Ros-dev

ros-dev@reactos.org

5791 discussions

BuildBot changes
by Colin Finck 24 Jan '11

24 Jan '11

Hello all, As you have certainly noticed, Amine, Olaf and me have upgraded the BuildBot some days ago. While doing this, we have added Olaf's build slaves which used to be available at http://reactos.ath.cx:8081. The new BuildBot version finally enables authenticated access for features like forcing a build. When you want to do this now, you have to enter your SVN credentials into the respective fields. If you are a tester and need these BuildBot features, please ask Aleksey or me for getting the appropriate credentials. In the next few days, we're probably going to make the testing results of his CMake bot available at http://reactos.org/testman and upload his builds to http://iso.reactos.org. I might also change the BuildBot Web page design. Now that the main part has got wider, we might remove the ReactOS design around it to get some space for the real important stuff. - Colin

4 4

Re: [ros-dev] [ros-diffs] [spetreolle] 50454: [CMAKE] Use cmake subversion capabilities to generate buildno.h and version.h.
by Colin Finck 21 Jan '11

21 Jan '11

spetreolle(a)svn.reactos.org wrote: > + subversion_wc_info(${CMAKE_CURRENT_SOURCE_DIR} SVNINFO) This will fail when we're not inside a working copy (see http://www.vtk.org/Bug/view.php?id=10200), because the macro outputs an error in this case. Please take a look into http://svn.reactos.org/svn/project-tools/trunk/rosev_ircsystem/src/CMakeLis… to see how you can easily reimplement this step properly. Apart from that, nice to see that we finally get rid of this old stuff! :-) - Colin

2 1

Re: [ros-dev] [ros-diffs] [tkreuzer] 50443: [GDI32] Fix broken portable implementation of _lrintf, 2nd try :)
by Aleksey Bragin 19 Jan '11

19 Jan '11

I bet _lrintf just needed more underscores! ;) On Jan 19, 2011, at 9:10 PM, tkreuzer(a)svn.reactos.org wrote: > Author: tkreuzer > Date: Wed Jan 19 18:10:50 2011 > New Revision: 50443 > > URL: http://svn.reactos.org/svn/reactos?rev=50443&view=rev > Log: > [GDI32] > Fix broken portable implementation of _lrintf, 2nd try :) > > Modified: > trunk/reactos/dll/win32/gdi32/include/gdi32p.h > > Modified: trunk/reactos/dll/win32/gdi32/include/gdi32p.h > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/dll/win32/ > gdi32/include/gdi32p.h?rev=50443&r1=50442&r2=50443&view=diff > ====================================================================== > ======== > --- trunk/reactos/dll/win32/gdi32/include/gdi32p.h [iso-8859-1] > (original) > +++ trunk/reactos/dll/win32/gdi32/include/gdi32p.h [iso-8859-1] Wed > Jan 19 18:10:50 2011 > @@ -413,7 +413,7 @@ > } > #else > /* slow, but portable */ > - return (int)(x >= 0 ? x+0.5 : x-0.5); > + return (int)(f >= 0 ? f+0.5 : f-0.5); > #endif > } > > >

1 0

Getting a Windows Server 2003 license for the project?
by Colin Finck 19 Jan '11

19 Jan '11

Hi everybody, I've been thinking about getting a license of an English Windows Server 2003 Standard 32-Bit for the project. It could be installed on one of our servers and be made available over RDP. This would enable project members to do development and testing work on our actual target platform. Considering that some developers even use a non-Windows platform for development work, it might simplify their work as well. We may as well use the license for other purposes (Buildslave, Testslave, whatever), but at least native building could be done by any Windows version. And in this case, I might be able to donate an XP Pro license myself (German though). As I don't know about the needs of the other members, I'd like to hear your opinion about my idea. It would also be nice to hear if anybody knows a cheap (but legal!) way to get such a license or can even donate one (e.g. unused OEM license shipped with a server, unused license after getting Server 2008, etc.) English Windows licenses are rare/expensive on German eBay, so this would only be a last resort :-) Cheers, Colin P.S.: If you have the opposite problem and actually need a Linux VM available over SSH/RDP (e.g. for testing build system changes), just let me know and I could set it up.

