gcl-devel
[Top][All Lists]
Advanced

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

RE: [Axiom-developer] RE: [Gcl-devel] ANSI test Windows OPEN.*failures


From: Mike Thomas
Subject: RE: [Axiom-developer] RE: [Gcl-devel] ANSI test Windows OPEN.*failures
Date: Thu, 2 Dec 2004 10:02:19 +1000

Hi Camm.

| > You'll be pleased to know I finally managed to set up a reasonably
| > stable DeMudi Linux system and checked out HEAD GCL's "open" - it does
| > not unexpectedly make non-existent directories per my earlier guess.
| > I think that the Windows Axiom build is producing harmless
| > messages/warnings during the build which are not produced on Linux,
| > which lead to the need for the directory "AHYP.erlib"; it was that
| > directory which caused the major barf on Windows.
| >
|
| Mike, are the warnings not issues on Linux, or is the warning
| reporting code failing on Windows?  If the latter, can we isolate to a
| lisp command?

Ultimately, hopefully, and not this week.

| > The database build bug looks like a pathname bug which I will also
| > have to track down when I have more time - until then I am copying the
| > daase files by hand halfway through the build.
| >
|
| I think this is the pathname-name case issue Tim reported.  Please let
| me know if not.

I suspect not as I had the same symptom 6 months ago with an earlier stable
GCL (perhaps 2.6.2?).  Perhaps some other bug since fixed which
coincidentally duplicates that symptom, but...

| > Putting all this aside, I've today built Axiom on two Windows boxes -
| > XP (PIV) and 2000 Pro (AMD64) and once built it runs like a 'ken
| > bought one' on both machines (text only).  Now that I'm understanding
| > the Axiom source code layout a little better I'm finding it relatively
| > easy to work with - it reminds me of Haskell.
| >
|
| Great!

Yes, Axiom was the last serious barrier in relation to application support
on  Windows relative to the Unix platforms.  Next on the plate is to return
to 2.6.5 and ensure that the recent patches haven't broken anything on
Windows.  I think I will also have to bring it up to scratch for the current
MinGW32 gcc version.

Frankly, I would like to dump 2.6.x ASAP.

Have a pleasant evening,

Mike Thomas.






reply via email to

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