2014-01-17 11:52:17 +09:00
from __future__ import unicode_literals
2013-06-24 04:55:53 +09:00
import json
import re
2013-11-03 03:48:39 +09:00
from . subtitles import SubtitlesInfoExtractor
2013-06-24 04:55:53 +09:00
2013-11-05 20:00:13 +09:00
from . . utils import (
2014-03-05 21:22:10 +09:00
compat_str ,
2013-11-05 20:00:13 +09:00
)
2014-01-17 11:52:17 +09:00
2013-11-03 03:48:39 +09:00
class TEDIE ( SubtitlesInfoExtractor ) :
2014-03-21 00:33:23 +09:00
_VALID_URL = r ''' (?x)
( ? P < proto > https ? : / / )
( ? P < type > www | embed ) ( ? P < urlmain > \. ted \. com /
2014-03-05 21:27:26 +09:00
(
( ? P < type_playlist > playlists ( ? : / \d + ) ? ) # We have a playlist
|
( ( ? P < type_talk > talks ) ) # We have a simple talk
2014-03-27 10:22:40 +09:00
|
( ? P < type_watch > watch ) / [ ^ / ] + / [ ^ / ] +
2014-03-05 21:27:26 +09:00
)
( / lang / ( . * ? ) ) ? # The url may contain the language
2014-03-27 10:22:40 +09:00
/ ( ? P < name > [ \w - ] + ) # Here goes the name and then ".html"
2014-03-21 00:33:23 +09:00
. * ) $
2014-03-05 21:27:26 +09:00
'''
2014-03-27 10:22:40 +09:00
_TESTS = [ {
2014-01-17 11:52:17 +09:00
' url ' : ' http://www.ted.com/talks/dan_dennett_on_our_consciousness.html ' ,
2014-01-17 11:54:54 +09:00
' md5 ' : ' 4ea1dada91e4174b53dac2bb8ace429d ' ,
2014-01-17 11:52:17 +09:00
' info_dict ' : {
2014-03-05 22:27:45 +09:00
' id ' : ' 102 ' ,
' ext ' : ' mp4 ' ,
2014-03-05 05:47:01 +09:00
' title ' : ' The illusion of consciousness ' ,
2014-03-05 21:27:26 +09:00
' description ' : ( ' Philosopher Dan Dennett makes a compelling '
' argument that not only don \' t we understand our own '
' consciousness, but that half the time our brains are '
' actively fooling us. ' ) ,
2014-03-05 05:47:01 +09:00
' uploader ' : ' Dan Dennett ' ,
2014-04-07 20:07:07 +09:00
' width ' : 854 ,
2013-06-28 03:46:46 +09:00
}
2014-03-27 10:22:40 +09:00
} , {
' url ' : ' http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms ' ,
' md5 ' : ' 226f4fb9c62380d11b7995efa4c87994 ' ,
' info_dict ' : {
' id ' : ' vishal-sikka-the-beauty-and-power-of-algorithms ' ,
' ext ' : ' mp4 ' ,
' title ' : ' Vishal Sikka: The beauty and power of algorithms ' ,
' thumbnail ' : ' re:^https?://.+ \ .jpg ' ,
' description ' : ' Adaptive, intelligent, and consistent, algorithms are emerging as the ultimate app for everything from matching consumers to products to assessing medical diagnoses. Vishal Sikka shares his appreciation for the algorithm, charting both its inherent beauty and its growing power. ' ,
}
2014-04-14 22:23:12 +09:00
} , {
' url ' : ' http://www.ted.com/talks/gabby_giffords_and_mark_kelly_be_passionate_be_courageous_be_your_best ' ,
' info_dict ' : {
' id ' : ' 1972 ' ,
' ext ' : ' flv ' ,
' title ' : ' Be passionate. Be courageous. Be your best. ' ,
' uploader ' : ' Gabby Giffords and Mark Kelly ' ,
' description ' : ' md5:d89e1d8ebafdac8e55df4c219ecdbfe9 ' ,
} ,
' params ' : {