12 32

Re: [ros-dev] [ros-diffs] [mkupfer] 50398: - Fix draw of cursors in static controls - Last part of fix for bug #4874
by Timo Kreuzer 16 Jan '11

16 Jan '11

Shouldn't that work the same way (at least if there is any hbmColor)? Cursors can have a different size as well, so maybe we should rather query the hbmMask, if hbmColor is 0(=) I didn't bother reading the context of the function though. Am 16.01.2011 13:51, schrieb mkupfer(a)svn.reactos.org: > Author: mkupfer > Date: Sun Jan 16 12:51:14 2011 > New Revision: 50398 > > URL: http://svn.reactos.org/svn/reactos?rev=50398&view=rev > Log: > - Fix draw of cursors in static controls > - Last part of fix for bug #4874 > > Modified: > trunk/reactos/dll/win32/user32/controls/static.c > > Modified: trunk/reactos/dll/win32/user32/controls/static.c > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/dll/win32/user32/controls/… > ============================================================================== > --- trunk/reactos/dll/win32/user32/controls/static.c [iso-8859-1] (original) > +++ trunk/reactos/dll/win32/user32/controls/static.c [iso-8859-1] Sun Jan 16 12:51:14 2011 > @@ -863,7 +863,15 @@ > else > { > BITMAP bm; > - if (!GetObjectW(info.hbmColor, sizeof(BITMAP),&bm)) return; > + if (info.fIcon) > + { > + GetObjectW(info.hbmColor, sizeof(BITMAP),&bm); > + } > + else > + { > + bm.bmWidth = GetSystemMetrics(SM_CXCURSOR); > + bm.bmHeight = GetSystemMetrics(SM_CYCURSOR); > + } > if (style& SS_CENTERIMAGE) > { > iconRect.left = (rc.right - rc.left) / 2 - bm.bmWidth / 2; > > >

1 0

Re: [ros-dev] [ros-diffs] [weiden] 50396: Portability bugfixes
by Ged Murphy 16 Jan '11

16 Jan '11

Thomas the 2nd?? On 16 January 2011 01:32, <weiden(a)svn.reactos.org> wrote: > Author: weiden > Date: Sun Jan 16 01:32:45 2011 > New Revision: 50396 > > URL: http://svn.reactos.org/svn/reactos?rev=50396&view=rev > Log: > Portability bugfixes > > Modified: > trunk/reactos/lib/rtl/srw.c > > Modified: trunk/reactos/lib/rtl/srw.c > URL: > http://svn.reactos.org/svn/reactos/trunk/reactos/lib/rtl/srw.c?rev=50396&r1… > > ============================================================================== > --- trunk/reactos/lib/rtl/srw.c [iso-8859-1] (original) > +++ trunk/reactos/lib/rtl/srw.c [iso-8859-1] Sun Jan 16 01:32:45 2011 > @@ -131,7 +131,7 @@ > } >

2 1

Re: [ros-dev] [ros-diffs] [gadamopoulos] 50381: [rosautotest] -Implement closing any dialog that shows and stays visible for some time. This way rosautotest can now continue if a test application crashes or asserts.
by Timo Kreuzer 13 Jan '11

13 Jan '11

Am 13.01.2011 10:58, schrieb gadamopoulos(a)svn.reactos.org: > Author: gadamopoulos > Date: Thu Jan 13 09:58:04 2011 > New Revision: 50381 > > URL: http://svn.reactos.org/svn/reactos?rev=50381&view=rev > Log: > [rosautotest] > -Implement closing any dialog that shows and stays visible for some time. This way rosautotest can now continue if a test application crashes or asserts. > Seems it causes a regression in testbot: *http://www.reactos.org/testman/compare.php?ids=4355,4356 *

1 0

Re: [ros-dev] [ros-diffs] [fireball] 50309: [HEAP] - Fix a typo (assignment) in the ASSERT, spotted by Johannes Anderwald.
by Ged Murphy 10 Jan '11

10 Jan '11

fireball(a)svn.reactos.org wrote: > if (Locked) RtlLeaveHeapLock(Heap->LockVariable); > + DbgBreakPoint(); > return FALSE; Doh

3 2

coordinate transformation
by Timo Kreuzer 06 Jan '11

06 Jan '11

Hi, I'm currently working on the coordinate transformation code in win32k. While Windows does this in win32k, I wonder why its not done in user mode and if we could depart from windows design here. The advantages would be decreased complexity in the kernel and no use of floating point emulation on x86 (better performance). Are there any things that would make it unreasonable? I don't plan to go and change this now, I just like to discuss whether we might or might not do that at some point. Timo

2 1

Re: [ros-dev] [ros-diffs] [tkreuzer] 50243: [CMAKE] Add generation of a depencency graph for shared libraries. The output is a graphml file. Can be enabled with GENERATE_DEPENDENCY_GRAPH switch.
by Ged Murphy 01 Jan '11

01 Jan '11

I didn't know about this feature. Will this help the effort towards a minwin effort in fixing the dependency hierarchy (mess)? Ged. On 31 December 2010 16:49, <tkreuzer(a)svn.reactos.org> wrote: > Author: tkreuzer > Date: Fri Dec 31 16:49:49 2010 > New Revision: 50243 > > URL: http://svn.reactos.org/svn/reactos?rev=50243&view=rev > Log: > [CMAKE] > Add generation of a depencency graph for shared libraries. > The output is a graphml file. Can be enabled with GENERATE_DEPENDENCY_GRAPH > switch. > >

3 3

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev