From 6db96268c521e945d42649607db1574f5d92e082 Mon Sep 17 00:00:00 2001 From: alard Date: Sat, 11 May 2024 23:58:15 +0200 Subject: [PATCH] [ie/TV5Monde] Fix extractor (#9143) Closes #9118 Authored by: alard, seproDev Co-authored-by: sepro <4618135+seproDev@users.noreply.github.com> --- yt_dlp/extractor/tv5mondeplus.py | 149 ++++++++++++++----------------- 1 file changed, 68 insertions(+), 81 deletions(-) diff --git a/yt_dlp/extractor/tv5mondeplus.py b/yt_dlp/extractor/tv5mondeplus.py index a445fae85..52ff230f2 100644 --- a/yt_dlp/extractor/tv5mondeplus.py +++ b/yt_dlp/extractor/tv5mondeplus.py @@ -2,85 +2,88 @@ from .common import InfoExtractor from ..utils import ( + clean_html, determine_ext, extract_attributes, + get_element_by_class, + get_element_html_by_class, int_or_none, - parse_duration, - traverse_obj, - try_get, url_or_none, ) +from ..utils.traversal import traverse_obj class TV5MondePlusIE(InfoExtractor): - IE_DESC = 'TV5MONDE+' - _VALID_URL = r'https?://(?:www\.)?(?:tv5mondeplus|revoir\.tv5monde)\.com/toutes-les-videos/[^/]+/(?P[^/?#]+)' + IE_NAME = 'TV5MONDE' + _VALID_URL = r'https?://(?:www\.)?tv5monde\.com/tv/video/(?P[^/?#]+)' _TESTS = [{ - # movie - 'url': 'https://revoir.tv5monde.com/toutes-les-videos/cinema/les-novices', - 'md5': 'c86f60bf8b75436455b1b205f9745955', + # documentary + 'url': 'https://www.tv5monde.com/tv/video/65931-baudouin-l-heritage-d-un-roi-baudouin-l-heritage-d-un-roi', + 'md5': 'd2a708902d3df230a357c99701aece05', 'info_dict': { - 'id': 'ZX0ipMyFQq_6D4BA7b', - 'display_id': 'les-novices', + 'id': '3FPa7JMu21_6D4BA7b', + 'display_id': '65931-baudouin-l-heritage-d-un-roi-baudouin-l-heritage-d-un-roi', 'ext': 'mp4', - 'title': 'Les novices', - 'description': 'md5:2e7c33ba3ad48dabfcc2a956b88bde2b', - 'upload_date': '20230821', - 'thumbnail': 'https://revoir.tv5monde.com/uploads/media/video_thumbnail/0738/60/01e952b7ccf36b7c6007ec9131588954ab651de9.jpeg', - 'duration': 5177, - 'episode': 'Les novices', + 'title': "Baudouin, l'héritage d'un roi", + 'thumbnail': 'https://psi.tv5monde.com/upsilon-images/960x540/6f/baudouin-f49c6b0e.jpg', + 'duration': 4842, + 'upload_date': '20240130', + 'timestamp': 1706641242, + 'episode': "BAUDOUIN, L'HERITAGE D'UN ROI", + 'description': 'md5:78125c74a5cac06d7743a2d09126edad', + 'series': "Baudouin, l'héritage d'un roi", }, }, { # series episode - 'url': 'https://revoir.tv5monde.com/toutes-les-videos/series-fictions/opj-les-dents-de-la-terre-2', + 'url': 'https://www.tv5monde.com/tv/video/52952-toute-la-vie-mardi-23-mars-2021', + 'md5': 'f5e09637cadd55639c05874e22eb56bf', 'info_dict': { - 'id': 'wJ0eeEPozr_6D4BA7b', - 'display_id': 'opj-les-dents-de-la-terre-2', + 'id': 'obRRZ8m6g9_6D4BA7b', + 'display_id': '52952-toute-la-vie-mardi-23-mars-2021', 'ext': 'mp4', - 'title': "OPJ - Les dents de la Terre (2)", - 'description': 'md5:288f87fd68d993f814e66e60e5302d9d', - 'upload_date': '20230823', - 'series': 'OPJ', - 'episode': 'Les dents de la Terre (2)', - 'duration': 2877, - 'thumbnail': 'https://dl-revoir.tv5monde.com/images/1a/5753448.jpg' + 'title': 'Toute la vie', + 'description': 'md5:a824a2e1dfd94cf45fa379a1fb43ce65', + 'thumbnail': 'https://psi.tv5monde.com/media/image/960px/5880553.jpg', + 'duration': 2526, + 'upload_date': '20230721', + 'timestamp': 1689971646, + 'series': 'Toute la vie', + 'episode': 'Mardi 23 mars 2021', }, }, { # movie - 'url': 'https://revoir.tv5monde.com/toutes-les-videos/cinema/ceux-qui-travaillent', - 'md5': '32fa0cde16a4480d1251502a66856d5f', + 'url': 'https://www.tv5monde.com/tv/video/8771-ce-fleuve-qui-nous-charrie-ce-fleuve-qui-nous-charrie-p001-ce-fleuve-qui-nous-charrie', + 'md5': '87cefc34e10a6bf4f7823cccd7b36eb2', 'info_dict': { - 'id': 'dc57a011-ec4b-4648-2a9a-4f03f8352ed3', - 'display_id': 'ceux-qui-travaillent', + 'id': 'DOcfvdLKXL_6D4BA7b', + 'display_id': '8771-ce-fleuve-qui-nous-charrie-ce-fleuve-qui-nous-charrie-p001-ce-fleuve-qui-nous-charrie', 'ext': 'mp4', - 'title': 'Ceux qui travaillent', - 'description': 'md5:570e8bb688036ace873b2d50d24c026d', - 'upload_date': '20210819', + 'title': 'Ce fleuve