plib-devel

RE: [Plib-devel] [PATCH] puListBox.cxx: don't crash with empty li st + setTopItem(0)

[Fay J. F Dr C. U. AFSEO/SK <joh...@eg...>]

Melchior,

	That is a good find, but I'm a little concerned about the fix.  I
would prefer the following code:

puListBox.cxx:74
  void puListBox::setTopItem( int item_index )
  {
    top = item_index ;
+   if ( top > num-1 )
+     top = num-1;
    if ( top < 0 )
      top = 0 ;
-   else if ( top > num-1 )
-     top = num-1;

    puPostRefresh () ;
  }

That will handle your case and will also handle a case in which "item_index"
is greater than the number of entries.  You might wind up crashing anyway,
though.

	I appear to be CVS-challenged again.  Would somebody else please put
this in for us?

John F. Fay
Technical Fellow, Jacobs/Sverdrup TEAS Group
850-729-6330
joh...@eg...
 
-----Original Message-----
From: pli...@li...
[mailto:pli...@li...] On Behalf Of Melchior FRANZ
Sent: Tuesday, April 18, 2006 10:19 AM
To: pli...@li...
Subject: [Plib-devel] [PATCH] puListBox.cxx: don't crash with empty list +
setTopItem(0)

The attached patch makes sure that plib doesn't crash if a puListBox was
created with an empty list and the top item is set to 0 (which happens on
"up" arrow in puList, for example). In this case

    else if ( top > num-1 )

evaluates to (0 > -1) -> true, which sets 

       top = num-1;   -> -1

which accesses invalid memory with list[-1] in puListBox.cxx:135.

m.

[Plib-devel] p-Guide update

[Fay J. F Dr C. U. AFSEO/SK <joh...@eg...>]

Gentlemen,

This is just a aquick note; I'm on travel right now and in my spare time I
am upgrading
p-Guide to support the new widget types.  I plan to put it out to the
community when I
get back and work off my backlog, probably in early May.

- John F. Fay

[Plib-devel] [PATCH] puListBox.cxx: don't crash with empty list + setTopItem(0)

[Melchior F. <mf...@us...>]

The attached patch makes sure that plib doesn't crash if a puListBox
was created with an empty list and the top item is set to 0 (which
happens on "up" arrow in puList, for example). In this case

    else if ( top > num-1 )

evaluates to (0 > -1) -> true, which sets 

       top = num-1;   -> -1

which accesses invalid memory with list[-1] in puListBox.cxx:135.

m.

Re: [Plib-devel] [PATCH] Memory leak in SSG SGI loader

[Jan R. <slo...@gm...>]

Am Sun, 16 Apr 2006 18:48:29 +0200 schrieb Bram Stolk <br...@sa...>:

> > inconsistencies. Can someone please apply it to ssgLoadSGI.cxx?
> 
> applied.

Thanks!
Jan

-- 
Jan Reucker
email: jan dot reucker at web dot de
web:   http://www.reucker-online.de

Re: [Plib-devel] [PATCH] Memory leak in SSG SGI loader

[Bram S. <br...@sa...>]

Jan Reucker wrote:
> Hello,
> 
> while valgrind'ing CRRCsim to find some nasty memory leaks I stumbled
> upon a leak in SSG's SGI texture loader. A temporary array is
> allocated but never free'd when loading RLE compressed textures.
> The attached patch fixes this issue and corrects some minor indentation
> inconsistencies. Can someone please apply it to ssgLoadSGI.cxx?
> Thanks a lot in advance!

applied.

   Bram

[Plib-devel] [PATCH] Memory leak in SSG SGI loader

[Jan R. <slo...@gm...>]

Hello,

while valgrind'ing CRRCsim to find some nasty memory leaks I stumbled
upon a leak in SSG's SGI texture loader. A temporary array is
allocated but never free'd when loading RLE compressed textures.
The attached patch fixes this issue and corrects some minor indentation
inconsistencies. Can someone please apply it to ssgLoadSGI.cxx?
Thanks a lot in advance!

Valgrind reported another leak in ssgInit() (ssg.cxx:128): 

  new ssgContext ;  /* Sets the current context with defaults */

Wouldn't it make sense to keep a pointer to this context and have some
function like ssgExit() clean it up before exiting from the application?

