Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ceskatelevize.py

   1 # -*- coding: utf-8 -*-
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_urllib_request,
   9     compat_urllib_parse,
  10     compat_urllib_parse_unquote,
  11     compat_urllib_parse_urlparse,
  12 )
  13 from ..utils import (
  14     ExtractorError,
  15     float_or_none,
  16 )
  17
  18
  19 class CeskaTelevizeIE(InfoExtractor):
  20     _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P<id>[^?#]+)'
  21
  22     _TESTS = [
  23         {
  24             'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
  25             'info_dict': {
  26                 'id': '214411058091220',
  27                 'ext': 'mp4',
  28                 'title': 'Hyde Park Civilizace',
  29                 'description': 'Věda a současná civilizace. Interaktivní pořad - prostor pro vaše otázky a komentáře',
  30                 'thumbnail': 're:^https?://.*\.jpg',
  31                 'duration': 3350,
  32             },
  33             'params': {
  34                 # m3u8 download
  35                 'skip_download': True,
  36             },
  37         },
  38         {
  39             'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina',
  40             'info_dict': {
  41                 'id': '14716',
  42                 'ext': 'mp4',
  43                 'title': 'První republika: Zpěvačka z Dupárny Bobina',
  44                 'description': 'Sága mapující atmosféru první republiky od r. 1918 do r. 1945.',
  45                 'thumbnail': 're:^https?://.*\.jpg',
  46                 'duration': 88.4,
  47             },
  48             'params': {
  49                 # m3u8 download
  50                 'skip_download': True,
  51             },
  52         },
  53     ]
  54
  55     def _real_extract(self, url):
  56         url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
  57
  58         mobj = re.match(self._VALID_URL, url)
  59         video_id = mobj.group('id')
  60
  61         webpage = self._download_webpage(url, video_id)
  62
  63         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
  64         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
  65             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
  66
  67         typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
  68         episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
  69
  70         data = {
  71             'playlist[0][type]': typ,
  72             'playlist[0][id]': episode_id,
  73             'requestUrl': compat_urllib_parse_urlparse(url).path,
  74             'requestSource': 'iVysilani',
  75         }
  76
  77         req = compat_urllib_request.Request(
  78             'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
  79             data=compat_urllib_parse.urlencode(data))
  80
  81         req.add_header('Content-type', 'application/x-www-form-urlencoded')
  82         req.add_header('x-addr', '127.0.0.1')
  83         req.add_header('X-Requested-With', 'XMLHttpRequest')
  84         req.add_header('Referer', url)
  85
  86         playlistpage = self._download_json(req, video_id)
  87
  88         playlist_url = playlistpage['url']
  89         if playlist_url == 'error_region':
  90             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
  91
  92         req = compat_urllib_request.Request(compat_urllib_parse_unquote(playlist_url))
  93         req.add_header('Referer', url)
  94
  95         playlist = self._download_json(req, video_id)
  96
  97         item = playlist['playlist'][0]
  98         formats = []
  99         for format_id, stream_url in item['streamUrls'].items():
 100             formats.extend(self._extract_m3u8_formats(stream_url, video_id, 'mp4'))
 101         self._sort_formats(formats)
 102
 103         title = self._og_search_title(webpage)
 104         description = self._og_search_description(webpage)
 105         duration = float_or_none(item.get('duration'))
 106         thumbnail = item.get('previewImageUrl')
 107
 108         subtitles = {}
 109         subs = item.get('subtitles')
 110         if subs:
 111             subtitles = self.extract_subtitles(episode_id, subs)
 112
 113         return {
 114             'id': episode_id,
 115             'title': title,
 116             'description': description,
 117             'thumbnail': thumbnail,
 118             'duration': duration,
 119             'formats': formats,
 120             'subtitles': subtitles,
 121         }
 122
 123     def _get_subtitles(self, episode_id, subs):
 124         original_subtitles = self._download_webpage(
 125             subs[0]['url'], episode_id, 'Downloading subtitles')
 126         srt_subs = self._fix_subtitles(original_subtitles)
 127         return {
 128             'cs': [{
 129                 'ext': 'srt',
 130                 'data': srt_subs,
 131             }]
 132         }
 133
 134     @staticmethod
 135     def _fix_subtitles(subtitles):
 136         """ Convert millisecond-based subtitles to SRT """
 137
 138         def _msectotimecode(msec):
 139             """ Helper utility to convert milliseconds to timecode """
 140             components = []
 141             for divider in [1000, 60, 60, 100]:
 142                 components.append(msec % divider)
 143                 msec //= divider
 144             return "{3:02}:{2:02}:{1:02},{0:03}".format(*components)
 145
 146         def _fix_subtitle(subtitle):
 147             for line in subtitle.splitlines():
 148                 m = re.match(r"^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$", line)
 149                 if m:
 150                     yield m.group(1)
 151                     start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
 152                     yield "{0} --> {1}".format(start, stop)
 153                 else:
 154                     yield line
 155
 156         return "\r\n".join(_fix_subtitle(subtitles))