aboutsummaryrefslogtreecommitdiffstats
path: root/youtube_dl/extractor/sandia.py
diff options
context:
space:
mode:
authorWolfgang Wiedmeyer <wolfgit@wiedmeyer.de>2018-01-14 00:50:11 +0100
committerWolfgang Wiedmeyer <wolfgit@wiedmeyer.de>2018-01-14 00:50:11 +0100
commitb382d0f71eeece4e6e1cc3662f3c89884d5ac05d (patch)
treee48661b9f1de2380d4bff94f416624827746abd7 /youtube_dl/extractor/sandia.py
parent5020cb7e8233683ef7845fea1233a95f3f775a7b (diff)
parenta86922c4702e2c8538337124c5bf02a4b5f9aa4a (diff)
downloadyoutube-dl-b382d0f71eeece4e6e1cc3662f3c89884d5ac05d.zip
youtube-dl-b382d0f71eeece4e6e1cc3662f3c89884d5ac05d.tar.gz
youtube-dl-b382d0f71eeece4e6e1cc3662f3c89884d5ac05d.tar.bz2
Merge branch 'upstream'
Diffstat (limited to 'youtube_dl/extractor/sandia.py')
-rw-r--r--youtube_dl/extractor/sandia.py65
1 files changed, 0 insertions, 65 deletions
diff --git a/youtube_dl/extractor/sandia.py b/youtube_dl/extractor/sandia.py
deleted file mode 100644
index 96e43af..0000000
--- a/youtube_dl/extractor/sandia.py
+++ /dev/null
@@ -1,65 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import json
-
-from .common import InfoExtractor
-from ..utils import (
- int_or_none,
- mimetype2ext,
-)
-
-
-class SandiaIE(InfoExtractor):
- IE_DESC = 'Sandia National Laboratories'
- _VALID_URL = r'https?://digitalops\.sandia\.gov/Mediasite/Play/(?P<id>[0-9a-f]+)'
- _TEST = {
- 'url': 'http://digitalops.sandia.gov/Mediasite/Play/24aace4429fc450fb5b38cdbf424a66e1d',
- 'md5': '9422edc9b9a60151727e4b6d8bef393d',
- 'info_dict': {
- 'id': '24aace4429fc450fb5b38cdbf424a66e1d',
- 'ext': 'mp4',
- 'title': 'Xyce Software Training - Section 1',
- 'description': 're:(?s)SAND Number: SAND 2013-7800.{200,}',
- 'upload_date': '20120409',
- 'timestamp': 1333983600,
- 'duration': 7794,
- }
- }
-
- def _real_extract(self, url):
- video_id = self._match_id(url)
-
- presentation_data = self._download_json(
- 'http://digitalops.sandia.gov/Mediasite/PlayerService/PlayerService.svc/json/GetPlayerOptions',
- video_id, data=json.dumps({
- 'getPlayerOptionsRequest': {
- 'ResourceId': video_id,
- 'QueryString': '',
- }
- }), headers={
- 'Content-Type': 'application/json; charset=utf-8',
- })['d']['Presentation']
-
- title = presentation_data['Title']
-
- formats = []
- for stream in presentation_data.get('Streams', []):
- for fd in stream.get('VideoUrls', []):
- formats.append({
- 'format_id': fd['MediaType'],
- 'format_note': fd['MimeType'].partition('/')[2],
- 'ext': mimetype2ext(fd['MimeType']),
- 'url': fd['Location'],
- 'protocol': 'f4m' if fd['MimeType'] == 'video/x-mp4-fragmented' else None,
- })
- self._sort_formats(formats)
-
- return {
- 'id': video_id,
- 'title': title,
- 'description': presentation_data.get('Description'),
- 'formats': formats,
- 'timestamp': int_or_none(presentation_data.get('UnixTime'), 1000),
- 'duration': int_or_none(presentation_data.get('Duration'), 1000),
- }