There's another leak in the AC3D loader. The allocation occurs in 
do_material(char*) (ssgLoadAC.cxx:292) and is never undone, but I've
not yet investigated how to fix this one.

Kind regards,
Jan R.

-- 
Jan Reucker
email: jan dot reucker at web dot de
web:   http://www.reucker-online.de

Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Bram S. <br...@sa...>]

James 'J.C.' Jones wrote:
> On Thu, 2006-04-13 at 09:18 -0500, Fay John F Dr CTR USAF AFSEO/SK
> wrote:
> 
>>Everybody else,
>>
>>        We are gathered here to remove the deprecated classes from the
>>CVS tree of PUI.  If any man know just cause as to why these classes
>>should not be removed, speak now, or else forever hold your peace.
> 
> 
> Put them to the sword!

Ok, they're dead.

Can someone please try to reproduce my problem with
examples/src/ssg/sky ?

When switching e.g. between sun and moon, the box is not updated
when toggled with the little arrow icons.

Also: we should check the docs for pua/pu issues.

  Bram


> 
> I'll get around to fixing P-Guide in the next month or so.
> 
> Cheers!
> 


-- 
Bram Stolk, VR Engineer SARA, Amsterdam.   tel +31 20 592 3000

"Windows is a 32-bit extension to a 16-bit graphical shell for an 8-bit 
 operating system originally coded for a 4-bit microprocessor by a 2-bit 
 company that can't stand 1 bit of competition."

RE: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[James 'J.C.' J. <jc...@uf...>]

On Thu, 2006-04-13 at 09:18 -0500, Fay John F Dr CTR USAF AFSEO/SK
wrote:
> Everybody else,
>=20
>         We are gathered here to remove the deprecated classes from the
> CVS tree of PUI.  If any man know just cause as to why these classes
> should not be removed, speak now, or else forever hold your peace.

Put them to the sword!

I'll get around to fixing P-Guide in the next month or so.

Cheers!

--=20
 James 'J.C.' Jones - <jc...@uf...> - http://pugsplace.net/
 "To err is humor." - Anonymous

RE: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Fay J. F Dr C. U. AFSEO/SK <joh...@eg...>]

Bram,

	The only difference between a deprecated "pu*****" class and its
"puAux" equivalent is the addition of the letter "a".

Everybody else,

	We are gathered here to remove the deprecated classes from the CVS
tree of PUI.  If any man know just cause as to why these classes should not
be removed, speak now, or else forever hold your peace.

John F. Fay
Technical Fellow, Jacobs/Sverdrup TEAS Group
850-729-6330
joh...@eg...
 
-----Original Message-----
From: pli...@li...
[mailto:pli...@li...] On Behalf Of Bram Stolk
Sent: Wednesday, April 12, 2006 12:06 PM
To: pli...@li...
Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput
Base commit?

Fay John F Dr CTR USAF AFSEO/SK wrote:
> OK, I think the time has come to cut the Gordian knot.  The "puComboBox" 
> class has been deprecated for more than two years and Bram finds that 
> it doesn't work when the "puAux" equivalent does work.  Let us delete it.
> 

Ok, I ripped obsolete classes from my working copy.
Do you want me to commit it to the repo?

I fixed all examples/ and most demos/
(Not yet demos/p-guide... as I wasnt sure how to handle that one)

Remaining issue: docs
Do all obsolete puFoo classes have a puaFoo equivalent?
If so, is the API the same?

   Bram

Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Bram S. <br...@sa...>]

Bram Stolk wrote:

> Remaining issue: docs


Also remaining:
the examples/src/ssg/sky example doesnt work properly:
the entryboxes do not update their contents when clicking
on the small arrows, even though I use pua classes, and
no longer the obsolete stuff from pu

  Bram

Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Bram S. <br...@sa...>]

Fay John F Dr CTR USAF AFSEO/SK wrote:
> OK, I think the time has come to cut the Gordian knot.  The "puComboBox" 
> class has been deprecated for more than two years and Bram finds that it 
> doesn't work when the "puAux" equivalent does work.  Let us delete it.
> 

Ok, I ripped obsolete classes from my working copy.
Do you want me to commit it to the repo?

