mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-08-22 11:21:32 +00:00
Code Issues Projects Releases Wiki Activity

[ie/NRKTVEpisode] Fix extractor (#14065)

Browse Source 
Closes #14054
Authored by: runarmod
This commit is contained in:
Runar Saur Modahl 2025-08-20 00:45:55 +02:00 committed by GitHub
parent 35da8df4f8
commit 7540aa1da1
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 17 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
yt_dlp/extractor/nrk.py
View File

@ -446,9 +446,10 @@ class NRKTVIE(InfoExtractor):
class NRKTVEpisodeIE(InfoExtractor): class NRKTVEpisodeIE(InfoExtractor):
_VALID_URL = r'https?://tv\.nrk\.no/serie/(?P<id>[^/]+/sesong/(?P<season_number>\d+)/episode/(?P<episode_number>\d+))' _VALID_URL = r'https?://tv\.nrk\.no/serie/(?P<id>[^/?#]+/sesong/(?P<season_number>\d+)/episode/(?P<episode_number>\d+))'
_TESTS = [{ _TESTS = [{
'url': 'https://tv.nrk.no/serie/hellums-kro/sesong/1/episode/2', 'url': 'https://tv.nrk.no/serie/hellums-kro/sesong/1/episode/2',
'add_ie': [NRKIE.ie_key()],
'info_dict': { 'info_dict': {
'id': 'MUHH36005220', 'id': 'MUHH36005220',
'ext': 'mp4', 'ext': 'mp4',
@ -485,26 +486,23 @@ class NRKTVEpisodeIE(InfoExtractor):
}] }]
def _real_extract(self, url): def _real_extract(self, url):
display_id, season_number, episode_number = self._match_valid_url(url).groups() display_id, season_number, episode_number = self._match_valid_url(url).group(
'id', 'season_number', 'episode_number')
webpage, urlh = self._download_webpage_handle(url, display_id)
webpage = self._download_webpage(url, display_id) if NRKTVIE.suitable(urlh.url):
nrk_id = NRKTVIE._match_id(urlh.url)
else:
nrk_id = self._search_json(
r'<script\b[^>]+\bid="pageData"[^>]*>', webpage,
'page data', display_id)['initialState']['selectedEpisodePrfId']
if not re.fullmatch(NRKTVIE._EPISODE_RE, nrk_id):
raise ExtractorError('Unable to extract NRK ID')
info = self._search_json_ld(webpage, display_id, default={}) return self.url_result(
nrk_id = info.get('@id') or self._html_search_meta( f'nrk:{nrk_id}', NRKIE, nrk_id,
'nrk:program-id', webpage, default=None) or self._search_regex( season_number=int(season_number),
rf'data-program-id=["\']({NRKTVIE._EPISODE_RE})', webpage, episode_number=int(episode_number))
'nrk id')
assert re.match(NRKTVIE._EPISODE_RE, nrk_id)
info.update({
'_type': 'url',
'id': nrk_id,
'url': f'nrk:{nrk_id}',
'ie_key': NRKIE.ie_key(),
'season_number': int(season_number),
'episode_number': int(episode_number),
})
return info
class NRKTVSerieBaseIE(NRKBaseIE): class NRKTVSerieBaseIE(NRKBaseIE):