axiom-developer
[Top][All Lists]
Advanced

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

Re: [Axiom-developer] Upper and lower case problems


From: root
Subject: Re: [Axiom-developer] Upper and lower case problems
Date: Wed, 3 Sep 2003 06:43:48 -0400

> Keith Harrison, HP Labs Bristol wrote:

> Gentlemen,
> 
> Firstly let me say that you are doing a great job.

Thanks, from all of us.

> 
> I downloaded the alpha version from Savannah and tried to install it on
> my laptop. (I'm running windows XP. I use Vmware to run a redhat client.
> Don't ask why :-))

Hey, at least you're trying for purity :-)

Actually, I have Windows XP running at my elbow. It is one of the
planned ports. "Hello, World" compiles and runs so I'm nearly there :-)

> 
> My first error was to root everything at /SPAD. There are a number of
> places in the code (and makefiles) where the file names are downshifted.
> Thus the compile could not find "/spad/...". The cure was to make sure
> the full pathname for the root directory did not contain upshifted
> characters. E.g. /axiom or /home/kah/axiom. I think the one that got me
> was in lsp/ccl/src/boot/Makefile  
>            "DEPS = (load (string-downcase (quote ${IN}/npextras.lisp))".
> OK it's part of GCL but it still burnt me.

Yeah, that code exists because DOS (remember DOS?) couldn't handle
dual case files. It's been a long time since I had it running on DOS.
I don't remember all of the places this affects but I'll find them.
I'll put this on the list as a bug and get it fixed. I thought I 
wrote it up in the docs but I guess not. 

> 
> My second error was to download the files to the windows host. I had
> problems with a number of input files where there are files with the
> same name (one upshifted and the other downshifted). EXPR.input.pamphlet
> and expr.input.pamphlet for example. This will be a problem when you try
> to compile on windows.

Ah. right. I'll fix that one also. Some systems CAN handle DuAl CaSe.
I was aware (but not paying attention to the fact ) that Windows
couldn't. Send M$ a bug report and maybe they'll fix it :-)
Except for the input files (which I haven't really spent much time
on except to box them up as pamphlets there shouldn't be any case 
issues. The input files will eventually be replaced by CATS (Computer
Algebra Test Suite).

> 
> My last comment is that I tried compiling as root. And crashed linux.

Boffo job, mate! How did you manage to do that? I always work as root
and have only crashed systems where I did something deliberately wrong
(of which compiling Axiom doesn't qualify). Working as root raises
your purity level +5, strength +3, handsomeness +7, and likelihood
of getting flamed +15. I haven't actually crashed a system in years.
 
> 
> Installing as myself was successful. It took nearly 4 hours - but it
> worked.
> Congratulations.

Great. Feedback is most welcome. Bitch loudly about what needs to
be fixed. No error is too small. We maintain high standards here.
We appreciate the effort you made to get it to run.

Once I figure out how to upload files to the file download area I'm
going to post a full executable version. I'm in the process of testing
it on various platforms. So far it appears to run on RedHat Linux 9,
Debian, and the vanilla Mandrake 9.1 install (but not compile due to a
missing XFree library. I have to find and add the rpm). I'm
downloading FreeBSD at the moment. When I'm done Axiom should run (and
build) on all the unixy systems.

I'm looking for contacts at all of the Linux distributions so we
can ship Axiom as part of Linux. That way everyone has the option
of having it on their desktop. That should save you hours of compiles.

It's a footrace (depending on what task ends first) whether the 0th
version of the CATS, the graphics, or the windows port is the next
major event.


Tim Daly
address@hidden
address@hidden





reply via email to

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