From a8f1d167f6741485b7cbec2cb355315c3774d5bc Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Wed, 23 Dec 2015 17:55:58 +0100 Subject: [PATCH] [arte] Prefer json URLs that contain the video id from the 'vid' parameter in the URL (fixes #7920) --- youtube_dl/extractor/arte.py | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index 2a00da3ee0..10301a8eac 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -68,9 +68,13 @@ class ArteTVPlus7IE(InfoExtractor): def _extract_url_info(cls, url): mobj = re.match(cls._VALID_URL, url) lang = mobj.group('lang') - # This is not a real id, it can be for example AJT for the news - # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal - video_id = mobj.group('id') + query = compat_parse_qs(compat_urllib_parse_urlparse(url).query) + if 'vid' in query: + video_id = query['vid'][0] + else: + # This is not a real id, it can be for example AJT for the news + # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal + video_id = mobj.group('id') return video_id, lang def _real_extract(self, url): @@ -79,9 +83,15 @@ class ArteTVPlus7IE(InfoExtractor): return self._extract_from_webpage(webpage, video_id, lang) def _extract_from_webpage(self, webpage, video_id, lang): + patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']') + ids = (video_id, '') + # some pages contain multiple videos (like + # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D), + # so we first try to look for json URLs that contain the video id from + # the 'vid' parameter. + patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates] json_url = self._html_search_regex( - [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'], - webpage, 'json vp url', default=None) + patterns, webpage, 'json vp url', default=None) if not json_url: iframe_url = self._html_search_regex( r']+src=(["\'])(?P.+\bjson_url=.+?)\1',