yt-dlp/yt_dlp/extractor/radiofrance.py

105 lines
4.7 KiB
Python
Raw Normal View History

import re
from .common import InfoExtractor
2022-06-10 19:03:54 +00:00
from ..utils import parse_duration, unified_strdate
class RadioFranceIE(InfoExtractor):
_VALID_URL = r'^https?://maison\.radiofrance\.fr/radiovisions/(?P<id>[^?#]+)'
2014-03-23 16:43:33 +00:00
IE_NAME = 'radiofrance'
_TEST = {
2014-03-23 16:43:33 +00:00
'url': 'http://maison.radiofrance.fr/radiovisions/one-one',
'md5': 'bdbb28ace95ed0e04faab32ba3160daf',
'info_dict': {
'id': 'one-one',
'ext': 'ogg',
2016-02-14 09:37:17 +00:00
'title': 'One to one',
'description': "Plutôt que d'imaginer la radio de demain comme technologie ou comme création de contenu, je veux montrer que quelles que soient ses évolutions, j'ai l'intime conviction que la radio continuera d'être un grand média de proximité pour les auditeurs.",
'uploader': 'Thomas Hercouët',
},
}
def _real_extract(self, url):
m = self._match_valid_url(url)
video_id = m.group('id')
webpage = self._download_webpage(url, video_id)
2014-03-23 16:43:33 +00:00
title = self._html_search_regex(r'<h1>(.*?)</h1>', webpage, 'title')
description = self._html_search_regex(
r'<div class="bloc_page_wrapper"><div class="text">(.*?)</div>',
2014-03-23 16:43:33 +00:00
webpage, 'description', fatal=False)
uploader = self._html_search_regex(
r'<div class="credit">&nbsp;&nbsp;&copy;&nbsp;(.*?)</div>',
2014-03-23 16:43:33 +00:00
webpage, 'uploader', fatal=False)
formats_str = self._html_search_regex(
r'class="jp-jplayer[^"]*" data-source="([^"]+)">',
2014-03-23 16:43:33 +00:00
webpage, 'audio URLs')
formats = [
{
2013-12-17 11:35:16 +00:00
'format_id': fm[0],
'url': fm[1],
'vcodec': 'none',
'quality': i,
}
2014-03-23 16:43:33 +00:00
for i, fm in
enumerate(re.findall(r"([a-z0-9]+)\s*:\s*'([^']+)'", formats_str))
]
return {
'id': video_id,
'title': title,
'formats': formats,
'description': description,
'uploader': uploader,
}
2022-06-10 19:03:54 +00:00
class FranceCultureIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?radiofrance\.fr/(?:franceculture|fip|francemusique|mouv|franceinter)/podcasts/(?:[^?#]+/)?(?P<display_id>[^?#]+)-(?P<id>\d+)($|[?#])'
2022-06-10 19:03:54 +00:00
_TESTS = [
{
'url': 'https://www.radiofrance.fr/franceculture/podcasts/science-en-questions/la-physique-d-einstein-aiderait-elle-a-comprendre-le-cerveau-8440487',
'info_dict': {
'id': '8440487',
'display_id': 'la-physique-d-einstein-aiderait-elle-a-comprendre-le-cerveau',
'ext': 'mp3',
'title': 'La physique dEinstein aiderait-elle à comprendre le cerveau ?',
'description': 'Existerait-il un pont conceptuel entre la physique de lespace-temps et les neurosciences ?',
'thumbnail': 'https://cdn.radiofrance.fr/s3/cruiser-production/2022/05/d184e7a3-4827-4494-bf94-04ed7b120db4/1200x630_gettyimages-200171095-001.jpg',
'upload_date': '20220514',
'duration': 2750,
},
},
{
'url': 'https://www.radiofrance.fr/franceinter/podcasts/la-rafle-du-vel-d-hiv-une-affaire-d-etat/les-racines-du-crime-episode-1-3715507',
'only_matching': True,
}
2022-06-10 19:03:54 +00:00
]
def _real_extract(self, url):
video_id, display_id = self._match_valid_url(url).group('id', 'display_id')
webpage = self._download_webpage(url, display_id)
# _search_json_ld doesn't correctly handle this. See https://github.com/yt-dlp/yt-dlp/pull/3874#discussion_r891903846
video_data = self._search_json('', webpage, 'audio data', display_id, contains_pattern=r'{\s*"@type"\s*:\s*"AudioObject".+}')
2022-06-10 19:03:54 +00:00
return {
'id': video_id,
'display_id': display_id,
'url': video_data['contentUrl'],
'ext': video_data.get('encodingFormat'),
'vcodec': 'none' if video_data.get('encodingFormat') == 'mp3' else None,
'duration': parse_duration(video_data.get('duration')),
'title': self._html_search_regex(r'(?s)<h1[^>]*itemprop="[^"]*name[^"]*"[^>]*>(.+?)</h1>',
webpage, 'title', default=self._og_search_title(webpage)),
'description': self._html_search_regex(
r'(?s)<meta name="description"\s*content="([^"]+)', webpage, 'description', default=None),
'thumbnail': self._og_search_thumbnail(webpage),
'uploader': self._html_search_regex(
r'(?s)<span class="author">(.*?)</span>', webpage, 'uploader', default=None),
'upload_date': unified_strdate(self._search_regex(
r'"datePublished"\s*:\s*"([^"]+)', webpage, 'timestamp', fatal=False))
}