X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/bddc9fc577d16b1428924bf8a5c37ef1d9295f14..2459bd35438081164499efc6ea2ea8cd89eceb04:/youtube_dl/extractor/spiegel.py diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py index 98a65b7..6955205 100644 --- a/youtube_dl/extractor/spiegel.py +++ b/youtube_dl/extractor/spiegel.py @@ -1,11 +1,26 @@ import re -import xml.etree.ElementTree from .common import InfoExtractor class SpiegelIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P[0-9]+)(?:\.html)?(?:#.*)?$' + _TESTS = [{ + u'url': u'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html', + u'file': u'1259285.mp4', + u'md5': u'2c2754212136f35fb4b19767d242f66e', + u'info_dict': { + u"title": u"Vulkanausbruch in Ecuador: Der \"Feuerschlund\" ist wieder aktiv" + } + }, + { + u'url': u'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html', + u'file': u'1309159.mp4', + u'md5': u'f2cdf638d7aa47654e251e1aee360af1', + u'info_dict': { + u'title': u'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers' + } + }] def _real_extract(self, url): m = re.match(self._VALID_URL, url) @@ -13,25 +28,36 @@ class SpiegelIE(InfoExtractor): webpage = self._download_webpage(url, video_id) - video_title = self._html_search_regex(r'
(.*?)
', - webpage, u'title') + video_title = self._html_search_regex( + r'
(.*?)
', webpage, u'title') xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml' - xml_code = self._download_webpage(xml_url, video_id, - note=u'Downloading XML', errnote=u'Failed to download XML') - - idoc = xml.etree.ElementTree.fromstring(xml_code) - last_type = idoc[-1] - filename = last_type.findall('./filename')[0].text - duration = float(last_type.findall('./duration')[0].text) + idoc = self._download_xml( + xml_url, video_id, + note=u'Downloading XML', errnote=u'Failed to download XML') + + formats = [ + { + 'format_id': n.tag.rpartition('type')[2], + 'url': u'http://video2.spiegel.de/flash/' + n.find('./filename').text, + 'width': int(n.find('./width').text), + 'height': int(n.find('./height').text), + 'abr': int(n.find('./audiobitrate').text), + 'vbr': int(n.find('./videobitrate').text), + 'vcodec': n.find('./codec').text, + 'acodec': 'MP4A', + } + for n in list(idoc) + # Blacklist type 6, it's extremely LQ and not available on the same server + if n.tag.startswith('type') and n.tag != 'type6' + ] + formats.sort(key=lambda f: f['vbr']) + duration = float(idoc[0].findall('./duration')[0].text) - video_url = 'http://video2.spiegel.de/flash/' + filename - video_ext = filename.rpartition('.')[2] info = { 'id': video_id, - 'url': video_url, - 'ext': video_ext, 'title': video_title, 'duration': duration, + 'formats': formats, } - return [info] + return info