February 2008 - Wine-devel - list.winehq.org

Help me(Ajudem-me)C compiler cannot create executables
by flavio melo 26 Feb '08

26 Feb '08

Hello I'm Flavio Melo(brazilian)! Well, i use the ubuntu 7.04, and there is any time that i try use the wine,but i no get install it, i can't for net because i haven't internet (repository), but i download the file wine-0.9.54 of your site in the other computer(in my job), and, i follow the instruccions(i think there are corrects, get it read manuals) and my terminal said: ---------------------------------------------------------------------------- flavio(a)Discipulo:~$ cd wine flavio(a)Discipulo:~/wine$ cd tools flavio(a)Discipulo:~/wine/tools$ ./wineinstall WINE Installer v0.75 ~/wine ~/wine/tools Running configure... configure: creating cache config.cache checking build system type... i686-pc-linux-gnulibc1 checking host system type... i686-pc-linux-gnulibc1 checking whether make sets $(MAKE)... yes checking for gcc... gcc checking for C compiler default output file name... configure: error: C compiler cannot create executables See `config.log' for more details. Configure failed, aborting install. flavio(a)Discipulo:~/wine/tools$ ----------------------------------------------------------------------------- I wrote to you for know if you can help me(Please), what i must make for my gcc compile? Thank for attencion! Meu testemunho... http://www.semearcristo.net/art/art_00007.aspx?idPag=1&idReg=1&idCat=6&strB… --------------------------------- Abra sua conta no Yahoo! Mail, o único sem limite de espaço para armazenamento!

3 2

Re: comctl32: rebar[1/4]: an implementation of RB_SIZETORECT (fixes bug #7774)
by John Klehm 26 Feb '08

26 Feb '08

Missing patch On Tue, Feb 26, 2008 at 3:36 PM, Mikołaj Zalewski <mikolaj(a)zalewski.pl> wrote: > I couldn't work out the algorithm to make it pixel-for-pixel identical > as on Windows but this function implements the general idea that I hope > will work for applications. > > >

1 0

re: gdi32: Protect against short sized LOGPALETTE in CreatePalette.
by Dan Kegel 26 Feb '08

26 Feb '08

Hi, your patch uses IsBadReadPtr(), but that function is deprecated in favor of using an exception handler. See http://www.winehq.org/pipermail/wine-devel/2006-April/046560.html Here's an example (sorry it's a messy one, it's just the first one google found): http://www.winehq.org/pipermail/wine-patches/2004-November/013791.html Please try recoding using that approach. If you have trouble getting this patch in, you might want to file a bug at http://bugs.winehq.org for your problem and link to your patch; that will record the problem so somebody else can pick it up later. - Dan

3 2

re: gdi32: Patch review request
by Dan Kegel 26 Feb '08

26 Feb '08

Daniel Mario Vega wrote: >A few days ago I sent 2 patches for gdi32 to wine-patches >that passed unnoticed. I would like if someone could give > me some feedback on them. OK. I replied to each of them just now with my guess. Don't worry, patches being ignored is normal, it just means you have to improve them if possible and resend. If you do resend them, be sure to give the same subject as last time, with '(resend)' appended if no changes made, or '(try 2)' appended if there were changes, and append a note about the changes to the description.

1 0

re: gdi32: StartDocA support for legacy apps with incomplete DOCINFOA parameter.
by Dan Kegel 26 Feb '08

26 Feb '08

This patch looks nice, but it could use a conformance test. Please add one and resend. And consider filing a bug and referring to the bug in the patch description, that never hurts. - Dan

1 0

Re: [try4] user32: cursoricon.c. Fix 16b BMP handling with BT_BITFIELD.
by Dmitry Timoshkov 26 Feb '08

26 Feb '08

"Alexander Demakin" <avd(a)etersoft.ru> wrote: > This patch fixes image header handling with BT_BITFIELD value in biCompression > field. 3 DWORDs of mask were included into image bits, it caused some color > troubles and image horizontal shift. Size of bitmap info structure includes > now this bytes. > Changelog: > user32.dll Fix 16b BMP handling with BT_BITFIELD. I wonder whether gcc complained loudly that 'result' may be used uninitialized. -- Dmitry.

