perl-win32-gui-hackers Mailing List for Win32::GUI (Page 2)

Does anyone ever use the Makefile.PL options USERESOURCE or BUILDENV?

I'm considering removing them to simplify the main Makefile.PL

If no-one shouts they'll be gone in the next release.

Rob.

On 01/02/2008, Robert May <ro...@th...> wrote:
> > > - Find out why my build on Cygwin results in an unloadable
> > > SciLexer.dll (permissions I think)
> >
> > Yes. On cygwin it is missing the x "permission", which is "Read &
> > Execute" on Windows ACLs.
>
> I'll investigate further.

I just raised a support request to add the execute bit to SciLexer.dll
in the repository.

Rob.

Reini,

Please stop moving threads that I start on the hackers list onto the
users list.  Users have little/no interest in this.


> > - Find out why my build on Cygwin results in an unloadable
> > SciLexer.dll (permissions I think)
>
> Yes. On cygwin it is missing the x "permission", which is "Read &
> Execute" on Windows ACLs.

I'll investigate further.

> Note the other cygwin permission problems:
> * GUI.rc is not writable, so the default make will fail.

I thought we fixed this wiath a patch to build_tools/UpdateRC.pl - was
this not enough/wrong?

> * samples/* are not writable, so the blib update for the demos will fail.
> No patch, I fix that in my build script.

You mean that blib/lib/Win32/GUI/demos/* is not writable?  I don't
need write permission on samples/* to copy out of that directory.

The copying of demos works fine on the initial build - it's only
subsequent builds that won't copy the samples.  Work around is to
'make clean' ....

> > - check that the new tests for Scintilla work under cygwin
>
> I'll do.

Thanks - let me know.  They seem ok for me.

Regards,
Rob.

> On 31/01/2008, Jeremy White  wrote:
>> Just testing the latest Win32-GUI code line and I'm getting a few odd er=
rors (mingw & 5.8.8)
>
> ActiveState Perl 5.8.8 build 822?
> What gcc version?

Vista, AS822, 3.4.5 (mingw special)

>> It could be my environment and I haven't got time to dig deeper, but tho=
ught I'd report
>> them just in case it's something simple.
>
> I hope it's your environment ... I can't get anything similar with
> WinXP, AS822, 3.4.5 (mingw special)
>
> Do the build test OK?

Yeah - a few issues surounding pod and BitmapInLine (about closing files - =
probably a Vista thing).

BTW - Installing V 1.5 back and the errors go away...

> I'd appreciate it if nothing else if you could send me a script
> exhibiting one (or more) of these problems so I can be sure whether
> I'm not seeing them too.

I should have some more time to look at this over the weekend.

Cheers,

jez.
_________________________________________________________________
Telly addicts unite!
http://www.searchgamesbox.com/tvtown.shtml=

On 31/01/2008, Jeremy White <jez...@ho...> wrote:
> Just testing the latest Win32-GUI code line and I'm getting a few odd errors (mingw & 5.8.8)

ActiveState Perl 5.8.8 build 822?
What gcc version?

> It could be my environment and I haven't got time to dig deeper, but thought I'd report
> them just in case it's something simple.

I hope it's your environment ... I can't get anything similar with
WinXP, AS822, 3.4.5 (mingw special)

Do the build test OK?

I'd appreciate it if nothing else if you could send me a script
exhibiting one (or more) of these problems so I can be sure whether
I'm not seeing them too.

Thanks for the report.
Rob.

> When I exit my app I get lots error messages:
>
> Can't call method "FETCH" on an undefined value during global destruction.
> ...
> Can't call method "FETCH" on an undefined value during global destruction.
>
> With the final line being:
>
> Callback called exit during global destruction.
>
> Also I'm using Scintilla and it seems when the scintilla window contains no text (perhaps I am passing undef?) I get the following error:
>
> Can't locate auto/Win32/GUI/Scintilla/SendMessage.al in @INC (@INC contains C:/Perl/site/lib C:/Perl/lib .) at C:/Perl/site/lib/Win32/GUI/Scintilla.pm at line 1430
>
> I've played with the Scintilla examples and they seem fine.
>
> Cheers,
>
> jez..
>
> _________________________________________________________________
> Get Hotmail on your mobile, text MSN to 63463!
> http://mobile.uk.msn.com/pc/mail.aspx


-- 
Please update your address book with my new email address:
ro...@th...

Just testing the latest Win32-GUI code line and I'm getting a few odd error=
s (mingw & 5.8.8) It could be my environment and I haven't got time to dig =
deeper, but thought I'd report them just in case it's something simple.

