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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import str_to_int
   7
   8
   9 class PressTVIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?presstv\.ir/[^/]+/(?P<y>[0-9]+)/(?P<m>[0-9]+)/(?P<d>[0-9]+)/(?P<id>[0-9]+)/'
  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             'ext': 'mp4',
  18             'title': 'Organic mattresses used to clean waste water',
  19             'upload_date': '20160409',
  20             'thumbnail': 're:^https?://.*\.jpg',
  21             'description': 'md5:20002e654bbafb6908395a5c0cfcd125'
  22         }
  23     }
  24
  25     def _real_extract(self, url):
  26         video_id = self._match_id(url)
  27         webpage = self._download_webpage(url, video_id)
  28
  29         # extract video URL from webpage
  30         video_url = self._html_search_regex(r'<input type="hidden" id="inpPlayback" value="([^"]+)" />', webpage,
  31                                             'Video URL')
  32
  33         # build list of available formats
  34         # specified in http://www.presstv.ir/Scripts/playback.js
  35         base_url = 'http://192.99.219.222:82/presstv'
  36         _formats = [
  37             ("180p", "_low200.mp4"),
  38             ("360p", "_low400.mp4"),
  39             ("720p", "_low800.mp4"),
  40             ("1080p", ".mp4")
  41         ]
  42
  43         formats = []
  44         for fmt in _formats:
  45             format_id, extension = fmt
  46             formats.append({
  47                 'url': base_url + video_url[:-4] + extension,
  48                 'format_id': format_id
  49             })
  50
  51         # extract video metadata
  52         title = self._html_search_meta('title', webpage, 'Title', True)
  53         title = title.partition('-')[2].strip()
  54
  55         thumbnail = self._og_search_thumbnail(webpage)
  56         description = self._og_search_description(webpage)
  57
  58         match = re.match(PressTVIE._VALID_URL, url)
  59         upload_date = '%04d%02d%02d' % (
  60             str_to_int(match.group('y')),
  61             str_to_int(match.group('m')),
  62             str_to_int(match.group('d'))
  63         )
  64
  65         return {
  66             'id': video_id,
  67             'title': title,
  68             'formats': formats,
  69             'thumbnail': thumbnail,
  70             'upload_date': upload_date,
  71             'description': description
  72         }