[PATCH] + Re: [ANNOUNCEMENT] Server Test 57

[Jehan]

Harold Hunt wrote:
> Jehan,
> 
> That last message fired off before I was done... my Ctrl key is sticking so
> when I hit Enter it sent a Ctrl+Enter, which sends the message.
> 
> 
>>I don't think you need to manage the cursor in WM_ACTIVATEAPP, managing
>>it in WM_ACTIVATE should be enough because, if I'm not wrong, XWin will
>>always receive a WM_ACTIVATE with a WM_ACTIVATEAPP.
> 
> 
> Then again, your version gets trapped in a state where the cursor is shown,
> whereas I haven't seen that problem with Test57 yet :)

Actually, I'm now using Test57, and I haven't seen my problem either.
And I was not saying that this was related to my problem, I'm just 
saying that it looks like unecessary code.


> I remember spending a *lot* of time reading the docs about WM_ACTIVATE and
> WM_ACTIVATEAPP to figure out where to show and hide the cursor.  There were
> constant bug repots until I got it right, so I'm going to tend to believe
> that the show/hide in WM_ACTIVATEAPP is required, until someone presents me
> with some sample code and documentation exerpts that shows me that 1)
> they've spent more time on the problem than I have and 2) that the show/hide
> code in WM_ACTIVATEAPP is indeed not needed.

But the code changed since and got simplified a lot, especially 
everything about hiding/showing the cursor the previous screen and all 
that stuff is gone. So maybe the bug has been flushed with the code :).
But that's true that it doesn't cost much to have it anyway and that 
you're the boss (^_^)

By the way, here is another patch that should prevent some problems with 
multiroot systems (not tested).

	Jehan
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: active.patch
URL: <http://cygwin.com/pipermail/cygwin-xfree/attachments/20020610/a5e405da/attachment.ksh>