mirror of
https://github.com/yt-dlp/yt-dlp.git
synced 2024-12-17 16:43:12 +00:00
64 lines
2.1 KiB
Python
64 lines
2.1 KiB
Python
from __future__ import unicode_literals
|
|
|
|
import re
|
|
|
|
from .common import InfoExtractor
|
|
|
|
|
|
class MDRIE(InfoExtractor):
|
|
_VALID_URL = r'^(?P<domain>https?://(?:www\.)?mdr\.de)/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)(?:_|\.html)'
|
|
|
|
# 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)
|
|
video_id = m.group('video_id')
|
|
domain = m.group('domain')
|
|
|
|
# determine title and media streams from webpage
|
|
html = self._download_webpage(url, video_id)
|
|
|
|
title = self._html_search_regex(r'<h[12]>(.*?)</h[12]>', html, 'title')
|
|
xmlurl = self._search_regex(
|
|
r'dataURL:\'(/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, 'XML URL')
|
|
|
|
doc = self._download_xml(domain + xmlurl, video_id)
|
|
formats = []
|
|
for a in doc.findall('./assets/asset'):
|
|
url_el = a.find('.//progressiveDownloadUrl')
|
|
if url_el is None:
|
|
continue
|
|
abr = int(a.find('bitrateAudio').text) // 1000
|
|
media_type = a.find('mediaType').text
|
|
format = {
|
|
'abr': abr,
|
|
'filesize': int(a.find('fileSize').text),
|
|
'url': url_el.text,
|
|
}
|
|
|
|
vbr_el = a.find('bitrateVideo')
|
|
if vbr_el is None:
|
|
format.update({
|
|
'vcodec': 'none',
|
|
'format_id': '%s-%d' % (media_type, abr),
|
|
})
|
|
else:
|
|
vbr = int(vbr_el.text) // 1000
|
|
format.update({
|
|
'vbr': vbr,
|
|
'width': int(a.find('frameWidth').text),
|
|
'height': int(a.find('frameHeight').text),
|
|
'format_id': '%s-%d' % (media_type, vbr),
|
|
})
|
|
formats.append(format)
|
|
self._sort_formats(formats)
|
|
|
|
return {
|
|
'id': video_id,
|
|
'title': title,
|
|
'formats': formats,
|
|
}
|