gitlab-com-dstftw/youtube-dl
1
0
Fork 0
mirror of https://gitlab.com/dstftw/youtube-dl.git synced 2020-11-16 09:42:26 +00:00
Code Issues Releases Wiki Activity

[mdr] Simplify

Browse source
This commit is contained in:
Philipp Hagemeister 2013-12-16 05:44:34 +01:00
parent b2ae513586
commit 09dacfa57f
1 changed files with 47 additions and 46 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

93
youtube_dl/extractor/mdr.py
View file

@ -5,35 +5,28 @@ from ..utils import (
ExtractorError, ExtractorError,
) )
class MDRIE(InfoExtractor): 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)/mediathek/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)_.*'
_TITLE = r'<h2>(?P<title1>[^<]+)<span>(?P<title2>[^<]+)</span></h2>'
_MEDIA_XML = r'(?P<xmlurl>/mediathek/(.+)/(video|audio)([0-9]+)-avCustom.xml)'
_MEDIA_STREAM_VIDEO = r'<asset>.*<frameWidth>(?P<frameWidth>[0-9]+)</frameWidth>.*<flashMediaServerApplicationURL>(?P<flashMediaServerApplicationURL>[^<]+)</flashMediaServerApplicationURL><flashMediaServerURL>(?P<flashMediaServerURL>[^<]+)</flashMediaServerURL>.*<progressiveDownloadUrl>(?P<progressiveDownloadUrl>[^<]+)</progressiveDownloadUrl></asset>'
_MEDIA_STREAM_AUDIO = r'<asset>.*<mediaType>(?P<mediaType>[A-Z0-9]+)</mediaType><bitrateAudio>(?P<bitrateAudio>[0-9]+)</bitrateAudio>.*<flashMediaServerApplicationURL>(?P<flashMediaServerApplicationURL>[^<]+)</flashMediaServerApplicationURL><flashMediaServerURL>(?P<flashMediaServerURL>[^<]+)</flashMediaServerURL>.*<progressiveDownloadUrl>(?P<progressiveDownloadUrl>[^<]+)</progressiveDownloadUrl></asset>'
_TESTS = [{ _TESTS = [{
u'url': u'http://www.mdr.de/mediathek/themen/nachrichten/video165624_zc-c5c7de76_zs-3795826d.html', u'url': u'http://www.mdr.de/mediathek/themen/nachrichten/video165624_zc-c5c7de76_zs-3795826d.html',
u'file': u'165624.mp4', u'file': u'165624.mp4',
u'md5': u'95165945756198b8fa2dea10f0b04614', u'md5': u'ae785f36ecbf2f19b42edf1bc9c85815',
u'info_dict': { u'info_dict': {
u"title": u"MDR aktuell Eins30 09.12.2013, 22:48 Uhr" u"title": u"MDR aktuell Eins30 09.12.2013, 22:48 Uhr"
}, },
#u'skip': u'Requires rtmpdump' # rtmp is optional
}, },
{ {
u'url': u' http://www.mdr.de/mediathek/radio/mdr1-radio-sachsen/audio718370_zc-67b21197_zs-1b9b2483.html', u'url': u'http://www.mdr.de/mediathek/radio/mdr1-radio-sachsen/audio718370_zc-67b21197_zs-1b9b2483.html',
u'file': u'718370.mp4', u'file': u'718370.mp3',
u'md5': u'4a5b1fbb5519fb0d929c384b6ff7cb8b', u'md5': u'a9d21345a234c7b45dee612f290fd8d7',
u'info_dict': { u'info_dict': {
u"title": u"MDR 1 RADIO SACHSEN 10.12.2013, 05:00 Uhr" u"title": u"MDR 1 RADIO SACHSEN 10.12.2013, 05:00 Uhr"
}, },
#u'skip': u'Requires rtmpdump' # rtmp is optional
}] }]
def _real_extract(self, url): def _real_extract(self, url):
# determine video id from url
m = re.match(self._VALID_URL, url) m = re.match(self._VALID_URL, url)
video_id = m.group('video_id') video_id = m.group('video_id')
domain = m.group('domain') domain = m.group('domain')
@ -41,38 +34,46 @@ class MDRIE(InfoExtractor):
# determine title and media streams from webpage # determine title and media streams from webpage
html = self._download_webpage(url, video_id) html = self._download_webpage(url, video_id)
t = re.search(self._TITLE, html)
if not t:
raise ExtractorError(u'no title found')
title = t.group('title1') + t.group('title2')
m = re.search(self._MEDIA_XML, html)
if not m:
raise ExtractorError(u'no xml found')
xmlurl = m.group('xmlurl')
xml = self._download_webpage(domain+xmlurl, video_id, 'download XML').replace('\n','').replace('\r','').replace('<asset>','\n<asset>').replace('</asset>','</asset>\n')
if(mediatype == "video"):
streams = [mo.groupdict() for mo in re.finditer(self._MEDIA_STREAM_VIDEO, xml)]
if not streams:
raise ExtractorError(u'no media found')
# choose default media type and highest quality for now
stream = max([s for s in streams if s["progressiveDownloadUrl"].startswith("http://") ],
key=lambda s: int(s["frameWidth"]))
else:
streams = [mo.groupdict() for mo in re.finditer(self._MEDIA_STREAM_AUDIO, xml)]
if not streams:
raise ExtractorError(u'no media found')
# choose default media type (MP4) and highest quality for now
stream = max([s for s in streams if s["progressiveDownloadUrl"].startswith("http://") and s["mediaType"] == "MP4" ],
key=lambda s: int(s["bitrateAudio"]))
# there's two possibilities: RTMP stream or HTTP download title = self._html_search_regex(r'<h2>(.*?)</h2>', html, u'title')
info = {'id': video_id, 'title': title, 'ext': 'mp4'} xmlurl = self._search_regex(
if not stream["progressiveDownloadUrl"]: r'(/mediathek/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, u'XML URL')
self.to_screen(u'RTMP download detected')
assert stream['flashMediaServerURL'].startswith('mp4:') doc = self._download_xml(domain + xmlurl, video_id)
info["url"] = stream["flashMediaServerApplicationURL"] formats = []
info["play_path"] = stream['flashMediaServerURL'] for a in doc.findall('./assets/asset'):
else: url_el = a.find('.//progressiveDownloadUrl')
assert stream["progressiveDownloadUrl"].endswith('.mp4') if url_el is None:
info["url"] = stream["progressiveDownloadUrl"] continue
return [info] 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': u'%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': u'%s-%d' % (media_type, vbr),
})
formats.append(format)
formats.sort(key=lambda f: (f.get('vbr'), f['abr']))
if not formats:
raise ValueError('Could not find any valid formats')
return {
'id': video_id,
'title': title,
'formats': formats,
}