2014-09-13 09:47:19 -06:00
# coding: utf-8
from __future__ import unicode_literals
import re
from . common import InfoExtractor
from . . utils import url_basename
class BehindKinkIE ( InfoExtractor ) :
2014-09-15 07:09:17 -06:00
_VALID_URL = r ' http://(?:www \ .)?behindkink \ .com/(?P<year>[0-9] {4} )/(?P<month>[0-9] {2} )/(?P<day>[0-9] {2} )/(?P<id>[^/#?_]+) '
2014-09-13 09:47:19 -06:00
_TEST = {
' url ' : ' http://www.behindkink.com/2014/08/14/ab1576-performers-voice-finally-heard-the-bill-is-killed/ ' ,
' md5 ' : ' 41ad01222b8442089a55528fec43ec01 ' ,
' info_dict ' : {
' id ' : ' 36370 ' ,
' ext ' : ' mp4 ' ,
' title ' : ' AB1576 - PERFORMERS VOICE FINALLY HEARD - THE BILL IS KILLED! ' ,
' description ' : ' The adult industry voice was finally heard as Assembly Bill 1576 remained \xa0 in suspense today at the Senate Appropriations Hearing. AB1576 was, among other industry damaging issues, a condom mandate... ' ,
' upload_date ' : ' 20140814 ' ,
' thumbnail ' : ' http://www.behindkink.com/wp-content/uploads/2014/08/36370_AB1576_Win.jpg ' ,
' age_limit ' : 18 ,
}
}
def _real_extract ( self , url ) :
mobj = re . match ( self . _VALID_URL , url )
display_id = mobj . group ( ' id ' )
year = mobj . group ( ' year ' )
month = mobj . group ( ' month ' )
day = mobj . group ( ' day ' )
upload_date = year + month + day
2014-09-15 07:09:17 -06:00
webpage = self . _download_webpage ( url , display_id )
2014-09-13 09:47:19 -06:00
video_url = self . _search_regex (
r " ' file ' : \ s* ' ([^ ' ]+) ' " ,
webpage , ' URL base ' )
video_id = url_basename ( video_url )
video_id = video_id . split ( ' _ ' ) [ 0 ]
return {
' id ' : video_id ,
' url ' : video_url ,
' ext ' : ' mp4 ' ,
' title ' : self . _og_search_title ( webpage ) ,
' display_id ' : display_id ,
' thumbnail ' : self . _og_search_thumbnail ( webpage ) ,
' description ' : self . _og_search_description ( webpage ) ,
' upload_date ' : upload_date ,
' age_limit ' : 18 ,
}