plib-devel

Re: [Plib-devel] puDialogBox error

[Eero P. <epa...@ko...>]

Fay John F Contr AAC/WMG wrote:

> You are absolutely right, deleting the widget with "puDeleteObject" 
> makes all the difference.  We need to fix the web page there.  In fact, 
> this problem was exactly the reason for writing "puDeleteObject".
> 

I wonder if the real fix would require making the problematic destructor
private and puDeleteObject a friend function ... No actually that
would be too ugly...

It might be sufficient to remove puCleanUpJunk() from puInterface
destructor.

(Keep in mind I have still not properly demonstrated this with
the latest sources)


		Eero

RE: [Plib-devel] puDialogBox error

[Fay J. F C. AAC/W. <joh...@eg...>]

You are absolutely right, deleting the widget with "puDeleteObject" makes
all the difference.  We need to fix the web page there.  In fact, this
problem was exactly the reason for writing "puDeleteObject".

John F. Fay
joh...@eg...



-----Original Message-----
From: pli...@li...
[mailto:pli...@li...]On Behalf Of Eero Pajarre
Sent: Thursday, April 08, 2004 3:56 AM
To: pli...@li...
Subject: Re: [Plib-devel] puDialogBox error


Fay John F Contr AAC/WMG wrote:

> Eero,
> 
>         I've tried to reproduce this but can't.  Can you send me a 
> sample program please?
> 

Notice that this error does not typically crash the program,
I only noticed it by using a debugging malloc library.
Other alternative would be following the order of object
destruction and puOneShot::dohit in debugger.

I did some checking on the plib sources and noticed
that puDialogBox was used in p-guide, but there
it was not deleted using delete. Instead
puDeleteObject was used, and I think this avoids the problem.

So maybe the problem only exists in
http://plib.sourceforge.net/pui/index.html#puDialogBox
which uses delete in getting rid of the dialog box.


I will see if I have time for making a proper test case using
the current pui sources.


		Eero





-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
plib-devel mailing list
pli...@li...
https://lists.sourceforge.net/lists/listinfo/plib-devel

RE: [Plib-devel] pwSetAutoRepeatKey...and pwMacOSX.

[Fay J. F C. AAC/W. <joh...@eg...>]

Since a tenet of windowing systems (at least of the GLUT type) is platform
independence, I guess we will want to change the Windows code to match the
Linux behavior.  Here is a patch that does this:

pwWindows.cxx:122:
    case WM_KEYDOWN:
      /* If the key is already down, we are on auto-repeat.  Break if the
autorepeat is disabled. */
-       if ( ( updown == PW_DOWN ) && ( int(wParam) == old_key ) && (
!autoRepeat ) )
+       if ( ( updown == PW_DOWN ) && ( int(wParam) == old_key ) )
+       {
+         if ( autoRepeat )
+         {
+           /* Disable CTRL, SHIFT, CapsLock keys from making a callback */
+           if ( ( key == VK_CONTROL ) || ( key == VK_SHIFT ) || ( key ==
VK_CAPITAL ) )
+             break ;
+
+           if ( key != -1 && kbCB )  // Autorepeat enabled, call the
callback with an "up" setting
+             (*kbCB) ( key, PW_UP, lastx, lasty ) ;
+         }
+         else
+           break ;  // Autorepeat not enabled, just break
+       }

        updown = PW_DOWN ;
        old_key = wParam ;


Regarding the bug in the X11 version (about how if you press shift-W,
release the shift, then release the w, you don't get an up-callback for the
"W"):  Windows behaves in exactly the same manner if the keys are pressed
long enough for the autorepeat to come on (whether PW passes them through or
not).  If everything is done quickly, then we get a key press event for the
capital letter and a key release event for the lowercase letter.  By moving
the test for CTRL/SHIFT/CapsLock to above the assignment for "key = wParam"
I can remove the second key press callback invocation (with the lowercase
"w") but we still have the phenomenon of the key press callback being called
with one key and the key release with another key.  It may need just to be
documented as a "feature."

