From: Sergey M․ Date: Thu, 19 Jul 2018 19:39:20 +0000 (+0700) Subject: [vrtnu] Relax title extraction and extract JSON-LD (closes #17018) X-Git-Url: http://git.oshgnacknak.de/?a=commitdiff_plain;h=fd62b36680ff7d7bea789ac0031a33fc2d9270ad;p=youtube-dl [vrtnu] Relax title extraction and extract JSON-LD (closes #17018) --- diff --git a/youtube_dl/extractor/canvas.py b/youtube_dl/extractor/canvas.py index 8ac62c1a6..174fd9e2b 100644 --- a/youtube_dl/extractor/canvas.py +++ b/youtube_dl/extractor/canvas.py @@ -11,6 +11,7 @@ from ..utils import ( strip_or_none, float_or_none, int_or_none, + merge_dicts, parse_iso8601, ) @@ -248,9 +249,13 @@ class VrtNUIE(GigyaBaseIE): webpage, urlh = self._download_webpage_handle(url, display_id) - title = self._html_search_regex( + info = self._search_json_ld(webpage, display_id, default={}) + + # title is optional here since it may be extracted by extractor + # that is delegated from here + title = strip_or_none(self._html_search_regex( r'(?ms)

(.+?)

', - webpage, 'title').strip() + webpage, 'title', default=None)) description = self._html_search_regex( r'(?ms)
(.+?)
', @@ -295,7 +300,7 @@ class VrtNUIE(GigyaBaseIE): # the first one video_id = list(video.values())[0].get('videoid') - return { + return merge_dicts(info, { '_type': 'url_transparent', 'url': 'https://mediazone.vrt.be/api/v1/vrtvideo/assets/%s' % video_id, 'ie_key': CanvasIE.ie_key(), @@ -307,4 +312,4 @@ class VrtNUIE(GigyaBaseIE): 'season_number': season_number, 'episode_number': episode_number, 'release_date': release_date, - } + })