[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: behaviour different when loaded as toolbox

From: Mike Miller
Subject: Re: behaviour different when loaded as toolbox
Date: Thu, 15 Jan 2015 22:42:03 -0500
User-agent: Mutt/1.5.23 (2014-03-12)

On Thu, Jan 15, 2015 at 15:20:58 -0800, withaar wrote:
> Hi all,
> I am working on the quaternion toolbox and an running into odd behaviour. I
> like making my own folder somewhere and work on functions, possibly
> shadowing the original. In my case I originally did not have the quaternion
> toolbox installed - at the moment I do but the result is the same.
> Question: does pkg does something special when loading/installing a toolbox?

In general, yes a package may have a PKG_ADD script that is executed
when it is loaded each time using "pkg load". And installing a package
may do several things as well, such as compiling oct-files.

> q = quaternion(randn(2,3,4),randn(2,3,4),randn(2,3,4),randn(2,3,4));
> q.w(1,1,1)
> q.w(1,1,1) = 0
> error: a cs-list cannot be further indexed
> error: invalid assignment to cs-list outside multiple assignment
> #------------------------------------------------
> Where does the difference in behaviour come from? For the alternate version
> subsref/subsassgn never get called.

I can't reproduce this, attempting to do the same thing that I think you
are, so I'm not sure. I guess the obvious question is, is this the same
version of the quaternion package? Or is it possibly an older version
and you are simply seeing a bug that has been fixed in 2.2.2?

Here's what I did:

  $ hg clone
  $ cd quaternion/src
  $ mkoctfile
  $ cd ..
  $ octave
  octave:1> addpath ([pwd "/inst"]);
  octave:2> addpath ([pwd "/src"]);
  octave:3> q = quaternion(randn(2,3,4),randn(2,3,4),randn(2,3,4),randn(2,3,4));
  octave:4> q.w(1,1,1)
  ans = -0.033714
  octave:5> q.w(1,1,1) = 0;
  octave:6> q.w(1,1,1)
  ans = 0


Attachment: signature.asc
Description: Digital signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]