From 481c5c5137baab386c1cc4a4163bae7404a25f61 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Fri, 29 Jul 2016 21:43:17 +0700 Subject: [PATCH] [tv2:article] Fix extraction (Closes #10188) --- youtube_dl/extractor/tv2.py | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) diff --git a/youtube_dl/extractor/tv2.py b/youtube_dl/extractor/tv2.py index 86bb7915db..e4b4ac0e7a 100644 --- a/youtube_dl/extractor/tv2.py +++ b/youtube_dl/extractor/tv2.py @@ -8,6 +8,7 @@ determine_ext, int_or_none, float_or_none, + js_to_json, parse_iso8601, remove_end, ) @@ -105,7 +106,7 @@ class TV2ArticleIE(InfoExtractor): 'url': 'http://www.tv2.no/2015/05/16/nyheter/alesund/krim/pingvin/6930542', 'info_dict': { 'id': '6930542', - 'title': 'Russen hetses etter pingvintyveri – innrømmer å ha åpnet luken på buret', + 'title': 'Russen hetses etter pingvintyveri - innrømmer å ha åpnet luken på buret', 'description': 'md5:339573779d3eea3542ffe12006190954', }, 'playlist_count': 2, @@ -119,9 +120,23 @@ def _real_extract(self, url): webpage = self._download_webpage(url, playlist_id) + # Old embed pattern (looks unused nowadays) + assets = re.findall(r'data-assetid=["\'](\d+)', webpage) + + if not assets: + # New embed pattern + for v in re.findall('TV2ContentboxVideo\(({.+?})\)', webpage): + video = self._parse_json( + v, playlist_id, transform_source=js_to_json, fatal=False) + if not video: + continue + asset = video.get('assetId') + if asset: + assets.append(asset) + entries = [ - self.url_result('http://www.tv2.no/v/%s' % video_id, 'TV2') - for video_id in re.findall(r'data-assetid="(\d+)"', webpage)] + self.url_result('http://www.tv2.no/v/%s' % asset_id, 'TV2') + for asset_id in assets] title = remove_end(self._og_search_title(webpage), ' - TV2.no') description = remove_end(self._og_search_description(webpage), ' - TV2.no')