1 from __future__
import unicode_literals
6 from .subtitles
import SubtitlesInfoExtractor
13 class TEDIE(SubtitlesInfoExtractor
):
16 (?P<type>www|embed)(?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',
43 'url': 'http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms',
44 'md5': '226f4fb9c62380d11b7995efa4c87994',
46 'id': 'vishal-sikka-the-beauty-and-power-of-algorithms',
48 'title': 'Vishal Sikka: The beauty and power of algorithms',
49 'thumbnail': 're:^https?://.+\.jpg',
50 '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.',
53 'url': 'http://www.ted.com/talks/gabby_giffords_and_mark_kelly_be_passionate_be_courageous_be_your_best',
54 'md5': '49144e345a899b8cb34d315f3b9cfeeb',
58 'title': 'Be passionate. Be courageous. Be your best.',
59 'uploader': 'Gabby Giffords and Mark Kelly',
60 'description': 'md5:5174aed4d0f16021b704120360f72b92',
65 'low': {'preference': 1, 'width': 320, 'height': 180}
,
66 'medium': {'preference': 2, 'width': 512, 'height': 288}
,
67 'high': {'preference': 3, 'width': 854, 'height': 480}
,
70 def _extract_info(self
, webpage
):
71 info_json
= self
._search
_regex
(r
'q\("\w+.init",({.+})\)</script>',
73 return json
.loads(info_json
)
75 def _real_extract(self
, url
):
76 m
= re
.match(self
._VALID
_URL
, url
, re
.VERBOSE
)
77 if m
.group('type') == 'embed':
78 desktop_url
= m
.group('proto') + 'www' + m
.group('urlmain')
79 return self
.url_result(desktop_url
, 'TED')
80 name
= m
.group('name')
81 if m
.group('type_talk'):
82 return self
._talk
_info
(url
, name
)
83 elif m
.group('type_watch'):
84 return self
._watch
_info
(url
, name
)
86 return self
._playlist
_videos
_info
(url
, name
)
88 def _playlist_videos_info(self
, url
, name
):
89 '''Returns the videos of the playlist'''
91 webpage
= self
._download
_webpage
(url
, name
,
92 'Downloading playlist webpage')
93 info
= self
._extract
_info
(webpage
)
94 playlist_info
= info
['playlist']
97 self
.url_result('http://www.ted.com/talks/' + talk
['slug'], self
.ie_key())
98 for talk
in info
['talks']
100 return self
.playlist_result(
102 playlist_id
=compat_str(playlist_info
['id']),
103 playlist_title
=playlist_info
['title'])
105 def _talk_info(self
, url
, video_name
):
106 webpage
= self
._download
_webpage
(url
, video_name
)
107 self
.report_extraction(video_name
)
109 talk_info
= self
._extract
_info
(webpage
)['talks'][0]
113 'format_id': format_id
,
115 } for (format_id
, format_url
) in talk_info
['nativeDownloads'].items() if format_url
is not None]
118 finfo
= self
._NATIVE
_FORMATS
.get(f
['format_id'])
124 'format_id': f
['name'],
125 'url': talk_info
['streamer'],
126 'play_path': f
['file'],
129 'height': f
['height'],
131 } for f
in talk_info
['resources']['rtmp']]
132 self
._sort
_formats
(formats
)
134 video_id
= compat_str(talk_info
['id'])
136 video_subtitles
= self
.extract_subtitles(video_id
, talk_info
)
137 if self
._downloader
.params
.get('listsubtitles', False):
138 self
._list
_available
_subtitles
(video_id
, talk_info
)
141 thumbnail
= talk_info
['thumb']
142 if not thumbnail
.startswith('http'):
143 thumbnail
= 'http://' + thumbnail
146 'title': talk_info
['title'],
147 'uploader': talk_info
['speaker'],
148 'thumbnail': thumbnail
,
149 'description': self
._og
_search
_description
(webpage
),
150 'subtitles': video_subtitles
,
154 def _get_available_subtitles(self
, video_id
, talk_info
):
155 languages
= [lang
['languageCode'] for lang
in talk_info
.get('languages', [])]
159 url
= 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/srt' % (video_id
, l
)
160 sub_lang_list
[l
] = url
163 self
._downloader
.report_warning('video doesn\'t have subtitles')
166 def _watch_info(self
, url
, name
):
167 webpage
= self
._download
_webpage
(url
, name
)
169 config_json
= self
._html
_search
_regex
(
170 r
"data-config='([^']+)", webpage
, 'config')
171 config
= json
.loads(config_json
)
172 video_url
= config
['video']['url']
173 thumbnail
= config
.get('image', {}).get('url')
175 title
= self
._html
_search
_regex
(
176 r
"(?s)<h1(?:\s+class='[^']+')?>(.+?)</h1>", webpage
, 'title')
177 description
= self
._html
_search
_regex
(
179 r
'(?s)<h4 class="[^"]+" id="h3--about-this-talk">.*?</h4>(.*?)</div>',
180 r
'(?s)<p><strong>About this talk:</strong>\s+(.*?)</p>',
182 webpage
, 'description', fatal
=False)
188 'thumbnail': thumbnail
,
189 'description': description
,