X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/0cf0312991a54458a07e903da2e47e9f3c8855ae..d967b1a589a4ead1c13f0c9e17c77cb9a906e556:/youtube_dl/extractor/ceskatelevize.py diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py index 97feb67..b27b4e6 100644 --- a/youtube_dl/extractor/ceskatelevize.py +++ b/youtube_dl/extractor/ceskatelevize.py @@ -4,72 +4,94 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import ( - compat_urllib_request, +from ..compat import ( compat_urllib_parse, + compat_urllib_parse_unquote, compat_urllib_parse_urlparse, +) +from ..utils import ( ExtractorError, + float_or_none, + sanitized_Request, ) class CeskaTelevizeIE(InfoExtractor): - _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P[^?#]+)' - - _TESTS = [ - { - 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/213512120230004-spanelska-chripka', - 'info_dict': { - 'id': '213512120230004', - 'ext': 'flv', - 'title': 'První republika: Španělská chřipka', - 'duration': 3107.4, - }, - 'params': { - 'skip_download': True, # requires rtmpdump - }, - 'skip': 'Works only from Czech Republic.', + _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', + 'ext': 'mp4', + 'title': 'Hyde Park Civilizace', + 'description': 'md5:fe93f6eda372d150759d11644ebbfb4a', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 3350, }, - { - 'url': 'http://www.ceskatelevize.cz/ivysilani/1030584952-tsatsiki-maminka-a-policajt', - 'info_dict': { - 'id': '20138143440', - 'ext': 'flv', - 'title': 'Tsatsiki, maminka a policajt', - 'duration': 6754.1, - }, - 'params': { - 'skip_download': True, # requires rtmpdump - }, - 'skip': 'Works only from Czech Republic.', + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina', + 'info_dict': { + 'id': '61924494876844374', + '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, + }, + 'params': { + # m3u8 download + 'skip_download': True, }, - { - 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina', + }, { + # video with 18+ caution trailer + 'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/', + 'info_dict': { + 'id': '215562210900007-bogotart', + 'title': 'Queer: Bogotart', + 'description': 'Alternativní průvodce současným queer světem', + }, + 'playlist': [{ 'info_dict': { - 'id': '14716', - 'ext': 'flv', - 'title': 'První republika: Zpěvačka z Dupárny Bobina', - 'duration': 90, + 'id': '61924494876844842', + 'ext': 'mp4', + 'title': 'Queer: Bogotart (Varování 18+)', + 'duration': 10.2, }, - 'params': { - 'skip_download': True, # requires rtmpdump + }, { + 'info_dict': { + 'id': '61924494877068022', + 'ext': 'mp4', + 'title': 'Queer: Bogotart (Queer)', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 1558.3, }, + }], + 'params': { + # m3u8 download + 'skip_download': True, }, - ] + }] def _real_extract(self, url): url = url.replace('/porady/', '/ivysilani/').replace('/video/', '') mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + playlist_id = mobj.group('id') - webpage = self._download_webpage(url, video_id) + webpage = self._download_webpage(url, playlist_id) NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.' if '%s

' % NOT_AVAILABLE_STRING in webpage: raise ExtractorError(NOT_AVAILABLE_STRING, expected=True) - typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type') - episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id') + typ = self._html_search_regex( + r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type') + episode_id = self._html_search_regex( + r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id') data = { 'playlist[0][type]': typ, @@ -78,49 +100,93 @@ class CeskaTelevizeIE(InfoExtractor): 'requestSource': 'iVysilani', } - req = compat_urllib_request.Request('http://www.ceskatelevize.cz/ivysilani/ajax/get-playlist-url', - data=compat_urllib_parse.urlencode(data)) + 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, video_id) + playlistpage = self._download_json(req, playlist_id) - req = compat_urllib_request.Request(compat_urllib_parse.unquote(playlistpage['url'])) - req.add_header('Referer', url) + playlist_url = playlistpage['url'] + if playlist_url == 'error_region': + raise ExtractorError(NOT_AVAILABLE_STRING, expected=True) - playlist = self._download_xml(req, video_id) - - formats = [] - for i in playlist.find('smilRoot/body'): - if 'AD' not in i.attrib['id']: - base_url = i.attrib['base'] - parsedurl = compat_urllib_parse_urlparse(base_url) - duration = i.attrib['duration'] - - for video in i.findall('video'): - if video.attrib['label'] != 'AD': - format_id = video.attrib['label'] - play_path = video.attrib['src'] - vbr = int(video.attrib['system-bitrate']) - - formats.append({ - 'format_id': format_id, - 'url': base_url, - 'vbr': vbr, - 'play_path': play_path, - 'app': parsedurl.path[1:] + '?' + parsedurl.query, - 'rtmp_live': True, - 'ext': 'flv', - }) - - self._sort_formats(formats) + 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, + }) + + return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) + + def _get_subtitles(self, episode_id, subs): + original_subtitles = self._download_webpage( + subs[0]['url'], episode_id, 'Downloading subtitles') + srt_subs = self._fix_subtitles(original_subtitles) return { - 'id': episode_id, - 'title': self._html_search_regex(r'(.+?) — iVysílání — Česká televize', webpage, 'title'), - 'duration': float(duration), - 'formats': formats, + 'cs': [{ + 'ext': 'srt', + 'data': srt_subs, + }] } + + @staticmethod + def _fix_subtitles(subtitles): + """ Convert millisecond-based subtitles to SRT """ + + def _msectotimecode(msec): + """ Helper utility to convert milliseconds to timecode """ + components = [] + for divider in [1000, 60, 60, 100]: + components.append(msec % divider) + msec //= divider + 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) + if m: + yield m.group(1) + start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:]) + yield '{0} --> {1}'.format(start, stop) + else: + yield line + + return '\r\n'.join(_fix_subtitle(subtitles))