aflevering 1', - 'description': 'De Nachtwacht krijgt te maken met een parasiet', - 'thumbnail': r're:^https?://.*\.jpg$', - 'duration': 1468.02, - 'timestamp': 1609225200, - 'upload_date': '20201229', - 'series': 'Nachtwacht', - 'season': 'Reeks 3', - 'episode': 'De Greystook', - 'episode_number': 1, - }, - 'expected_warnings': ['is not a supported codec', 'Unknown MIME type'], - }, { - 'url': 'https://www.ketnet.be/themas/karrewiet/jaaroverzicht-20200/karrewiet-het-jaar-van-black-mamba', - 'only_matching': True, - }] - - def _real_extract(self, url): - display_id = self._match_id(url) - - video = self._download_json( - 'https://senior-bff.ketnet.be/graphql', display_id, query={ - 'query': '''{ - video(id: "content/ketnet/nl/%s.model.json") { - description - episodeNr - imageUrl - mediaReference - programTitle - publicationDate - seasonTitle - subtitleVideodetail - titleVideodetail - } -}''' % display_id, - })['data']['video'] - - mz_id = compat_urllib_parse_unquote(video['mediaReference']) - - return { - '_type': 'url_transparent', - 'id': mz_id, - 'title': video['titleVideodetail'], - 'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/' + mz_id, - 'thumbnail': video.get('imageUrl'), - 'description': video.get('description'), - 'timestamp': parse_iso8601(video.get('publicationDate')), - 'series': video.get('programTitle'), - 'season': video.get('seasonTitle'), - 'episode': video.get('subtitleVideodetail'), - 'episode_number': int_or_none(video.get('episodeNr')), - 'ie_key': CanvasIE.ie_key(), - } diff --git a/yt_dlp/extractor/vrt.py b/yt_dlp/extractor/vrt.py index 26f48bf67..bacd3df29 100644 --- a/yt_dlp/extractor/vrt.py +++ b/yt_dlp/extractor/vrt.py @@ -1,45 +1,137 @@ -from .common import InfoExtractor +import functools +import json +import time +import urllib.error +import urllib.parse + +from .gigya import GigyaBaseIE from ..utils import ( + ExtractorError, + clean_html, extract_attributes, float_or_none, get_element_by_class, + get_element_html_by_class, + int_or_none, + join_nonempty, + jwt_encode_hs256, + make_archive_id, + parse_age_limit, + parse_iso8601, + str_or_none, strip_or_none, - unified_timestamp, + traverse_obj, + url_or_none, + urlencode_postdata, ) -class VRTIE(InfoExtractor): +class VRTBaseIE(GigyaBaseIE): + _GEO_BYPASS = False + _PLAYER_INFO = { + 'platform': 'desktop', + 'app': { + 'type': 'browser', + 'name': 'Chrome', + }, + 'device': 'undefined (undefined)', + 'os': { + 'name': 'Windows', + 'version': 'x86_64' + }, + 'player': { + 'name': 'VRT web player', + 'version': '2.7.4-prod-2023-04-19T06:05:45' + } + } + # From https://player.vrt.be/vrtnws/js/main.js & https://player.vrt.be/ketnet/js/main.fd1de01a40a1e3d842ea.js + _JWT_KEY_ID = '0-0Fp51UZykfaiCJrfTE3+oMI8zvDteYfPtR+2n1R+z8w=' + _JWT_SIGNING_KEY = '2a9251d782700769fb856da5725daf38661874ca6f80ae7dc2b05ec1a81a24ae' + + def _extract_formats_and_subtitles(self, data, video_id): + if traverse_obj(data, 'drm'): + self.report_drm(video_id) + + formats, subtitles = [], {} + for target in traverse_obj(data, ('targetUrls', lambda _, v: url_or_none(v['url']) and v['type'])): + format_type = target['type'].upper() + format_url = target['url'] + if format_type in ('HLS', 'HLS_AES'): + fmts, subs = self._extract_m3u8_formats_and_subtitles( + format_url, video_id, 'mp4', m3u8_id=format_type, fatal=False) + formats.extend(fmts) + self._merge_subtitles(subs, target=subtitles) + elif format_type == 'HDS': + formats.extend(self._extract_f4m_formats( + format_url, video_id, f4m_id=format_type, fatal=False)) + elif format_type == 'MPEG_DASH': + fmts, subs = self._extract_mpd_formats_and_subtitles( + format_url, video_id, mpd_id=format_type, fatal=False) + formats.extend(fmts) + self._merge_subtitles(subs, target=subtitles) + elif format_type == 'HSS': + fmts, subs = self._extract_ism_formats_and_subtitles( + format_url, video_id, ism_id='mss', fatal=False) + formats.extend(fmts) + self._merge_subtitles(subs, target=subtitles) + else: + formats.append({ + 'format_id': format_type, + 'url': format_url, + }) + + for sub in traverse_obj(data, ('subtitleUrls', lambda _, v: v['url'] and v['type'] == 'CLOSED')): + subtitles.setdefault('nl', []).append({'url': sub['url']}) + + return formats, subtitles + + def _call_api(self, video_id, client='null', id_token=None, version='v2'): + player_info = {'exp': (round(time.time(), 3) + 900), **self._PLAYER_INFO} + player_token = self._download_json( + 'https://media-services-public.vrt.be/vualto-video-aggregator-web/rest/external/v2/tokens', + video_id, 'Downloading player token', headers={ + **self.geo_verification_headers(), + 'Content-Type': 'application/json', + }, data=json.dumps({ + 'identityToken': id_token or {}, + 'playerInfo': jwt_encode_hs256(player_info, self._JWT_SIGNING_KEY, headers={ + 'kid': self._JWT_KEY_ID + }).decode() + }, separators=(',', ':')).encode())['vrtPlayerToken'] + + return self._download_json( + f'https://media-services-public.vrt.be/media-aggregator/{version}/media-items/{video_id}', + video_id, 'Downloading API JSON', query={ + 'vrtPlayerToken': player_token, + 'client': client, + }, expected_status=400) + + +class VRTIE(VRTBaseIE): IE_DESC = 'VRT NWS, Flanders News, Flandern Info and Sporza' _VALID_URL = r'https?://(?:www\.)?(?Pvrt\.be/vrtnws|sporza\.be)/[a-z]{2}/\d{4}/\d{2}/\d{2}/(?P[^/?&#]+)' _TESTS = [{ 'url': 'https://www.vrt.be/vrtnws/nl/2019/05/15/beelden-van-binnenkant-notre-dame-een-maand-na-de-brand/', - 'md5': 'e1663accf5cf13f375f3cd0d10476669', 'info_dict': { 'id': 'pbs-pub-7855fc7b-1448-49bc-b073-316cb60caa71$vid-2ca50305-c38a-4762-9890-65cbd098b7bd', 'ext': 'mp4', 'title': 'Beelden van binnenkant Notre-Dame, één maand na de brand', - 'description': 'Op maandagavond 15 april ging een deel van het dakgebinte van de Parijse kathedraal in vlammen op.', - 'timestamp': 1557924660, - 'upload_date': '20190515', + 'description': 'md5:6fd85f999b2d1841aa5568f4bf02c3ff', 'duration': 31.2, + 'thumbnail': 'https://images.vrt.be/orig/2019/05/15/2d914d61-7710-11e9-abcc-02b7b76bf47f.jpg', }, + 'params': {'skip_download': 'm3u8'}, }, { 'url': 'https://sporza.be/nl/2019/05/15/de-belgian-cats-zijn-klaar-voor-het-ek/', - 'md5': '910bba927566e9ab992278f647eb4b75', 'info_dict': { 'id': 'pbs-pub-f2c86a46-8138-413a-a4b9-a0015a16ce2c$vid-1f112b31-e58e-4379-908d-aca6d80f8818', 'ext': 'mp4', - 'title': 'De Belgian Cats zijn klaar voor het EK mét Ann Wauters', - 'timestamp': 1557923760, - 'upload_date': '20190515', + 'title': 'De Belgian Cats zijn klaar voor het EK', + 'description': 'Video: De Belgian Cats zijn klaar voor het EK mét Ann Wauters | basketbal, sport in het journaal', 'duration': 115.17, + 'thumbnail': 'https://images.vrt.be/orig/2019/05/15/11c0dba3-770e-11e9-abcc-02b7b76bf47f.jpg', }, - }, { - 'url': 'https://www.vrt.be/vrtnws/en/2019/05/15/belgium_s-eurovision-entry-falls-at-the-first-hurdle/', - 'only_matching': True, - }, { - 'url': 'https://www.vrt.be/vrtnws/de/2019/05/15/aus-fuer-eliott-im-halbfinale-des-eurosongfestivals/', - 'only_matching': True, + 'params': {'skip_download': 'm3u8'}, }] _CLIENT_MAP = { 'vrt.be/vrtnws': 'vrtnieuws', @@ -49,34 +141,285 @@ class VRTIE(InfoExtractor): def _real_extract(self, url): site, display_id = self._match_valid_url(url).groups() webpage = self._download_webpage(url, display_id) - attrs = extract_attributes(self._search_regex( - r'(<[^>]+class="vrtvideo( [^"]*)?"