[Top][All Lists]

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

[Gcl-devel] Re: new xgcl

From: Camm Maguire
Subject: [Gcl-devel] Re: new xgcl
Date: 22 Jun 2006 17:16:03 -0400
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2

Greetings, and my apologies for being so confusing!

In this image, you should be able to load gcl_dwtest by hand and then
test with (wtesta) ... (wtestk).  You will see the segfault at wtestc.

As for the display fault, wtesta when run under an image with no
DISPLAY in the environment illustrates this.  Should be able to simply
'unset DISPLAY' at the shell prompt before running saved_xgcl.

I'd be more than happy to make the xgcl autoload (or any other you may
desire) work as you wish.  There is no initialization of the xgcl
software needed in the base image.  I thought once about autoloading
some compiled module, but the way it stands now is that all is
pre-linked into the main image.  So (xgcl) would then just print out a
message about issuing some other command.  Perhaps we could put an
entry in the (help) command?  Or maybe call it (xgcl-demo)?

Take care,

"Gordon Shaw Novak" <address@hidden> writes:

> Camm,
> I see a new saved_xgcl in your directory, but I get an error on
> executing (xgcl) :
> address@hidden:/home/camm/gcl-2.6.8pre/xgcl-2$ ./saved_xgcl
> GCL (GNU Common Lisp)  2.6.8 ANSI    Jun 20 2006 16:07:27
> >(xgcl)
> Error in EVAL [or a callee]: Lookup failed in 
> directories:("/home/camm/gcl-2.6.8pre/gcl-tk/"
> "/home/camm/gcl-2.6.8pre/lsp/") for name SYSTEM::|gcl_dwtest| with extensions 
> (".o"
>                                                             ".lsp"
>                                                             ".lisp")
> What I would recommend is that doing (xgcl) should enable the xgcl
> software but not load the gcl_dwtest.lsp file, but instead print
> a message such as:  Enter (load-window-test) to see examples/
> Then load-window-test could load gcl_dwtest, call the (wtesta)
> and (wtestb), and print the message as you do now.
> Then, if you can tell me the conditions that generate a segfault
> when the display is not defined, I can find out where that is
> happening and put in a better error test and break.
> Best regards, Gordon

Camm Maguire                                            address@hidden
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah

reply via email to

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