IIRC, it's a hack because neither x nor y is usually -32000, so those are flag values instead of fixed actual values. You have enough physical screens laid out in a line it's possible (16 1080p + a 1280 wide one), but still unlikely on one system given it would take multiple video cards to drive that many. The desktop as a decorated window hides the resize frame and title bar by positioning outside the effective area of a multiple screen layout and using a width and height that just encompasses the layout in Display Properties as the client area dimensions, with (0,0) of the viewports being one of the corners of the primary screen's client area. This forces at least one position coordinate to be negative by at least the width of the resize frame. I forget whether this is documented in the DDK or SDK, sorry, and which edition.
------ Original message------From: James TaborDate: Sat, Feb 13, 2016 12:03 AMTo: ReactOS Development List;Cc: Subject:[ros-dev] [ros-diffs] [hbelusca] 70713: [WIN32SS] - Fix orthograph - Mention in the code that some stuff are hacks & need to be fixed. - Add parentheses around bit checks.
Hello,
Why is this a hack?
Thanks,
James
Modified: trunk/reactos/win32ss/user/ntuser/winpos.c
URL: http://svn.reactos.org/svn/reactos/trunk/reactos/win32ss/user/ntuser/winpos…
==============================================================================
--- trunk/reactos/win32ss/user/ntuser/winpos.c [iso-8859-1] (original)
+++ trunk/reactos/win32ss/user/ntuser/winpos.c [iso-8859-1] Fri Feb 12
16:40:36 2016
@@ -724,8 +724,8 @@
pwndParent = Window->spwndParent;
if (pwndParent == UserGetDesktopWindow())
{
- ERR("Parent is Desktop, Min off screen!\n");
- /* ReactOS doesn't support iconic minimize to desktop */
+ ERR("FIXME: Parent is Desktop, Min off screen!\n");
+ /* FIXME: ReactOS doesn't support iconic minimize to desktop */
Pos->x = Pos->y = -32000;
Window->InternalPos.flags |= WPF_MININIT;
Window->InternalPos.IconPos.x = Pos->x;
_______________________________________________
Ros-dev mailing list
Ros-dev(a)reactos.org
http://www.reactos.org/mailman/listinfo/ros-dev
On Wed, Feb 10, 2016 at 5:28 PM, <hbelusca(a)svn.reactos.org> wrote:
> else // if (SecurityData->ReferenceCount <= 1)
Is the intent to free the data while SecurityData->RefCount will remain at 1?!
Best regards,
Alex Ionescu
I think you guys took more time to debug and mess with the DPRINT then
it would've taken to implement the function. Come on guys.
Best regards,
Alex Ionescu
On Mon, Feb 1, 2016 at 3:10 PM, <hbelusca(a)svn.reactos.org> wrote:
> Author: hbelusca
> Date: Mon Feb 1 23:10:38 2016
> New Revision: 70674
>
> URL: http://svn.reactos.org/svn/reactos?rev=70674&view=rev
> Log:
> [CMLIB]: Demote the DPRINT1 saying that we leak the security block descriptor to a DPRINT. People (Alex & me) working on cmlib already know this. "Fixes" timeout problems of the testbots due to spamming this dprint.
>
> Modified:
> trunk/reactos/lib/cmlib/cmkeydel.c
>
> Modified: trunk/reactos/lib/cmlib/cmkeydel.c
> URL: http://svn.reactos.org/svn/reactos/trunk/reactos/lib/cmlib/cmkeydel.c?rev=7…
> ==============================================================================
> --- trunk/reactos/lib/cmlib/cmkeydel.c [iso-8859-1] (original)
> +++ trunk/reactos/lib/cmlib/cmkeydel.c [iso-8859-1] Mon Feb 1 23:10:38 2016
> @@ -215,7 +215,7 @@
> }
>
> /* FIXME: This leaks the security desriptor! */
> - DPRINT1("Potentially leaking key security descriptor. Please call CmpFreeSecurityDescriptor\n");
> + DPRINT("Potentially leaking key security descriptor. Please call CmpFreeSecurityDescriptor\n");
>
> /* Free the key body itself, and then return our status */
> if (!CmpFreeKeyBody(Hive, Cell)) return STATUS_INSUFFICIENT_RESOURCES;
>
>
On 2016-01-24 19.00, ros-dev-request(a)reactos.org wrote:
> + if (Length != Length)
Now that's what we call an unlikely condition :-D
The CPU branch predictor will have no problem.
Magic indeed.. Hope it's not indicative..
L.
For everyone who thinks this looks terrible: don't view it in your
browser. Instead download it and open it with a proper image viewer.
Then your eyes stop bleeding and it actually looks good :)
Timo
Am 06.01.2016 um 10:31 schrieb cfinck(a)svn.reactos.org:
> Author: cfinck
> Date: Wed Jan 6 09:31:53 2016
> New Revision: 70507
>
> URL: http://svn.reactos.org/svn/reactos?rev=70507&view=rev
> Log:
> [BOOTVID]
> Change the Blue Screen Font hardcoded into bootvid.dll to the Open Source "Anonymous Pro" font, which makes a nice 8x13 font.
> Blue Screens now look like this: http://fs5.directupload.net/images/160106/ehv6245t.png
>
> [BOOTVID_FONT_GENERATOR]
> In case you find an even better font, you now have a nice little tool to test that font and generate the corresponding FontData array for bootvid.
>
> Added:
> trunk/rosapps/applications/devutils/bootvid_font_generator/
> trunk/rosapps/applications/devutils/bootvid_font_generator/CMakeLists.txt (with props)
> trunk/rosapps/applications/devutils/bootvid_font_generator/bootvid_font_generator.c (with props)
> Modified:
> trunk/reactos/drivers/base/bootvid/arm/bootdata.c
> trunk/reactos/drivers/base/bootvid/i386/bootdata.c
> trunk/rosapps/applications/devutils/CMakeLists.txt
>
> [This mail would be too long, it was shortened to contain the URLs only.]
>
> Modified: trunk/reactos/drivers/base/bootvid/arm/bootdata.c
> URL: http://svn.reactos.org/svn/reactos/trunk/reactos/drivers/base/bootvid/arm/b…
>
> Modified: trunk/reactos/drivers/base/bootvid/i386/bootdata.c
> URL: http://svn.reactos.org/svn/reactos/trunk/reactos/drivers/base/bootvid/i386/…
>
> Modified: trunk/rosapps/applications/devutils/CMakeLists.txt
> URL: http://svn.reactos.org/svn/reactos/trunk/rosapps/applications/devutils/CMak…
>
> Added: trunk/rosapps/applications/devutils/bootvid_font_generator/CMakeLists.txt
> URL: http://svn.reactos.org/svn/reactos/trunk/rosapps/applications/devutils/boot…
>
> Added: trunk/rosapps/applications/devutils/bootvid_font_generator/bootvid_font_generator.c
> URL: http://svn.reactos.org/svn/reactos/trunk/rosapps/applications/devutils/boot…
>
>
>
Can I have commit access to !trunk branches for not forking to Git?
(for reviving the POSIX subsystem with the Newlib C library and adding
pthreads support)
Thanks,