gcl-devel
[Top][All Lists]
Advanced

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

[Gcl-devel] Re: var types


From: Camm Maguire
Subject: [Gcl-devel] Re: var types
Date: 22 Jun 2006 17:29:59 -0400
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2

Greetings!

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

> Camm, the following is my list of global vars used in dwindow.lsp
> and their types, as determined by looking at the values of these
> vars in xgcl on a 32-bit machine.  Basically 'fixnum' denotes the
> address of a record, while 'int' denotes a smaller integer.
> 
> (int    *mouse-x* )
> (int    *mouse-y* )
> (int    *mouse-window* )
> (fixnum *window-display* nil)
> (int    *window-screen* nil)
> (int    *root-window*)
> (int    *black-pixel*) 
> (int    *white-pixel*)
> (int    *default-fg-color*)
> (int    *default-bg-color*)
> (fixnum *default-size-hints*)
> (fixnum *default-GC*)
> (int    *default-colormap*)
> (fixnum *window-event*)
> (int    *window-default-pos-x* 10)
> (int    *window-default-pos-y* 20)
> (int    *window-default-border* 1)
> (int    *window-default-cursor* 68)
> (int    *window-save-foreground*)
> (int    *window-save-function*)
> (fixnum *window-attributes*)
> (fixnum *window-attr*)
> (fixnum -> int *root-return*         (int-array 1)) ; unused
> (fixnum -> int *child-return*        (int-array 1))
> (fixnum -> int *root-x-return*       (int-array 1))
> (fixnum -> int *root-y-return*       (int-array 1))
> (fixnum -> int *win-x-return*        (int-array 1)) ; unused
> (fixnum -> int *win-y-return*        (int-array 1)) ; unused
> (fixnum -> int *mask-return*         (int-array 1)) ; unused
> (fixnum -> int *x-return*            (int-array 1))
> (fixnum -> int *y-return*            (int-array 1))
> (fixnum -> int *width-return*        (int-array 1))
> (fixnum -> int *height-return*       (int-array 1))
> (fixnum -> int *depth-return*        (int-array 1)) ; unused
> (fixnum -> int *border-width-return* (int-array 1))
> (fixnum -> int *text-width-return*   (int-array 1)) ; very unused
> (fixnum -> int *direction-return*    (int-array 1)) ; unused
> (fixnum -> int *ascent-return*       (int-array 1))
> (fixnum -> int *descent-return*      (int-array 1))
> (fixnum -> int *overall-return*      (int-array 1)) ; unused
> (fixnum *GC-Values*)
> (fixnum *window-xcolor* nil)
> (int    *window-menu-code* nil)
> (fixnum -> int *min-keycodes-return*   (int-array 1))
> (fixnum -> int *max-keycodes-return*   (int-array 1))
> (fixnum -> int *keycodes-return*       (int-array 1)) ; unused
> 

Thanks for this!

> The comment 'unused' means the parameter is required by X, but
> I don't use its value.
> 
> It appears to me that all the existing uses of int-pos should
> in fact be returning 32-bit integers; thus, there would be no  
> need for a function fixnum-pos.
> 

Well, in gdb, the size of Window on harper is currently 8 bytes,
meaning that *child-return* for example needs fixnum-pos, AFAICS (I
think you read this value in the mouse pointer routine).  The
unused ones requiring 8 bytes just need enough space to write into --
currently int-array allocates a fixnum array, but reads a short or
long integer therefrom in int-pos and fixnum-pos respectively.  Other
pointers of which I do not yet know the size are Keysym*, etc.  

> If you got (wtesta) and (wtestb) to work, that is most of it.

Great!

> The remainder hangs on a single function, window-track-mouse.

Appears so.

> If you can give me a saved_xgcl that (preferably) does not
> execute the test file functions when you say (xgcl) and is
> sufficient to run (wtesta) and (wtestb), I should be able to
> do some debugging with that.
> 

I think we crossed in the email -- the current saved_xgcl should allow
direct wtest testing, unless I've messed something else up :-).

>(load "gcl_dwtest")
>(wtesta)
>(wtestb)
>(wtestc) ;segfault here.

Take care,

> Many thanks, 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]