Re: [Harbour] Ghost cursor with GTWVT

2010-04-05 Thread Xavi
Hi Viktor, Ok. Best regards, Xavi El 05/04/2010 0:12, Viktor Szakáts escribió: Hi, Me neither, I guess that Tortoise always makes the patch with the latest version... I correct it manually, no problem. Maybe it's better to do all the changes in the repository or don't change any in a file

Re: [Harbour] Ghost cursor with GTWVT

2010-04-04 Thread Viktor Szakáts
Hi, > Me neither, I guess that Tortoise always makes the patch with the latest > version... I correct it manually, no problem. > Maybe it's better to do all the changes in the repository or don't change any > in a file under discussion. I don't know. The best is to send .diffs against latest so

Re: [Harbour] Ghost cursor with GTWVT

2010-04-04 Thread Xavi
Hi Viktor, I don't understand this patch, it's against an older version of the files (missing my last minor update), you added Me neither, I guess that Tortoise always makes the patch with the latest version... I correct it manually, no problem. Maybe it's better to do all the changes in the

Re: [Harbour] Ghost cursor with GTWVT

2010-04-04 Thread Viktor Szakáts
Hi Xavi, I don't understand this patch, it's against an older version of the files (missing my last minor update), you added conditional code for layered mode (why?) and you still try to force your setcancel hack. It also employs trick to track focus, and I'm not sure I like it. IMO the issue to

Re: [Harbour] Ghost cursor with GTWVT

2010-04-01 Thread Viktor Szakáts
Hi, > El 01/04/2010 19:26, Viktor Szakáts escribió: >> BTW there were a lot of screen painting irregularities without >> WS_EX_LAYERED, when the window was behind another >> window and foremost window was moved. It also fixes painting >> issue when app is doing a loop, or otherwise unresponsive. >

Re: [Harbour] Ghost cursor with GTWVT

2010-04-01 Thread Xavi
... In typical code like loops with pure RDDs you can force this with SetPos .- nCursor := SetCursor( 0 ) WHILE !EOF() SetPos( 0, 0 ) ... SKIP() ENDDO SetCursor( nCursor ) Of course it supports many optimizations. if( ++n % 1 == 0 ); SetPos( 0, 0 ); endif Sorry, ... and I th

Re: [Harbour] Ghost cursor with GTWVT

2010-04-01 Thread Xavi
El 01/04/2010 19:26, Viktor Szakáts escribió: BTW there were a lot of screen painting irregularities without WS_EX_LAYERED, when the window was behind another window and foremost window was moved. It also fixes painting issue when app is doing a loop, or otherwise unresponsive. Thanks but this

Re: [Harbour] Ghost cursor with GTWVT

2010-04-01 Thread Viktor Szakáts
Hi, > And added HB_TRACEs to see the behavior of the caret. > Note that, with the current code, try creating the caret without focus > (window is the background) and always show caret if exist. All of this can > affect in the tests. > Attached gtwvt.zip (gtwvt.h,gtwvt.c) with the changes, I think

Re: [Harbour] Ghost cursor with GTWVT

2010-03-30 Thread Przemysław Czerpak
On Tue, 30 Mar 2010, Xavi wrote: Hi, > >ShowCaret() optimization and maybe it will help you to find when the > >cursor is wrongly refreshed. > I think I've found the root of the issue. > Please don't nothing about it. Let me do some checking, ...tonight. OK and thank you. best regards, Przemek

Re: [Harbour] Ghost cursor with GTWVT

2010-03-30 Thread Xavi
Hi Przemek, ShowCaret() optimization and maybe it will help you to find when the cursor is wrongly refreshed. I think I've found the root of the issue. Please don't nothing about it. Let me do some checking, ...tonight. Best regards, Xavi ___ Harbou

Re: [Harbour] Ghost cursor with GTWVT

2010-03-30 Thread Xavi
Hi Przemek, [...] Anyhow I always prefer to detect real reason of problem first to check if such modification is really necessary and the problem is not side effect of some other code or bugs. Otherwise each of us will add some new peaces of code which only mask real bugs making the whole code

Re: [Harbour] Ghost cursor with GTWVT

2010-03-30 Thread Przemysław Czerpak
On Tue, 30 Mar 2010, Xavi wrote: Hi, > Sometimes GTWVT leaves the previous mark of cursor. Attached picture > to illustrate the issue. > This happens because the pre-cursor position not is marked with > InvalidateRect (RedrawDiff not find differences) getting knocked out > of WM_PAINT. So it mea

Re: [Harbour] Ghost cursor with GTWVT

2010-03-29 Thread Viktor Szakáts
Hi Xavi, My only comment is that some sort of "setcancel" patch is mixed into your attached patches, which should definitely be deleted. Brgds, Viktor On 2010 Mar 30, at 05:28, Xavi wrote: > Hi all, > > Sometimes GTWVT leaves the previous mark of cursor. Attached picture to > illustrate the

[Harbour] Ghost cursor with GTWVT

2010-03-29 Thread Xavi
Hi all, Sometimes GTWVT leaves the previous mark of cursor. Attached picture to illustrate the issue. This happens because the pre-cursor position not is marked with InvalidateRect (RedrawDiff not find differences) getting knocked out of WM_PAINT. In GTWVG not happen because overwrite PutChar