I fixed all examples/ and most demos/
(Not yet demos/p-guide... as I wasnt sure how to handle that one)

Remaining issue: docs
Do all obsolete puFoo classes have a puaFoo equivalent?
If so, is the API the same?

   Bram



> John F. Fay
> Technical Fellow, Jacobs/Sverdrup TEAS Group
> 850-729-6330
> joh...@eg...
>  
> -----Original Message-----
> From: pli...@li... 
> [mailto:pli...@li...] On Behalf Of Bram Stolk
> Sent: Monday, April 10, 2006 4:34 PM
> To: pli...@li...
> Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent 
> puInput Base commit?
> 
> Fay John F Dr CTR USAF AFSEO/SK wrote:
>  >
>  >
>  > Bram,
>  >
>  >         The only difference between your "combobox.cxx" and the one I
>  > used is that your last argument in the creation of the puComboBox
>  > widget is "true" while mine was "false".  When I changed mine to
>  > "true" it didn't make any difference as far as the input box being
>  > updated is concerned.
>  >
>  >         On a whim, I changed the "puCOmboBox" to "puaComboBox" and the
>  > program quit working.  What happened is that I hadn't recompiled the
>  > "puAux_d.lib" file (I'm running under Windows).  When I did this, the
>  > program started working again.
> 
> Man, this is so confusing.
> 
> For me, it is the opposite:
> When changed from puComboBox to puaComboBox it *started* working.
> (Note: I had to change the last parm value from true to int for this).
> 
> puaComboBox works, puComboBox does not.
> 
> I made sure that
> - I did a make clean
> - I did not forget the make install
> - I checked all my filesystems for old plib installs: not there
> - Re-ran all the configure.in's and autogens (in root dir, and
>    examples dir).
> - Checked which '-L' dirs are used on the link command
> - etc, etc.
> 
> Is there anyone who had a working puComboBox on linux?
> Because I, as sure as hell, can't get it to work.
> 
> I'll check the fix into CVS though, as it does seem to fix the puA 
> version, which was broken also according to melchior.
> 
> 
> 
>   Bram
> 

RE: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Fay J. F Dr C. U. AFSEO/SK <joh...@eg...>]

OK, I think the time has come to cut the Gordian knot.  The "puComboBox"
class has been deprecated for more than two years and Bram finds that it
doesn't work when the "puAux" equivalent does work.  Let us delete it.

John F. Fay
Technical Fellow, Jacobs/Sverdrup TEAS Group
850-729-6330
joh...@eg...
 
-----Original Message-----
From: pli...@li...
[mailto:pli...@li...] On Behalf Of Bram Stolk
Sent: Monday, April 10, 2006 4:34 PM
To: pli...@li...
Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput
Base commit?

Fay John F Dr CTR USAF AFSEO/SK wrote:
> 
> 
> Bram,
> 
>         The only difference between your "combobox.cxx" and the one I 
> used is that your last argument in the creation of the puComboBox 
> widget is "true" while mine was "false".  When I changed mine to 
> "true" it didn't make any difference as far as the input box being 
> updated is concerned.
> 
>         On a whim, I changed the "puCOmboBox" to "puaComboBox" and the 
> program quit working.  What happened is that I hadn't recompiled the 
> "puAux_d.lib" file (I'm running under Windows).  When I did this, the 
> program started working again.

Man, this is so confusing.

For me, it is the opposite:
When changed from puComboBox to puaComboBox it *started* working.
(Note: I had to change the last parm value from true to int for this).

puaComboBox works, puComboBox does not.

I made sure that
- I did a make clean
- I did not forget the make install
- I checked all my filesystems for old plib installs: not there
- Re-ran all the configure.in's and autogens (in root dir, and
   examples dir).
- Checked which '-L' dirs are used on the link command
- etc, etc.

Is there anyone who had a working puComboBox on linux?
Because I, as sure as hell, can't get it to work.

I'll check the fix into CVS though, as it does seem to fix the puA version,
which was broken also according to melchior.



