yt_dlp/extractor/presstv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4
   5 from .common import InfoExtractor
   6 from ..utils import remove_start
   7
   8
   9 class PressTVIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?presstv\.ir/[^/]+/(?P<y>\d+)/(?P<m>\d+)/(?P<d>\d+)/(?P<id>\d+)/(?P<display_id>[^/]+)?'
  11
  12     _TEST = {
  13         'url': 'http://www.presstv.ir/Detail/2016/04/09/459911/Australian-sewerage-treatment-facility-/',
  14         'md5': '5d7e3195a447cb13e9267e931d8dd5a5',
  15         'info_dict': {
  16             'id': '459911',
  17             'display_id': 'Australian-sewerage-treatment-facility-',
  18             'ext': 'mp4',
  19             'title': 'Organic mattresses used to clean waste water',
  20             'upload_date': '20160409',
  21             'thumbnail': r're:^https?://.*\.jpg',
  22             'description': 'md5:20002e654bbafb6908395a5c0cfcd125'
  23         }
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = self._match_valid_url(url)
  28         video_id = mobj.group('id')
  29         display_id = mobj.group('display_id') or video_id
  30
  31         webpage = self._download_webpage(url, display_id)
  32
  33         # extract video URL from webpage
  34         video_url = self._hidden_inputs(webpage)['inpPlayback']
  35
  36         # build list of available formats
  37         # specified in http://www.presstv.ir/Scripts/playback.js
  38         base_url = 'http://192.99.219.222:82/presstv'
  39         _formats = [
  40             (180, '_low200.mp4'),
  41             (360, '_low400.mp4'),
  42             (720, '_low800.mp4'),
  43             (1080, '.mp4')
  44         ]
  45
  46         formats = [{
  47             'url': base_url + video_url[:-4] + extension,
  48             'format_id': '%dp' % height,
  49             'height': height,
  50         } for height, extension in _formats]
  51
  52         # extract video metadata
  53         title = remove_start(
  54             self._html_search_meta('title', webpage, fatal=True), 'PressTV-')
  55
  56         thumbnail = self._og_search_thumbnail(webpage)
  57         description = self._og_search_description(webpage)
  58
  59         upload_date = '%04d%02d%02d' % (
  60             int(mobj.group('y')),
  61             int(mobj.group('m')),
  62             int(mobj.group('d')),
  63         )
  64
  65         return {
  66             'id': video_id,
  67             'display_id': display_id,
  68             'title': title,
  69             'formats': formats,
  70             'thumbnail': thumbnail,
  71             'upload_date': upload_date,
  72             'description': description
  73         }