From 73d93f948ee71b2a07d46fdc4d446255d8ddcc9f Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Thu, 14 Apr 2016 15:08:01 +0800 Subject: [PATCH] [lecture2go] Fix extraction RTSP stream fails to download. Seems it's a mpv bug as direct playback works well: $ mpv --ytdl-format rtsp https://lecture2go.uni-hamburg.de/veranstaltungen/-/v/17473 --- youtube_dl/extractor/lecture2go.py | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/youtube_dl/extractor/lecture2go.py b/youtube_dl/extractor/lecture2go.py index 40a3d2346..81b5d41be 100644 --- a/youtube_dl/extractor/lecture2go.py +++ b/youtube_dl/extractor/lecture2go.py @@ -6,6 +6,7 @@ import re from .common import InfoExtractor from ..utils import ( determine_ext, + determine_protocol, parse_duration, int_or_none, ) @@ -18,10 +19,14 @@ class Lecture2GoIE(InfoExtractor): 'md5': 'ac02b570883020d208d405d5a3fd2f7f', 'info_dict': { 'id': '17473', - 'ext': 'flv', + 'ext': 'mp4', 'title': '2 - Endliche Automaten und reguläre Sprachen', 'creator': 'Frank Heitmann', 'duration': 5220, + }, + 'params': { + # m3u8 download + 'skip_download': True, } } @@ -32,14 +37,18 @@ class Lecture2GoIE(InfoExtractor): title = self._html_search_regex(r']+class="title">(.+)', webpage, 'title') formats = [] - for url in set(re.findall(r'"src","([^"]+)"', webpage)): + for url in set(re.findall(r'var\s+playerUri\d+\s*=\s*"([^"]+)"', webpage)): ext = determine_ext(url) + protocol = determine_protocol({'url': url}) if ext == 'f4m': - formats.extend(self._extract_f4m_formats(url, video_id)) + formats.extend(self._extract_f4m_formats(url, video_id, f4m_id='hds')) elif ext == 'm3u8': - formats.extend(self._extract_m3u8_formats(url, video_id)) + formats.extend(self._extract_m3u8_formats(url, video_id, ext='mp4', m3u8_id='hls')) else: + if protocol == 'rtmp': + continue # XXX: currently broken formats.append({ + 'format_id': protocol, 'url': url, })