[ydl-gen] Scilab for linux ppc

[Derick Centeno]

On Thu, 07 Dec 2006 16:45:00 -0800
Warren Nagourney <warren at phys.washington.edu> wrote:

> Hi Derick,
> 
> I downloaded the sources from www.scilab.org and attempted to compile  
> them under OS X (using gcc 4). They claimed to be configured for a  
> general unix distribution, so it should have worked (I ran configure  
> first and it seemed to detect the system fine). I can try the same  
> thing on my linux box and see whether it would work (the problem was  
> an incompatibility between some Apple header and a scilab header file).
> 
> Could it be obtained using yum? I thought scilab was a little too  
> arcane for the standard yum repositories.
> 
> Thanks.
> 
> -wn
> 

Hi Warren:

I've pretty much abandoned OS X as a development platform.  I'm not interested
in Intel.  OS X runs awfully slow on a PowerPC system and gets in the way of
an otherwise robust architecture.  I'm not the fellow to discuss details of
OS X or Darwin programming development. If that is what you are interested in
you could participate with Apple's Developer program and their developer's mail
list. The questions regarding OS X are discussed in detail there.

I'm committed to the PowerPC platform, and have become rather focused on Linux
and YDL in particular.

Regarding yum and using it to search for scilab; my own view was that you
really never know until you try.  

If it's not found by yum you may want to try a Debian variant known as Ubuntu
here:

http://packages.ubuntulinux.org/cgi-bin/download.pl?arch=powerpc&file=pool%2Fmultiverse%2Fs%2Fscilab%2Fscilab_2.7-13_powerpc.deb&md5sum=d60c05f374539645187b5e6a042ec2bb&arch=powerpc&type=main

The only difficulty is that you will have to be familiar with unpacking Debian
packages.

I got the source from here:

http://www.scilab.org/download/index_download.php?page=release.html

A possible reason sources can't or won't compile within YDL are:

* dependencies may be processor (Intel) specific.
* library calls may be processor (Intel) specific.

Unfortunately, one won't know which is the case until one attempts to compile
within the YDL environment one uses.  Read the output as the program attempts
to  compile.  Anything missing will be listed with warning statements etc.;
remember to read the documentation included with the source for associated
programs which should be with the program but which were not included with the
source.  This is all tedious stuff, but you've come this far so why stop at
this point.

It may not hurt to ask TSS to include it as a package, in it's mirrors in the
future. Or that scilab.org develop a package for YDL.  It might be even better
to get the two entities to acknowledge each other and do us, the users, a favor!

Best wishes... Derick.