]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/rice.py
[adn] fix subtitle extraction(#12724)
[youtube-dl.git] / youtube_dl / extractor / rice.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import compat_parse_qs
8 from ..utils import (
9     xpath_text,
10     xpath_element,
11     int_or_none,
12     parse_iso8601,
13     ExtractorError,
14 )
15
16
17 class RICEIE(InfoExtractor):
18     _VALID_URL = r'https?://mediahub\.rice\.edu/app/[Pp]ortal/video\.aspx\?(?P<query>.+)'
19     _TEST = {
20         'url': 'https://mediahub.rice.edu/app/Portal/video.aspx?PortalID=25ffd62c-3d01-4b29-8c70-7c94270efb3e&DestinationID=66bc9434-03bd-4725-b47e-c659d8d809db&ContentID=YEWIvbhb40aqdjMD1ALSqw',
21         'md5': '9b83b4a2eead4912dc3b7fac7c449b6a',
22         'info_dict': {
23             'id': 'YEWIvbhb40aqdjMD1ALSqw',
24             'ext': 'mp4',
25             'title': 'Active Learning in Archeology',
26             'upload_date': '20140616',
27             'timestamp': 1402926346,
28         }
29     }
30     _NS = 'http://schemas.datacontract.org/2004/07/ensembleVideo.Data.Service.Contracts.Models.Player.Config'
31
32     def _real_extract(self, url):
33         qs = compat_parse_qs(re.match(self._VALID_URL, url).group('query'))
34         if not qs.get('PortalID') or not qs.get('DestinationID') or not qs.get('ContentID'):
35             raise ExtractorError('Invalid URL', expected=True)
36
37         portal_id = qs['PortalID'][0]
38         playlist_id = qs['DestinationID'][0]
39         content_id = qs['ContentID'][0]
40
41         content_data = self._download_xml('https://mediahub.rice.edu/api/portal/GetContentTitle', content_id, query={
42             'portalId': portal_id,
43             'playlistId': playlist_id,
44             'contentId': content_id
45         })
46         metadata = xpath_element(content_data, './/metaData', fatal=True)
47         title = xpath_text(metadata, 'primaryTitle', fatal=True)
48         encodings = xpath_element(content_data, './/encodings', fatal=True)
49         player_data = self._download_xml('https://mediahub.rice.edu/api/player/GetPlayerConfig', content_id, query={
50             'temporaryLinkId': xpath_text(encodings, 'temporaryLinkId', fatal=True),
51             'contentId': content_id,
52         })
53
54         common_fmt = {}
55         dimensions = xpath_text(encodings, 'dimensions')
56         if dimensions:
57             wh = dimensions.split('x')
58             if len(wh) == 2:
59                 common_fmt.update({
60                     'width': int_or_none(wh[0]),
61                     'height': int_or_none(wh[1]),
62                 })
63
64         formats = []
65         rtsp_path = xpath_text(player_data, self._xpath_ns('RtspPath', self._NS))
66         if rtsp_path:
67             fmt = {
68                 'url': rtsp_path,
69                 'format_id': 'rtsp',
70             }
71             fmt.update(common_fmt)
72             formats.append(fmt)
73         for source in player_data.findall(self._xpath_ns('.//Source', self._NS)):
74             video_url = xpath_text(source, self._xpath_ns('File', self._NS))
75             if not video_url:
76                 continue
77             if '.m3u8' in video_url:
78                 formats.extend(self._extract_m3u8_formats(video_url, content_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
79             else:
80                 fmt = {
81                     'url': video_url,
82                     'format_id': video_url.split(':')[0],
83                 }
84                 fmt.update(common_fmt)
85                 rtmp = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', video_url)
86                 if rtmp:
87                     fmt.update({
88                         'url': rtmp.group('url'),
89                         'play_path': rtmp.group('playpath'),
90                         'app': rtmp.group('app'),
91                         'ext': 'flv',
92                     })
93                 formats.append(fmt)
94         self._sort_formats(formats)
95
96         thumbnails = []
97         for content_asset in content_data.findall('.//contentAssets'):
98             asset_type = xpath_text(content_asset, 'type')
99             if asset_type == 'image':
100                 image_url = xpath_text(content_asset, 'httpPath')
101                 if not image_url:
102                     continue
103                 thumbnails.append({
104                     'id': xpath_text(content_asset, 'ID'),
105                     'url': image_url,
106                 })
107
108         return {
109             'id': content_id,
110             'title': title,
111             'description': xpath_text(metadata, 'abstract'),
112             'duration': int_or_none(xpath_text(metadata, 'duration')),
113             'timestamp': parse_iso8601(xpath_text(metadata, 'dateUpdated')),
114             'thumbnails': thumbnails,
115             'formats': formats,
116         }