From f1f6f5aa5e2a6d66fa54d35bf3e8b3626e85ee73 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Tom=C3=A1=C5=A1=20=C4=8Cech?= Date: Sat, 7 May 2016 20:15:49 +0200 Subject: [PATCH] [ceskatelevize] Add support for live streams Live streams has no playlist title, use title of the stream containing TV channel name. Internal m3u8 handler doesn't seem to handle well continuous streams. Add test for live stream. Remove no longer reachable test. --- youtube_dl/extractor/ceskatelevize.py | 35 +++++++++++++++++++-------- 1 file changed, 25 insertions(+), 10 deletions(-) diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py index 6652c8e42a..b418885317 100644 --- a/youtube_dl/extractor/ceskatelevize.py +++ b/youtube_dl/extractor/ceskatelevize.py @@ -33,14 +33,13 @@ class CeskaTelevizeIE(InfoExtractor): 'skip_download': True, }, }, { - 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina', + # live stream + 'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/', 'info_dict': { - 'id': '61924494876844374', + 'id': 402, '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': 're:ČT Sport.*', + 'is_live': True, }, 'params': { # m3u8 download @@ -118,19 +117,21 @@ def _real_extract(self, url): 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_title = self._og_search_title(webpage, default=None) + playlist_description = self._og_search_description(webpage, default=None) playlist = self._download_json(req, playlist_id)['playlist'] playlist_len = len(playlist) entries = [] for item in playlist: + is_live = item['type'] == 'LIVE' 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', fatal=False)) + entry_protocol='m3u8' if is_live else 'm3u8_native', + fatal=False)) self._sort_formats(formats) item_id = item.get('id') or item['assetId'] @@ -145,14 +146,28 @@ def _real_extract(self, url): if subs: subtitles = self.extract_subtitles(episode_id, subs) + if playlist_len == 1: + if is_live: + # live streams has channel name in title + final_title = self._live_title(title) + elif playlist_title: + # title is always set (no KeyError caught) + # and gives good fallback + final_title = title + else: + final_title = playlist_title + else: + final_title = '%s (%s)' % (playlist_title, title) + entries.append({ 'id': item_id, - 'title': playlist_title if playlist_len == 1 else '%s (%s)' % (playlist_title, title), + 'title': final_title, 'description': playlist_description if playlist_len == 1 else None, 'thumbnail': thumbnail, 'duration': duration, 'formats': formats, 'subtitles': subtitles, + 'is_live': is_live, }) return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)