On Mon, Feb 18, 2002 at 08:14:53PM +0100, Tim van Erven <[EMAIL PROTECTED]> wrote:
> On Fri, Feb 15, 2002 at 04:04:09PM -0500, Andriy Gapon <[EMAIL PROTECTED]> wrote:
>>
>>> Date: Fri, 15 Feb 2002 06:42:40 +0530 (IST)
>>> From: Kingsly John <[EMAIL PROTECTED]>
>>> Subject: Re: [Licq-devel] proposal or rather just a hint
>>>
>>>
>>> All my floaties do stay right on top... without any of the elaborate steps
>>> you have mentioned.
>>
>> I wish I could say the same for myself.
>> which version of Qt do you have ? I have 2.3.1 here (FreeBSD port) and
>> looking into the source:
>> ...
>> qwidget_x11.cpp:276
>> } else { // Style_NoBorder
>> setWFlags( WX11BypassWM ); // ### compatibility
>> }
>> ...
>> qwidget_x11.cpp:290
>> if ( testWFlags( WX11BypassWM ) ) {
>> wsa.override_redirect = TRUE;
>> wsa_mask |= CWOverrideRedirect;
>> }
>> ...
>> as I understand, override_redirect means that when this window is mapped
>> window manager will not receive notification event and will not start
>> to manage this window, so it will be able to put other managed windows
>> above floaty windows. At least that's true for afterstep.
>>
>>> I'm using fvwm and it works perfectly without any extra configuration ...
>>> have you tried with any of the other windows managers ?? maybe it's an
>>> afterstep issue.
>>
>> I had the same problem with windowmaker and judging from the code snippet
>> above I can not understand how fvwm does it.
>>
>>
>>> I don't think anyone would be interested in a floaty that gets hidden ...
>>> my desktop is mostly so cluttered with xterms that if a floaty were to go
>>> anyplace other than on top... I'll probably not see it again for a long
>>> long time!
>>
>> that's the reason why I wanted to share this. if we have other unlucky
>> souls stuck with afterstep and having the same issue, this information may be
>> worth including into some README.afterstep
>
> I'm using afterstep too and it appears I'm experiencing the same
> symptoms. Now I never use floaties, so it's not much of a problem for
> me, but that doesn't change the fact that it's not the right behaviour.
I should add to that that I'm running Linux (Debian woody) with qt
2.3.1 on X 4.1.0.
--
Tim van Erven <[EMAIL PROTECTED]>
_______________________________________________
Licq-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/licq-devel
