[spiegel] Implement format selection

This commit is contained in:
Philipp Hagemeister 2013-11-16 01:33:12 +01:00
parent 91c7271aab
commit 7150858d49
2 changed files with 44 additions and 16 deletions

View File

@ -786,13 +786,19 @@ class YoutubeDL(object):
return fdict['format_note'] return fdict['format_note']
res = u'' res = u''
if fdict.get('vcodec') is not None: if fdict.get('vcodec') is not None:
res += fdict['vcodec'] res += u'%-5s' % fdict['vcodec']
elif fdict.get('vbr') is not None:
res += u'video'
if fdict.get('vbr') is not None: if fdict.get('vbr') is not None:
res += u'@%4dk' % fdict['vbr'] res += u'@%4dk' % fdict['vbr']
if fdict.get('acodec') is not None: if fdict.get('acodec') is not None:
if res: if res:
res += u', ' res += u', '
res += fdict['acodec'] res += u'%-5s' % fdict['acodec']
elif fdict.get('abr') is not None:
if res:
res += u', '
res += 'audio'
if fdict.get('abr') is not None: if fdict.get('abr') is not None:
res += u'@%3dk' % fdict['abr'] res += u'@%3dk' % fdict['abr']
return res return res

View File

@ -2,18 +2,27 @@ import re
import xml.etree.ElementTree import xml.etree.ElementTree
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import determine_ext
class SpiegelIE(InfoExtractor): class SpiegelIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$' _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$'
_TEST = { _TESTS = [{
u'url': u'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html', u'url': u'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
u'file': u'1259285.mp4', u'file': u'1259285.mp4',
u'md5': u'2c2754212136f35fb4b19767d242f66e', u'md5': u'2c2754212136f35fb4b19767d242f66e',
u'info_dict': { u'info_dict': {
u"title": u"Vulkanausbruch in Ecuador: Der \"Feuerschlund\" ist wieder aktiv" 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): def _real_extract(self, url):
m = re.match(self._VALID_URL, url) m = re.match(self._VALID_URL, url)
@ -21,25 +30,38 @@ class SpiegelIE(InfoExtractor):
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
video_title = self._html_search_regex(r'<div class="module-title">(.*?)</div>', video_title = self._html_search_regex(
webpage, u'title') r'<div class="module-title">(.*?)</div>', webpage, u'title')
xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml' xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml'
xml_code = self._download_webpage(xml_url, video_id, xml_code = self._download_webpage(
xml_url, video_id,
note=u'Downloading XML', errnote=u'Failed to download XML') note=u'Downloading XML', errnote=u'Failed to download XML')
idoc = xml.etree.ElementTree.fromstring(xml_code) 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)
video_url = 'http://video2.spiegel.de/flash/' + filename formats = [
video_ext = filename.rpartition('.')[2] {
'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)
info = { info = {
'id': video_id, 'id': video_id,
'url': video_url,
'ext': video_ext,
'title': video_title, 'title': video_title,
'duration': duration, 'duration': duration,
'formats': formats,
} }
return [info] return info