From fb97809e64a145768346345ff2b5e5539880c014 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 1 Oct 2015 22:44:51 +0600 Subject: [PATCH] [videolecturesnet] Improve playlist extraction --- youtube_dl/extractor/videolecturesnet.py | 34 +++++++++++++++--------- 1 file changed, 21 insertions(+), 13 deletions(-) diff --git a/youtube_dl/extractor/videolecturesnet.py b/youtube_dl/extractor/videolecturesnet.py index eadff8d180..7041654060 100644 --- a/youtube_dl/extractor/videolecturesnet.py +++ b/youtube_dl/extractor/videolecturesnet.py @@ -3,8 +3,14 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urlparse -from ..utils import parse_duration +from ..compat import ( + compat_HTTPError, + compat_urlparse, +) +from ..utils import ( + ExtractorError, + parse_duration, +) class VideoLecturesNetIE(InfoExtractor): @@ -28,17 +34,19 @@ class VideoLecturesNetIE(InfoExtractor): video_id = self._match_id(url) smil_url = 'http://videolectures.net/%s/video/1/smil.xml' % video_id - smil = self._download_smil(smil_url, video_id, fatal=False) - - # Probably a playlist - if smil is False: - webpage = self._download_webpage(url, video_id) - entries = [ - self.url_result(compat_urlparse.urljoin(url, video_url), 'VideoLecturesNet') - for _, video_url in re.findall(r']+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', webpage)] - playlist_title = self._html_search_meta('title', webpage, 'title', fatal=True) - playlist_description = self._html_search_meta('description', webpage, 'description') - return self.playlist_result(entries, video_id, playlist_title, playlist_description) + + try: + smil = self._download_smil(smil_url, video_id) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: + # Probably a playlist + webpage = self._download_webpage(url, video_id) + entries = [ + self.url_result(compat_urlparse.urljoin(url, video_url), 'VideoLecturesNet') + for _, video_url in re.findall(r']+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', webpage)] + playlist_title = self._html_search_meta('title', webpage, 'title', fatal=True) + playlist_description = self._html_search_meta('description', webpage, 'description') + return self.playlist_result(entries, video_id, playlist_title, playlist_description) info = self._parse_smil(smil, smil_url, video_id)