From 3ae02d4a626b6966ba9678827c5bfc0e8804a2aa Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sat, 12 Dec 2020 04:48:56 +0700 Subject: [PATCH] [wdr] Extent subtitles extraction and improve overall extraction (closes #22672, closes #22723) --- youtube_dl/extractor/wdr.py | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/youtube_dl/extractor/wdr.py b/youtube_dl/extractor/wdr.py index ba97d983b..2903d189e 100644 --- a/youtube_dl/extractor/wdr.py +++ b/youtube_dl/extractor/wdr.py @@ -17,6 +17,7 @@ from ..utils import ( unified_strdate, update_url_query, urlhandle_detect_ext, + url_or_none, ) @@ -42,16 +43,20 @@ class WDRIE(InfoExtractor): is_live = metadata.get('mediaType') == 'live' tracker_data = metadata['trackerData'] + title = tracker_data['trackerClipTitle'] + media_resource = metadata['mediaResource'] formats = [] # check if the metadata contains a direct URL to a file - for kind, media_resource in media_resource.items(): + for kind, media in media_resource.items(): + if not isinstance(media, dict): + continue if kind not in ('dflt', 'alt'): continue - for tag_name, medium_url in media_resource.items(): + for tag_name, medium_url in media.items(): if tag_name not in ('videoURL', 'audioURL'): continue @@ -88,8 +93,16 @@ class WDRIE(InfoExtractor): 'url': caption_url, 'ext': 'ttml', }] - - title = tracker_data['trackerClipTitle'] + captions_hash = media_resource.get('captionsHash') + if isinstance(captions_hash, dict): + for ext, format_url in captions_hash.items(): + format_url = url_or_none(format_url) + if not format_url: + continue + subtitles.setdefault('de', []).append({ + 'url': format_url, + 'ext': determine_ext(format_url, None) or ext, + }) return { 'id': tracker_data.get('trackerClipId', video_id),