From 878d11ec29d63b2c8ca32163f38762d1b12a52c6 Mon Sep 17 00:00:00 2001 From: Philipp Hagemeister Date: Mon, 24 Mar 2014 21:36:26 +0100 Subject: [PATCH] [arte] Add support for multiple formats --- youtube_dl/extractor/arte.py | 38 ++++++++++++++++-------------------- 1 file changed, 17 insertions(+), 21 deletions(-) diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index edf627e3c..49fb64077 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -21,7 +21,7 @@ from ..utils import ( class ArteTvIE(InfoExtractor): - _VALID_URL = r'(?:http://)?videos\.arte\.tv/(?Pfr|de)/.*-(?P.*?)\.html' + _VALID_URL = r'http://videos\.arte\.tv/(?Pfr|de)/.*-(?P.*?)\.html' IE_NAME = 'arte.tv' def _real_extract(self, url): @@ -31,28 +31,24 @@ class ArteTvIE(InfoExtractor): ref_xml_url, video_id, note='Downloading metadata') config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang) config_xml_url = config_node.attrib['ref'] - config_xml = self._download_webpage( + config = self._download_xml( config_xml_url, video_id, note='Downloading configuration') - video_urls = list(re.finditer(r'(?P.*?)', config_xml)) - def _key(m): - quality = m.group('quality') - if quality == 'hd': - return 2 - else: - return 1 - # We pick the best quality - video_urls = sorted(video_urls, key=_key) - video_url = list(video_urls)[-1].group('url') - - title = self._html_search_regex(r'(.*?)', config_xml, 'title') - thumbnail = self._html_search_regex(r'(.*?)', - config_xml, 'thumbnail') - return {'id': video_id, - 'title': title, - 'thumbnail': thumbnail, - 'url': video_url, - 'ext': 'flv', + formats = [{ + 'forma_id': q.attrib['quality'], + 'url': q.text, + 'quality': 2 if q.attrib['quality'] == 'hd' else 1, + } for q in config.findall('.//quality')] + self._sort_formats(formats) + + title = config.find('.//name').text + thumbnail = config.find('.//firstThumbnailUrl').text + return { + 'id': video_id, + 'title': title, + 'thumbnail': thumbnail, + 'url': video_url, + 'ext': 'flv', }