Server : Apache/2.4.6 (CentOS) OpenSSL/1.0.2k-fips PHP/7.4.20 System : Linux st2.domain.com 3.10.0-1127.10.1.el7.x86_64 #1 SMP Wed Jun 3 14:28:03 UTC 2020 x86_64 User : apache ( 48) PHP Version : 7.4.20 Disable Function : NONE Directory : /usr/lib/python2.7/site-packages/youtube_dl/extractor/ |
# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
class EchoMskIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?echo\.msk\.ru/sounds/(?P<id>\d+)'
_TEST = {
'url': 'http://www.echo.msk.ru/sounds/1464134.html',
'md5': '2e44b3b78daff5b458e4dbc37f191f7c',
'info_dict': {
'id': '1464134',
'ext': 'mp3',
'title': 'Особое мнение - 29 декабря 2014, 19:08',
},
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
audio_url = self._search_regex(
r'<a rel="mp3" href="([^"]+)">', webpage, 'audio URL')
title = self._html_search_regex(
r'<a href="/programs/[^"]+" target="_blank">([^<]+)</a>',
webpage, 'title')
air_date = self._html_search_regex(
r'(?s)<div class="date">(.+?)</div>',
webpage, 'date', fatal=False, default=None)
if air_date:
air_date = re.sub(r'(\s)\1+', r'\1', air_date)
if air_date:
title = '%s - %s' % (title, air_date)
return {
'id': video_id,
'url': audio_url,
'title': title,
}