When I exit my app I get lots error messages:

Can't call method "FETCH" on an undefined value during global destruction.
...
Can't call method "FETCH" on an undefined value during global destruction.

With the final line being:

Callback called exit during global destruction.

Also I'm using Scintilla and it seems when the scintilla window contains no=
 text (perhaps I am passing undef?) I get the following error:

Can't locate auto/Win32/GUI/Scintilla/SendMessage.al in @INC (@INC contains=
 C:/Perl/site/lib C:/Perl/lib .) at C:/Perl/site/lib/Win32/GUI/Scintilla.pm=
 at line 1430

I've played with the Scintilla examples and they seem fine.

Cheers,

jez..
 =20
_________________________________________________________________
Get Hotmail on your mobile, text MSN to 63463!
http://mobile.uk.msn.com/pc/mail.aspx=

I think I'm making good progress.  I just cimmitted everything that I
had outstanding in my local repository.  Here's what I have left to do
before a 1.06 release:

- Check that some 'redefined definitions' warnings have been removed
by the updates to ppport.h
- Find out why my build on Cygwin results in an unloadable
SciLexer.dll (permissions I think)
- Add to all Makefiles a suitable addition of the -mms-bitfields flag
for gcc, to allow building under 5.10 with GCC against a VC++ compiled
Perl to work.
- check that the new tests for Scintilla work under cygwin
- update $VERSION of modules that have changes
- update Win23::GUI $VERSION to 1.05_99 and make a beta distribution available.

As long as I don't find any more problems this shouldn't take long.

Regards,
Rob.

>> As this looked so serious I downloaded 5.10.0 (1002) to test this and I =
can see what you see.
>> This is a major problem in general, not just for Win32::GUI (my own XS m=
odules fail their
>> test suit due to this issue).
>
> Jan (ActiveState) confirmed and explained the problem (follow the
> rt.perl.org link above).
>
>> Looks like I wont be upgrading to 5.10 any time soon:)
>
> Turns out there is a simple solution - add the -mms_bitfields option
> to your cflags to get gcc to use the same bitfield-alignment as MSVC.
>
> I've submitted a patch to ActiveState, which I hope will make it into
> their next build. In the meantime be *VERY* wary of any modules for
> ActiveState 5.10 that are built with gcc.

Tested with the -mms_bitfields option and the problem goes away. All my XS =
modules now pass their tests.

Good find.

Cheers,

jez.
_________________________________________________________________
Who's friends with who and co-starred in what?
http://www.searchgamesbox.com/celebrityseparation.shtml=

On 30/01/2008, Jeremy White <jez...@ho...> wrote:
> > After further investigation I have submitted the following perlbug report:
> >
> > http://rt.perl.org/rt3/Public/Bug/Display.html?id=50386
> >
> > which is also tracked by ActiveState here:
> >
> > http://bugs.activestate.com/show_bug.cgi?id=74532
> >
>
> As this looked so serious I downloaded 5.10.0 (1002) to test this and I can see what you see.
> This is a major problem in general, not just for Win32::GUI (my own XS modules fail their
> test suit due to this issue).

Jan (ActiveState) confirmed and explained the problem (follow the
rt.perl.org link above).

> Looks like I wont be upgrading to 5.10 any time soon:)

Turns out there is a simple solution - add the -mms_bitfields option
to your cflags to get gcc to use the same bitfield-alignment as MSVC.

I've submitted a patch to ActiveState, which I hope will make it into
their next build.  In the meantime be *VERY* wary of any modules for
ActiveState 5.10 that are built with gcc.

Regards,
Rob.

> After further investigation I have submitted the following perlbug report=
:
>
> http://rt.perl.org/rt3/Public/Bug/Display.html?id=3D50386
>
> which is also tracked by ActiveState here:
>
> http://bugs.activestate.com/show_bug.cgi?id=3D74532
>
> I'd appreciate anyone with a working 5.10.0 installation and mingw
> environment to download the attachment from the rt.perl.org link
> above, unpack it and try building and testing the module to see if
> there's something screwy with my environment/mingw install.


As this looked so serious I downloaded 5.10.0 (1002) to test this and I can=
 see what you see. This is a major problem in general, not just for Win32::=
GUI (my own XS modules fail their test suit due to this issue).

