2014-01-16 18:07:01 -08:00
from __future__ import unicode_literals
2013-06-23 13:31:12 -07:00
import re
from . common import InfoExtractor
2015-04-22 10:24:14 -07:00
from . . compat import compat_urllib_request
2013-06-23 13:31:12 -07:00
from . . utils import (
ExtractorError ,
unescapeHTML ,
)
class FlickrIE ( InfoExtractor ) :
2014-09-24 01:46:33 -07:00
_VALID_URL = r ' https?://(?:www \ .|secure \ .)?flickr \ .com/photos/(?P<uploader_id>[ \ w \ -_@]+)/(?P<id> \ d+).* '
2013-06-27 11:46:46 -07:00
_TEST = {
2014-01-16 18:07:01 -08:00
' url ' : ' http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/ ' ,
' md5 ' : ' 6fdc01adbc89d72fc9c4f15b4a4ba87b ' ,
' info_dict ' : {
2014-09-24 01:46:33 -07:00
' id ' : ' 5645318632 ' ,
' ext ' : ' mp4 ' ,
2014-11-23 11:41:03 -08:00
" description " : " Waterfalls in the Springtime at Dark Hollow Waterfalls. These are located just off of Skyline Drive in Virginia. They are only about 6/10 of a mile hike but it is a pretty steep hill and a good climb back up. " ,
" uploader_id " : " forestwander-nature-pictures " ,
2014-01-16 18:07:01 -08:00
" title " : " Dark Hollow Waterfalls "
2013-06-27 11:46:46 -07:00
}
}
2013-06-23 13:31:12 -07:00
def _real_extract ( self , url ) :
mobj = re . match ( self . _VALID_URL , url )
video_id = mobj . group ( ' id ' )
video_uploader_id = mobj . group ( ' uploader_id ' )
webpage_url = ' http://www.flickr.com/photos/ ' + video_uploader_id + ' / ' + video_id
2015-04-22 10:24:14 -07:00
req = compat_urllib_request . Request ( webpage_url )
req . add_header (
' User-Agent ' ,
# it needs a more recent version
' Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20150101 Firefox/38.0 (Chrome) ' )
webpage = self . _download_webpage ( req , video_id )
secret = self . _search_regex ( r ' secret " \ s*: \ s* " ( \ w+) " ' , webpage , ' secret ' )
2013-06-23 13:31:12 -07:00
first_url = ' https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id= ' + video_id + ' &secret= ' + secret + ' &bitrate=700&target=_self '
first_xml = self . _download_webpage ( first_url , video_id , ' Downloading first data webpage ' )
node_id = self . _html_search_regex ( r ' <Item id= " id " >( \ d+- \ d+)</Item> ' ,
2014-11-23 12:39:15 -08:00
first_xml , ' node_id ' )
2013-06-23 13:31:12 -07:00
second_url = ' https://secure.flickr.com/video_playlist.gne?node_id= ' + node_id + ' &tech=flash&mode=playlist&bitrate=700&secret= ' + secret + ' &rd=video.yahoo.com&noad=1 '
second_xml = self . _download_webpage ( second_url , video_id , ' Downloading second data webpage ' )
self . report_extraction ( video_id )
mobj = re . search ( r ' <STREAM APP= " (.+?) " FULLPATH= " (.+?) " ' , second_xml )
if mobj is None :
2014-01-16 18:07:01 -08:00
raise ExtractorError ( ' Unable to extract video url ' )
2013-06-23 13:31:12 -07:00
video_url = mobj . group ( 1 ) + unescapeHTML ( mobj . group ( 2 ) )
2014-09-24 01:46:33 -07:00
return {
' id ' : video_id ,
' url ' : video_url ,
' ext ' : ' mp4 ' ,
' title ' : self . _og_search_title ( webpage ) ,
2013-07-12 10:00:19 -07:00
' description ' : self . _og_search_description ( webpage ) ,
2014-09-24 01:46:33 -07:00
' thumbnail ' : self . _og_search_thumbnail ( webpage ) ,
2013-06-23 13:31:12 -07:00
' uploader_id ' : video_uploader_id ,
2014-09-24 01:46:33 -07:00
}