Re: [ros-dev] [ros-diffs] [tkreuzer] 50605: [WIN32K] Fix a bug in calculation of the background color for 32bpp -> 1bpp color translations. Fixes bug 2372

[Gabriel ilardi]

What about the patch in that bug http://www.reactos.org/bugzilla/attachment.cgi?id=1568 :) ?
> Date: Fri, 4 Feb 2011 13:53:50 +0000
> To: [e-mail]
> From: [e-mail]
> Subject: [ros-diffs] [tkreuzer] 50605: [WIN32K] Fix a bug in calculation of the background color for 32bpp -> 1bpp color translations. Fixes bug 2372
>
> Author: tkreuzer
> Date: Fri Feb  4 13:53:49 2011
> New Revision: 50605
>
> URL: http://svn.reactos.org/svn/reactos?rev=50605&view=rev
> Log:
> [WIN32K]
> Fix a bug in calculation of the background color for 32bpp -> 1bpp color translations.
> Fixes bug 2372
>
> Modified:
>     trunk/reactos/subsystems/win32/win32k/eng/xlate.c
>
> Modified: trunk/reactos/subsystems/win32/win32k/eng/xlate.c
> URL: http://svn.reactos.org/svn/reactos/trunk/reactos/subsystems/win32/win32k/eng...
> ==============================================================================
> --- trunk/reactos/subsystems/win32/win32k/eng/xlate.c [iso-8859-1] (original)
> +++ trunk/reactos/subsystems/win32/win32k/eng/xlate.c [iso-8859-1] Fri Feb  4 13:53:49 2011
> @@ -91,7 +91,7 @@
>      iColor >>= 13;
>      iNewColor |= iColor & 0x3E0;
> /* Copy green */
> /* Copy blue */
> iColor >>= 13;
> iNewColor |= iColor & 0x1F;
>
> @@ -417,11 +417,11 @@
>              pexlo->aulXlate[0] =
>                  PALETTE_ulGetNearestPaletteIndex(ppalSrc, crSrcBackColor);
>          }
>    else if (ppalSrc->flFlags & PAL_RGB)
>    {
>        pexlo->aulXlate[0] = crSrcBackColor;
>    }
>    else if (ppalSrc->flFlags & PAL_BGR)
>    {
>        pexlo->aulXlate[0] = crSrcBackColor;
>    }
>    else if (ppalSrc->flFlags & PAL_RGB)
>    {
>        pexlo->aulXlate[0] = RGB(GetBValue(crSrcBackColor),
>                                 GetGValue(crSrcBackColor),

[Timo Kreuzer]

Am 04.02.2011 14:59, schrieb Gabriel ilardi:
> What about the patch in that bug 
> http://www.reactos.org/bugzilla/attachment.cgi?id=1568 :) ?

I tested the patch.  It doesn't look as good as it looks currently. To 
make it look properly we need alpha channel.
If someone could make a proper alpha bitmap, I can add the neccessary 
code to sysdm.

I also just noticed that there is another drawing bug, which prevents 
the current code from working properly (while the attached testcase 
works). I'll try to fix that one as well.

Timo