Looks like I wont be upgrading to 5.10 any time soon:)

Cheers,

jez.
_________________________________________________________________
Free games, great prizes - get gaming at Gamesbox.=20
http://www.searchgamesbox.com=

On 27/01/2008, Robert May <rob...@us...> wrote:
> (2) MinGW 5.1.3.  Perl 5.6.1. 5.8.8, 5.10.0
> runtime=mingw-runtime-3.13.tar.gz
> w32api=w32api-3.10.tar.gz
> binutils=binutils-2.17.50-20060824-1.tar.gz
> core=gcc-core-3.4.5-20060117-1.tar.gz
> gpp=gcc-g++-3.4.5-20060117-1.tar.gz
>
> - build and test is clean with the exception of perl 5.10.0:
> Activestate build 1001 is fine, but there is a non-fatal warning
> thrown up in AS build 1002 during the NotifyIcon tests.  It appears to
> be that Win32::GUI::GetDllVersion() doesn't see the list context, and
> returns the string variant.  This needs further investigation, as it
> would be a bad thing if this manifests itself elsewhere.  It's OK with
> the Microsoft compiler, so I assume its a perl problem and not a
> Win32::GUI problem ....

After further investigation I have submitted the following perlbug report:

http://rt.perl.org/rt3/Public/Bug/Display.html?id=50386

which is also tracked by ActiveState here:

http://bugs.activestate.com/show_bug.cgi?id=74532

I'd appreciate anyone with a working 5.10.0 installation and mingw
environment to download the attachment from the rt.perl.org link
above, unpack it and try building and testing the module to see if
there's something screwy with my environment/mingw install.

Regards,
Rob.

On 28/01/2008, Reini Urban <ru...@x-...> wrote:
>
> Attached are my tests.

Many thanks.  I'll tidy them up and apply them.

Knowing what the problem was, with a bit of code inspection I was able
to get the crash to happen reliably with this one-liner:

perl -mWin32::GUI -mWin32::GUI::Scintilla -e
'Win32::GUI::Window->new->AddScintilla->SetFocus'

I'll use that as the actual test for the crash, allowing me to
simplify the other tests somewhat.

Rob.

2008/1/27, Robert May <rob...@us...>:
> On 13/01/2008, Reini Urban <ru...@x-...> wrote:
> > [Sorry for TOFU]
> > I found and fixed the error. (Scintilla crash)
> >
> > Remove the line with #define PERL_GET_NO_CONTEXT at the top of Scintilla.xs
> >
> > The reason why you cannot reproduce it, it because MSVC obviously is
> > compiled with PERL_OBJECT (see GUI.h for the perlud struct),
> > Without PERL_OBJECT you may not change the setting of
> > PERL_GET_NO_CONTEXT to the one in GUI.xs. Otherwise you are accessing
> > the struct with and without the my_perl field, causing Scintilla to
> > crash.
> >
> > It was not the stack corrupt, just perlud.
> >
> > When I got a decent internet connection tomorrow I'll post my full patch
> > with added testcase which catches this problem.
>
> I'll apply your fix.  If you have the time to write a test case that
> would be much appreciated.

Attached are my tests.

-- 
Reini Urban
http://phpwiki.org/              http://murbreak.at/
http://spacemovie.mur.at/   http://helsinki.at/

Time is well overdue for a new release.  I had hoped to do this around
September of last year, but I have spend the last few months adjusting
to living on a new continent.

I think I am finally able to spend some time looking at Win32::GUI.
I have a few outstanding changes in my local repository to commit,
which I will do as soon as I am happy that they are OK.  Once I have
done that I will make a new release - hopefully sometime in Feb.

The code is in pretty good shape.  I spent some of today compiling it
in different environments:

(1) WinXP(SP2), VS98/VC6 (Sp6 with Feb 2003 SDK).  Perl 5.6.1, 5.8.8, 5.10.0
- I have a small change to the BitmapInline changes I made recently to
make them work properly on 5.6.1

otherwise build and test all loks clean.

(2) MinGW 5.1.3.  Perl 5.6.1. 5.8.8, 5.10.0
runtime=mingw-runtime-3.13.tar.gz
w32api=w32api-3.10.tar.gz
binutils=binutils-2.17.50-20060824-1.tar.gz
core=gcc-core-3.4.5-20060117-1.tar.gz
gpp=gcc-g++-3.4.5-20060117-1.tar.gz

- build and test is clean with the exception of perl 5.10.0:
Activestate build 1001 is fine, but there is a non-fatal warning
thrown up in AS build 1002 during the NotifyIcon tests.  It appears to
be that Win32::GUI::GetDllVersion() doesn't see the list context, and
returns the string variant.  This needs further investigation, as it
would be a bad thing if this manifests itself elsewhere.  It's OK with
the Microsoft compiler, so I assume its a perl problem and not a
Win32::GUI problem ....

