_ Git - youtube-dl/blob - youtube_dl/extractor/presstv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import remove_start
   8
   9
  10 class PressTVIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?presstv\.ir/[^/]+/(?P<y>\d+)/(?P<m>\d+)/(?P<d>\d+)/(?P<id>\d+)/(?P<display_id>[^/]+)?'
  12
  13     _TEST = {
  14         'url': 'http://www.presstv.ir/Detail/2016/04/09/459911/Australian-sewerage-treatment-facility-/',
  15         'md5': '5d7e3195a447cb13e9267e931d8dd5a5',
  16         'info_dict': {
  17             'id': '459911',
  18             'display_id': 'Australian-sewerage-treatment-facility-',
  19             'ext': 'mp4',
  20             'title': 'Organic mattresses used to clean waste water',
  21             'upload_date': '20160409',
  22             'thumbnail': r're:^https?://.*\.jpg',
  23             'description': 'md5:20002e654bbafb6908395a5c0cfcd125'
  24         }
  25     }
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29         video_id = mobj.group('id')
  30         display_id = mobj.group('display_id') or video_id
  31
  32         webpage = self._download_webpage(url, display_id)
  33
  34         # extract video URL from webpage
  35         video_url = self._hidden_inputs(webpage)['inpPlayback']
  36
  37         # build list of available formats
  38         # specified in http://www.presstv.ir/Scripts/playback.js
  39         base_url = 'http://192.99.219.222:82/presstv'
  40         _formats = [
  41             (180, '_low200.mp4'),
  42             (360, '_low400.mp4'),
  43             (720, '_low800.mp4'),
  44             (1080, '.mp4')
  45         ]
  46
  47         formats = [{
  48             'url': base_url + video_url[:-4] + extension,
  49             'format_id': '%dp' % height,
  50             'height': height,
  51         } for height, extension in _formats]
  52
  53         # extract video metadata
  54         title = remove_start(
  55             self._html_search_meta('title', webpage, fatal=True), 'PressTV-')
  56
  57         thumbnail = self._og_search_thumbnail(webpage)
  58         description = self._og_search_description(webpage)
  59
  60         upload_date = '%04d%02d%02d' % (
  61             int(mobj.group('y')),
  62             int(mobj.group('m')),
  63             int(mobj.group('d')),
  64         )
  65
  66         return {
  67             'id': video_id,
  68             'display_id': display_id,
  69             'title': title,
  70             'formats': formats,
  71             'thumbnail': thumbnail,
  72             'upload_date': upload_date,
  73             'description': description
  74         }