]>
Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ted.py
1 from __future__
import unicode_literals
6 from . subtitles
import SubtitlesInfoExtractor
13 class TEDIE ( SubtitlesInfoExtractor
):
16 (?P<type>www|embed(?:-ssl)?)(?P<urlmain>\.ted\.com/
18 (?P<type_playlist>playlists(?:/\d+)?) # We have a playlist
20 ((?P<type_talk>talks)) # We have a simple talk
22 (?P<type_watch>watch)/[^/]+/[^/]+
24 (/lang/(.*?))? # The url may contain the language
25 /(?P<name>[\w-]+) # Here goes the name and then ".html"
29 'url' : 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html' ,
30 'md5' : 'fc94ac279feebbce69f21c0c6ee82810' ,
34 'title' : 'The illusion of consciousness' ,
35 'description' : ( 'Philosopher Dan Dennett makes a compelling '
36 'argument that not only don \' t we understand our own '
37 'consciousness, but that half the time our brains are '
38 'actively fooling us.' ),
39 'uploader' : 'Dan Dennett' ,
44 'url' : 'http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms' ,
45 'md5' : '226f4fb9c62380d11b7995efa4c87994' ,
47 'id' : 'vishal-sikka-the-beauty-and-power-of-algorithms' ,
49 'title' : 'Vishal Sikka: The beauty and power of algorithms' ,
50 'thumbnail' : 're:^https?://.+\.jpg' ,
51 '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.' ,
54 'url' : 'http://www.ted.com/talks/gabby_giffords_and_mark_kelly_be_passionate_be_courageous_be_your_best' ,
58 'title' : 'Be passionate. Be courageous. Be your best.' ,
59 'uploader' : 'Gabby Giffords and Mark Kelly' ,
60 'description' : 'md5:5174aed4d0f16021b704120360f72b92' ,
64 'url' : 'http://www.ted.com/playlists/who_are_the_hackers' ,
67 'title' : 'Who are the hackers?' ,
69 'playlist_mincount' : 6 ,
71 # contains a youtube video
72 'url' : 'https://www.ted.com/talks/douglas_adams_parrots_the_universe_and_everything' ,
73 'add_ie' : [ 'Youtube' ],
77 'title' : 'Douglas Adams: Parrots the Universe and Everything' ,
78 'description' : 'md5:01ad1e199c49ac640cb1196c0e9016af' ,
79 'uploader' : 'University of California Television (UCTV)' ,
80 'uploader_id' : 'UCtelevision' ,
81 'upload_date' : '20080522' ,
84 'skip_download' : True ,
89 'low' : { 'preference' : 1 , 'width' : 320 , 'height' : 180 },
90 'medium' : { 'preference' : 2 , 'width' : 512 , 'height' : 288 },
91 'high' : { 'preference' : 3 , 'width' : 854 , 'height' : 480 },
94 def _extract_info ( self
, webpage
):
95 info_json
= self
._ search
_ regex
( r
'q\("\w+.init",({.+})\)</script>' ,
97 return json
. loads ( info_json
)
99 def _real_extract ( self
, url
):
100 m
= re
. match ( self
._ VALID
_U RL
, url
, re
. VERBOSE
)
101 if m
. group ( 'type' ). startswith ( 'embed' ):
102 desktop_url
= m
. group ( 'proto' ) + 'www' + m
. group ( 'urlmain' )
103 return self
. url_result ( desktop_url
, 'TED' )
104 name
= m
. group ( 'name' )
105 if m
. group ( 'type_talk' ):
106 return self
._ talk
_ info
( url
, name
)
107 elif m
. group ( 'type_watch' ):
108 return self
._ watch
_ info
( url
, name
)
110 return self
._ playlist
_ videos
_ info
( url
, name
)
112 def _playlist_videos_info ( self
, url
, name
):
