youtube_dl/extractor/dvtv.py

   1 # coding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     js_to_json,
   8     unescapeHTML
   9 )
  10
  11
  12 class DVTVIE(InfoExtractor):
  13     IE_NAME = 'dvtv'
  14     IE_DESC = 'http://video.aktualne.cz/dvtv/'
  15
  16     _VALID_URL = r'http://video\.aktualne\.cz/dvtv/(?P<id>[a-z0-9-]+/r~[0-9a-f]{32})/?'
  17
  18     _TESTS = [{
  19         'url': 'http://video.aktualne.cz/dvtv/vondra-o-ceskem-stoleti-pri-pohledu-na-havla-mi-bylo-trapne/r~e5efe9ca855511e4833a0025900fea04/',
  20         'md5': '75800f964fa0f82939a2914563301f72',
  21         'info_dict': {
  22             'id': 'e5efe9ca855511e4833a0025900fea04',
  23             'ext': 'webm',
  24             'title': 'Vondra o Českém století: Při pohledu na Havla mi bylo trapně'
  25         }
  26     }, {
  27         'url': 'http://video.aktualne.cz/dvtv/stropnicky-policie-vrbetice-preventivne-nekontrolovala/r~82ed4322849211e4a10c0025900fea04/',
  28         'md5': '6388f1941b48537dbd28791f712af8bf',
  29         'info_dict': {
  30             'id': '82ed4322849211e4a10c0025900fea04',
  31             'ext': 'mp4',
  32             'title': 'Stropnický: Policie Vrbětice preventivně nekontrolovala'
  33         }
  34     }]
  35
  36     def _real_extract(self, url):
  37         video_id = self._match_id(url)
  38         webpage = self._download_webpage(url, video_id)
  39
  40         code = self._search_regex(
  41             r'(?s)embedData[0-9a-f]{32}\[\'asset\'\] = (\{.+?\});',
  42             webpage, 'video JSON')
  43         payload = self._parse_json(code, video_id, transform_source=js_to_json)
  44         formats = []
  45         for source in payload['sources']:
  46             ext = source['type'][6:]
  47             formats.append({
  48                 'url': source['file'],
  49                 'ext': ext,
  50                 'format': '%s %s' % (ext, source['label']),
  51                 'format_id': '%s-%s' % (ext, source['label']),
  52                 'height': int(source['label'].rstrip('p')),
  53                 'fps': 25,
  54             })
  55         self._sort_formats(formats)
  56
  57         return {
  58             'id': video_id[-32:],
  59             'display_id': video_id[:-35],
  60             'title': unescapeHTML(payload['title']),
  61             'thumbnail': 'http:%s' % payload['image'],
  62             'formats': formats
  63         }