  Bram

Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Bram S. <br...@sa...>]

Fay John F Dr CTR USAF AFSEO/SK wrote:
> 
> 
> Bram,
> 
>         The only difference between your "combobox.cxx" and the one I 
> used is that your last argument in the creation of the puComboBox widget 
> is "true" while mine was "false".  When I changed mine to "true" it 
> didn't make any difference as far as the input box being updated is 
> concerned.
> 
>         On a whim, I changed the "puCOmboBox" to "puaComboBox" and the 
> program quit working.  What happened is that I hadn't recompiled the 
> "puAux_d.lib" file (I'm running under Windows).  When I did this, the 
> program started working again.

Man, this is so confusing.

For me, it is the opposite:
When changed from puComboBox to puaComboBox it *started* working.
(Note: I had to change the last parm value from true to int for this).

puaComboBox works, puComboBox does not.

I made sure that
- I did a make clean
- I did not forget the make install
- I checked all my filesystems for old plib installs: not there
- Re-ran all the configure.in's and autogens (in root dir, and
   examples dir).
- Checked which '-L' dirs are used on the link command
- etc, etc.

Is there anyone who had a working puComboBox on linux?
Because I, as sure as hell, can't get it to work.

I'll check the fix into CVS though, as it does seem to fix
the puA version, which was broken also according to melchior.



  Bram


> 
> John F. Fay
> Technical Fellow, Jacobs/Sverdrup TEAS Group
> 850-729-6330
> joh...@eg...
>  
> -----Original Message-----
> From: pli...@li... 
> [mailto:pli...@li...] On Behalf Of Bram Stolk
> Sent: Monday, April 10, 2006 9:41 AM
> To: pli...@li...
> Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent 
> puInput Base commit?
> 
> Fay John F Dr CTR USAF AFSEO/SK wrote:
>  > OK, then, I will echo Melchior's confusion.  It works for me, it works
>  > for him, and it doesn't work for you.  How about you send me your
>  > actual source files and I will make sure that we are actually "singing
>  > from the same sheet of music."
>  >
>  > John F. Fay
>  > Technical Fellow, Jacobs/Sverdrup TEAS Group 850-729-6330
>  > joh...@eg...
>  > 
>  > -----Original Message-----
>  > From: pli...@li...
>  > [mailto:pli...@li...] On Behalf Of Bram
>  > Stolk
>  > Sent: Monday, April 10, 2006 9:20 AM
>  > To: pli...@li...
>  > Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent
>  > puInput Base commit?
>  >
>  > Fay John F Dr CTR USAF AFSEO/SK wrote:
>  >> Bram,
>  >>
>  >>         It is indeed applied to "pu.h".  Can you send me your sample
>  >> program please?  I will take a quick look into it.
>  >
>  >
>  > It was simply the one posted by Melchior on this list, so the same you
>  > already tested yourself, I guess.
>  >
>  >  Bram
>  >
>  >
> 
> 
> -- 
> Bram Stolk, VR Engineer SARA, Amsterdam.   tel +31 20 592 3000
> 
> "Windows is a 32-bit extension to a 16-bit graphical shell for an 8-bit  
> operating system originally coded for a 4-bit microprocessor by a 2-bit  
> company that can't stand 1 bit of competition."
> 

RE: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Fay J. F Dr C. U. AFSEO/SK <joh...@eg...>]

Bram,

	The only difference between your "combobox.cxx" and the one I used
is that your last argument in the creation of the puComboBox widget is
"true" while mine was "false".  When I changed mine to "true" it didn't make
any difference as far as the input box being updated is concerned.

	On a whim, I changed the "puCOmboBox" to "puaComboBox" and the
program quit working.  What happened is that I hadn't recompiled the
"puAux_d.lib" file (I'm running under Windows).  When I did this, the
program started working again.

John F. Fay
Technical Fellow, Jacobs/Sverdrup TEAS Group
850-729-6330
joh...@eg...
 
-----Original Message-----
From: pli...@li...
[mailto:pli...@li...] On Behalf Of Bram Stolk
Sent: Monday, April 10, 2006 9:41 AM
To: pli...@li...
Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput
Base commit?