John F. Fay
joh...@eg...



-----Original Message-----
From: pli...@li...
[mailto:pli...@li...]On Behalf Of Steve Baker
Sent: Wednesday, April 07, 2004 8:13 PM
To: pli...@li...
Subject: Re: [Plib-devel] pwSetAutoRepeatKey...and pwMacOSX.


Fay John F Contr AAC/WMG wrote:
> Gentlemen,
> 
> 	I've just sent Steve privately a "pwWindows.cxx" file with an
> auto-repeat patch.  I've not been able to test it but I think it will
work.
> Correctly, even.  It compiles and was not much of a patch.
> 
> 	I tested the auto-repeat feature using the "freeglut" library and
> found that Windows gives a WM_KEYDOWN message when a key is pressed.  Then
> it gives a string of WM_KEYDOWN messages when a key is held down.  It
gives
> a WM_KEYUP message when the key is released.

Hmmm - in Linux/X11, you get rapidly alternating key-up/key-down messages
when
you are holding down a key with autorepeat enabled.

> 	When the user presses two keys one after the other, Windows will
> send a WM_KEYDOWN message for each one.  If the first one is
auto-repeating,
> its auto-repeat will stop and the second key's auto-repeat will eventually
> start.  When a key is released (either the first or the second), Windows
> sends a WM_KEYUP message for each key.  My patch does not support this
> behavior, and I don't know that the X11 code does either.

The X11 code does this OK.

There is one bug in the X11 version that I can see.

If you push (and hold) SHIFT plus a letter key until it starts repeating,
then release the SHIFT key without releasing the letter key - then you
get a key-down event for the lowercase key - but no key-up event for the
uppercase key.  When you finally release the letter key, you get a
lower-case
key-up event - but still no uppercase key-up event.

This could confuse programs like pw_demo that track the up/down events to
try to know which keys are being held down.

It's actually kinda nasty to fix this.  It might seem easy enough to track
when the lowercase version of a key is hit whilst the uppercase version
has not yet been released, this isn't so simple for things like punctuation
characters that are on the same key on some keyboards - but different keys
on others.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
plib-devel mailing list
pli...@li...
https://lists.sourceforge.net/lists/listinfo/plib-devel

Learning SSG (was RE: [Plib-devel] tb ssg dependancy...)

[Fay J. F C. AAC/W. <joh...@eg...>]

Hmmm ... I see myself learning SSG sooner or later.  I've been shying away
from it because it's so big and I have my hands full.  Steve mentioned some
time ago the possibility of merging PUI into SSG, and his mention of a 3-d
widget strikes a chord.  (If anybody has looked at the puaCompass widget ...
it's supposed to be 3-d but I kludged it into 2-d.)  Between that and
learning my way around the Triereis code ... I think my days of SSG
ignorance may be numbered.

John F. Fay
joh...@eg...



-----Original Message-----
From: pli...@li...
[mailto:pli...@li...]On Behalf Of Steve Baker
Sent: Friday, April 09, 2004 11:53 AM
To: pli...@li...
Subject: Re: [Plib-devel] tb ssg dependancy...

<snip>

If we go the route of adding texture to PUI buttons, I think it's a small
step to
saying that PUI should use SSG's ssgState so that buttons can be lit as 3D
objects,
and (eventually) support stuff like shaders when we eventually get that into
SSG.

<snip>

Re: [Plib-devel] tb ssg dependancy...

[Steve B. <sjb...@ai...>]

Fay John F Contr AAC/WMG wrote:
> This was, I think, a suggestion to add a "texture button" to PUI.  A texture
> button would behave just like a one-shot but would have a textured picture
> drawn on it instead of having a legend written in it.  Since the present
> "puButton" can have its own rendering callback instead of the default, I
> would be inclined to say "Thank you, but no thank you" to this offer.  I
> don't recall whether we answered this--I can't find anything in my e-mail
> archive--but if anybody has any other comment I invite it.