# rtmp download
' skip_download ' : True ,
} ,
2014-03-27 10:22:40 +09:00
} ]
2013-06-24 04:55:53 +09:00
2014-04-07 20:07:07 +09:00
_NATIVE_FORMATS = {
' low ' : { ' preference ' : 1 , ' width ' : 320 , ' height ' : 180 } ,
' medium ' : { ' preference ' : 2 , ' width ' : 512 , ' height ' : 288 } ,
' high ' : { ' preference ' : 3 , ' width ' : 854 , ' height ' : 480 } ,
2014-03-05 05:47:01 +09:00
}
2013-06-24 04:55:53 +09:00
2014-03-05 21:22:10 +09:00
def _extract_info ( self , webpage ) :
2014-03-05 21:27:26 +09:00
info_json = self . _search_regex ( r ' q \ ( " \ w+.init " ,( { .+}) \ )</script> ' ,
webpage , ' info json ' )
2014-03-05 21:22:10 +09:00
return json . loads ( info_json )
2013-06-24 04:55:53 +09:00
def _real_extract ( self , url ) :
2014-03-05 21:27:26 +09:00
m = re . match ( self . _VALID_URL , url , re . VERBOSE )
2014-03-21 00:33:23 +09:00
if m . group ( ' type ' ) == ' embed ' :
desktop_url = m . group ( ' proto ' ) + ' www ' + m . group ( ' urlmain ' )
return self . url_result ( desktop_url , ' TED ' )
2014-03-05 21:27:26 +09:00
name = m . group ( ' name ' )
2013-06-24 04:55:53 +09:00
if m . group ( ' type_talk ' ) :
2014-03-05 21:27:26 +09:00
return self . _talk_info ( url , name )
2014-03-27 10:22:40 +09:00
elif m . group ( ' type_watch ' ) :
return self . _watch_info ( url , name )
2014-03-05 21:27:26 +09:00
else :
2014-03-05 21:22:10 +09:00
return self . _playlist_videos_info ( url , name )
2013-06-24 04:55:53 +09:00
2014-03-05 21:22:10 +09:00
def _playlist_videos_info ( self , url , name ) :
2013-06-24 04:55:53 +09:00
''' Returns the videos of the playlist '''
2013-11-15 22:33:51 +09:00
2014-03-05 21:22:10 +09:00
webpage = self . _download_webpage ( url , name ,
' Downloading playlist webpage ' )
info = self . _extract_info ( webpage )
playlist_info = info [ ' playlist ' ]
2013-06-24 04:55:53 +09:00
2013-11-15 22:33:51 +09:00
playlist_entries = [
2014-04-21 19:34:32 +09:00
self . url_result ( ' http://www.ted.com/talks/ ' + talk [ ' slug ' ] , self . ie_key ( ) )
2014-03-05 21:22:10 +09:00
for talk in info [ ' talks ' ]
2013-11-15 22:33:51 +09:00
]
return self . playlist_result (
2014-03-05 21:22:10 +09:00
playlist_entries ,
playlist_id = compat_str ( playlist_info [ ' id ' ] ) ,
playlist_title = playlist_info [ ' title ' ] )
2013-06-24 04:55:53 +09:00
2014-03-05 21:27:26 +09:00
def _talk_info ( self , url , video_name ) :
webpage = self . _download_webpage ( url , video_name )
2013-06-24 04:55:53 +09:00
self . report_extraction ( video_name )
2013-11-03 03:48:39 +09:00
2014-03-05 21:22:10 +09:00
talk_info = self . _extract_info ( webpage ) [ ' talks ' ] [ 0 ]
2013-11-03 03:48:39 +09:00
2014-03-05 05:47:01 +09:00
formats = [ {
' url ' : format_url ,
' format_id ' : format_id ,
' format ' : format_id ,
2014-04-14 22:23:12 +09:00
} for ( format_id , format_url ) in talk_info [ ' nativeDownloads ' ] . items ( ) if format_url is not None ]
if formats :