Fay John F Dr CTR USAF AFSEO/SK wrote:
> OK, then, I will echo Melchior's confusion.  It works for me, it works 
> for him, and it doesn't work for you.  How about you send me your 
> actual source files and I will make sure that we are actually "singing 
> from the same sheet of music."
> 
> John F. Fay
> Technical Fellow, Jacobs/Sverdrup TEAS Group 850-729-6330 
> joh...@eg...
>  
> -----Original Message-----
> From: pli...@li...
> [mailto:pli...@li...] On Behalf Of Bram 
> Stolk
> Sent: Monday, April 10, 2006 9:20 AM
> To: pli...@li...
> Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent 
> puInput Base commit?
> 
> Fay John F Dr CTR USAF AFSEO/SK wrote:
>> Bram,
>>
>>         It is indeed applied to "pu.h".  Can you send me your sample 
>> program please?  I will take a quick look into it.
> 
> 
> It was simply the one posted by Melchior on this list, so the same you 
> already tested yourself, I guess.
> 
>  Bram
> 
> 


-- 
Bram Stolk, VR Engineer SARA, Amsterdam.   tel +31 20 592 3000

"Windows is a 32-bit extension to a 16-bit graphical shell for an 8-bit
operating system originally coded for a 4-bit microprocessor by a 2-bit
company that can't stand 1 bit of competition."

Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Bram S. <br...@sa...>]

Fay John F Dr CTR USAF AFSEO/SK wrote:
> OK, then, I will echo Melchior's confusion.  It works for me, it works
> for him, and it doesn't work for you.  How about you send me your actual
> source files and I will make sure that we are actually "singing from the
> same sheet of music."
> 
> John F. Fay
> Technical Fellow, Jacobs/Sverdrup TEAS Group
> 850-729-6330
> joh...@eg...
>  
> -----Original Message-----
> From: pli...@li...
> [mailto:pli...@li...] On Behalf Of Bram Stolk
> Sent: Monday, April 10, 2006 9:20 AM
> To: pli...@li...
> Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent
> puInput Base commit?
> 
> Fay John F Dr CTR USAF AFSEO/SK wrote:
>> Bram,
>>
>>         It is indeed applied to "pu.h".  Can you send me your sample
>> program please?  I will take a quick look into it.
> 
> 
> It was simply the one posted by Melchior on this list, so the same you
> already tested yourself, I guess.
> 
>  Bram
> 
> 


-- 
Bram Stolk, VR Engineer SARA, Amsterdam.   tel +31 20 592 3000

"Windows is a 32-bit extension to a 16-bit graphical shell for an 8-bit 
 operating system originally coded for a 4-bit microprocessor by a 2-bit 
 company that can't stand 1 bit of competition."

RE: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Fay J. F Dr C. U. AFSEO/SK <joh...@eg...>]

OK, then, I will echo Melchior's confusion.  It works for me, it works for
him, and it doesn't work for you.  How about you send me your actual source
files and I will make sure that we are actually "singing from the same sheet
of music."

John F. Fay
Technical Fellow, Jacobs/Sverdrup TEAS Group
850-729-6330
joh...@eg...
 
-----Original Message-----
From: pli...@li...
[mailto:pli...@li...] On Behalf Of Bram Stolk
Sent: Monday, April 10, 2006 9:20 AM
To: pli...@li...
Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput
Base commit?

Fay John F Dr CTR USAF AFSEO/SK wrote:
> Bram,
> 
>         It is indeed applied to "pu.h".  Can you send me your sample 
> program please?  I will take a quick look into it.


It was simply the one posted by Melchior on this list, so the same you
already tested yourself, I guess.

 Bram

Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Bram S. <br...@sa...>]

Fay John F Dr CTR USAF AFSEO/SK wrote:
> Bram,
> 
>         It is indeed applied to "pu.h".  Can you send me your sample
> program please?  I will take a quick look into it.


It was simply the one posted by Melchior on this list, so the
same you already tested yourself, I guess.