I agree that this 'texture button' patch should be rejected.

It's not an entirely unreasonable request.  However, doing it as a specific
puTexturedButton is a terrible way to do it.   If we were to do it, it should
be built into the lowest level PUI rendering object (puBox? IIRC) so that all
widgets can be textured - and what's more, you'd be able to specify a default
texture style just as we define a default colour or font.

In order to make this work, we'd have to move the SSG texture loading classes
out into the Utility library (I guess) so they could be shared between SSG and
PUI.

This is all quite a bit of work.

So:  No - I don't want a half-assed implementation of a puTexturedButton.
      Yes - I do think that textured rendering of PUI widgets in general is OK.
      ...and to do it PROPERLY requires careful thought.
      In the meantime, you don't NEED textured buttons because there is a rendering
      callback 'hook'.  I have textured buttons in my Triereis game - they work
      just fine without any additional PUI functionality.

If we go the route of adding texture to PUI buttons, I think it's a small step to
saying that PUI should use SSG's ssgState so that buttons can be lit as 3D objects,
and (eventually) support stuff like shaders when we eventually get that into SSG.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

RE: [Plib-devel] Re: Olivier's Joystick Code

[Fay J. F C. AAC/W. <joh...@eg...>]

Gentlemen,

	I have gone through Olivier's code a bit more thoroughly and now it
compiles.  I have looked at all the changes and (except for the MacOSX file,
which I think Olivier tested) they look like they shouldn't affect
functionality.  I compiled the "jsWindows" file on my machine and it
compiles now.  I also added all the "#if defined (UL_<operating system>)"
back to the files.

	I can send a copy of the files to any interested party.  While we
are doing patch releases we probably shouldn't put them into CVS yet.

John F. Fay
joh...@eg...



-----Original Message-----
From: pli...@li...
[mailto:pli...@li...]On Behalf Of Olivier A
Sent: Tuesday, April 06, 2004 5:17 PM
To: pli...@li...
Subject: [Plib-devel] Re: Olivier's Joystick Code



    John Fay wrote:
>
> ...
> (1) jsNone.cxx:25 - MSVC objects to the statement "struct os_specific 
> {}".
> I changed it to "struct os_specific ;" and the file compiled just fine.

   I di not try to compile this file...

> (2) jsWindows.cxx:42 - Apparently "error" is a private member of
> "jsJoystick".  We need to add an accessor method to "js.h" and use it 
> here.

   ?? getOEMProductName is a member of "struct os_specific_s" which is 
declared friend of jsJoystick: it should have access to all members of 
jsJoystick...

> (3) jsWindows.cxx:57, 68 - Apparently "os" is also a private member of
> 'jsJoystick".  Again we need to add an accessor method to "js.h" and 
> use it
> here.

   Same remark as above.
   I have not these errors with gcc 3.3 an an old Metrowerks compiler.
   Perhaps with some compiler you can access only protected members with 
friend functions ...

> (4) jsWindows.cxx:120 - The variable "jsCaps" is actually a member of 
> "os".

   Yes you're right: "os->" is missing !

> (5) jsWindows.cxx:162 - The variable "num_hats" is not a member of
> "jsJoystick".

   Remove the line: at some time I added a "num_hats" member to 
jsJoystick, but forgot to remove it in some files...
(I can neither test nor compile on windows systems.)

Olivier A.



-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
plib-devel mailing list
pli...@li...
https://lists.sourceforge.net/lists/listinfo/plib-devel

RE: [Plib-devel] tb ssg dependancy...

[Fay J. F C. AAC/W. <joh...@eg...>]

This was, I think, a suggestion to add a "texture button" to PUI.  A texture
button would behave just like a one-shot but would have a textured picture
drawn on it instead of having a legend written in it.  Since the present
"puButton" can have its own rendering callback instead of the default, I
would be inclined to say "Thank you, but no thank you" to this offer.  I
don't recall whether we answered this--I can't find anything in my e-mail
archive--but if anybody has any other comment I invite it.

