June 2005 - Wine-devel - list.winehq.org

Re: d3d9: Correct number of params for ValidateVertexShader
by Oliver Stieber 28 Jun '05

28 Jun '05

The first parameter can be named (it's pFunction, a pointer to the shader function to be validated), and d3d8 also needs to be patched. The scrapland demo can be used to test that d3d8 is correct. (http://www.download.com/Scrapland-demo/3000-7563_4-10325931.html) --- James Hawkins <truiken(a)gmail.com> wrote: > Hi, > > Changelog > * Correct number of params for ValidateVertexShader. > > -- > James Hawkins > > Index: dlls/d3d9/d3d9.spec > =================================================================== > RCS file: /home/wine/wine/dlls/d3d9/d3d9.spec,v > retrieving revision 1.1 > diff -p -U3 -r1.1 d3d9.spec > --- dlls/d3d9/d3d9.spec 1 Jul 2003 01:09:17 -0000 > 1.1 > +++ dlls/d3d9/d3d9.spec 28 Jun 2005 05:35:43 -0000 > @@ -1,5 +1,5 @@ > @ stdcall D3D9GetSWInfo() > @ stdcall DebugSetMute() > @ stdcall Direct3DCreate9(long) > -@ stdcall ValidatePixelShader(ptr ptr) > +@ stdcall ValidatePixelShader(ptr ptr ptr ptr ptr) > @ stdcall ValidateVertexShader(ptr ptr) > Index: dlls/d3d9/d3d9_main.c > ================================================================== > RCS file: /home/wine/wine/dlls/d3d9/d3d9_main.c,v > retrieving revision 1.7 > diff -p -U3 -r1.7 d3d9_main.c > --- dlls/d3d9/d3d9_main.c 6 May 2005 19:38:50 > -0000 1.7 > +++ dlls/d3d9/d3d9_main.c 28 Jun 2005 05:35:43 > -0000 > @@ -72,9 +72,9 @@ BOOL WINAPI DllMain(HINSTANCE > hInstDLL, > > /*********************************************************************** > * ValidateVertexShader (D3D9.@) > */ > -BOOL WINAPI ValidateVertexShader(LPVOID what, > LPVOID toto) { > - FIXME("(void): stub: %p %p\n", what, toto); > - return TRUE; > +BOOL WINAPI ValidateVertexShader(LPVOID x1, LPVOID > x2, LPVOID x3, LPVOID x4, LPVOID x5) { > + FIXME("(%p,%p,%p,%p,%p) stub\n", x1, x2, x3, > x4, x5); > + return TRUE; > } > > /*********************************************************************** ___________________________________________________________ How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos http://uk.photos.yahoo.com

2 2

AppDB
by Mitchell Mebane 27 Jun '05

27 Jun '05

Hi, My Summer of Code proposal to work on the AppDB wasn't accepted, but I still wish to do some of the things I listed. Anyway, I was looking at the code for image handling, and I see it is written for GD 1. Thus, any image which has to be resized gets cut down to 256 colors, and is simply resized instead of being resampled, which makes the image blocky. According to the PHP manual, these better functions require PHP 4.0.6 and GD 2.0.1 or higher. Is the AppDB server really running such an old version of PHP that this is required? --Mitchell Mebane -- I find that a great part of the information I have was acquired by looking up something and finding something else on the way. -- Franklin P. Adams

4 3

Soliciting suggestions for AppDB
by Mitchell Mebane 27 Jun '05

27 Jun '05

Hi all, I'm putting together a Summer of Code proposal for working on the AppDB. I've been talking with Chris Morgan, and he has a few suggestions, but I was looking for more. Does anybody have any features they'd like added to the AppDB, quirks they'd like worked out, or things of that nature? Thanks, Mitchell Mebane -- The roots of education are bitter, but the fruit is sweet. --Aristotle