 Bram


> 
> John F. Fay
> Technical Fellow, Jacobs/Sverdrup TEAS Group
> 850-729-6330
> joh...@eg...
>  
> -----Original Message-----
> From: pli...@li...
> [mailto:pli...@li...] On Behalf Of Bram Stolk
> Sent: Saturday, April 08, 2006 8:46 AM
> To: pli...@li...
> Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent
> puInput Base commit?
> 
> Melchior FRANZ wrote:
>> * Fay John F Dr CTR USAF AFSEO/SK -- Thursday 30 March 2006 23:35:
>>
>>>To fix the "puaComboBox" widget, please add the following methods to
>>>the "puInput" widget definition:
>>
>> [...]
>>
>>>This fixes Melchior's and Bram's test case on my Windows box.  If
>>>somebody would please put it into CVS I would appreciate it.
>>
>>
>> Indeed! Works for me. Thanks.
> 
> That's weird.
> It is applied to pu.h, right?
> Because it does not work for me: although the entry is changed, the
> visible contents is not refreshed, and needs a mouseclick in the
> entrybox before the new entry contents is shown.
> 
> stolk@suske:~/src/plib/src/pui$ cvs diff br...@cv...'s
> password:
> ? .deps
> ? Makefile
> ? Makefile.in
> cvs diff: Diffing .
> Index: pu.h
> ===================================================================
> RCS file: /cvsroot/plib/plib/src/pui/pu.h,v retrieving revision 1.155
> diff -r1.155 pu.h
> 1661a1662,1691
>  >
>  >   virtual void setValue ( puValue *pv )
>  >   {
>  >     puValue::setValue ( pv ) ;
>  >     delete displayed_text ;  displayed_text = NULL ;
>  >   }
>  >
>  >   virtual void setValue ( int   i )
>  >   {
>  >     puValue::setValue ( i ) ;
>  >     delete displayed_text ;  displayed_text = NULL ;
>  >   }
>  >
>  >   virtual void setValue ( float f )
>  >   {
>  >     puValue::setValue ( f ) ;
>  >     delete displayed_text ;  displayed_text = NULL ;
>  >   }
>  >
>  >   virtual void setValue ( const char *s )
>  >   {
>  >     puValue::setValue ( s ) ;
>  >     delete displayed_text ;  displayed_text = NULL ;
>  >   }
>  >
>  >   virtual void setValue ( bool b )
>  >   {
>  >     puValue::setValue ( b ) ;
>  >     delete displayed_text ;  displayed_text = NULL ;
>  >   }
> 
> 
>>
>> m.
>>
>>
>> -------------------------------------------------------
>> This SF.Net email is sponsored by xPML, a groundbreaking scripting
>> language that extends applications into web and mobile media. Attend
>> the live webcast and join the prime developer group breaking into this
> new coding territory!
>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=1216
> <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=1216>
>> 42 _______________________________________________
>> plib-devel mailing list
>> pli...@li...
>> https://lists.sourceforge.net/lists/listinfo/plib-devel
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting
> language that extends applications into web and mobile media. Attend the
> live webcast and join the prime developer group breaking into this new
> coding territory!
> 
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642>
> _______________________________________________
> plib-devel mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-devel
> 

RE: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Fay J. F Dr C. U. AFSEO/SK <joh...@eg...>]

Bram,

	It is indeed applied to "pu.h".  Can you send me your sample program
please?  I will take a quick look into it.

John F. Fay
Technical Fellow, Jacobs/Sverdrup TEAS Group
850-729-6330
joh...@eg...
 
-----Original Message-----
From: pli...@li...
[mailto:pli...@li...] On Behalf Of Bram Stolk
Sent: Saturday, April 08, 2006 8:46 AM
To: pli...@li...
Subject: Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput
Base commit?

Melchior FRANZ wrote:
> * Fay John F Dr CTR USAF AFSEO/SK -- Thursday 30 March 2006 23:35:
> 
>>To fix the "puaComboBox" widget, please add the following methods to 
>>the "puInput" widget definition:
> 
> [...]
> 
>>This fixes Melchior's and Bram's test case on my Windows box.  If 
>>somebody would please put it into CVS I would appreciate it.
> 
> 
> Indeed! Works for me. Thanks.

