1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-11-25 03:32:05 +00:00
youtube-dl/youtube_dl/extractor/krasview.py

60 lines
1.9 KiB
Python
Raw Normal View History

# encoding: utf-8
from __future__ import unicode_literals
import json
import re
from .common import InfoExtractor
from ..utils import (
int_or_none,
unescapeHTML,
)
class KrasViewIE(InfoExtractor):
IE_DESC = 'Красвью'
_VALID_URL = r'https?://krasview\.ru/video/(?P<id>\d+)'
_TEST = {
'url': 'http://krasview.ru/video/512228',
'md5': '3b91003cf85fc5db277870c8ebd98eae',
'info_dict': {
'id': '512228',
'ext': 'mp4',
'title': 'Снег, лёд, заносы',
'description': 'Снято в городе Нягань, в Ханты-Мансийском автономном округе.',
'duration': 27,
'thumbnail': 're:^https?://.*\.jpg',
},
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id)
flashvars = json.loads(self._search_regex(
r'flashvars\s*:\s*({.+?})\s*}\);', webpage, 'flashvars'))
video_url = flashvars['url']
title = unescapeHTML(flashvars['title'])
description = unescapeHTML(flashvars.get('subtitle') or self._og_search_description(webpage, default=None))
thumbnail = flashvars['image']
duration = int(flashvars['duration'])
filesize = int(flashvars['size'])
width = int_or_none(self._og_search_property('video:width', webpage, 'video width'))
height = int_or_none(self._og_search_property('video:height', webpage, 'video height'))
return {
'id': video_id,
'url': video_url,
'title': title,
'description': description,
'thumbnail': thumbnail,
'duration': duration,
'filesize': filesize,
'width': width,
'height': height,
}