From: Jaime Marquínez Ferrándiz Date: Tue, 3 Jun 2014 19:56:49 +0000 (+0200) Subject: [ard] Fix format extraction (fixes #3006 and #3032) X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=commitdiff_plain;h=6a3fa81ffba194c9fdcb30b61f72935d63f30a53;p=youtube-dl [ard] Fix format extraction (fixes #3006 and #3032) --- diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py index a87b32b22..c6d22c029 100644 --- a/youtube_dl/extractor/ard.py +++ b/youtube_dl/extractor/ard.py @@ -39,16 +39,18 @@ class ARDIE(InfoExtractor): title = self._html_search_regex( [r'(.*?)', + r'', r'

(.*?)

'], webpage, 'title') description = self._html_search_meta( 'dcterms.abstract', webpage, 'description') thumbnail = self._og_search_thumbnail(webpage) - streams = [ - mo.groupdict() - for mo in re.finditer( - r'mediaCollection\.addMediaStream\((?P\d+), (?P\d+), "(?P[^"]*)", "(?P[^"]*)", "[^"]*"\)', webpage)] + + media_info = self._download_json( + 'http://www.ardmediathek.de/play/media/%s' % video_id, video_id) + # The second element of the _mediaArray contains the standard http urls + streams = media_info['_mediaArray'][1]['_mediaStreamArray'] if not streams: if '"fsk"' in webpage: raise ExtractorError('This video is only available after 20:00') @@ -56,21 +58,12 @@ class ARDIE(InfoExtractor): formats = [] for s in streams: format = { - 'quality': int(s['quality']), + 'quality': s['_quality'], + 'url': s['_stream'], } - if s.get('rtmp_url'): - format['protocol'] = 'rtmp' - format['url'] = s['rtmp_url'] - format['playpath'] = s['video_url'] - else: - format['url'] = s['video_url'] - - quality_name = self._search_regex( - r'[,.]([a-zA-Z0-9_-]+),?\.mp4', format['url'], - 'quality name', default='NA') - format['format_id'] = '%s-%s-%s-%s' % ( - determine_ext(format['url']), quality_name, s['media_type'], - s['quality']) + + format['format_id'] = '%s-%s' % ( + determine_ext(format['url']), format['quality']) formats.append(format)