X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/f19349c26118065acbe21509383c63465df794fe..97a8fc3ae80fb363c69c2e6b8c29b5373ac72aea:/youtube_dl/extractor/ceskatelevize.py?ds=sidebyside diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py index 6f7b2a7..dd2529a 100644 --- a/youtube_dl/extractor/ceskatelevize.py +++ b/youtube_dl/extractor/ceskatelevize.py @@ -1,11 +1,10 @@ -# -*- coding: utf-8 -*- +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, compat_urllib_parse_unquote, compat_urllib_parse_urlparse, ) @@ -13,19 +12,21 @@ from ..utils import ( ExtractorError, float_or_none, sanitized_Request, + urlencode_postdata, + USER_AGENTS, ) class CeskaTelevizeIE(InfoExtractor): - _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(?:[^/]+/)*(?P[^/#?]+)/*(?:[#?].*)?$' + _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/(porady|ivysilani)/(?:[^/]+/)*(?P[^/#?]+)/*(?:[#?].*)?$' _TESTS = [{ 'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220', 'info_dict': { - 'id': '61924494876951776', + 'id': '61924494877246241', 'ext': 'mp4', - 'title': 'Hyde Park Civilizace', - 'description': 'md5:fe93f6eda372d150759d11644ebbfb4a', - 'thumbnail': 're:^https?://.*\.jpg', + 'title': 'Hyde Park Civilizace: Život v Grónsku', + 'description': 'md5:3fec8f6bb497be5cdb0c9e8781076626', + 'thumbnail': r're:^https?://.*\.jpg', 'duration': 3350, }, 'params': { @@ -33,19 +34,33 @@ class CeskaTelevizeIE(InfoExtractor): 'skip_download': True, }, }, { - 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina', + 'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en', 'info_dict': { - 'id': '61924494876844374', + 'id': '61924494877028507', 'ext': 'mp4', - 'title': 'První republika: Zpěvačka z Dupárny Bobina', - 'description': 'Sága mapující atmosféru první republiky od r. 1918 do r. 1945.', - 'thumbnail': 're:^https?://.*\.jpg', - 'duration': 88.4, + 'title': 'Hyde Park Civilizace: Bonus 01 - En', + 'description': 'English Subtittles', + 'thumbnail': r're:^https?://.*\.jpg', + 'duration': 81.3, }, 'params': { # m3u8 download 'skip_download': True, }, + }, { + # live stream + 'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/', + 'info_dict': { + 'id': 402, + 'ext': 'mp4', + 'title': r're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$', + 'is_live': True, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + 'skip': 'Georestricted to Czech Republic', }, { # video with 18+ caution trailer 'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/', @@ -66,7 +81,7 @@ class CeskaTelevizeIE(InfoExtractor): 'id': '61924494877068022', 'ext': 'mp4', 'title': 'Queer: Bogotart (Queer)', - 'thumbnail': 're:^https?://.*\.jpg', + 'thumbnail': r're:^https?://.*\.jpg', 'duration': 1558.3, }, }], @@ -100,59 +115,99 @@ class CeskaTelevizeIE(InfoExtractor): 'requestSource': 'iVysilani', } - req = sanitized_Request( - 'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist', - data=compat_urllib_parse.urlencode(data)) - - req.add_header('Content-type', 'application/x-www-form-urlencoded') - req.add_header('x-addr', '127.0.0.1') - req.add_header('X-Requested-With', 'XMLHttpRequest') - req.add_header('Referer', url) - - playlistpage = self._download_json(req, playlist_id) - - playlist_url = playlistpage['url'] - if playlist_url == 'error_region': - raise ExtractorError(NOT_AVAILABLE_STRING, expected=True) - - req = sanitized_Request(compat_urllib_parse_unquote(playlist_url)) - req.add_header('Referer', url) - - playlist_title = self._og_search_title(webpage) - playlist_description = self._og_search_description(webpage) - - playlist = self._download_json(req, playlist_id)['playlist'] - playlist_len = len(playlist) - entries = [] - for item in playlist: - formats = [] - for format_id, stream_url in item['streamUrls'].items(): - formats.extend(self._extract_m3u8_formats( - stream_url, playlist_id, 'mp4', entry_protocol='m3u8_native')) - self._sort_formats(formats) - - item_id = item.get('id') or item['assetId'] - title = item['title'] - - duration = float_or_none(item.get('duration')) - thumbnail = item.get('previewImageUrl') - - subtitles = {} - if item.get('type') == 'VOD': - subs = item.get('subtitles') - if subs: - subtitles = self.extract_subtitles(episode_id, subs) - - entries.append({ - 'id': item_id, - 'title': playlist_title if playlist_len == 1 else '%s (%s)' % (playlist_title, title), - 'description': playlist_description if playlist_len == 1 else None, - 'thumbnail': thumbnail, - 'duration': duration, - 'formats': formats, - 'subtitles': subtitles, - }) + + for user_agent in (None, USER_AGENTS['Safari']): + req = sanitized_Request( + 'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist', + data=urlencode_postdata(data)) + + req.add_header('Content-type', 'application/x-www-form-urlencoded') + req.add_header('x-addr', '127.0.0.1') + req.add_header('X-Requested-With', 'XMLHttpRequest') + if user_agent: + req.add_header('User-Agent', user_agent) + req.add_header('Referer', url) + + playlistpage = self._download_json(req, playlist_id, fatal=False) + + if not playlistpage: + continue + + playlist_url = playlistpage['url'] + if playlist_url == 'error_region': + raise ExtractorError(NOT_AVAILABLE_STRING, expected=True) + + req = sanitized_Request(compat_urllib_parse_unquote(playlist_url)) + req.add_header('Referer', url) + + playlist_title = self._og_search_title(webpage, default=None) + playlist_description = self._og_search_description(webpage, default=None) + + playlist = self._download_json(req, playlist_id, fatal=False) + if not playlist: + continue + + playlist = playlist.get('playlist') + if not isinstance(playlist, list): + continue + + playlist_len = len(playlist) + + for num, item in enumerate(playlist): + is_live = item.get('type') == 'LIVE' + formats = [] + for format_id, stream_url in item.get('streamUrls', {}).items(): + if 'playerType=flash' in stream_url: + stream_formats = self._extract_m3u8_formats( + stream_url, playlist_id, 'mp4', 'm3u8_native', + m3u8_id='hls-%s' % format_id, fatal=False) + else: + stream_formats = self._extract_mpd_formats( + stream_url, playlist_id, + mpd_id='dash-%s' % format_id, fatal=False) + # See https://github.com/rg3/youtube-dl/issues/12119#issuecomment-280037031 + if format_id == 'audioDescription': + for f in stream_formats: + f['source_preference'] = -10 + formats.extend(stream_formats) + + if user_agent and len(entries) == playlist_len: + entries[num]['formats'].extend(formats) + continue + + item_id = item.get('id') or item['assetId'] + title = item['title'] + + duration = float_or_none(item.get('duration')) + thumbnail = item.get('previewImageUrl') + + subtitles = {} + if item.get('type') == 'VOD': + subs = item.get('subtitles') + if subs: + subtitles = self.extract_subtitles(episode_id, subs) + + if playlist_len == 1: + final_title = playlist_title or title + if is_live: + final_title = self._live_title(final_title) + else: + final_title = '%s (%s)' % (playlist_title, title) + + entries.append({ + 'id': item_id, + 'title': final_title, + 'description': playlist_description if playlist_len == 1 else None, + 'thumbnail': thumbnail, + 'duration': duration, + 'formats': formats, + 'subtitles': subtitles, + 'is_live': is_live, + }) + + for e in entries: + self._sort_formats(e['formats']) return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) @@ -177,16 +232,16 @@ class CeskaTelevizeIE(InfoExtractor): for divider in [1000, 60, 60, 100]: components.append(msec % divider) msec //= divider - return "{3:02}:{2:02}:{1:02},{0:03}".format(*components) + return '{3:02}:{2:02}:{1:02},{0:03}'.format(*components) def _fix_subtitle(subtitle): for line in subtitle.splitlines(): - m = re.match(r"^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$", line) + m = re.match(r'^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$', line) if m: yield m.group(1) start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:]) - yield "{0} --> {1}".format(start, stop) + yield '{0} --> {1}'.format(start, stop) else: yield line - return "\r\n".join(_fix_subtitle(subtitles)) + return '\r\n'.join(_fix_subtitle(subtitles))