qui nous charrie', + 'description': 'md5:62ba3f875343c7fc4082bdfbbc1be992', + 'thumbnail': 'https://psi.tv5monde.com/media/image/960px/5476617.jpg', + 'duration': 5300, + 'upload_date': '20210822', + 'timestamp': 1629594105, + 'episode': 'CE FLEUVE QUI NOUS CHARRIE-P001-CE FLEUVE QUI NOUS CHARRIE', + 'series': 'Ce fleuve qui nous charrie', }, - 'skip': 'no longer available', }, { - # series episode - 'url': 'https://revoir.tv5monde.com/toutes-les-videos/series-fictions/vestiaires-caro-actrice', + # news + 'url': 'https://www.tv5monde.com/tv/video/70402-tv5monde-le-journal-edition-du-08-05-24-11h', + 'md5': 'c62977d6d10754a2ecebba70ad370479', 'info_dict': { - 'id': '9e9d599e-23af-6915-843e-ecbf62e97925', - 'display_id': 'vestiaires-caro-actrice', + 'id': 'LgQFrOCNsc_6D4BA7b', + 'display_id': '70402-tv5monde-le-journal-edition-du-08-05-24-11h', 'ext': 'mp4', - 'title': "Vestiaires - Caro actrice", - 'description': 'md5:db15d2e1976641e08377f942778058ea', - 'upload_date': '20210819', - 'series': "Vestiaires", - 'episode': 'Caro actrice', + 'title': 'TV5MONDE, le journal', + 'description': 'md5:777dc209eaa4423b678477c36b0b04a8', + 'thumbnail': 'https://psi.tv5monde.com/media/image/960px/6184105.jpg', + 'duration': 854, + 'upload_date': '20240508', + 'timestamp': 1715159640, + 'series': 'TV5MONDE, le journal', + 'episode': 'EDITION DU 08/05/24 - 11H', }, - 'params': { - 'skip_download': True, - }, - 'skip': 'no longer available', - }, { - 'url': 'https://revoir.tv5monde.com/toutes-les-videos/series-fictions/neuf-jours-en-hiver-neuf-jours-en-hiver', - 'only_matching': True, - }, { - 'url': 'https://revoir.tv5monde.com/toutes-les-videos/info-societe/le-journal-de-la-rts-edition-du-30-01-20-19h30', - 'only_matching': True, }] _GEO_BYPASS = False @@ -98,7 +101,6 @@ def _real_extract(self, url): if ">Ce programme n'est malheureusement pas disponible pour votre zone géographique.<" in webpage: self.raise_geo_restricted(countries=['FR']) - title = episode = self._html_search_regex(r'

([^<]+)', webpage, 'title') vpl_data = extract_attributes(self._search_regex( r'(<[^>]+class="video_player_loader"[^>]+>)', webpage, 'video player loader')) @@ -147,26 +149,7 @@ def process_video_files(v): process_video_files(video_files) metadata = self._parse_json( - vpl_data['data-metadata'], display_id) - duration = (int_or_none(try_get(metadata, lambda x: x['content']['duration'])) - or parse_duration(self._html_search_meta('duration', webpage))) - - description = self._html_search_regex( - r'(?s)]+class=["\']episode-texte[^>]+>(.+?)', webpage, - 'description', fatal=False) - - series = self._html_search_regex( - r']+class=["\']episode-emission[^>]+>([^<]+)', webpage, - 'series', default=None) - - if series and series != title: - title = '%s - %s' % (series, title) - - upload_date = self._search_regex( - r'(?:date_publication|publish_date)["\']\s*:\s*["\'](\d{4}_\d{2}_\d{2})', - webpage, 'upload date', default=None) - if upload_date: - upload_date = upload_date.replace('_', '') + vpl_data.get('data-metadata') or '{}', display_id, fatal=False) if not video_id: video_id = self._search_regex( @@ -175,16 +158,20 @@ def process_video_files(v): default=display_id) return { + **traverse_obj(metadata, ('content', { + 'id': ('id', {str}), + 'title': ('title', {str}), + 'episode': ('title', {str}), + 'series': ('series', {str}), + 'timestamp': ('publishDate_ts', {int_or_none}), + 'duration': ('duration', {int_or_none}), + })), 'id': video_id, 'display_id': display_id, - 'title': title, - 'description': description, - 'thumbnail': vpl_data.get('data-image'), - 'duration': duration, - 'upload_date': upload_date, + 'title': clean_html(get_element_by_class('main-title', webpage)), + 'description': clean_html(get_element_by_class('text', get_element_html_by_class('ep-summary', webpage) or '')), + 'thumbnail': url_or_none(vpl_data.get('data-image')), 'formats': formats, 'subtitles': self._extract_subtitles(self._parse_json( traverse_obj(vpl_data, ('data-captions', {str}), default='{}'), display_id, fatal=False)), - 'series': series, - 'episode': episode, }