I'm glad to see I am not the only one to use garnet these days...!
| >>>>> "bu" == bernard URBAN <Bernard.Urban@...> writes:
|
| tl> (lisp-implementation-version) -> "2000-03-09 (March 2000)"
| tl> with new-clx module on a gnu/linux RedHat 6.0 system (intel)
|
| bu> This may be your problem. With mit-clx, I can compile and load
| bu> Garnet without problem.
|
| actually, I have the opposite problem: Garnet more-or-less works with
| nclx but not at all with mit-clx. Here is a message I sent here in
| March:
|
| -----------------------------------------------------------------------
| It doesn't work correctly with CLISP: with nclx it seems to miss some
| refresh events (windows sometimes stay undrawn), and the logo in the
| demos section doesn't display in color. It doesn't work at all with
^^^^^^^^^^^^^^^^^^^^^^^^
This may be caused by a problem with "true-color" screen support. This
problem is fixed in Fred Gilham <gilham@... version
(thanks to Paolo for this information. It did not solved my problem
with clisp but it did solve the color problem.)
url: ftp://ftp.csl.sri.com/pub/users/gilham/garnet/
In this version (maybe it should be called garnet-3.0.x) the pixmap
problem is partially solved. Garnet was not able to load pixmaps with
depth > 8. Now It displays correctly the first half of the pixmap!
| mit-clx (nor indeed do basic CLX demos): there are problems with setf
| functions being hidden by setf-expanders:
Actually I am using garnet with allegro, this was my first try with
nclx/clisp. I will try mit-clx and see if I get the same problem.
|
| when loading garnet:
|
| WARNING:
| The function (SETF XLIB:GCONTEXT-FUNCTION) is hidden by a SETF expander.
| WARNING:
| The function (SETF XLIB:GCONTEXT-FOREGROUND) is hidden by a SETF expander.
| WARNING:
| The function (SETF XLIB:GCONTEXT-BACKGROUND) is hidden by a SETF expander.
| [...]
|
| then when running:
|
| *** - FUNCALL: the function #:|(SETF XLIB:WINDOW-PLIST)| is undefined
|
|
| I think the problem is in the mit-clx/depdefs.lsp:def-clx-class macro,
| which seems to be emulating CLOS with structures, but it's too hairy
| for me to understand.
|
Thibault Langlois
|