_ Git - youtube-dl/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_parse_unquote,
   9     compat_urllib_parse_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     float_or_none,
  14     sanitized_Request,
  15     urlencode_postdata,
  16 )
  17
  18
  19 class CeskaTelevizeIE(InfoExtractor):
  20     _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(?:[^/]+/)*(?P<id>[^/#?]+)/*(?:[#?].*)?$'
  21     _TESTS = [{
  22         'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
  23         'info_dict': {
  24             'id': '61924494876951776',
  25             'ext': 'mp4',
  26             'title': 'Hyde Park Civilizace',
  27             'description': 'md5:fe93f6eda372d150759d11644ebbfb4a',
  28             'thumbnail': 're:^https?://.*\.jpg',
  29             'duration': 3350,
  30         },
  31         'params': {
  32             # m3u8 download
  33             'skip_download': True,
  34         },
  35     }, {
  36         # live stream
  37         'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
  38         'info_dict': {
  39             'id': 402,
  40             'ext': 'mp4',
  41             'title': 're:ČT Sport.*',
  42             'is_live': True,
  43         },
  44         'params': {
  45             # m3u8 download
  46             'skip_download': True,
  47         },
  48     }, {
  49         # video with 18+ caution trailer
  50         'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
  51         'info_dict': {
  52             'id': '215562210900007-bogotart',
  53             'title': 'Queer: Bogotart',
  54             'description': 'Alternativní průvodce současným queer světem',
  55         },
  56         'playlist': [{
  57             'info_dict': {
  58                 'id': '61924494876844842',
  59                 'ext': 'mp4',
  60                 'title': 'Queer: Bogotart (Varování 18+)',
  61                 'duration': 10.2,
  62             },
  63         }, {
  64             'info_dict': {
  65                 'id': '61924494877068022',
  66                 'ext': 'mp4',
  67                 'title': 'Queer: Bogotart (Queer)',
  68                 'thumbnail': 're:^https?://.*\.jpg',
  69                 'duration': 1558.3,
  70             },
  71         }],
  72         'params': {
  73             # m3u8 download
  74             'skip_download': True,
  75         },
  76     }]
  77
  78     def _real_extract(self, url):
  79         url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
  80
  81         mobj = re.match(self._VALID_URL, url)
  82         playlist_id = mobj.group('id')
  83
  84         webpage = self._download_webpage(url, playlist_id)
  85
  86         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
  87         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
  88             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
  89
  90         typ = self._html_search_regex(
  91             r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
  92         episode_id = self._html_search_regex(
  93             r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
  94
  95         data = {
  96             'playlist[0][type]': typ,
  97             'playlist[0][id]': episode_id,
  98             'requestUrl': compat_urllib_parse_urlparse(url).path,
  99             'requestSource': 'iVysilani',
 100         }
 101
 102         req = sanitized_Request(
 103             'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
 104             data=urlencode_postdata(data))
 105
 106         req.add_header('Content-type', 'application/x-www-form-urlencoded')
 107         req.add_header('x-addr', '127.0.0.1')
 108         req.add_header('X-Requested-With', 'XMLHttpRequest')
 109         req.add_header('Referer', url)
 110
 111         playlistpage = self._download_json(req, playlist_id)
 112
 113         playlist_url = playlistpage['url']
 114         if playlist_url == 'error_region':
 115             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 116
 117         req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
 118         req.add_header('Referer', url)
 119
 120         playlist_title = self._og_search_title(webpage, default=None)
 121         playlist_description = self._og_search_description(webpage, default=None)
 122
 123         playlist = self._download_json(req, playlist_id)['playlist']
 124         playlist_len = len(playlist)
 125
 126         entries = []
 127         for item in playlist:
 128             is_live = item['type'] == 'LIVE'
 129             formats = []
 130             for format_id, stream_url in item['streamUrls'].items():
 131                 formats.extend(self._extract_m3u8_formats(
 132                     stream_url, playlist_id, 'mp4',
 133                     entry_protocol='m3u8' if is_live else 'm3u8_native',
 134                     fatal=False))
 135             self._sort_formats(formats)
 136
 137             item_id = item.get('id') or item['assetId']
 138             title = item['title']
 139
 140             duration = float_or_none(item.get('duration'))
 141             thumbnail = item.get('previewImageUrl')
 142
 143             subtitles = {}
 144             if item.get('type') == 'VOD':
 145                 subs = item.get('subtitles')
 146                 if subs:
 147                     subtitles = self.extract_subtitles(episode_id, subs)
 148
 149             if playlist_len == 1:
 150                 if is_live:
 151                     # live streams has channel name in title
 152                     final_title = self._live_title(title)
 153                 elif playlist_title:
 154                     # title is always set (no KeyError caught)
 155                     # and gives good fallback
 156                     final_title = title
 157                 else:
 158                     final_title = playlist_title
 159             else:
 160                 final_title = '%s (%s)' % (playlist_title, title)
 161
 162             entries.append({
 163                 'id': item_id,
 164                 'title': final_title,
 165                 'description': playlist_description if playlist_len == 1 else None,
 166                 'thumbnail': thumbnail,
 167                 'duration': duration,
 168                 'formats': formats,
 169                 'subtitles': subtitles,
 170                 'is_live': is_live,
 171             })
 172
 173         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 174
 175     def _get_subtitles(self, episode_id, subs):
 176         original_subtitles = self._download_webpage(
 177             subs[0]['url'], episode_id, 'Downloading subtitles')
 178         srt_subs = self._fix_subtitles(original_subtitles)
 179         return {
 180             'cs': [{
 181                 'ext': 'srt',
 182                 'data': srt_subs,
 183             }]
 184         }
 185
 186     @staticmethod
 187     def _fix_subtitles(subtitles):
 188         """ Convert millisecond-based subtitles to SRT """
 189
 190         def _msectotimecode(msec):
 191             """ Helper utility to convert milliseconds to timecode """
 192             components = []
 193             for divider in [1000, 60, 60, 100]:
 194                 components.append(msec % divider)
 195                 msec //= divider
 196             return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components)
 197
 198         def _fix_subtitle(subtitle):
 199             for line in subtitle.splitlines():
 200                 m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line)
 201                 if m:
 202                     yield m.group(1)
 203                     start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
 204                     yield '{0} --> {1}'.format(start, stop)
 205                 else:
 206                     yield line
 207
 208         return '\r\n'.join(_fix_subtitle(subtitles))