+from __future__ import unicode_literals
+
import re
from .common import InfoExtractor
-from ..utils import (
- ExtractorError,
-)
class MDRIE(InfoExtractor):
- _VALID_URL = r'^(?P<domain>(?:https?://)?(?:www\.)?mdr\.de)/mediathek/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)_.*'
+ _VALID_URL = r'^(?P<domain>https?://(?:www\.)?mdr\.de)/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)(?:_|\.html)'
- _TESTS = [{
- u'url': u'http://www.mdr.de/mediathek/themen/nachrichten/video165624_zc-c5c7de76_zs-3795826d.html',
- u'file': u'165624.mp4',
- u'md5': u'ae785f36ecbf2f19b42edf1bc9c85815',
- u'info_dict': {
- u"title": u"MDR aktuell Eins30 09.12.2013, 22:48 Uhr"
- },
- },
- {
- u'url': u'http://www.mdr.de/mediathek/radio/mdr1-radio-sachsen/audio718370_zc-67b21197_zs-1b9b2483.html',
- u'file': u'718370.mp3',
- u'md5': u'a9d21345a234c7b45dee612f290fd8d7',
- u'info_dict': {
- u"title": u"MDR 1 RADIO SACHSEN 10.12.2013, 05:00 Uhr"
- },
- }]
+ # No tests, MDR regularily deletes its videos
+ _TEST = {
+ 'url': 'http://www.mdr.de/fakt/video189002.html',
+ 'only_matching': True,
+ }
def _real_extract(self, url):
m = re.match(self._VALID_URL, url)
# determine title and media streams from webpage
html = self._download_webpage(url, video_id)
- title = self._html_search_regex(r'<h2>(.*?)</h2>', html, u'title')
+ title = self._html_search_regex(r'<h[12]>(.*?)</h[12]>', html, 'title')
xmlurl = self._search_regex(
- r'(/mediathek/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, u'XML URL')
+ r'dataURL:\'(/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, 'XML URL')
doc = self._download_xml(domain + xmlurl, video_id)
formats = []
if vbr_el is None:
format.update({
'vcodec': 'none',
- 'format_id': u'%s-%d' % (media_type, abr),
+ 'format_id': '%s-%d' % (media_type, abr),
})
else:
vbr = int(vbr_el.text) // 1000
'vbr': vbr,
'width': int(a.find('frameWidth').text),
'height': int(a.find('frameHeight').text),
- 'format_id': u'%s-%d' % (media_type, vbr),
+ 'format_id': '%s-%d' % (media_type, vbr),
})
formats.append(format)
- formats.sort(key=lambda f: (f.get('vbr'), f['abr']))
- if not formats:
- raise ExtractorError(u'Could not find any valid formats')
+ self._sort_formats(formats)
return {
'id': video_id,