John F. Fay 
joh...@eg... 

-----Original Message-----
From: pli...@li...
[mailto:pli...@li...]On Behalf Of Simon
Sent: Wednesday, March 17, 2004 10:09 AM
To: pli...@li...
Subject: [Plib-devel] tb ssg dependancy...



Hi, I should really point out that it's been a while since I played with pui
but here goes...The button does not react differently at all from a
puOneShot.  The only difference is that a texture can be applied to it.  I
didn't want to put all the texture loading into the button as I thought it
would make it overly complex for such a simple thing.  I did this also
because I wasn't sure if I would stick with using ssg to load the textures.
It seems pretty easy to me to swap the texture controls over to whatever
system you might be using.  All that happens is that a texture handle is
passed into glBindTexture.

Re: [Plib-devel] SG bug.

[Steve B. <sjb...@ai...>]

Fay John F Contr AAC/WMG wrote:
> OK, and in view of your announcement of 1.8.3, do you mean that they need to
> be fixed or that you have fixed them?

Sorry - I meant to say that I fixed them...before the 1.8.3 release.

I'd like to keep knocking off these kinds of issues in consecutive 1.8.x
releases for a while.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

RE: [Plib-devel] SG bug.

[Fay J. F C. AAC/W. <joh...@eg...>]

OK, and in view of your announcement of 1.8.3, do you mean that they need to
be fixed or that you have fixed them?

John F. Fay
joh...@eg...



-----Original Message-----
From: pli...@li...
[mailto:pli...@li...]On Behalf Of Steve Baker
Sent: Thursday, April 08, 2004 8:55 PM
To: pli...@li...
Subject: [Plib-devel] SG bug.


The following routines are completely broken in SG - and as far as
I can tell, they've always been broken:

   sgDistSquaredToLineSegmentVec3
   sgDistToLineSegmentVec3
   sgdDistSquaredToLineSegmentVec3
   sgdDistToLineSegmentVec3

They aren't used anywhere else in PLIB - but probably some applications
will need this fix.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
plib-devel mailing list
pli...@li...
https://lists.sourceforge.net/lists/listinfo/plib-devel

[Plib-devel] ANNOUNCING PLIB 1.8.3

[Steve B. <sjb...@ai...>]

