|
|
|
@ -1,9 +1,9 @@
|
|
|
|
|
# coding: utf-8
|
|
|
|
|
from __future__ import unicode_literals
|
|
|
|
|
|
|
|
|
|
from .common import InfoExtractor
|
|
|
|
|
import time
|
|
|
|
|
|
|
|
|
|
from .common import InfoExtractor
|
|
|
|
|
from ..utils import encode_dict
|
|
|
|
|
from ..compat import (
|
|
|
|
|
compat_urllib_request,
|
|
|
|
@ -60,10 +60,10 @@ class VidtoIE(InfoExtractor):
|
|
|
|
|
time.sleep(7)
|
|
|
|
|
post_result = self._download_webpage(
|
|
|
|
|
req, video_id,
|
|
|
|
|
note='Proceed to video...', errnote='unable to proceed', fatal=True)
|
|
|
|
|
note='Proceed to video...', errnote='unable to proceed')
|
|
|
|
|
|
|
|
|
|
file_link_regex = r'file_link\s*=\s*\'(https?:\/\/[0-9a-zA-z.\/\-_]+)'
|
|
|
|
|
file_link = self._search_regex(file_link_regex, post_result, 'file_link', fatal=True)
|
|
|
|
|
file_link = self._search_regex(file_link_regex, post_result, 'file_link')
|
|
|
|
|
|
|
|
|
|
return {
|
|
|
|
|
'id': video_id,
|
|
|
|
|