]>
jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/viidea.py
1 from __future__
import unicode_literals
5 from .common
import InfoExtractor
17 class ViideaIE(InfoExtractor
):
18 _VALID_URL
= r
'''(?x)http://(?:www\.)?(?:
20 flexilearn\.viidea\.net|
21 presentations\.ocwconsortium\.org|
22 video\.travel-zoom\.si|
23 video\.pomp-forum\.si|
29 video\.kiberpipa\.org|
32 edemokracija\.viidea\.com
33 )(?:/lecture)?/(?P<id>[^/]+)(?:/video/(?P<part>\d+))?/*(?:[#?].*)?$'''
36 'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/',
39 'display_id': 'promogram_igor_mekjavic_eng',
41 'title': 'Automatics, robotics and biocybernetics',
42 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
43 'thumbnail': 're:http://.*\.jpg',
44 'timestamp': 1372349289,
45 'upload_date': '20130627',
50 'skip_download': True,
53 # video with invalid direct format links (HTTP 403)
54 'url': 'http://videolectures.net/russir2010_filippova_nlp/',
57 'display_id': 'russir2010_filippova_nlp',
59 'title': 'NLP at Google',
60 'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3',
61 'thumbnail': 're:http://.*\.jpg',
62 'timestamp': 1284375600,
63 'upload_date': '20100913',
68 'skip_download': True,
72 'url': 'http://videolectures.net/deeplearning2015_montreal/',
75 'title': 'Deep Learning Summer School, Montreal 2015',
76 'description': 'md5:0533a85e4bd918df52a01f0e1ebe87b7',
77 'thumbnail': 're:http://.*\.jpg',
78 'timestamp': 1438560000,
83 'url': 'http://videolectures.net/mlss09uk_bishop_ibi/',
86 'display_id': 'mlss09uk_bishop_ibi',
87 'title': 'Introduction To Bayesian Inference',
88 'thumbnail': 're:http://.*\.jpg',
89 'timestamp': 1251622800,
94 'display_id': 'mlss09uk_bishop_ibi_part1',
96 'title': 'Introduction To Bayesian Inference (Part 1)',
97 'thumbnail': 're:http://.*\.jpg',
99 'timestamp': 1251622800,
100 'upload_date': '20090830',
105 'display_id': 'mlss09uk_bishop_ibi_part2',
107 'title': 'Introduction To Bayesian Inference (Part 2)',
108 'thumbnail': 're:http://.*\.jpg',
110 'timestamp': 1251622800,
111 'upload_date': '20090830',
117 def _real_extract(self
, url
):
118 lecture_slug
, explicit_part_id
= re
.match(self
._VALID
_URL
, url
).groups()
120 webpage
= self
._download
_webpage
(url
, lecture_slug
)
122 cfg
= self
._parse
_json
(self
._search
_regex
(
123 [r
'cfg\s*:\s*({.+?})\s*,\s*[\da-zA-Z_]+\s*:\s*\(?\s*function',
124 r
'cfg\s*:\s*({[^}]+})'],
125 webpage
, 'cfg'), lecture_slug
, js_to_json
)
127 lecture_id
= compat_str(cfg
['obj_id'])
129 base_url
= self
._proto
_relative
_url
(cfg
['livepipe'], 'http:')
131 lecture_data
= self
._download
_json
(
132 '%s/site/api/lecture/%s?format=json' % (base_url
, lecture_id
),
133 lecture_id
)['lecture'][0]
137 'display_id': lecture_slug
,
138 'title': lecture_data
['title'],
139 'timestamp': parse_iso8601(lecture_data
.get('time')),
140 'description': lecture_data
.get('description_wiki'),
141 'thumbnail': lecture_data
.get('thumb'),
144 playlist_entries
= []
145 lecture_type
= lecture_data
.get('type')
146 parts
= [compat_str(video
) for video
in cfg
.get('videos', [])]
148 multipart
= len(parts
) > 1
150 def extract_part(part_id
):
151 smil_url
= '%s/%s/video/%s/smil.xml' % (base_url
, lecture_slug
, part_id
)
152 smil
= self
._download
_smil
(smil_url
, lecture_id
)
153 info
= self
._parse
_smil
(smil
, smil_url
, lecture_id
)
154 info
['id'] = lecture_id
if not multipart
else '%s_part%s' % (lecture_id
, part_id
)
155 info
['display_id'] = lecture_slug
if not multipart
else '%s_part%s' % (lecture_slug
, part_id
)
157 info
['title'] += ' (Part %s)' % part_id
158 switch
= smil
.find('.//switch')
159 if switch
is not None:
160 info
['duration'] = parse_duration(switch
.attrib
.get('dur'))
161 item_info
= lecture_info
.copy()
162 item_info
.update(info
)
165 if explicit_part_id
or not multipart
:
166 result
= extract_part(explicit_part_id
or parts
[0])
169 '_type': 'multi_video',
170 'entries': [extract_part(part
) for part
in parts
],
172 result
.update(lecture_info
)
174 # Immediately return explicitly requested part or non event item
175 if explicit_part_id
or lecture_type
!= 'evt':
178 playlist_entries
.append(result
)
180 # It's probably a playlist
181 if not parts
or lecture_type
== 'evt':
182 playlist_webpage
= self
._download
_webpage
(
183 '%s/site/ajax/drilldown/?id=%s' % (base_url
, lecture_id
), lecture_id
)
185 self
.url_result(compat_urlparse
.urljoin(url
, video_url
), 'Viidea')
186 for _
, video_url
in re
.findall(
187 r
'<a[^>]+href=(["\'])(.+?
)\
1[^
>]+id=["\']lec=\d+', playlist_webpage)]
188 playlist_entries.extend(entries)
190 playlist = self.playlist_result(playlist_entries, lecture_id)
191 playlist.update(lecture_info)