That's weird.
It is applied to pu.h, right?
Because it does not work for me: although the entry is changed, the visible
contents is not refreshed, and needs a mouseclick in the entrybox before the
new entry contents is shown.

stolk@suske:~/src/plib/src/pui$ cvs diff br...@cv...'s
password:
? .deps
? Makefile
? Makefile.in
cvs diff: Diffing .
Index: pu.h
===================================================================
RCS file: /cvsroot/plib/plib/src/pui/pu.h,v retrieving revision 1.155 diff
-r1.155 pu.h
1661a1662,1691
 >
 >   virtual void setValue ( puValue *pv )
 >   {
 >     puValue::setValue ( pv ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }
 >
 >   virtual void setValue ( int   i )
 >   {
 >     puValue::setValue ( i ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }
 >
 >   virtual void setValue ( float f )
 >   {
 >     puValue::setValue ( f ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }
 >
 >   virtual void setValue ( const char *s )
 >   {
 >     puValue::setValue ( s ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }
 >
 >   virtual void setValue ( bool b )
 >   {
 >     puValue::setValue ( b ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }


> 
> m.
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting 
> language that extends applications into web and mobile media. Attend 
> the live webcast and join the prime developer group breaking into this new
coding territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=1216
> 42 _______________________________________________
> plib-devel mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-devel



-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
plib-devel mailing list
pli...@li...
https://lists.sourceforge.net/lists/listinfo/plib-devel

[Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Melchior F. <mf...@us...>]

* Bram Stolk -- Saturday 08 April 2006 15:46:
> It is applied to pu.h, right?

It is to be applied to pu.h, yes.



> Because it does not work for me: although the entry is changed,
> the visible contents is not refreshed, and needs a mouseclick in
> the entrybox before the new entry contents is shown.

That would indeed be exactly the old bug. But I checked again (just
in case I had mixed up two versions on my disk), and it does definitely
work. I attached the diff.

m.

Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Bram S. <br...@sa...>]

Melchior FRANZ wrote:
> * Fay John F Dr CTR USAF AFSEO/SK -- Thursday 30 March 2006 23:35:
> 
>>To fix the "puaComboBox" widget, please add the following methods to
>>the "puInput" widget definition:
> 
> [...]
> 
>>This fixes Melchior's and Bram's test case on my Windows box.  If somebody
>>would please put it into CVS I would appreciate it.
> 
> 
> Indeed! Works for me. Thanks.

That's weird.
It is applied to pu.h, right?
Because it does not work for me: although the entry is changed,
the visible contents is not refreshed, and needs a mouseclick in
the entrybox before the new entry contents is shown.

stolk@suske:~/src/plib/src/pui$ cvs diff
br...@cv...'s password:
? .deps
? Makefile
? Makefile.in
cvs diff: Diffing .
Index: pu.h
===================================================================
RCS file: /cvsroot/plib/plib/src/pui/pu.h,v
retrieving revision 1.155
diff -r1.155 pu.h
1661a1662,1691
 >
 >   virtual void setValue ( puValue *pv )
 >   {
 >     puValue::setValue ( pv ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }
 >
 >   virtual void setValue ( int   i )
 >   {
 >     puValue::setValue ( i ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }
 >
 >   virtual void setValue ( float f )
 >   {
 >     puValue::setValue ( f ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }
 >
 >   virtual void setValue ( const char *s )
 >   {
 >     puValue::setValue ( s ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }
 >
 >   virtual void setValue ( bool b )
 >   {
 >     puValue::setValue ( b ) ;
 >     delete displayed_text ;  displayed_text = NULL ;
 >   }


> 
> m.
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting language
> that extends applications into web and mobile media. Attend the live webcast
> and join the prime developer group breaking into this new coding territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> plib-devel mailing list
> pli...@li...
> https://lists.sourceforge.net/lists/listinfo/plib-devel

Re: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Bram S. <br...@sa...>]

Fay John F Dr CTR USAF AFSEO/SK wrote:

> This fixes Melchior's and Bram's test case on my Windows box.  If 
> somebody would please put it into CVS I would appreciate it.

I wish I could... cvs is stone dead at the moment.
SF has had very unreliable cvs service, for at least a year now.

[Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Melchior F. <mf...@us...>]

* Fay John F Dr CTR USAF AFSEO/SK -- Thursday 30 March 2006 23:35:
> To fix the "puaComboBox" widget, please add the following methods to
> the "puInput" widget definition:
[...]
> This fixes Melchior's and Bram's test case on my Windows box.  If somebody
> would please put it into CVS I would appreciate it.