ANNOUNCING PLIB 1.8.3
~~~~~~~~~~~~~~~~~~~~~

This is just a maintenance release - a few small fixes:

   * NET now allows a NULL argument to netSocket::accept

   * Added #include <stdio.h> to all top level header files
     in order to ensure that NULL is defined.

   * Added ssgaScreenDump into ssgAux.dsp

   * Cleaned up some warnings from MSVC.

   * Added Christos Dimitrakakis' fixes to SL to prevent
     crackling in certain envelope conditions.

   * Added pwAutoRepeatKey to control keyboard auto-repeat.

   * Fixed a horrific bug in the sgDistToLineSegment routines.

Enjoy!

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

[Plib-devel] SG bug.

[Steve B. <sjb...@ai...>]

The following routines are completely broken in SG - and as far as
I can tell, they've always been broken:

   sgDistSquaredToLineSegmentVec3
   sgDistToLineSegmentVec3
   sgdDistSquaredToLineSegmentVec3
   sgdDistToLineSegmentVec3

They aren't used anywhere else in PLIB - but probably some applications
will need this fix.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

[Plib-devel] Re: Olivier's Joystick Code

[Olivier A <oli...@wa...>]

   About joystick code and hats: Hats are implemented as two axes per 
hat in jsWindows and jsBsd. I do the same for jsMacOSX. But I am not 
very sure of the values that are reported. Can someone help me ?

   In jsMacOSX, I use the following pattern:

"north-south" axe:		1, 1,  0, -1, -1, -1,  0,  1     0
"east-west" axe:		0, 1,  1,  1,  0, -1, -1, -1     0
hat position:			N  NE  E  SE   S  SW   W   NW   idle

   Is it the good one (when I push the hat ahead, I get the rear view in 
FlightGear...) ?

   Thank you.

Olivier A.

[Plib-devel] pwInit and OpenGL options

[Olivier A <oli...@wa...>]

   The only OpenGL option that can be set with pw is the "multisample" 
option. Is it possible, in a future version, to add more options, such 
as a stencil buffer ? I am not an OpenGL specialist, so I do not know 
what actually are these options for.
   FlightGear is also in the process to remove GLUT dependencies. They 
have put os dependencies in a file (fg_os.cxx if I remember) which is 
very similar to pw, except in the options at init time.

Olivier A.

RE: [Plib-devel] Re: pwMacOSX.cxx

[Fay J. F C. AAC/W. <joh...@eg...>]

They are in the "jsXXX.cxx" files in CVS, but not in Olivier's modified
versions.  Obviously they need to go into Olivier's copies before those
files go into CVS.

John F. Fay
joh...@eg...



-----Original Message-----
From: pli...@li...
[mailto:pli...@li...]On Behalf Of Steve Baker
Sent: Wednesday, April 07, 2004 7:37 PM
To: pli...@li...
Subject: Re: [Plib-devel] Re: pwMacOSX.cxx


Olivier A wrote:

>   Oops, I am sorry. I modified my own version and forgot to add the 
> conditional compilation... (these conditional compilations are not in 
> the jsXXX.cxx files either...)

Yes they are!

RE: [Plib-devel] pwSetAutoRepeatKey...and pwMacOSX.

[Fay J. F C. AAC/W. <joh...@eg...>]

OK, I have just sent Steve another "pwWindows.cxx" file that fixes things so
that pressing the control, shift, or caps lock keys will not trigger a
callback.

John F. Fay
joh...@eg...



-----Original Message-----
From: pli...@li...
[mailto:pli...@li...]On Behalf Of Steve Baker
Sent: Wednesday, April 07, 2004 7:39 PM
To: pli...@li...
Subject: Re: [Plib-devel] pwSetAutoRepeatKey...and pwMacOSX.


Fay John F Contr AAC/WMG wrote:
> An interesting note on the key messages in PW:  the control and shift keys
> apparently also generate messages under Windows.  I don't know whether
this
> was planned for or desired; I'm just noting that it happens.

No - it's not desirable (well - at least - I was copying the style of GLUT
and it doesn't do it.  I'm also pretty sure you can't get X11 to do that.).

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
plib-devel mailing list
pli...@li...
https://lists.sourceforge.net/lists/listinfo/plib-devel

Re: [Plib-devel] puDialogBox error

[Eero P. <epa...@ko...>]

Fay John F Contr AAC/WMG wrote:

> Eero,
> 
>         I've tried to reproduce this but can't.  Can you send me a 
> sample program please?
> 

Notice that this error does not typically crash the program,
I only noticed it by using a debugging malloc library.
Other alternative would be following the order of object
destruction and puOneShot::dohit in debugger.

I did some checking on the plib sources and noticed
