[Axiom-developer] RE: generalized makefile rules

[Weiss, Juergen]

Tim,

> -----Original Message-----
> From: root [mailto:address@hidden 
> Sent: Sunday, August 17, 2003 6:49 PM
> To: Weiss, Juergen
> Cc: address@hidden; address@hidden; address@hidden
> Subject: Re: generalized makefile rules
> 
> 
> Juergen,
> 
> Well, except for a few domains I've finished the lattice.
> I've diffed your algebra versions with mine and I'm applying
> your patches. I've found them quite useful as they fixed some
> bugs I intended to chase. Excellent work on your part.

About 10 years ago I tried to figure out the algebra lattice
myself. I logged the files loaded by each compile and later
on tried to use tsort or something similar. It did not work
out and I gave up (too many circular dependencies). So I 
can at least partially appreciate what you have achieved. 

I got the impression, that in certain circumstances, the
content of the databases is enough to compile references
to some modules. In most cases the modules have to be
loaded. But I did not figure out, what the difference was.

> I see that you have rewritten interval.as to interval.spad.
> I've rewritten the \author tags on all of the files to indicate
> the actual authors wherever I know the authors. Did you write
> interval.spad.pamphlet?

I actually just changed the syntax of the original .as file.
So I would not call me the author. I have a newer version
of the file (two changes), which I will send you tonight.
With current database files, two function calls, which did
not compile, finally do.

> I also have algebra that has been recently written that I'm 
> working on merging. 
> 
> I'm trying at the moment to rebuild the database but I'm getting
> a value stack overflow in GCL while loading the domains. (Daly's
> law: there is no such thing as a simple job.) Other than that
> problem and a few remaining domains that I need to fix, the whole
> Axiom world builds from scratch. I never thought I'd see the day
> that would happen. 

This is strange indeed. I used the most recent cvs files and 
compiled under Debian (with the supplied gcl). I think I only
fixed the category vector copying. Have you rebuild interpsys
without any algebra files preloaded or with more recent ones? 

> Once I fix the value stack overflow I'll concentrate on getting
> the world imported onto Savannah
>
> Tim
> address@hidden
> address@hidden

Juergen 

Juergen Weiss     | Universitaet Mainz, Zentrum fuer Datenverarbeitung,
address@hidden| 55099 Mainz, Tel: +49(6131)39-26361, FAX:
+49(6131)39-26407


>