Indeed! Works for me. Thanks.

m.

RE: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Fay J. F Dr C. U. AFSEO/SK <joh...@eg...>]

Gentlemen,

	To fix the "puaComboBox" widget, please add the following methods to
the "puInput" widget definition:

<begin quote>
  virtual void setValue ( puValue *pv )
  {
    puValue::setValue ( pv ) ;
    delete displayed_text ;  displayed_text = NULL ;
  }

  virtual void setValue ( int   i )
  {
    puValue::setValue ( i ) ;
    delete displayed_text ;  displayed_text = NULL ;
  }

  virtual void setValue ( float f )
  {
    puValue::setValue ( f ) ;
    delete displayed_text ;  displayed_text = NULL ;
  }

  virtual void setValue ( const char *s )
  {
    puValue::setValue ( s ) ;
    delete displayed_text ;  displayed_text = NULL ;
  }

  virtual void setValue ( bool b )
  {
    puValue::setValue ( b ) ;
    delete displayed_text ;  displayed_text = NULL ;
  }
<end quote>

This fixes Melchior's and Bram's test case on my Windows box.  If somebody
would please put it into CVS I would appreciate it.

John F. Fay
Technical Fellow, Jacobs/Sverdrup TEAS Group
850-729-6330
joh...@eg...
 
-----Original Message-----
From: pli...@li...
[mailto:pli...@li...] On Behalf Of Melchior FRANZ
Sent: Tuesday, March 28, 2006 10:52 AM
To: pli...@li...
Subject: [Plib-devel] Re: [BUG] puComboBox broken since recent puInputBase
commit?

* Bram Stolk -- Tuesday 28 March 2006 18:19:
> Melchior FRANZ wrote:
> > It looks like the recent puInput{,Base} commit broke comboboxes?
> > Can someone check (and fix :-)?  Selecting a puPopupMenu entry 
> > doesn't update the puInput any more.

> Do you have a test sample?

Attached. It's derived from simple.cxx. With a plib from before the
puInputBase changes (about two or three weeks back), the input field is
updated if you select an entry from the popup. With HEAD it isn't.



> I never understood "widget_list" from examples/src/pui If I start 
> that, it opens a gazzillion windows, none of which are responsive.

I've well understood that it's meant to be as useless as it is. I really
wonder why there's no application that demonstrates all available widgets.

m.

[Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Melchior F. <mf...@us...>]

* Fay John F Dr CTR USAF AFSEO/SK -- Thursday 30 March 2006 20:31:
> Hang on, I got it in an e-mail you sent earlier this week.  I'm getting onto
> it.

Too late. I'm just too quick. The version from that message used puComboBox,
the new one uses puaComboBox. Both expose the bug. You said "send me a
current PLIB and [...]" and that's what I sent -- 3.5MB. No idea if it
even got through. As I said in that message, but I'd like to repeat that:

"[...] the html files in doc/ don't have that horrible & illegible
dark green with dark yellow font, but a combination that doesn't make
you cringe. Apart from that it's pure CVS/HEAD."

Any chance that this attack on good taste gets fixed some day?
Makes plib look bad at first sight.  :-}

m.

RE: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base commit?

[Fay J. F Dr C. U. AFSEO/SK <joh...@eg...>]

Hang on, I got it in an e-mail you sent earlier this week.  I'm getting onto
it. 


John F. Fay
Technical Fellow, Jacobs/Sverdrup TEAS Group
850-729-6330
joh...@eg...
 
-----Original Message-----
From: pli...@li...
[mailto:pli...@li...] On Behalf Of Melchior FRANZ
Sent: Thursday, March 30, 2006 12:27 PM
To: pli...@li...
Subject: [Plib-devel] Re: [BUG] puComboBox broken since recent puInput Base
commit?

* Fay John F Dr CTR USAF AFSEO/SK -- Thursday 30 March 2006 20:10:
> Unfortunately I am CVS-challenged

Ahh. OK, that explains it.  :-)



> Can one of you please send me a current PLIB and test case that shows 
> what the bug is?

I'll send in a minute to the sender address from your message.

m.


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
plib-devel mailing list
pli...@li...
https://lists.sourceforge.net/lists/listinfo/plib-devel