for f in formats :
finfo = self . _NATIVE_FORMATS . get ( f [ ' format_id ' ] )
if finfo :
f . update ( finfo )
else :
# Use rtmp downloads
formats = [ {
' format_id ' : f [ ' name ' ] ,
' url ' : talk_info [ ' streamer ' ] ,
' play_path ' : f [ ' file ' ] ,
' ext ' : ' flv ' ,
' width ' : f [ ' width ' ] ,
' height ' : f [ ' height ' ] ,
' tbr ' : f [ ' bitrate ' ] ,
} for f in talk_info [ ' resources ' ] [ ' rtmp ' ] ]
2014-03-05 05:47:01 +09:00
self . _sort_formats ( formats )
2014-03-05 22:27:45 +09:00
video_id = compat_str ( talk_info [ ' id ' ] )
2013-11-03 03:48:39 +09:00
# subtitles
2014-03-05 05:47:01 +09:00
video_subtitles = self . extract_subtitles ( video_id , talk_info )
2013-11-03 03:48:39 +09:00
if self . _downloader . params . get ( ' listsubtitles ' , False ) :
2014-03-05 05:47:01 +09:00
self . _list_available_subtitles ( video_id , talk_info )
2013-11-03 03:48:39 +09:00
return
2014-03-16 19:24:11 +09:00
thumbnail = talk_info [ ' thumb ' ]
if not thumbnail . startswith ( ' http ' ) :
thumbnail = ' http:// ' + thumbnail
2013-11-15 22:06:38 +09:00
return {
2013-11-03 03:48:39 +09:00
' id ' : video_id ,
2014-03-05 05:47:01 +09:00
' title ' : talk_info [ ' title ' ] ,
' uploader ' : talk_info [ ' speaker ' ] ,
2014-03-16 19:24:11 +09:00
' thumbnail ' : thumbnail ,
2014-03-05 05:47:01 +09:00
' description ' : self . _og_search_description ( webpage ) ,
2013-11-03 03:48:39 +09:00
' subtitles ' : video_subtitles ,
2013-10-04 17:32:34 +09:00
' formats ' : formats ,
}
2014-03-05 05:47:01 +09:00
def _get_available_subtitles ( self , video_id , talk_info ) :
languages = [ lang [ ' languageCode ' ] for lang in talk_info . get ( ' languages ' , [ ] ) ]
if languages :
sub_lang_list = { }
for l in languages :
url = ' http://www.ted.com/talks/subtitles/id/ %s /lang/ %s /format/srt ' % ( video_id , l )
sub_lang_list [ l ] = url
return sub_lang_list
else :
2014-04-21 19:34:32 +09:00
self . _downloader . report_warning ( ' video doesn \' t have subtitles ' )
2014-03-05 05:47:01 +09:00
return { }
2014-03-27 10:22:40 +09:00
def _watch_info ( self , url , name ) :
webpage = self . _download_webpage ( url , name )
config_json = self . _html_search_regex (
r " data-config= ' ([^ ' ]+) " , webpage , ' config ' )
config = json . loads ( config_json )
video_url = config [ ' video ' ] [ ' url ' ]
thumbnail = config . get ( ' image ' , { } ) . get ( ' url ' )
title = self . _html_search_regex (
r " (?s)<h1(?: \ s+class= ' [^ ' ]+ ' )?>(.+?)</h1> " , webpage , ' title ' )
description = self . _html_search_regex (
2014-04-21 19:37:16 +09:00
[
r ' (?s)<h4 class= " [^ " ]+ " id= " h3--about-this-talk " >.*?</h4>(.*?)</div> ' ,
r ' (?s)<p><strong>About this talk:</strong> \ s+(.*?)</p> ' ,
] ,
2014-03-27 10:22:40 +09:00
webpage , ' description ' , fatal = False )
return {
' id ' : name ,
' url ' : video_url ,
' title ' : title ,
' thumbnail ' : thumbnail ,
' description ' : description ,
}