From e18f1da97a8840522b0cefac3af0995751733d67 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sat, 8 Apr 2017 19:41:14 +0700 Subject: [PATCH] [ceskateleveize] Improve extraction and remove URL replacement hacks --- youtube_dl/extractor/ceskatelevize.py | 61 +++++++++++---------------- 1 file changed, 24 insertions(+), 37 deletions(-) diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py index dd2529a6d..0daee313f 100644 --- a/youtube_dl/extractor/ceskatelevize.py +++ b/youtube_dl/extractor/ceskatelevize.py @@ -18,7 +18,7 @@ from ..utils import ( class CeskaTelevizeIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/(porady|ivysilani)/(?:[^/]+/)*(?P[^/#?]+)/*(?:[#?].*)?$' + _VALID_URL = r'https?://(?:www\.)?ceskatelevize\.cz/ivysilani/(?:[^/?#&]+/)*(?P[^/#?]+)' _TESTS = [{ 'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220', 'info_dict': { @@ -62,40 +62,12 @@ class CeskaTelevizeIE(InfoExtractor): }, 'skip': 'Georestricted to Czech Republic', }, { - # 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': '61924494876844842', - 'ext': 'mp4', - 'title': 'Queer: Bogotart (Varování 18+)', - 'duration': 10.2, - }, - }, { - 'info_dict': { - 'id': '61924494877068022', - 'ext': 'mp4', - 'title': 'Queer: Bogotart (Queer)', - 'thumbnail': r're:^https?://.*\.jpg', - 'duration': 1558.3, - }, - }], - 'params': { - # m3u8 download - 'skip_download': True, - }, + 'url': 'http://www.ceskatelevize.cz/ivysilani/embed/iFramePlayer.php?hash=d6a3e1370d2e4fa76296b90bad4dfc19673b641e&IDEC=217 562 22150/0004&channelID=1&width=100%25', + 'only_matching': True, }] def _real_extract(self, url): - url = url.replace('/porady/', '/ivysilani/').replace('/video/', '') - - mobj = re.match(self._VALID_URL, url) - playlist_id = mobj.group('id') + playlist_id = self._match_id(url) webpage = self._download_webpage(url, playlist_id) @@ -103,13 +75,28 @@ class CeskaTelevizeIE(InfoExtractor): 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') + type_ = None + episode_id = None + + playlist = self._parse_json( + self._search_regex( + r'getPlaylistUrl\(\[({.+?})\]', webpage, 'playlist', + default='{}'), playlist_id) + if playlist: + type_ = playlist.get('type') + episode_id = playlist.get('id') + + if not type_: + type_ = self._html_search_regex( + r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', + webpage, 'type') + if not episode_id: + episode_id = self._html_search_regex( + r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', + webpage, 'episode_id') data = { - 'playlist[0][type]': typ, + 'playlist[0][type]': type_, 'playlist[0][id]': episode_id, 'requestUrl': compat_urllib_parse_urlparse(url).path, 'requestSource': 'iVysilani',