[ie/TV5Monde] Fix extractor (#9143)

Closes #9118
Authored by: alard, seproDev

Co-authored-by: sepro <4618135+seproDev@users.noreply.github.com>
This commit is contained in:
alard 2024-05-11 23:58:15 +02:00 committed by GitHub
parent 800a43983e
commit 6db96268c5
No known key found for this signature in database
GPG key ID: B5690EEEBB952194

View file

@ -2,85 +2,88 @@ import urllib.parse
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
clean_html,
determine_ext, determine_ext,
extract_attributes, extract_attributes,
get_element_by_class,
get_element_html_by_class,
int_or_none, int_or_none,
parse_duration,
traverse_obj,
try_get,
url_or_none, url_or_none,
) )
from ..utils.traversal import traverse_obj
class TV5MondePlusIE(InfoExtractor): class TV5MondePlusIE(InfoExtractor):
IE_DESC = 'TV5MONDE+' IE_NAME = 'TV5MONDE'
_VALID_URL = r'https?://(?:www\.)?(?:tv5mondeplus|revoir\.tv5monde)\.com/toutes-les-videos/[^/]+/(?P<id>[^/?#]+)' _VALID_URL = r'https?://(?:www\.)?tv5monde\.com/tv/video/(?P<id>[^/?#]+)'
_TESTS = [{ _TESTS = [{
# movie # documentary
'url': 'https://revoir.tv5monde.com/toutes-les-videos/cinema/les-novices', 'url': 'https://www.tv5monde.com/tv/video/65931-baudouin-l-heritage-d-un-roi-baudouin-l-heritage-d-un-roi',
'md5': 'c86f60bf8b75436455b1b205f9745955', 'md5': 'd2a708902d3df230a357c99701aece05',
'info_dict': { 'info_dict': {
'id': 'ZX0ipMyFQq_6D4BA7b', 'id': '3FPa7JMu21_6D4BA7b',
'display_id': 'les-novices', 'display_id': '65931-baudouin-l-heritage-d-un-roi-baudouin-l-heritage-d-un-roi',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Les novices', 'title': "Baudouin, l'héritage d'un roi",
'description': 'md5:2e7c33ba3ad48dabfcc2a956b88bde2b', 'thumbnail': 'https://psi.tv5monde.com/upsilon-images/960x540/6f/baudouin-f49c6b0e.jpg',
'upload_date': '20230821', 'duration': 4842,
'thumbnail': 'https://revoir.tv5monde.com/uploads/media/video_thumbnail/0738/60/01e952b7ccf36b7c6007ec9131588954ab651de9.jpeg', 'upload_date': '20240130',
'duration': 5177, 'timestamp': 1706641242,
'episode': 'Les novices', 'episode': "BAUDOUIN, L'HERITAGE D'UN ROI",
'description': 'md5:78125c74a5cac06d7743a2d09126edad',
'series': "Baudouin, l'héritage d'un roi",
}, },
}, { }, {
# series episode # 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': { 'info_dict': {
'id': 'wJ0eeEPozr_6D4BA7b', 'id': 'obRRZ8m6g9_6D4BA7b',
'display_id': 'opj-les-dents-de-la-terre-2', 'display_id': '52952-toute-la-vie-mardi-23-mars-2021',
'ext': 'mp4', 'ext': 'mp4',
'title': "OPJ - Les dents de la Terre (2)", 'title': 'Toute la vie',
'description': 'md5:288f87fd68d993f814e66e60e5302d9d', 'description': 'md5:a824a2e1dfd94cf45fa379a1fb43ce65',
'upload_date': '20230823', 'thumbnail': 'https://psi.tv5monde.com/media/image/960px/5880553.jpg',
'series': 'OPJ', 'duration': 2526,
'episode': 'Les dents de la Terre (2)', 'upload_date': '20230721',
'duration': 2877, 'timestamp': 1689971646,
'thumbnail': 'https://dl-revoir.tv5monde.com/images/1a/5753448.jpg' 'series': 'Toute la vie',
'episode': 'Mardi 23 mars 2021',
}, },
}, { }, {
# movie # movie
'url': 'https://revoir.tv5monde.com/toutes-les-videos/cinema/ceux-qui-travaillent', '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': '32fa0cde16a4480d1251502a66856d5f', 'md5': '87cefc34e10a6bf4f7823cccd7b36eb2',
'info_dict': { 'info_dict': {
'id': 'dc57a011-ec4b-4648-2a9a-4f03f8352ed3', 'id': 'DOcfvdLKXL_6D4BA7b',
