From 180d2d1a9b2a1aaa04bcf7892c2dfe15b4bb5623 Mon Sep 17 00:00:00 2001 From: InvalidUsernameException Date: Thu, 19 Sep 2024 21:49:49 +0200 Subject: [PATCH] Update outdated API URL --- yt_dlp/extractor/zdf.py | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/yt_dlp/extractor/zdf.py b/yt_dlp/extractor/zdf.py index 345e21f75..ca5856b50 100644 --- a/yt_dlp/extractor/zdf.py +++ b/yt_dlp/extractor/zdf.py @@ -24,6 +24,11 @@ class ZDFBaseIE(InfoExtractor): _GEO_COUNTRIES = ['DE'] _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd', 'fhd', 'uhd') + def _download_mediathekv2_document(self, document_id): + return self._download_json( + f'https://zdf-prod-futura.zdf.de/mediathekV2/document/{document_id}', + document_id) + def _call_api(self, url, video_id, item, api_token=None, referrer=None): headers = {} if api_token: @@ -319,9 +324,7 @@ def _extract_regular(self, url, player, video_id): return self._extract_entry(player['content'], player, content, video_id) def _extract_mobile(self, video_id): - video = self._download_json( - f'https://zdf-cdn.live.cellular.de/mediathekV2/document/{video_id}', - video_id) + video = self._download_mediathekv2_document(video_id) formats = [] formitaeten = try_get(video, lambda x: x['document']['formitaeten'], list) @@ -450,9 +453,7 @@ def _real_extract(self, url): document_id = self._extract_document_id(webpage) if document_id is not None: - data = self._download_json( - f'https://zdf-prod-futura.zdf.de/mediathekV2/document/{document_id}', - document_id) + data = self._download_mediathekv2_document(document_id) for cluster in data['cluster']: for teaser in cluster['teaser']: