youtube_dl/extractor/sandia.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import json
   6 import re
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_urllib_request,
  11     compat_urlparse,
  12 )
  13 from ..utils import (
  14     int_or_none,
  15     js_to_json,
  16     mimetype2ext,
  17     unified_strdate,
  18 )
  19
  20
  21 class SandiaIE(InfoExtractor):
  22     IE_DESC = 'Sandia National Laboratories'
  23     _VALID_URL = r'https?://digitalops\.sandia\.gov/Mediasite/Play/(?P<id>[0-9a-f]+)'
  24     _TEST = {
  25         'url': 'http://digitalops.sandia.gov/Mediasite/Play/24aace4429fc450fb5b38cdbf424a66e1d',
  26         'md5': '9422edc9b9a60151727e4b6d8bef393d',
  27         'info_dict': {
  28             'id': '24aace4429fc450fb5b38cdbf424a66e1d',
  29             'ext': 'mp4',
  30             'title': 'Xyce Software Training - Section 1',
  31             'description': 're:(?s)SAND Number: SAND 2013-7800.{200,}',
  32             'upload_date': '20120904',
  33             'duration': 7794,
  34         }
  35     }
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         req = compat_urllib_request.Request(url)
  41         req.add_header('Cookie', 'MediasitePlayerCaps=ClientPlugins=4')
  42         webpage = self._download_webpage(req, video_id)
  43
  44         js_path = self._search_regex(
  45             r'<script type="text/javascript" src="(/Mediasite/FileServer/Presentation/[^"]+)"',
  46             webpage, 'JS code URL')
  47         js_url = compat_urlparse.urljoin(url, js_path)
  48
  49         js_code = self._download_webpage(
  50             js_url, video_id, note='Downloading player')
  51
  52         def extract_str(key, **args):
  53             return self._search_regex(
  54                 r'Mediasite\.PlaybackManifest\.%s\s*=\s*(.+);\s*?\n' % re.escape(key),
  55                 js_code, key, **args)
  56
  57         def extract_data(key, **args):
  58             data_json = extract_str(key, **args)
  59             if data_json is None:
  60                 return data_json
  61             return self._parse_json(
  62                 data_json, video_id, transform_source=js_to_json)
  63
  64         formats = []
  65         for i in itertools.count():
  66             fd = extract_data('VideoUrls[%d]' % i, default=None)
  67             if fd is None:
  68                 break
  69             formats.append({
  70                 'format_id': '%s' % i,
  71                 'format_note': fd['MimeType'].partition('/')[2],
  72                 'ext': mimetype2ext(fd['MimeType']),
  73                 'url': fd['Location'],
  74                 'protocol': 'f4m' if fd['MimeType'] == 'video/x-mp4-fragmented' else None,
  75             })
  76         self._sort_formats(formats)
  77
  78         slide_baseurl = compat_urlparse.urljoin(
  79             url, extract_data('SlideBaseUrl'))
  80         slide_template = slide_baseurl + re.sub(
  81             r'\{0:D?([0-9+])\}', r'%0\1d', extract_data('SlideImageFileNameTemplate'))
  82         slides = []
  83         last_slide_time = 0
  84         for i in itertools.count(1):
  85             sd = extract_str('Slides[%d]' % i, default=None)
  86             if sd is None:
  87                 break
  88             timestamp = int_or_none(self._search_regex(
  89                 r'^Mediasite\.PlaybackManifest\.CreateSlide\("[^"]*"\s*,\s*([0-9]+),',
  90                 sd, 'slide %s timestamp' % i, fatal=False))
  91             slides.append({
  92                 'url': slide_template % i,
  93                 'duration': timestamp - last_slide_time,
  94             })
  95             last_slide_time = timestamp
  96         formats.append({
  97             'format_id': 'slides',
  98             'protocol': 'slideshow',
  99             'url': json.dumps(slides),
 100             'preference': -10000,  # Downloader not yet written
 101         })
 102         self._sort_formats(formats)
 103
 104         title = extract_data('Title')
 105         description = extract_data('Description', fatal=False)
 106         duration = int_or_none(extract_data(
 107             'Duration', fatal=False), scale=1000)
 108         upload_date = unified_strdate(extract_data('AirDate', fatal=False))
 109
 110         return {
 111             'id': video_id,
 112             'title': title,
 113             'description': description,
 114             'formats': formats,
 115             'upload_date': upload_date,
 116             'duration': duration,
 117         }