yt-dlc/youtube_dlc/extractor/canalc2.py

74 lines
2.3 KiB
Python
Raw Normal View History

2013-08-22 13:54:23 +02:00
# coding: utf-8
2014-02-22 14:27:09 +01:00
from __future__ import unicode_literals
2013-08-22 13:54:23 +02:00
import re
from .common import InfoExtractor
2015-10-18 15:27:05 +02:00
from ..utils import parse_duration
2013-08-22 13:54:23 +02:00
2013-08-27 10:35:20 +02:00
2013-08-22 13:54:23 +02:00
class Canalc2IE(InfoExtractor):
2013-09-10 12:13:22 +02:00
IE_NAME = 'canalc2.tv'
_VALID_URL = r'https?://(?:(?:www\.)?canalc2\.tv/video/|archives-canalc2\.u-strasbg\.fr/video\.asp\?.*\bidVideo=)(?P<id>\d+)'
2013-08-22 13:54:23 +02:00
_TESTS = [{
2015-09-21 16:52:36 +02:00
'url': 'http://www.canalc2.tv/video/12163',
2014-02-22 14:27:09 +01:00
'md5': '060158428b650f896c542dfbb3d6487f',
'info_dict': {
'id': '12163',
2017-04-27 22:07:42 +02:00
'ext': 'mp4',
2015-10-18 15:27:22 +02:00
'title': 'Terrasses du Numérique',
'duration': 122,
2015-09-21 16:52:36 +02:00
},
}, {
'url': 'http://archives-canalc2.u-strasbg.fr/video.asp?idVideo=11427&voir=oui',
'only_matching': True,
}]
2013-08-22 13:54:23 +02:00
def _real_extract(self, url):
2015-09-21 16:52:36 +02:00
video_id = self._match_id(url)
webpage = self._download_webpage(
'http://www.canalc2.tv/video/%s' % video_id, video_id)
title = self._html_search_regex(
r'(?s)class="[^"]*col_description[^"]*">.*?<h3>(.+?)</h3>',
webpage, 'title')
formats = []
for _, video_url in re.findall(r'file\s*=\s*(["\'])(.+?)\1', webpage):
if video_url.startswith('rtmp://'):
rtmp = re.search(
r'^(?P<url>rtmp://[^/]+/(?P<app>.+/))(?P<play_path>mp4:.+)$', video_url)
formats.append({
'url': rtmp.group('url'),
'format_id': 'rtmp',
'ext': 'flv',
'app': rtmp.group('app'),
'play_path': rtmp.group('play_path'),
'page_url': url,
})
else:
formats.append({
'url': video_url,
'format_id': 'http',
})
if formats:
info = {
'formats': formats,
}
else:
info = self._parse_html5_media_entries(url, webpage, url)[0]
self._sort_formats(info['formats'])
2014-02-22 14:27:09 +01:00
info.update({
2014-02-22 14:27:09 +01:00
'id': video_id,
'title': title,
'duration': parse_duration(self._search_regex(
r'id=["\']video_duree["\'][^>]*>([^<]+)',
webpage, 'duration', fatal=False)),
})
return info