[^>]*>)', webpage, 'vrt video')) + attrs = extract_attributes(get_element_html_by_class('vrtvideo', webpage) or '') - asset_id = attrs['data-video-id'] - publication_id = attrs.get('data-publication-id') + asset_id = attrs.get('data-video-id') or attrs['data-videoid'] + publication_id = traverse_obj(attrs, 'data-publication-id', 'data-publicationid') if publication_id: - asset_id = publication_id + '$' + asset_id - client = attrs.get('data-client-code') or self._CLIENT_MAP[site] + asset_id = f'{publication_id}${asset_id}' + client = traverse_obj(attrs, 'data-client-code', 'data-client') or self._CLIENT_MAP[site] + + data = self._call_api(asset_id, client) + formats, subtitles = self._extract_formats_and_subtitles(data, asset_id) - title = strip_or_none(get_element_by_class( - 'vrt-title', webpage) or self._html_search_meta( - ['og:title', 'twitter:title', 'name'], webpage)) description = self._html_search_meta( ['og:description', 'twitter:description', 'description'], webpage) if description == '…': description = None - timestamp = unified_timestamp(self._html_search_meta( - 'article:published_time', webpage)) return { - '_type': 'url_transparent', 'id': asset_id, - 'display_id': display_id, - 'title': title, + 'formats': formats, + 'subtitles': subtitles, 'description': description, - 'thumbnail': attrs.get('data-posterimage'), - 'timestamp': timestamp, + 'thumbnail': url_or_none(attrs.get('data-posterimage')), 'duration': float_or_none(attrs.get('data-duration'), 1000), - 'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (client, asset_id), - 'ie_key': 'Canvas', + '_old_archive_ids': [make_archive_id('Canvas', asset_id)], + **traverse_obj(data, { + 'title': ('title', {str}), + 'description': ('shortDescription', {str}), + 'duration': ('duration', {functools.partial(float_or_none, scale=1000)}), + 'thumbnail': ('posterImageUrl', {url_or_none}), + }), + } + + +class VrtNUIE(VRTBaseIE): + IE_DESC = 'VRT MAX' + _VALID_URL = r'https?://(?:www\.)?vrt\.be/vrtnu/a-z/(?:[^/]+/){2}(?P[^/?#&]+)' + _TESTS = [{ + # CONTENT_IS_AGE_RESTRICTED + 'url': 'https://www.vrt.be/vrtnu/a-z/de-ideale-wereld/2023-vj/de-ideale-wereld-d20230116/', + 'info_dict': { + 'id': 'pbs-pub-855b00a8-6ce2-4032-ac4f-1fcf3ae78524$vid-d2243aa1-ec46-4e34-a55b-92568459906f', + 'ext': 'mp4', + 'title': 'Tom Waes', + 'description': 'Satirisch actualiteitenmagazine met Ella Leyers. Tom Waes is te gast.', + 'timestamp': 1673905125, + 'release_timestamp': 1673905125, + 'series': 'De ideale wereld', + 'season_id': '1672830988794', + 'episode': 'Aflevering 1', + 'episode_number': 1, + 'episode_id': '1672830988861', + 'display_id': 'de-ideale-wereld-d20230116', + 'channel': 'VRT', + 'duration': 1939.0, + 'thumbnail': 'https://images.vrt.be/orig/2023/01/10/1bb39cb3-9115-11ed-b07d-02b7b76bf47f.jpg', + 'release_date': '20230116', + 'upload_date': '20230116', + 'age_limit': 12, + }, + }, { + 'url': 'https://www.vrt.be/vrtnu/a-z/buurman--wat-doet-u-nu-/6/buurman--wat-doet-u-nu--s6-trailer/', + 'info_dict': { + 'id': 'pbs-pub-ad4050eb-d9e5-48c2-9ec8-b6c355032361$vid-0465537a-34a8-4617-8352-4d8d983b4eee', + 'ext': 'mp4', + 'title': 'Trailer seizoen 6 \'Buurman, wat doet u nu?