Re: bizzare bug

Figured it out.

C.discard event causes the problem.

I am making the (incorrect ?) assumption that :

C.Get.sint (S_'XKeyEvent.f_type keyEvent)

is making a "local" copy of the value.  The dependency of the correct  
operation on C.discard makes me think otherwise.  Certainly the  
question as to why adding a print statement works is not solved yet  
either.

The real question, is, how do I force the code to make a persistent  
copy of the variable value.  It seems as though it must be using the  
pointer to the memory in some fashion.  Yuch.  I knew there was a  
reason I gave up C.

Brian


On Jan 1, 2007, at 11:01 AM, brian wrote:

> Needless to say I, either I am completely missing something here, or
> I found an SML bug.
>
> Here's the code:
>
>    fun keyEvents (display, win) =
>        let val event = C.new U__XEvent.typ
>            fun getEvent() =
>                let val _ = F_XNextEvent.f (display, C.Ptr.|&| event)
>                    val etype = C.Get.sint (U__XEvent.f_type event)
>                    val keyEvent = (U__XEvent.f_xkey event)
> 		  val ewin = C.Get.ulong (S_'XKeyEvent.f_window keyEvent)
>                    (* key press or release *)
> 		  val ektype = C.Get.sint (S_'XKeyEvent.f_type keyEvent)
> 		  val ekstate = C.Get.uint (S_'XKeyEvent.f_state keyEvent)
> 		  val ekcode = C.Get.uint (S_'XKeyEvent.f_keycode keyEvent)
>                in
>
>                    print ("ewin="^Word32.toString(ewin)^
>                           " etype="^Int32.toString(etype)^"\n");
>                    if etype = XDefs.KeyPress andalso ewin = win then
>                        (print ("keyPress type:"^Int32.toString 
> (ektype)^
>                                " state:"^Word32.toString(ekstate)^
>                                " kcode:"^Word32.toString(ekcode)^
>                                "\n");
>                         SOME(ektype, ekstate, ekcode))
>                    else
>                        NONE
>                end
>            fun loop () =
>                let val e = getEvent()
>                in
>                    case e of
>                        SOME(x,y,z) => (C.discard event;
>                                    (x,y,z))
>                      | NONE => loop()
>                end
>
> The problem is this:
>
> If I remove the print statement in  if etype... then ... SOME(..)
> else NONE , the calling routine gets the wrong answers.  As you can
> imagine this is rather impossible to figure it out.  With the print
> statement there it works fine, i.e. I get correct keypress codes, but
> using :
>
>                    if etype = XDefs.KeyPress andalso ewin = win then
>                         SOME(ektype, ekstate, ekcode)
>                    else
>                        NONE
>
> Does not work, in that I get the same answer over and over again.
>
> ???
>
> Brian
>
>
>
> ---------------------------------------------------------------------- 
> ---
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to  
> share your
> opinions on IT & business topics through brief surveys - and earn cash
> http://www.techsay.com/default.php? 
> page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Smlnj-list mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/smlnj-list


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV