octave-maintainers
[Top][All Lists]
Advanced

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

Re: Octave/Win32 update


From: David Bateman
Subject: Re: Octave/Win32 update
Date: Mon, 29 Jan 2007 23:22:09 +0100
User-agent: Thunderbird 1.5.0.7 (X11/20060921)

Michael Goffioul wrote:
> John W. Eaton a écrit :
>> On 29-Jan-2007, David Bateman wrote:
>>
>> | All of the packages that replace existing functionality of octave
>> are in
>> | extra/. These functions are disabled by default at the install stage. I
>> | didn't check but I believe there is bubble help on each of the install
>> | menu items and if the information you request isn't already there, then
>> | it shouldn't be hard to add it..
>>
>> OK.
>>
>> I think people need to understand what they are doing when they
>> install a package like NaN.  That package may provide the behavior
>> that they want, but it replaces a number of core functions in Octave
>> with and so could produce surprising results.
>>   
> 
> As John stated, extra/ packages are not installed by default. Moreover,
> octave-forge path
> is appended to normal load-path, such that it doesn't shadow any normal
> octave function.
> (I don't know if it's good for all those packages, but I'm not really
> aware of what's exactly
> in those package, so I choose what I considered to be the safest case).
> 
> Michael.
> 
> 
> 

Michael,

This won't work.. The functions in NaN in particular are meant to shadow
the octave functions and treat the NaN values as missing data. This is a
very useful behavior in some circumstances, though not all. However, if
they come last in the path they don't shadow the octave core functions
and so they can't function as expected..

The addpath command by default adds to the front of the path (except the
directory . which is always first), and is what is used in "pkg load
<pkg>". I'd suggest doing something similar..

D.



reply via email to

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