\'', + 'description': 'md5:197424726c61384b4e5c519f16c0cf02', + 'timestamp': 1652940000, + 'release_timestamp': 1652940000, + 'series': 'Buurman, wat doet u nu?', + 'season': 'Seizoen 6', + 'season_number': 6, + 'season_id': '1652344200907', + 'episode': 'Aflevering 0', + 'episode_number': 0, + 'episode_id': '1652951873524', + 'display_id': 'buurman--wat-doet-u-nu--s6-trailer', + 'channel': 'VRT', + 'duration': 33.13, + 'thumbnail': 'https://images.vrt.be/orig/2022/05/23/3c234d21-da83-11ec-b07d-02b7b76bf47f.jpg', + 'release_date': '20220519', + 'upload_date': '20220519', + }, + 'params': {'skip_download': 'm3u8'}, + }] + _NETRC_MACHINE = 'vrtnu' + _authenticated = False + + def _perform_login(self, username, password): + auth_info = self._gigya_login({ + 'APIKey': '3_0Z2HujMtiWq_pkAjgnS2Md2E11a1AwZjYiBETtwNE-EoEHDINgtnvcAOpNgmrVGy', + 'targetEnv': 'jssdk', + 'loginID': username, + 'password': password, + 'authMode': 'cookie', + }) + + if auth_info.get('errorDetails'): + raise ExtractorError(f'Unable to login. VrtNU said: {auth_info["errorDetails"]}', expected=True) + + # Sometimes authentication fails for no good reason, retry + for retry in self.RetryManager(): + if retry.attempt > 1: + self._sleep(1, None) + try: + self._request_webpage( + 'https://token.vrt.be/vrtnuinitlogin', None, note='Requesting XSRF Token', + errnote='Could not get XSRF Token', query={ + 'provider': 'site', + 'destination': 'https://www.vrt.be/vrtnu/', + }) + self._request_webpage( + 'https://login.vrt.be/perform_login', None, + note='Performing login', errnote='Login failed', + query={'client_id': 'vrtnu-site'}, data=urlencode_postdata({ + 'UID': auth_info['UID'], + 'UIDSignature': auth_info['UIDSignature'], + 'signatureTimestamp': auth_info['signatureTimestamp'], + '_csrf': self._get_cookies('https://login.vrt.be').get('OIDCXSRF').value, + })) + except ExtractorError as e: + if isinstance(e.cause, urllib.error.HTTPError) and e.cause.code == 401: + retry.error = e + continue + raise + + self._authenticated = True + + def _real_extract(self, url): + display_id = self._match_id(url) + parsed_url = urllib.parse.urlparse(url) + details = self._download_json( + f'{parsed_url.scheme}://{parsed_url.netloc}{parsed_url.path.rstrip("/")}.model.json', + display_id, 'Downloading asset JSON', 'Unable to download asset JSON')['details'] + + watch_info = traverse_obj(details, ( + 'actions', lambda _, v: v['type'] == 'watch-episode', {dict}), get_all=False) or {} + video_id = join_nonempty( + 'episodePublicationId', 'episodeVideoId', delim='$', from_dict=watch_info) + if '$' not in video_id: + raise ExtractorError('Unable to extract video ID') + + vrtnutoken = self._download_json( + 'https://token.vrt.be/refreshtoken', video_id, note='Retrieving vrtnutoken', + errnote='Token refresh failed')['vrtnutoken'] if self._authenticated else None + + video_info = self._call_api(video_id, 'vrtnu-web@PROD', vrtnutoken) + + if 'title' not in video_info: + code = video_info.get('code') + if code in ('AUTHENTICATION_REQUIRED', 'CONTENT_IS_AGE_RESTRICTED'): + self.raise_login_required(code, method='password') + elif code in ('INVALID_LOCATION', 'CONTENT_AVAILABLE_ONLY_IN_BE'): + self.raise_geo_restricted(countries=['BE']) + elif code == 'CONTENT_AVAILABLE_ONLY_FOR_BE_RESIDENTS_AND_EXPATS': + if not self._authenticated: + self.raise_login_required(code, method='password') + self.raise_geo_restricted(countries=['BE']) + raise ExtractorError(code, expected=True) + + formats, subtitles = self._extract_formats_and_subtitles(video_info, video_id) + + return { + **traverse_obj(details, { + 'title': 'title', + 'description': ('description', {clean_html}), + 'timestamp': ('data', 'episode', 'onTime', 'raw', {parse_iso8601}), + 'release_timestamp': ('data', 'episode', 'onTime', 'raw', {parse_iso8601}), + 'series': ('data', 'program', 'title'), + 'season': ('data', 'season', 'title', 'value'), + 'season_number': ('data', 