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