|
From: | Michael D Godfrey |
Subject: | Re: 3.4.1 release |
Date: | Wed, 06 Apr 2011 17:51:04 -0700 |
User-agent: | Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.15) Gecko/20110307 Fedora/3.1.9-0.39.b3pre.fc14 Thunderbird/3.1.9 |
On 04/06/2011 05:34 PM, John W. Eaton wrote:
I would add 2 others:On 4-Apr-2011, Rik wrote: | After looking at the current list of bugs on the tracker, I think this one | (https://savannah.gnu.org/bugs/?32840).) is the only one I would nominate | for fixing before the next point release. | | The issue is that Octave packaging is placing dynamically linked functions | (.oct files) in $libexecdir. According to the GNU standard | (http://www.gnu.org/software/autoconf/manual/autoconf.html#Installation-Directory-Variables), | $libexecdir is for binary executables run by programs (whereas $bindir is | for binary executables run by users). The reporter attached a patch, but I | haven't verified it. I commented on the tracker. jwe 32682 This is in fltk/OpenGL code, so it could be claimed that it is not a stable part of the release. However, the fltk backend (except for the next bug) works pretty well, but seg faulting makes it really dangerous to use. 32980 This bug is very important if fltk is to be advertised as working at all. Until it is fixed, very misleading plots may be silently produced. If what I said in the comments is at all correct, the fix is easy and pretty safe. I also think that axes calls with Inf or NaN as values should not be allowed at all. I tried a few cases and these values seem to confuse the code pretty badly for both gnuplot and fltk. So, the fix should do a validity check in axes for all backends, and an additional check for fltk. ================================= Michael |
[Prev in Thread] | Current Thread | [Next in Thread] |