v/youtube-dl
1
0
Fork 0
mirror of https://github.com/l1ving/youtube-dl synced 2020-11-18 19:53:54 -08:00
Code Issues Releases Wiki Activity

[regiotv] Fix broken extractor (closes #15348)

Browse Source
This commit is contained in:
Alexander Seiler 2018-12-07 05:02:11 +01:00
parent 8c5879715f
commit a13c91a722
1 changed files with 13 additions and 34 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

47
youtube_dl/extractor/regiotv.py
View File

@ -3,59 +3,38 @@ from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import (
sanitized_Request,
xpath_text,
xpath_with_ns,
)
class RegioTVIE(InfoExtractor): class RegioTVIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?regio-tv\.de/video/(?P<id>[0-9]+)' _VALID_URL = r'https?://(?:www\.)?regio-tv\.de/.*vidid,(?P<id>[0-9]+)'
_TESTS = [{ _TEST = {
'url': 'http://www.regio-tv.de/video/395808.html', 'url': 'https://www.regio-tv.de/video_video,-nikolaus-besucht-sindelfinger-markt-_vidid,151732.html',
'info_dict': { 'info_dict': {
'id': '395808', 'id': '151732',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Wir in Ludwigsburg', 'title': 'Nikolaus besucht Sindelfinger Markt',
'description': 'Mit unseren zuckersüßen Adventskindern, außerdem besuchen wir die Abendsterne!', 'description': 'md5:205f8c1aa68fcc110c54a4e82b4ce43b',
} }
}, { }
'url': 'http://www.regio-tv.de/video/395808',
'only_matching': True,
}]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
key = self._search_regex( key = self._search_regex(
r'key\s*:\s*(["\'])(?P<key>.+?)\1', webpage, 'key', group='key') r'key\s*=\s*(?P<key>[0-9a-f]+)', webpage, 'key', group='key')
title = self._og_search_title(webpage) title = self._og_search_title(webpage)
SOAP_TEMPLATE = '<?xml version="1.0" encoding="utf-8"?><soap:Envelope xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns:soap="http://schemas.xmlsoap.org/soap/envelope/"><soap:Body><{0} xmlns="http://v.telvi.de/"><key xsi:type="xsd:string">{1}</key></{0}></soap:Body></soap:Envelope>' video_url = 'https://vimp.schwaebische.de/getMedia.php?key=%s&type=mp4' % key
request = sanitized_Request(
'http://v.telvi.de/',
SOAP_TEMPLATE.format('GetHTML5VideoData', key).encode('utf-8'))
video_data = self._download_xml(request, video_id, 'Downloading video XML')
NS_MAP = {
'xsi': 'http://www.w3.org/2001/XMLSchema-instance',
'soap': 'http://schemas.xmlsoap.org/soap/envelope/',
}
video_url = xpath_text(
video_data, xpath_with_ns('.//video', NS_MAP), 'video url', fatal=True)
thumbnail = xpath_text(
video_data, xpath_with_ns('.//image', NS_MAP), 'thumbnail')
description = self._og_search_description( description = self._og_search_description(
webpage) or self._html_search_meta('description', webpage) webpage) or self._html_search_meta('description', webpage)
thumbnail = self._og_search_thumbnail(webpage)
if thumbnail and thumbnail.startswith('/'):
thumbnail = 'https://regio-tv.de' + thumbnail
return { return {
'id': video_id, 'id': video_id,
'url': video_url, 'url': video_url,
'ext': 'mp4',
'title': title, 'title': title,
'description': description, 'description': description,
'thumbnail': thumbnail, 'thumbnail': thumbnail,