************************************************************************ *** WARNING -- WARNING -- WARNING -- WARNING -- WARNING -- WARNING *** ************************************************************************ *
* The e-mail-message appended below is list- or mailserver-related *
* That type of mail will by definition not be processed on this system
* and in general is not accepted in the fidonet.org-domain. *
* To avoid these bouncings the listkeeper is kindly requested to
* remove any subscription coming from a z2.fidonet.org-style address. *
* If you have reason to believe your message was removed without due
* reason then kindly inform ''.
*
************************************************************************

From: Javier Ros @2:292/862
To: fpc-develop @2:5020/1305.201
Subj.: [fpc-devel]PASCAL record vs C struct alignement (GTK related). Dated: 04 Sep 00 19:09:50
[processed here: 06 Sep 00 09:58:31]

@TOPT 201
@INTL 2:5020/1305 2:292/862
@MSGID: unavarra.es 346ab7ec
@REPLYADDR:
@REPLYTO: UUCP
@PID: Mozilla 4.72 [en] (X11; I; Linux 2.2.5-15 i586) @CHRS: LATIN-1 2
@RFC-Received: from vekoll.saturnus.vein.hu (
[193.6.40.1]) by infomag.iguana.be (8.9.3/8.9.3) with ESMTP id UAA16299 for ; Tue, 5 Sep 2000 20:46:07 +0200
@RFC-Received: from vekoll.saturnus.vein.hu ( [127.0.0.1]) by vekoll.saturnus.vein.hu (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA03408; Tue, 5 Sep 2000 18:55:34 +0200
@RFC-Received: from cartero.unavarra.es (cartero.unavarra.es [130.206.166.80]) by vekoll.saturnus.vein.hu (8.9.3/8.9.3/Debian 8.9.3-21) with ESMTP id SAA03391
for ; Tue, 5 Sep 2000 18:55:27 +0200
@RFC-Received: from si.unavarra.es (si.unavarra.es [130.206.166.108]) by cartero.unavarra.es (8.9.3/8.9.1) with ESMTP id SAA29081 for ; Tue, 5 Sep 2000 18:55:21 +0200 (MET DST)
@RFC-Received: from unavarra.es (s169m138.unavarra.es [130.206.169.138]) by si.unavarra.es (8.9.3/8.9.1) with ESMTP id SAA06805 for ; Tue, 5 Sep 2000 18:55:10 +0200 (MET DST)
@RFC-Message-ID:
@RFC-X-Accept-Language: en
@RFC-Sender:
@RFC-Errors-To:
@RFC-X-Mailman-Version: 1.0rc2
@RFC-List-Id: FPC developers' list <fpc-devel.lists.freepascal.org> @RFC-X-BeenThere:
From: Javier Ros
To: ""
Reply-To:

Still trying to port a GTK FPC linux aplication to FPC (close to finish)...

I have noticed that the gtk file selection dialog doesn't work in win32.

I've made a simplistic C program finding the same behaviour.

The solution in the case of the C program is to use the -fnative-struct GCC flag
(mingw version of the compiler). This flag makes that the C struct alignement used
by gcc agrees to the one of the microsoft compilers.

The answer to the pascal case aparently translates directly to a problem of
record alignement. I suppouse that up to now some FPC guru is using Tor's GTK
libraries troublelessly, and that this is done trought the use of some magical
(at least for me) compiler $mode directive.

Does anybody know which?

Another somehow related question

When using recent versions of the GTK 1.3 dll I'm getting some missing functions, that stops the program from working. For example, i suppouse that some
of you have get the complaint

missing gtk_paned_set_gutter_size

arosed by the windows dinamical linker (of course the function is not explicitly called from my program).

I know that this is a question related to a incompatibility of the versions of the gtk import units and the tor's gtk win32 dll version
(1.2.? vs 1.3.?).
But diging into the gtk import library I've found a comment saying

{ Always use smartlinking for win32, this solves some undefined
functions
  in the development gtk versions which change often (PFV) }
{$ifdef win32}
  {$smartlink on}
{$endif}

would this directive solve the missing function problem?

Some thinkings...

I hope that some of the thinkings presented avobe, and others, with their answers
can be taken as a good starting point to a FPC-GTK-WIN32-FAQ. The readme file
at the root of the fpc gtk directory would be a nice place to write down such
instructions or considerations...........

Thanks a lot