'display_id': 'ceux-qui-travaillent', 'display_id': '8771-ce-fleuve-qui-nous-charrie-ce-fleuve-qui-nous-charrie-p001-ce-fleuve-qui-nous-charrie',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Ceux qui travaillent', 'title': 'Ce fleuve qui nous charrie',
'description': 'md5:570e8bb688036ace873b2d50d24c026d', 'description': 'md5:62ba3f875343c7fc4082bdfbbc1be992',
'upload_date': '20210819', '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 # news
'url': 'https://revoir.tv5monde.com/toutes-les-videos/series-fictions/vestiaires-caro-actrice', 'url': 'https://www.tv5monde.com/tv/video/70402-tv5monde-le-journal-edition-du-08-05-24-11h',
'md5': 'c62977d6d10754a2ecebba70ad370479',
'info_dict': { 'info_dict': {
'id': '9e9d599e-23af-6915-843e-ecbf62e97925', 'id': 'LgQFrOCNsc_6D4BA7b',
'display_id': 'vestiaires-caro-actrice', 'display_id': '70402-tv5monde-le-journal-edition-du-08-05-24-11h',
'ext': 'mp4', 'ext': 'mp4',
'title': "Vestiaires - Caro actrice", 'title': 'TV5MONDE, le journal',
'description': 'md5:db15d2e1976641e08377f942778058ea', 'description': 'md5:777dc209eaa4423b678477c36b0b04a8',
'upload_date': '20210819', 'thumbnail': 'https://psi.tv5monde.com/media/image/960px/6184105.jpg',
'series': "Vestiaires", 'duration': 854,
'episode': 'Caro actrice', '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 _GEO_BYPASS = False
@ -98,7 +101,6 @@ class TV5MondePlusIE(InfoExtractor):
if ">Ce programme n'est malheureusement pas disponible pour votre zone géographique.<" in webpage: if ">Ce programme n'est malheureusement pas disponible pour votre zone géographique.<" in webpage:
self.raise_geo_restricted(countries=['FR']) self.raise_geo_restricted(countries=['FR'])
title = episode = self._html_search_regex(r'<h1>([^<]+)', webpage, 'title')
vpl_data = extract_attributes(self._search_regex( vpl_data = extract_attributes(self._search_regex(
r'(<[^>]+class="video_player_loader"[^>]+>)', r'(<[^>]+class="video_player_loader"[^>]+>)',
webpage, 'video player loader')) webpage, 'video player loader'))
@ -147,26 +149,7 @@ class TV5MondePlusIE(InfoExtractor):
process_video_files(video_files) process_video_files(video_files)
metadata = self._parse_json( metadata = self._parse_json(
vpl_data['data-metadata'], display_id) vpl_data.get('data-metadata') or '{}', display_id, fatal=False)
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)<div[^>]+class=["\']episode-texte[^>]+>(.+?)</div>', webpage,
'description', fatal=False)
series = self._html_search_regex(
r'<p[^>]+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('_', '')
if not video_id: if not video_id:
video_id = self._search_regex( video_id = self._search_regex(
@ -175,16 +158,20 @@ class TV5MondePlusIE(InfoExtractor):
default=display_id) default=display_id)
return { 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, 'id': video_id,
'display_id': display_id, 'display_id': display_id,
'title': title, 'title': clean_html(get_element_by_class('main-title', webpage)),
'description': description, 'description': clean_html(get_element_by_class('text', get_element_html_by_class('ep-summary', webpage) or '')),
'thumbnail': vpl_data.get('data-image'), 'thumbnail': url_or_none(vpl_data.get('data-image')),
'duration': duration,
'upload_date': upload_date,
'formats': formats, 'formats': formats,
'subtitles': self._extract_subtitles(self._parse_json( 'subtitles': self._extract_subtitles(self._parse_json(
traverse_obj(vpl_data, ('data-captions', {str}), default='{}'), display_id, fatal=False)), traverse_obj(vpl_data, ('data-captions', {str}), default='{}'), display_id, fatal=False)),
'series': series,
'episode': episode,
} }