]>
Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/dispeak.py
1 from __future__
import unicode_literals
5 from .common
import InfoExtractor
15 class DigitallySpeakingIE(InfoExtractor
):
16 _VALID_URL
= r
'https?://(?:s?evt\.dispeak|events\.digitallyspeaking)\.com/(?:[^/]+/)+xml/(?P<id>[^.]+)\.xml'
19 # From http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface
20 'url': 'http://evt.dispeak.com/ubm/gdc/sf16/xml/840376_BQRC.xml',
21 'md5': 'a8efb6c31ed06ca8739294960b2dbabd',
25 'title': 'Tenacious Design and The Interface of \'Destiny\'',
28 # From http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC
29 'url': 'http://events.digitallyspeaking.com/gdc/sf11/xml/12396_1299111843500GMPX.xml',
30 'only_matching': True,
32 # From http://www.gdcvault.com/play/1013700/Advanced-Material
33 'url': 'http://sevt.dispeak.com/ubm/gdc/eur10/xml/11256_1282118587281VNIT.xml',
34 'only_matching': True,
37 def _parse_mp4(self
, metadata
):
41 mp4_video
= xpath_text(metadata
, './mp4video', default
=None)
42 if mp4_video
is not None:
43 mobj
= re
.match(r
'(?P<root>https?://.*?/).*', mp4_video
)
44 video_root
= mobj
.group('root')
45 if video_root
is None:
46 http_host
= xpath_text(metadata
, 'httpHost', default
=None)
48 video_root
= 'http://%s/' % http_host
49 if video_root
is None:
50 # Hard-coded in http://evt.dispeak.com/ubm/gdc/sf16/custom/player2.js
51 # Works for GPUTechConf, too
52 video_root
= 'http://s3-2u.digitallyspeaking.com/'
54 formats
= metadata
.findall('./MBRVideos/MBRVideo')
57 for a_format
in formats
:
58 stream_name
= xpath_text(a_format
, 'streamName', fatal
=True)
59 video_path
= re
.match(r
'mp4\:(?P<path>.*)', stream_name
).group('path')
60 url
= video_root
+ video_path
61 bitrate
= xpath_text(a_format
, 'bitrate')
62 tbr
= int_or_none(bitrate
)
63 vbr
= int_or_none(self
._search
_regex
(
64 r
'-(\d+)\.mp4', video_path
, 'vbr', default
=None))
65 abr
= tbr
- vbr
if tbr
and vbr
else None
66 video_formats
.append({
75 def _parse_flv(self
, metadata
):
77 akamai_url
= xpath_text(metadata
, './akamaiHost', fatal
=True)
78 audios
= metadata
.findall('./audios/audio')
81 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url
,
82 'play_path': remove_end(audio
.get('url'), '.flv'),
85 'format_id': audio
.get('code'),
87 slide_video_path
= xpath_text(metadata
, './slideVideo', fatal
=True)
89 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url
,
90 'play_path': remove_end(slide_video_path
, '.flv'),
92 'format_note': 'slide deck video',
95 'format_id': 'slides',
97 speaker_video_path
= xpath_text(metadata
, './speakerVideo', fatal
=True)
99 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url
,
100 'play_path': remove_end(speaker_video_path
, '.flv'),
102 'format_note': 'speaker video',
105 'format_id': 'speaker',
109 def _real_extract(self
, url
):
110 video_id
= self
._match
_id
(url
)
112 xml_description
= self
._download
_xml
(url
, video_id
)
113 metadata
= xpath_element(xml_description
, 'metadata')
115 video_formats
= self
._parse
_mp
4(metadata
)
116 if video_formats
is None:
117 video_formats
= self
._parse
_flv
(metadata
)
121 'formats': video_formats
,
122 'title': xpath_text(metadata
, 'title', fatal
=True),
123 'duration': parse_duration(xpath_text(metadata
, 'endTime')),
124 'creator': xpath_text(metadata
, 'speaker'),