(3) VC7 free C++ 2003 toolkit version and VS2003 (VC8) express edition
These distributions do not have any MFC/ATL support, and so can't
build AxWindow, and possible other modules.  I'm seeing test reports
from testers using the full versions (VS2003/VS2005/VS2008) that there
are errors with the ATL stuff.  I'm a bit out of my depth in the
modules that use it, and don't have the environment (or the wish to
purchase such an environment) to fix these problems.

Does anyone with the know-how and environment wnat to look at these
problems?  Alternative does anyone have a full version of Visual
Studio 2005 or 2008 that they'd like to donate to me to look into
this?

I need to think about what to do here - but if nothing else I should
write up what can and what can't be built with the various tools.

(4) I haven't looked at cygwin at all, but am assuming that as Reini
is (relatively) quite all is well there - I will try to do get an
environment set up this week.

Further issues:

I am still sat on Reini's CustomDraw patch.  It is too big for me to
review and understand everything.  I think I will just apply it, but
not until after a 1.06 release.

It is my intention to release 3 binary builds: 5.6, 5.8, and 5.10.   I
intend to stop supporting perl 5.6 after this release.

The more time I spend working on this code-base the more I find that I
want to change - there is too much simple stuff that just doesn't
happen easily, and adding features is brittle, as the modularisation
is not right.

I have in the past talked about a completely new code-base, and I
expect to devote some time to that this year.  As such I don't expect
to be spending quite so much time on the active code-base.  I'll write
more on this in the coming weeks.

Regards,
Rob.

On 13/01/2008, Reini Urban <ru...@x-...> wrote:
> [Sorry for TOFU]
> I found and fixed the error. (Scintilla crash)
>
> Remove the line with #define PERL_GET_NO_CONTEXT at the top of Scintilla.xs
>
> The reason why you cannot reproduce it, it because MSVC obviously is
> compiled with PERL_OBJECT (see GUI.h for the perlud struct),
> Without PERL_OBJECT you may not change the setting of
> PERL_GET_NO_CONTEXT to the one in GUI.xs. Otherwise you are accessing
> the struct with and without the my_perl field, causing Scintilla to
> crash.
>
> It was not the stack corrupt, just perlud.
>
> When I got a decent internet connection tomorrow I'll post my full patch
> with added testcase which catches this problem.

I'll apply your fix.  If you have the time to write a test case that
would be much appreciated.

Thanks,
Rob.

On 11/01/2008, Jeremy White <jez...@ho...> wrote:
> The doc's say it it should return a hash, but it actually returns an array. Bug or doco shortfall?
>
>     ###########################################################################
>     # (@)METHOD:GetItem(NODE)
>     # (@)METHOD:ItemInfo(NODE)
>     # Returns an associative array of information about the given NODE:
>     #     -children
>     #     -image
>     #     -parent
>     #     -selectedimage
>     #     -state
>     #     -text
[snip]
>         XSRETURN(12);
>     } else {
>         XSRETURN_UNDEF;
>     }

It does what is says.  It returns an associative array (actually a
list, I guess).  I assume this is down so that you can write:

my %info = $node->ItemInfo();

However, on failure the code above will warn about an uneven number of
elements in the hash assignment, as the returned list will contain a
single 'undef'.

I'll fix that last XSRETURN_UNDEF to XSRETURN_EMPTY.

Can you suggest a better wording for the docs?

Rob.

On 02/08/2007, Reini Urban <ru...@x-...> wrote:
> Attached is a Carp patch for 5.9.5.

Thanks, and apologies for the delay.  Applied to CVS.  Builds and
tests OK for me (MSVC6, WinXP, AS Perl build 1001), but I haven't
actually tried using it for anything more taxing yet.

Rob.