113 '''Returns the videos of the playlist'''
115 webpage
= self
._ download
_ webpage
( url
, name
,
116 'Downloading playlist webpage' )
117 info
= self
._ extract
_ info
( webpage
)
118 playlist_info
= info
[ 'playlist' ]
121 self
. url_result ( 'http://www.ted.com/talks/' + talk
[ 'slug' ], self
. ie_key ())
122 for talk
in info
[ 'talks' ]
124 return self
. playlist_result (
126 playlist_id
= compat_str ( playlist_info
[ 'id' ]),
127 playlist_title
= playlist_info
[ 'title' ])
129 def _talk_info ( self
, url
, video_name
):
130 webpage
= self
._ download
_ webpage
( url
, video_name
)
131 self
. report_extraction ( video_name
)
133 talk_info
= self
._ extract
_ info
( webpage
)[ 'talks' ][ 0 ]
135 if talk_info
. get ( 'external' ) is not None :
136 self
. to_screen ( 'Found video from %s ' % talk_info
[ 'external' ][ 'service' ])
139 'url' : talk_info
[ 'external' ][ 'uri' ],
144 'format_id' : format_id
,
146 } for ( format_id
, format_url
) in talk_info
[ 'nativeDownloads' ]. items () if format_url
is not None ]
149 finfo
= self
._ NATIVE
_ FORMATS
. get ( f
[ 'format_id' ])
155 'format_id' : f
[ 'name' ],
156 'url' : talk_info
[ 'streamer' ],
157 'play_path' : f
[ 'file' ],
160 'height' : f
[ 'height' ],
162 } for f
in talk_info
[ 'resources' ][ 'rtmp' ]]
163 self
._ sort
_ formats
( formats
)
165 video_id
= compat_str ( talk_info
[ 'id' ])
167 video_subtitles
= self
. extract_subtitles ( video_id
, talk_info
)
168 if self
._ downloader
. params
. get ( 'listsubtitles' , False ):
169 self
._l ist
_ available
_ subtitles
( video_id
, talk_info
)
172 thumbnail
= talk_info
[ 'thumb' ]
173 if not thumbnail
. startswith ( 'http' ):
174 thumbnail
= 'http://' + thumbnail
177 'title' : talk_info
[ 'title' ]. strip (),
178 'uploader' : talk_info
[ 'speaker' ],
179 'thumbnail' : thumbnail
,
180 'description' : self
._ og
_ search
_ description
( webpage
),
181 'subtitles' : video_subtitles
,
183 'duration' : talk_info
. get ( 'duration' ),
186 def _get_available_subtitles ( self
, video_id
, talk_info
):
187 languages
= [ lang
[ 'languageCode' ] for lang
in talk_info
. get ( 'languages' , [])]
191 url
= 'http://www.ted.com/talks/subtitles/id/ %s /lang/ %s /format/srt' % ( video_id
, l
)
192 sub_lang_list
[ l
] = url
195 self
._ downloader
. report_warning ( 'video doesn \' t have subtitles' )
198 def _watch_info ( self
, url
, name
):
199 webpage
= self
._ download
_ webpage
( url
, name
)
201 config_json
= self
._ html
_ search
_ regex
(
202 r
'"pages\.jwplayer"\s*,\s*({.+?})\s*\)\s*</script>' ,
204 config
= json
. loads ( config_json
)[ 'config' ]
205 video_url
= config
[ 'video' ][ 'url' ]
206 thumbnail
= config
. get ( 'image' , {}). get ( 'url' )
208 title
= self
._ html
_ search
_ regex
(
209 r
"(?s)<h1(?:\s+class='[^']+')?>(.+?)</h1>" , webpage
, 'title' )
210 description
= self
._ html
_ search
_ regex
(
212 r
'(?s)<h4 class="[^"]+" id="h3--about-this-talk">.*?</h4>(.*?)</div>' ,
213 r
'(?s)<p><strong>About this talk:</strong>\s+(.*?)</p>' ,
215 webpage
, 'description' , fatal
= False )
221 'thumbnail' : thumbnail
,
222 'description' : description
,