that puDialogBox was used in p-guide, but there
it was not deleted using delete. Instead
puDeleteObject was used, and I think this avoids the problem.

So maybe the problem only exists in
http://plib.sourceforge.net/pui/index.html#puDialogBox
which uses delete in getting rid of the dialog box.


I will see if I have time for making a proper test case using
the current pui sources.


		Eero

Re: [Plib-devel] pwSetAutoRepeatKey...and pwMacOSX.

[Steve B. <sjb...@ai...>]

Fay John F Contr AAC/WMG wrote:
> Gentlemen,
> 
> 	I've just sent Steve privately a "pwWindows.cxx" file with an
> auto-repeat patch.  I've not been able to test it but I think it will work.
> Correctly, even.  It compiles and was not much of a patch.
> 
> 	I tested the auto-repeat feature using the "freeglut" library and
> found that Windows gives a WM_KEYDOWN message when a key is pressed.  Then
> it gives a string of WM_KEYDOWN messages when a key is held down.  It gives
> a WM_KEYUP message when the key is released.

Hmmm - in Linux/X11, you get rapidly alternating key-up/key-down messages when
you are holding down a key with autorepeat enabled.

> 	When the user presses two keys one after the other, Windows will
> send a WM_KEYDOWN message for each one.  If the first one is auto-repeating,
> its auto-repeat will stop and the second key's auto-repeat will eventually
> start.  When a key is released (either the first or the second), Windows
> sends a WM_KEYUP message for each key.  My patch does not support this
> behavior, and I don't know that the X11 code does either.

The X11 code does this OK.

There is one bug in the X11 version that I can see.

If you push (and hold) SHIFT plus a letter key until it starts repeating,
then release the SHIFT key without releasing the letter key - then you
get a key-down event for the lowercase key - but no key-up event for the
uppercase key.  When you finally release the letter key, you get a lower-case
key-up event - but still no uppercase key-up event.

This could confuse programs like pw_demo that track the up/down events to
try to know which keys are being held down.

It's actually kinda nasty to fix this.  It might seem easy enough to track
when the lowercase version of a key is hit whilst the uppercase version
has not yet been released, this isn't so simple for things like punctuation
characters that are on the same key on some keyboards - but different keys
on others.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

Re: [Plib-devel] pwSetAutoRepeatKey...and pwMacOSX.

[Steve B. <sjb...@ai...>]

Fay John F Contr AAC/WMG wrote:
> An interesting note on the key messages in PW:  the control and shift keys
> apparently also generate messages under Windows.  I don't know whether this
> was planned for or desired; I'm just noting that it happens.

No - it's not desirable (well - at least - I was copying the style of GLUT
and it doesn't do it.  I'm also pretty sure you can't get X11 to do that.).

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

Re: [Plib-devel] Re: pwMacOSX.cxx

[Steve B. <sjb...@ai...>]

Olivier A wrote:

>   Oops, I am sorry. I modified my own version and forgot to add the 
> conditional compilation... (these conditional compilations are not in 
> the jsXXX.cxx files either...)

Yes they are!

---------------------------- Steve Baker -------------------------
HomeEmail: <sjb...@ai...>    WorkEmail: <sj...@li...>
HomePage : http://www.sjbaker.org
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net
-----BEGIN GEEK CODE BLOCK-----
GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M-
V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++
-----END GEEK CODE BLOCK-----

[Plib-devel] Re: pwMacOSX.cxx

[Olivier A <oli...@wa...>]

> PLEASE STOP REMOVING THE CONDITIONAL COMPILATION AROUND THIS FILE!!!
>
> This file *WILL* be compiled on every system.   The hassle involved
> in building all of the Makefile crap to make each architecture build
> just the exact set of files it needs - and yet ensuring that all of
> them are in the distro - is more than I want to maintain.
>
> It's MUCH easier to simply '#ifdef' out the entire contents of the
> file on machines that don't need it.

   Oops, I am sorry. I modified my own version and forgot to add the 
conditional compilation... (these conditional compilations are not in 
the jsXXX.cxx files either...)

Olivier A.