>
> t/01_load............................NOK 2/3#   at t/01_load.t line 38.
> #     Tried to use 'Win32::GUI::Constants'.
> #     Error:  syntax error at
> /cygdrive/r/PERL/Win32-GUI/Win32-GUI-Constants/../
> blib/lib/Win32/GUI/Constants.pm line 274, near "Carp::croak qq(Can't
> continue after import errors)"
>
> and so on.
> --
> Reini Urban
> http://phpwiki.org/  http://murbreak.at/
> http://helsinki.at/  http://spacemovie.mur.at/
>
> 2007-08-02  Reini Urban <ru...@x-...>
>             * no Carp prototypes since 5.9.x.
>
> diff -ub  ./GUI.pm.orig
> --- ./GUI.pm.orig       2007-07-28 21:01:37.015625000 +0000
> +++ ./GUI.pm    2007-08-02 06:34:00.531000000 +0000
> @@ -423,8 +423,8 @@
>
>     if(keys(%options) != 0) {
>         require Carp;
> -       Carp::carp "Animate: Unrecognised options ".join(", ", keys(%options));
> -       return undef
> +       Carp::carp("Animate: Unrecognised options ".join(", ", keys(%options)));
> +       return undef;
>     }
>
>     $show      = 1      unless defined $show;
> @@ -435,13 +435,13 @@
>
>     if($animation !~ /roll|slide|blend|center/) {
>         require Carp;
> -       Carp::carp "Animate: Unrecognised animation type: $animation";
> +       Carp::carp("Animate: Unrecognised animation type: $animation");
>         return undef;
>     }
>
>     if($direction !~ /lr|tlbr|tb|trbl|rl|brtl|bt|bltr/) {
>         require Carp;
> -       Carp::carp "Animate: Unrecognised direction: $direction";
> +       Carp::carp("Animate: Unrecognised direction: $direction");
>         return undef unless $direction eq 'blrt'; # blrt allowed for deprection cycle
>     }
>
> diff -ub  ./Win32-GUI-Constants/Constants.pm.orig
> --- ./Win32-GUI-Constants/Constants.pm.orig     2007-07-28 21:01:38.359375000 +0000
> +++ ./Win32-GUI-Constants/Constants.pm  2007-08-02 06:33:48.702875000 +0000
> @@ -271,7 +271,7 @@
>      if($oops) {
>          # only require Carp if we need it
>          require Carp;
> -        Carp::croak qq(Can't continue after import errors);
> +        Carp::croak(qq(Can't continue after import errors));
>      }
>  }
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> Perl-Win32-GUI-Hackers mailing list
> Per...@li...
> https://lists.sourceforge.net/lists/listinfo/perl-win32-gui-hackers
> http://perl-win32-gui.sourceforge.net/
>
>


-- 
Please update your address book with my new email address:
ro...@th...

On 09/12/2007, Robert May <rob...@us...> wrote:
> I have just submitted the following SF support request:
> https://sourceforge.net/tracker/index.php?func=detail&aid=1847204&group_id=1&atid=200001
>
> As part of fixing Win32::GUI::BitmapInline to use a writeable
> temporary directory (rather than always using the current directory) I
> am moving the code into its own subdirectory.

I have finally found the time to update the repository.  All should
now be in a compilable state again.  Due to carelessness I mixed an
extra update (to ListView.xs) in with the commit, but it's easier to
leave that as is, rather than try to fix it.

> If you have a check-out CVS copy, then you may need to do a fresh
> export following this change (I am not familiar enough with CVS to be
> sure).

I think that adding '-d' (create missing directories) to your CVS
update command will be sufficient - but don't take my word for it.

Regards,
Rob.

[Sorry for TOFU]
I found and fixed the error. (Scintilla crash)

Remove the line with #define PERL_GET_NO_CONTEXT at the top of Scintilla.xs

The reason why you cannot reproduce it, it because MSVC obviously is
compiled with PERL_OBJECT (see GUI.h for the perlud struct),
Without PERL_OBJECT you may not change the setting of
PERL_GET_NO_CONTEXT to the one in GUI.xs. Otherwise you are accessing
the struct with and without the my_perl field, causing Scintilla to
crash.

It was not the stack corrupt, just perlud.

When I got a decent internet connection tomorrow I'll post my full patch
with added testcase which catches this problem.

