[extractor/lumni] Add extractor (#6302)

Authored by: carusocr
Closes #6202
This commit is contained in:
Chris Caruso 2023-03-04 11:22:11 -08:00 committed by GitHub
parent ed4cc4ea79
commit 1f8489cccb
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 27 additions and 0 deletions

View file

@ -962,6 +962,9 @@ from .lrt import (
LRTVODIE,
LRTStreamIE
)
from .lumni import (
LumniIE
)
from .lynda import (
LyndaIE,
LyndaCourseIE

24
yt_dlp/extractor/lumni.py Normal file
View file

@ -0,0 +1,24 @@
from .common import InfoExtractor
from .francetv import FranceTVIE
class LumniIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?lumni\.fr/video/(?P<id>[\w-]+)'
_TESTS = [{
'url': 'https://www.lumni.fr/video/l-homme-et-son-environnement-dans-la-revolution-industrielle',
'md5': '960e8240c4f2c7a20854503a71e52f5e',
'info_dict': {
'id': 'd2b9a4e5-a526-495b-866c-ab72737e3645',
'ext': 'mp4',
'title': "L'homme et son environnement dans la révolution industrielle - L'ère de l'homme",
'thumbnail': 'https://assets.webservices.francetelevisions.fr/v1/assets/images/a7/17/9f/a7179f5f-63a5-4e11-8d4d-012ab942d905.jpg',
'duration': 230,
}
}]
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'<div[^>]+data-factoryid\s*=\s*["\']([^"\']+)', webpage, 'video id')
return self.url_result(f'francetv:{video_id}', FranceTVIE, video_id)