[Plib-devel] Re: Olivier's Joystick Code

[Olivier A <oli...@wa...>]

   Steve baker wrote:
>
>> 	I'm testing out Olivier's new joystick code that hasn't made it into
>> CVS yet.
>
> Oh - I see.   Well, then I guess we shouldn't put it into 1.8.3 
> because it
> won't have been sufficiently tested.

   Yes, only the Mac OS X implementation is tested.
   Before it is put in a release version, the other jsXXX.cxx files must 
be tested to see if there is no errors at compile time and if it still 
works: I could have broken  something in the modifications.
   These files can be downloaded from:
	<http://perso.wanadoo.fr/oliver77-htm/plib/liste.htm>

>>  We were holding off on putting it in until after the new
>> release--and good thing we did, too.  Olivier's idea (correct me if 
>> I'm
>> wrong, Olivier) was to remove the operating-system-dependent code 
>> from the
>> header file and conceal the implementations in the "cxx" files where 
>> the
>> application can't see it.
>
> That's OK.
>
>> He did this because of my repeated notes about
>> the redefinition of the "_JS_MAX_AXES" constant in the Windows 
>> version.
>
> Yep, I see.

   Well, that was not the true reason. I began to modify the 
jsMacOSX.cxx file to implement hats functionality. Doing this, I had to 
modify the js.h header. As there were some whishes to get rid of os 
dependencies in this header file, I think this was the time to do 
that... The redefinition of "_JS_MAX_AXES" was also fixed by the time.

Olivier A.

Re: [Plib-devel] Plib SL

[Christos D. <dim...@id...>]

I unfortunately cannot check the changes right now, as I am back home
for Easter. Sorry.

-- 
Christos Dimitrakakis
IDIAP (http://www.idiap.ch/~dimitrak/main.html)

RE: [Plib-devel] pwSetAutoRepeatKey

[Fay J. F C. AAC/W. <joh...@eg...>]

One more time ...

pwWindows.cxx:40-44:
  static int             size   [2]   = { 640, 480 } ;

+ static bool            autoRepeat   = false ;
+
  static pwResizeCB     *resizeCB     = NULL ;


pwWindows.cxx:73-78:
  LRESULT CALLBACK WndProc( HWND hWnd, UINT uMsg, WPARAM wParam, LPARAM
lParam )
  {
-   int key = -1 ;
+   static int key = -1 ;
-   int updown = PW_UP ;
+   static int updown = PW_UP ;
    int button = -1 ;
    static int mb = 0 ;



pwWindows.cxx:121-129:
      case WM_KEYDOWN:
+       /* If the key is already down, we are on auto-repeat.  Break if the
autorepeat is disabled. */
+       if ( ( updown == PW_DOWN ) && ( int(wParam) == key ) && !autoRepeat
) break ;

        updown = PW_DOWN ;
        /* FALLTHROUGH */
      case WM_KEYUP:
+       if ( uMsg == WM_KEYUP ) updown = PW_UP ;
        key = wParam ;




pwWindows.cxx:192-195:
  void pwSetAutoRepeatKey ( bool enable )
  {
-   fprintf(stderr,"pwSetAutoRepeatKey is not yet functional under
Windows.\n" ) ;
+   autoRepeat = enable ;
  }

John F. Fay
joh...@eg...

RE: [Plib-devel] pwSetAutoRepeatKey...and pwMacOSX.

[Fay J. F C. AAC/W. <joh...@eg...>]

An interesting note on the key messages in PW:  the control and shift keys
apparently also generate messages under Windows.  I don't know whether this
was planned for or desired; I'm just noting that it happens.

John F. Fay
joh...@eg...

RE: [Plib-devel] puDialogBox error

[Fay J. F C. AAC/W. <joh...@eg...>]

Eero,

	I've tried to reproduce this but can't.  Can you send me a sample
program please?

John F. Fay
joh...@eg...



-----Original Message-----
From: pli...@li...
[mailto:pli...@li...]On Behalf Of Eero Pajarre
Sent: Wednesday, April 07, 2004 6:08 AM
To: pli...@li...
Subject: [Plib-devel] puDialogBox error


I met this problem with the old pui version I am using,
but it looks like that the same problem exists
with the current code...

If used like in the Programmers Guide example
the puDialogBox and the puOneShot inside it
seem to cause a "freed memory modified" error.

What happens with the old version is that the puOneShot
inside the dialog deletes the dialog, which then deletes its
children including the OneShot button. After
this the OneShot still tries to change its own value,
which is not so ok.

I have not tested this with the current Pui, but it looks
that the puCleanUpJunk inside the puInterface destructor
will trigger the incorrect behavior. (and removing it
should fix this).

			Eero




-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
plib-devel mailing list
pli...@li...
https://lists.sourceforge.net/lists/listinfo/plib-devel