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_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>[^/#?]+)/*(?:[#?].*)?$'
  21     _TESTS = [{
  22         'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
  23         'info_dict': {
  24             'id': '61924494876951776',
  25             'ext': 'mp4',
  26             'title': 'Hyde Park Civilizace',
  27             'description': 'md5:fe93f6eda372d150759d11644ebbfb4a',
  28             'thumbnail': 're:^https?://.*\.jpg',
  29             'duration': 3350,
  30         },
  31         'params': {
  32             # m3u8 download
  33             'skip_download': True,
  34         },
  35     }, {
  36         'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en',
  37         'info_dict': {
  38             'id': '61924494877028507',
  39             'ext': 'mp4',
  40             'title': 'Hyde Park Civilizace: Bonus 01 - En',
  41             'description': 'English Subtittles',
  42             'thumbnail': 're:^https?://.*\.jpg',
  43             'duration': 81.3,
  44         },
  45         'params': {
  46             # m3u8 download
  47             'skip_download': True,
  48         },
  49     }, {
  50         # live stream
  51         'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
  52         'info_dict': {
  53             'id': 402,
  54             'ext': 'mp4',
  55             'title': 're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
  56             'is_live': True,
  57         },
  58         'params': {
  59             # m3u8 download
  60             'skip_download': True,
  61         },
  62         'skip': 'Georestricted to Czech Republic',
  63     }, {
  64         # video with 18+ caution trailer
  65         'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
  66         'info_dict': {
  67             'id': '215562210900007-bogotart',
  68             'title': 'Queer: Bogotart',
  69             'description': 'Alternativní průvodce současným queer světem',
  70         },
  71         'playlist': [{
  72             'info_dict': {
  73                 'id': '61924494876844842',
  74                 'ext': 'mp4',
  75                 'title': 'Queer: Bogotart (Varování 18+)',
  76                 'duration': 10.2,
  77             },
  78         }, {
  79             'info_dict': {
  80                 'id': '61924494877068022',
  81                 'ext': 'mp4',
  82                 'title': 'Queer: Bogotart (Queer)',
  83                 'thumbnail': 're:^https?://.*\.jpg',
  84                 'duration': 1558.3,
  85             },
  86         }],
  87         'params': {
  88             # m3u8 download
  89             'skip_download': True,
  90         },
  91     }]
  92
  93     def _real_extract(self, url):
  94         url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
  95
  96         mobj = re.match(self._VALID_URL, url)
  97         playlist_id = mobj.group('id')
  98
  99         webpage = self._download_webpage(url, playlist_id)
 100
 101         NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
 102         if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
 103             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 104
 105         typ = self._html_search_regex(
 106             r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
 107         episode_id = self._html_search_regex(
 108             r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
 109
 110         data = {
 111             'playlist[0][type]': typ,
 112             'playlist[0][id]': episode_id,
 113             'requestUrl': compat_urllib_parse_urlparse(url).path,
 114             'requestSource': 'iVysilani',
 115         }
 116
 117         req = sanitized_Request(
 118             'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
 119             data=urlencode_postdata(data))
 120
 121         req.add_header('Content-type', 'application/x-www-form-urlencoded')
 122         req.add_header('x-addr', '127.0.0.1')
 123         req.add_header('X-Requested-With', 'XMLHttpRequest')
 124         req.add_header('Referer', url)
 125
 126         playlistpage = self._download_json(req, playlist_id)
 127
 128         playlist_url = playlistpage['url']
 129         if playlist_url == 'error_region':
 130             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 131
 132         req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
 133         req.add_header('Referer', url)
 134
 135         playlist_title = self._og_search_title(webpage, default=None)
 136         playlist_description = self._og_search_description(webpage, default=None)
 137
 138         playlist = self._download_json(req, playlist_id)['playlist']
 139         playlist_len = len(playlist)
 140
 141         entries = []
 142         for item in playlist:
 143             is_live = item.get('type') == 'LIVE'
 144             formats = []
 145             for format_id, stream_url in item['streamUrls'].items():
 146                 formats.extend(self._extract_m3u8_formats(
 147                     stream_url, playlist_id, 'mp4',
 148                     entry_protocol='m3u8' if is_live else 'm3u8_native',
 149                     fatal=False))
 150             self._sort_formats(formats)
 151
 152             item_id = item.get('id') or item['assetId']
 153             title = item['title']
 154
 155             duration = float_or_none(item.get('duration'))
 156             thumbnail = item.get('previewImageUrl')
 157
 158             subtitles = {}
 159             if item.get('type') == 'VOD':
 160                 subs = item.get('subtitles')
 161                 if subs:
 162                     subtitles = self.extract_subtitles(episode_id, subs)
 163
 164             if playlist_len == 1:
 165                 final_title = playlist_title or title
 166                 if is_live:
 167                     final_title = self._live_title(final_title)
 168             else:
 169                 final_title = '%s (%s)' % (playlist_title, title)
 170
 171             entries.append({
 172                 'id': item_id,
 173                 'title': final_title,
 174                 'description': playlist_description if playlist_len == 1 else None,
 175                 'thumbnail': thumbnail,
 176                 'duration': duration,
 177                 'formats': formats,
 178                 'subtitles': subtitles,
 179                 'is_live': is_live,
 180             })
 181
 182         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 183
 184     def _get_subtitles(self, episode_id, subs):
 185         original_subtitles = self._download_webpage(
 186             subs[0]['url'], episode_id, 'Downloading subtitles')
 187         srt_subs = self._fix_subtitles(original_subtitles)
 188         return {
 189             'cs': [{
 190                 'ext': 'srt',
 191                 'data': srt_subs,
 192             }]
 193         }
 194
 195     @staticmethod
 196     def _fix_subtitles(subtitles):
 197         """ Convert millisecond-based subtitles to SRT """
 198
 199         def _msectotimecode(msec):
 200             """ Helper utility to convert milliseconds to timecode """
 201             components = []
 202             for divider in [1000, 60, 60, 100]:
 203                 components.append(msec % divider)
 204                 msec //= divider
 205             return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components)
 206
 207         def _fix_subtitle(subtitle):
 208             for line in subtitle.splitlines():
 209                 m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line)
 210                 if m:
 211                     yield m.group(1)
 212                     start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
 213                     yield '{0} --> {1}'.format(start, stop)
 214                 else:
 215                     yield line
 216
 217         return '\r\n'.join(_fix_subtitle(subtitles))