'season', 'title', 'raw', {int_or_none}), + 'season_id': ('data', 'season', 'id', {str_or_none}), + 'episode': ('data', 'episode', 'number', 'value', {str_or_none}), + 'episode_number': ('data', 'episode', 'number', 'raw', {int_or_none}), + 'episode_id': ('data', 'episode', 'id', {str_or_none}), + 'age_limit': ('data', 'episode', 'age', 'raw', {parse_age_limit}), + }), + 'id': video_id, + 'display_id': display_id, + 'channel': 'VRT', + 'formats': formats, + 'duration': float_or_none(video_info.get('duration'), 1000), + 'thumbnail': url_or_none(video_info.get('posterImageUrl')), + 'subtitles': subtitles, + '_old_archive_ids': [make_archive_id('Canvas', video_id)], + } + + +class KetnetIE(VRTBaseIE): + _VALID_URL = r'https?://(?:www\.)?ketnet\.be/(?P(?:[^/]+/)*[^/?#&]+)' + _TESTS = [{ + 'url': 'https://www.ketnet.be/kijken/m/meisjes/6/meisjes-s6a5', + 'info_dict': { + 'id': 'pbs-pub-39f8351c-a0a0-43e6-8394-205d597d6162$vid-5e306921-a9aa-4fa9-9f39-5b82c8f1028e', + 'ext': 'mp4', + 'title': 'Meisjes', + 'episode': 'Reeks 6: Week 5', + 'season': 'Reeks 6', + 'series': 'Meisjes', + 'timestamp': 1685251800, + 'upload_date': '20230528', + }, + 'params': {'skip_download': 'm3u8'}, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + video = self._download_json( + 'https://senior-bff.ketnet.be/graphql', display_id, query={ + 'query': '''{ + video(id: "content/ketnet/nl/%s.model.json") { + description + episodeNr + imageUrl + mediaReference + programTitle + publicationDate + seasonTitle + subtitleVideodetail + titleVideodetail + } +}''' % display_id, + })['data']['video'] + + video_id = urllib.parse.unquote(video['mediaReference']) + data = self._call_api(video_id, 'ketnet@PROD', version='v1') + formats, subtitles = self._extract_formats_and_subtitles(data, video_id) + + return { + 'id': video_id, + 'formats': formats, + 'subtitles': subtitles, + '_old_archive_ids': [make_archive_id('Canvas', video_id)], + **traverse_obj(video, { + 'title': ('titleVideodetail', {str}), + 'description': ('description', {str}), + 'thumbnail': ('thumbnail', {url_or_none}), + 'timestamp': ('publicationDate', {parse_iso8601}), + 'series': ('programTitle', {str}), + 'season': ('seasonTitle', {str}), + 'episode': ('subtitleVideodetail', {str}), + 'episode_number': ('episodeNr', {int_or_none}), + }), + } + + +class DagelijkseKostIE(VRTBaseIE): + IE_DESC = 'dagelijksekost.een.be' + _VALID_URL = r'https?://dagelijksekost\.een\.be/gerechten/(?P[^/?#&]+)' + _TESTS = [{ + 'url': 'https://dagelijksekost.een.be/gerechten/hachis-parmentier-met-witloof', + 'info_dict': { + 'id': 'md-ast-27a4d1ff-7d7b-425e-b84f-a4d227f592fa', + 'ext': 'mp4', + 'title': 'Hachis parmentier met witloof', + 'description': 'md5:9960478392d87f63567b5b117688cdc5', + 'display_id': 'hachis-parmentier-met-witloof', + }, + 'params': {'skip_download': 'm3u8'}, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + video_id = self._html_search_regex( + r'data-url=(["\'])(?P(?:(?!\1).)+)\1', webpage, 'video id', group='id') + + data = self._call_api(video_id, 'dako@prod', version='v1') + formats, subtitles = self._extract_formats_and_subtitles(data, video_id) + + return { + 'id': video_id, + 'formats': formats, + 'subtitles': subtitles, + 'display_id': display_id, + 'title': strip_or_none(get_element_by_class( + 'dish-metadata__title', webpage) or self._html_search_meta('twitter:title', webpage)), + 'description': clean_html(get_element_by_class( + 'dish-description', webpage)) or self._html_search_meta( + ['description', 'twitter:description', 'og:description'], webpage), + '_old_archive_ids': [make_archive_id('Canvas', video_id)], }