1 1

marshalled ITypeComp
by Guillermo Winkler 26 Feb '08

26 Feb '08

Hi, I'm having some problems regarding ITypeInfo/ITypeComp implementation. IDispatch* pdisp; ITypeInfo* ptinfo; HRESULT hr = pdisp->GetTypeInfo(0, LOCALE_SYSTEM_DEFAULT, &typeinfo); hr = ptinfo->GetTypeComp(&ptcomp); pdisp is created from hr = CoCreateInstance(clsid, NULL, CLSCTX_SERVER, IID_IDispatch, (void**)&pdisp); That is, the server is an outprocess exe. In this case marshalling of ITypeComp fails in first instance because ITypeComp_fnQueryInterface (oleaut32\typelib.c) resolves on -> ITypeInfo_QueryInterface (oleaut32\typelib.c) That is not considering IID_ITypeComp as a valid alternative(but it is) if(IsEqualIID(riid, &IID_IUnknown) || IsEqualIID(riid,&IID_ITypeInfo)|| IsEqualIID(riid,&IID_ITypeInfo2)) *ppvObject = This; When I fixed this, I have a different problem now: fixme:ole:DllGetClassObject CLSID: {00020425-0000-0000-C000-000000000046} IID: {D5F569D0-593B-101A-B569-08002B2DBF7A} That is PSTypeComp to be marshalled asking for factory buffer IID_IPSFactoryBuffer. I would like to follow up with someone (if available) just to be sure I'm not wasting time on the wrong track. Thanks, Guillermo

2 2

gdi32: Patch review request
by Daniel Mario Vega 26 Feb '08

26 Feb '08

Hi! A few days ago I sent 2 patches for gdi32 to wine-patches that passed unnoticed. I would like if someone could give me some feedback on them. Let me first give you some background about these patches. There is an application called "Compumap" that is a street map of some Argentine cities. It is rather popular here because some years ago it was distributed with a newspaper. The program works quite well under Wine (although the installer still has some problems), but when you try to print a map it crashes. A few weeks ago I had some free time and I decided to try to find out why it crashes. I run it through the Wine Debugger and also under a debugger in Windows XP. The problem happened to be two API calls with incomplete/invalid parameters. Windows XP returned NULL or ignore the wrong data, while Wine crashes. 1) gdi32: Protect against short sized LOGPALETTE in CreatePalette. http://www.winehq.org/pipermail/wine-patches/2008-February/050354.html This crash was caused by "CreatePalette(HDC, LOGPALETTE)" being called with a short allocated palette. Windows XP returned NULL in that case. I tried to reproduce that behavior by first checking the parameter with "IsBadReadPtr()" function. 2) gdi32: StartDocA support for legacy apps with incomplete DOCINFOA parameter. http://www.winehq.org/pipermail/wine-patches/2008-February/050357.html This issue was a call to "StartDocA(HDC, DOCINFOA)" with structure DOCINFOA not completely initialized (there is garbage in the last 2 fields). I debugged this API and found that Windows XP ignored those fields when the data "seems to be invalid". I think that the last 2 fields were introduced in Win32 and some applications ported from Win16 "forget" to clear them. I simulated Windows XP behaviour by putting some safe checks on those fields. With those 2 fixes, the application is able to print under Wine. Both patches are short and rather easy to review. I will appreciate if someone could give me some feedback on them. Thanks ... -- Daniel Mario Vega

1 0

Re: opengl32/tests: Add the trailing '\n' to an ok() call and make a spelling fix.
by Paul Vriens 26 Feb '08

26 Feb '08

Francois Gouget wrote: > --- > dlls/opengl32/tests/opengl.c | 4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/dlls/opengl32/tests/opengl.c b/dlls/opengl32/tests/opengl.c > index 3836ba7..344e2c6 100644 > --- a/dlls/opengl32/tests/opengl.c > +++ b/dlls/opengl32/tests/opengl.c > @@ -384,9 +384,9 @@ START_TEST(opengl) > > /* We shouldn't be able to create a context from a hdc which doesn't have a pixel format set */ > hglrc = wglCreateContext(hdc); > - ok(hglrc == NULL, "wglCreateContext should fail when no pixel format has been set, but it passed"); > + ok(hglrc == NULL, "wglCreateContext should fail when no pixel format has been set, but it passed\n"); > error = GetLastError(); > - ok(error == ERROR_INVALID_PIXEL_FORMAT, "expected ERROR_INVALID_PIXEL_FORMAT for wglCreateContext without a pixelformat set, but recevied %#x", error); > + ok(error == ERROR_INVALID_PIXEL_FORMAT, "expected ERROR_INVALID_PIXEL_FORMAT for wglCreateContext without a pixelformat set, but recieved %#x\n", error); So where is the spelling fix ;-) Shouldn't that be "received"? -- Cheers, Paul.

1 0

Re: d3dx9: Add IMPORTLIB to every Makefile.in [PATCH 4/9]
by Alexandre Julliard 26 Feb '08

26 Feb '08

tony.wasserka(a)freenet.de writes: >>From 881b8de1e044ff90bcb543501208a9cd879cbc16 Mon Sep 17 00:00:00 2001 > From: Tony Wasserka <tony.wasserka(a)freenet.de> > Date: Sat, 23 Feb 2008 13:55:40 +0100 > Subject: [PATCH] d3dx9: Add IMPORTLIB to every Makefile.in That shouldn't be needed, there are no such import libs on Windows. -- Alexandre Julliard julliard(a)winehq.org

1 0