2008/1/7, Jeremy White <jez...@ho...>:
>
> I can't seem to reproduce:
>
> perl-5.8.8.822 on Vista
> perl-5.8.7.813 on XP SP2
>
> Running but the demos.pl and Editor.pl with Win32-GUI-1.05. Do you need to do anything special?
>
>
>
> > Date: Sun, 6 Jan 2008 21:21:59 -0600
> > From: ru...@x-...
> > To: per...@li...
> > Subject: [perl-win32-gui-hackers] Scintilla crashes at DoEvent_Scintilla
> >
> > cygwin, though I don't believe it is related.
> > perl-5.8.8 and perl-5.10.0
> > Win32-GUI-1.05
> >
> > scripts\win32-gui-demos.pl and Win32-GUI-Scintilla\demos\Editor.pl
> > crash. The first after starting an app, the 2nd immediately.
> > This came from the cygwin list, with Vista, but I and others can reproduce
> > it with XP2. Looks like an Scintilla.xs problem.
> >
> > Backtrace:
> >
> > $ gdb perl
> > GNU gdb 6.5.50.20060706-cvs (cygwin-special)
> > Copyright (C) 2006 Free Software Foundation, Inc.
> > GDB is free software, covered by the GNU General Public License, and you are
> > welcome to change it and/or distribute copies of it under certain conditions.
> > Type "show copying" to see the conditions.
> > There is absolutely no warranty for GDB. Type "show warranty" for details.
> > This GDB was configured as "i686-pc-cygwin"...
> > (gdb) run Editor.pl
> > Starting program: /usr/bin/perl.exe Editor.pl
> > Loaded symbols for /cygdrive/c/WINDOWS/system32/ntdll.dll
> > Loaded symbols for /cygdrive/c/WINDOWS/system32/kernel32.dll
> > Loaded symbols for /usr/bin/cygwin1.dll
> > Loaded symbols for /cygdrive/c/WINDOWS/system32/advapi32.dll
> > Loaded symbols for /cygdrive/c/WINDOWS/system32/rpcrt4.dll
> > Loaded symbols for /cygdrive/c/WINDOWS/system32/secur32.dll
> > Loaded symbols for /usr/bin/cygperl5_10.dll
> > Loaded symbols for /usr/bin/cygcrypt-0.dll
> >
> > Program received signal SIGSEGV, Segmentation fault.
> > 0x67503375 in Perl_gv_fetchpvn_flags (my_perl=0x17700e8,
> > nambeg=0xc5c17c "main::or_Notify", full_len=15, flags=0,
> > sv_type=13) at gv.c:909
> > 909 if (!stash || !SvREFCNT(stash)) /* symbol table
> > under destruction */
> > (gdb) bt
> > #0 0x67503375 in Perl_gv_fetchpvn_flags (my_perl=0x17700e8,
> > nambeg=0xc5c17c "main::or_Notify", full_len=15, flags=0,
> > sv_type=13) at gv.c:909
> > #1 0x6758dd2d in Perl_get_cvn_flags (my_perl=0x17700e8, name=0xc5c17c
> > "main::or_Notify", len=15, flags=0)
> > at perl.c:2479
> > #2 0x6758de63 in Perl_get_cv (my_perl=0x17700e8, name=0xc5c17c
> > "main::or_Notify", flags=0) at perl.c:2498
> > #3 0x6d541270 in DoEvent_Scintilla (my_perl=0x17700e8, Name=0xc5c17c
> > "main::or_Notify", code=2007, evt=0xc5c3bc)
> > at Scintilla.xs:67
> > #4 0x6d54286d in CallWndProc (nCode=0, wParam=0, lParam=12960480) at
> > Scintilla.xs:213
> > #5 0x7e4318e3 in UnhookWinEvent ()
> > #6 0x00000000 in ?? ()
> > (gdb) x stash
> > 0x1: Cannot access memory at address 0x1
> > (gdb)
> >
> > Reason: Someone is overwriting the stash with 1,
> > and the perlud name is also wrong.
> > It should be "Editor", not "or"
> > So I believe the stack is corrupt at all.
-- 
Reini Urban
http://phpwiki.org/              http://murbreak.at/
http://spacemovie.mur.at/   http://helsinki.at/

On Fri, 2008-01-11 at 10:55 +0000, Jeremy White wrote:
> The doc's say it it should return a hash, but it actually returns an array. Bug or doco shortfall?
> 

doco.

IANAL but it actually returns a LIST, (an "associative" one) that can be
assigned to a HASH or an ARRAY, so its a wording detail:

	%info = $w->ItemInfo($item);  # "return" a hash

Salvador.

The doc's say it it should return a hash, but it actually returns an array.=
 Bug or doco shortfall?


    #######################################################################=
####
    # (@)METHOD:GetItem(NODE)
    # (@)METHOD:ItemInfo(NODE)
    # Returns an associative array of information about the given NODE:
    #     -children
    #     -image
    #     -parent
    #     -selectedimage
    #     -state
    #     -text
void
GetItem(handle,item)
    HWND handle
    HTREEITEM item
ALIAS:
    Win32::GUI::TreeView::ItemInfo =3D 1
PREINIT:
    TV_ITEM tv_item;
    char pszText[1024];
PPCODE:
    ZeroMemory(&tv_item, sizeof(TV_ITEM));
    tv_item.hItem =3D item;
    tv_item.mask =3D TVIF_CHILDREN | TVIF_HANDLE | TVIF_IMAGE
                 | TVIF_PARAM | TVIF_SELECTEDIMAGE
                 | TVIF_TEXT | TVIF_STATE;
    tv_item.pszText =3D pszText;
    tv_item.cchTextMax =3D 1024;
    if(TreeView_GetItem(handle, &tv_item)) {
        EXTEND(SP, 12);
        XST_mPV(0, "-text");
        XST_mPV(1, tv_item.pszText);
        XST_mPV(2, "-image");
        XST_mIV(3, tv_item.iImage);
        XST_mPV(4, "-selectedimage");
        XST_mIV(5, tv_item.iSelectedImage);
        XST_mPV(6, "-children");
        XST_mIV(7, tv_item.cChildren);
        XST_mPV(8, "-parent");
        XST_mIV(9, (long) TreeView_GetParent(handle, item));
        XST_mPV(10, "-state");
        XST_mIV(11, tv_item.state);
        XSRETURN(12);
    } else {
        XSRETURN_UNDEF;
    }

_________________________________________________________________
Free games, great prizes - get gaming at Gamesbox.=20
http://www.searchgamesbox.com=

I can't seem to reproduce:

perl-5.8.8.822 on Vista
perl-5.8.7.813 on XP SP2

Running but the demos.pl and Editor.pl with Win32-GUI-1.05. Do you need to =
do anything special?



> Date: Sun, 6 Jan 2008 21:21:59 -0600
> From: ru...@x-...
> To: per...@li...
> Subject: [perl-win32-gui-hackers] Scintilla crashes at DoEvent_Scintilla
>
> cygwin, though I don't believe it is related.
> perl-5.8.8 and perl-5.10.0
> Win32-GUI-1.05
>
> scripts\win32-gui-demos.pl and Win32-GUI-Scintilla\demos\Editor.pl
> crash. The first after starting an app, the 2nd immediately.
> This came from the cygwin list, with Vista, but I and others can reproduc=
e
> it with XP2. Looks like an Scintilla.xs problem.
>
> Backtrace:
>
> $ gdb perl
> GNU gdb 6.5.50.20060706-cvs (cygwin-special)
> Copyright (C) 2006 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you =
are
> welcome to change it and/or distribute copies of it under certain conditi=
ons.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB. Type "show warranty" for details=
.
> This GDB was configured as "i686-pc-cygwin"...
> (gdb) run Editor.pl
> Starting program: /usr/bin/perl.exe Editor.pl
> Loaded symbols for /cygdrive/c/WINDOWS/system32/ntdll.dll
> Loaded symbols for /cygdrive/c/WINDOWS/system32/kernel32.dll
> Loaded symbols for /usr/bin/cygwin1.dll
> Loaded symbols for /cygdrive/c/WINDOWS/system32/advapi32.dll
> Loaded symbols for /cygdrive/c/WINDOWS/system32/rpcrt4.dll
> Loaded symbols for /cygdrive/c/WINDOWS/system32/secur32.dll
> Loaded symbols for /usr/bin/cygperl5_10.dll
> Loaded symbols for /usr/bin/cygcrypt-0.dll
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x67503375 in Perl_gv_fetchpvn_flags (my_perl=3D0x17700e8,
> nambeg=3D0xc5c17c "main::or_Notify", full_len=3D15, flags=3D0,
> sv_type=3D13) at gv.c:909
> 909 if (!stash || !SvREFCNT(stash)) /* symbol table
> under destruction */
> (gdb) bt
> #0 0x67503375 in Perl_gv_fetchpvn_flags (my_perl=3D0x17700e8,
> nambeg=3D0xc5c17c "main::or_Notify", full_len=3D15, flags=3D0,
> sv_type=3D13) at gv.c:909
> #1 0x6758dd2d in Perl_get_cvn_flags (my_perl=3D0x17700e8, name=3D0xc5c17c
> "main::or_Notify", len=3D15, flags=3D0)
> at perl.c:2479
> #2 0x6758de63 in Perl_get_cv (my_perl=3D0x17700e8, name=3D0xc5c17c
> "main::or_Notify", flags=3D0) at perl.c:2498
> #3 0x6d541270 in DoEvent_Scintilla (my_perl=3D0x17700e8, Name=3D0xc5c17c
> "main::or_Notify", code=3D2007, evt=3D0xc5c3bc)
> at Scintilla.xs:67
> #4 0x6d54286d in CallWndProc (nCode=3D0, wParam=3D0, lParam=3D12960480) a=
t
> Scintilla.xs:213
> #5 0x7e4318e3 in UnhookWinEvent ()
> #6 0x00000000 in ?? ()
> (gdb) x stash
> 0x1: Cannot access memory at address 0x1
> (gdb)
>
> Reason: Someone is overwriting the stash with 1,
> and the perlud name is also wrong.
> It should be "Editor", not "or"
> So I believe the stack is corrupt at all.
> --
> Reini Urban
> http://phpwiki.org/ http://murbreak.at/
> http://spacemovie.mur.at/ http://helsinki.at/
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Perl-Win32-GUI-Hackers mailing list
> Per...@li...
> https://lists.sourceforge.net/lists/listinfo/perl-win32-gui-hackers
> http://perl-win32-gui.sourceforge.net/

_________________________________________________________________
Who's friends with who and co-starred in what?
http://www.searchgamesbox.com/celebrityseparation.shtml=

cygwin, though I don't believe it is related.
perl-5.8.8 and perl-5.10.0
Win32-GUI-1.05

scripts\win32-gui-demos.pl and Win32-GUI-Scintilla\demos\Editor.pl
crash. The first after starting an app, the 2nd immediately.
This came from the cygwin list, with Vista, but I and others can reproduce
it with XP2. Looks like an Scintilla.xs problem.

Backtrace:

$ gdb perl
GNU gdb 6.5.50.20060706-cvs (cygwin-special)
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i686-pc-cygwin"...
(gdb) run Editor.pl
Starting program: /usr/bin/perl.exe Editor.pl
Loaded symbols for /cygdrive/c/WINDOWS/system32/ntdll.dll
Loaded symbols for /cygdrive/c/WINDOWS/system32/kernel32.dll
Loaded symbols for /usr/bin/cygwin1.dll
Loaded symbols for /cygdrive/c/WINDOWS/system32/advapi32.dll
Loaded symbols for /cygdrive/c/WINDOWS/system32/rpcrt4.dll
Loaded symbols for /cygdrive/c/WINDOWS/system32/secur32.dll
Loaded symbols for /usr/bin/cygperl5_10.dll
Loaded symbols for /usr/bin/cygcrypt-0.dll

Program received signal SIGSEGV, Segmentation fault.
0x67503375 in Perl_gv_fetchpvn_flags (my_perl=0x17700e8,
nambeg=0xc5c17c "main::or_Notify", full_len=15, flags=0,
    sv_type=13) at gv.c:909
909                 if (!stash || !SvREFCNT(stash)) /* symbol table
under destruction */
(gdb) bt
#0  0x67503375 in Perl_gv_fetchpvn_flags (my_perl=0x17700e8,
nambeg=0xc5c17c "main::or_Notify", full_len=15, flags=0,
    sv_type=13) at gv.c:909
#1  0x6758dd2d in Perl_get_cvn_flags (my_perl=0x17700e8, name=0xc5c17c
"main::or_Notify", len=15, flags=0)
    at perl.c:2479
#2  0x6758de63 in Perl_get_cv (my_perl=0x17700e8, name=0xc5c17c
"main::or_Notify", flags=0) at perl.c:2498
#3  0x6d541270 in DoEvent_Scintilla (my_perl=0x17700e8, Name=0xc5c17c
"main::or_Notify", code=2007, evt=0xc5c3bc)
    at Scintilla.xs:67
#4  0x6d54286d in CallWndProc (nCode=0, wParam=0, lParam=12960480) at
Scintilla.xs:213
#5  0x7e4318e3 in UnhookWinEvent ()
#6  0x00000000 in ?? ()
(gdb) x stash
0x1:    Cannot access memory at address 0x1
(gdb)

Reason: Someone is overwriting the stash with 1,
and the perlud name is also wrong.
It should be "Editor", not "or"
So I believe the stack is corrupt at all.
-- 
Reini Urban
http://phpwiki.org/              http://murbreak.at/
http://spacemovie.mur.at/   http://helsinki.at/

Thanks for the update - once you've updated the source I'll do some testing=
.

Cheers,

jez.

> Subject: [perl-win32-gui-hackers] CVS repository re-organisation
>
> I have just submitted the following SF support request:
> https://sourceforge.net/tracker/index.php?func=3Ddetail&aid=3D1847204&gro=
up_id=3D1&atid=3D200001
>
> As part of fixing Win32::GUI::BitmapInline to use a writeable
> temporary directory (rather than always using the current directory) I
> am moving the code into its own subdirectory.
>
> If you have a check-out CVS copy, then you may need to do a fresh
> export following this change (I am not familr enough with CVS to be
> sure). Further, until I check in some further changes the build will
> be broken temporarily following this repository change.
>
> Regards,
> Rob.
>

_________________________________________________________________
Get free emoticon packs and customisation from Windows Live.=20
http://www.pimpmylive.co.uk=

I have just submitted the following SF support request:
https://sourceforge.net/tracker/index.php?func=detail&aid=1847204&group_id=1&atid=200001

As part of fixing Win32::GUI::BitmapInline to use a writeable
temporary directory (rather than always using the current directory) I
am moving the code into its own subdirectory.

If you have a check-out CVS copy, then you may need to do a fresh
export following this change (I am not familr enough with CVS to be
sure).  Further, until I check in some further changes the build will
be broken temporarily following this repository change.

Regards,
Rob.