6 10

Version Info on user.exe
by Uwe Bonnes 27 Jun '05

27 Jun '05

Hallo, the online banking part of the T-Online Suite was an often demanded application on Linuxtag. The installer at http://service.t-online.de/c/06/50/11/650114.html is a 72 MByte download. Current CVS displays a blank screen on some splash screen, making navigating impossible. Older CVS versions (some days ago) displayed fine. I'll investigate this later, keeping the list informed. The real problem however is that the final installer asked for the Country version of user.exe. The installer expects 0407 for Germany. Wine doesn't supply that information, so the installer aborts. How should wine handle that situation? We probably don't want to compile contry specific versions of user.exe. -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- trace:ver:GetFileVersionInfoSizeW (L"c:\\windows\\system\\user.exe",0x7fd4e9b0) trace:ver:VERSION_GetFileVersionInfo_PE L"c:\\windows\\system\\user.exe" warn:ver:VERSION_GetFileVersionInfo_PE Could not load L"c:\\windows\\system\\user.exe" trace:ver:VERSION_GetFileVersionInfo_16 "c:\\windows\\system\\user.exe" trace:ver:print_vffi_debug structversion=1.0, fileversion=10.0.0.0, productversion=1.0.0.0, flagmask=0x0, flags= trace:ver:print_vffi_debug ( OS=0x0.0x0 UNKNOWN(0x0),BASE) trace:ver:print_vffi_debug filetype=DLL, trace:ver:print_vffi_debug filedata=0x0.0x0 trace:ver:GetFileVersionInfoA ("c:\\windows\\system\\user.exe",0,size=448,data=0xb3100b58) trace:ver:VERSION_GetFileVersionInfo_PE L"c:\\windows\\system\\user.exe" warn:ver:VERSION_GetFileVersionInfo_PE Could not load L"c:\\windows\\system\\user.exe" trace:ver:VERSION_GetFileVersionInfo_16 "c:\\windows\\system\\user.exe" trace:ver:print_vffi_debug structversion=1.0, fileversion=10.0.0.0, productversion=1.0.0.0, flagmask=0x0, flags= trace:ver:print_vffi_debug ( OS=0x0.0x0 UNKNOWN(0x0),BASE) trace:ver:print_vffi_debug filetype=DLL, trace:ver:print_vffi_debug filedata=0x0.0x0 trace:ver:VerQueryValueA (0xb3100b58,"\\VarFileInfo\\Translation",0x7fd4e9ac,0x7fd4e9a8) fixme:dialog:MSGBOX_OnInit task modal msgbox ! Not modal yet.

3 3

Re: msi: prevent locking the cd
by Alexandre Julliard 27 Jun '05

27 Jun '05

Aric Stewart <aric(a)codeweavers.com> writes: > Copy the msi file to a temp file to prevent locking a CD during an > install. This allows multi disc installs to eject the first volume. This fails make test: fixme:msi:MSI_OpenDatabaseW open failed r = 80030050! format.c:84: Test failed: Failed to open package format.c:102: Test failed: Failed to close package fixme:msi:MSI_OpenDatabaseW open failed r = 80030050! format.c:84: Test failed: Failed to open package format.c:668: Test failed: size wrong(16) etc... -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] FCI work for cabinet.dll [cabinet-fci-patch-03.diff]
by Alexandre Julliard 27 Jun '05

27 Jun '05

"Gerold J. Wucherpfennig" <gjwucherpfennig(a)gmx.net> writes: > +static cab_ULONG fci_set_little_endian_ulong( cab_ULONG i ) { > + unsigned char v[4]; > + v[0]=i; > + v[1]=i>>8; > + v[2]=i>>16; > + v[3]=i>>24; > + return *((cab_ULONG*)(v)); > +} > + > +static cab_ULONG fci_get_little_endian_ulong( cab_ULONG i ) { > + unsigned char v[4]; > + cab_ULONG r=0; > + memcpy(v,&i,4); > + r|=v[3]; > + r<<=8; > + r|=v[2]; > + r<<=8; > + r|=v[1]; > + r<<=8; > + r|=v[0]; > + return r; > +} You should use the Rtl*ByteSwap functions with the appropriate #ifdef WORDS_BIGENDIAN, it would be a lot more efficient, particularly on little endian platforms that represent 99.9% of the real world cases. -- Alexandre Julliard julliard(a)winehq.org

1 0

Battlefield 1942 Regression
by Stefan Dösinger 27 Jun '05

27 Jun '05

Hello, The patch http://cvs.winehq.org/patch.py?id=18375 from 2005/06/22 13:30:30 breaks Battlefield to a certain extend. Battlefield used to have major texture problems(Units partially not visible, wrong colors sometimes) which went away approximately 2 Weeks ago. Now they came back with this patch. It seems to me that this patch fixes some functionality which was broken 2 weeks ago and which causes the problems and the root cause is somewhere else. Can someone give me a hint/documentation links what might be wrong here? Is it worth trying Oliviers DirectX patches? It looks to me that he's just committing them to CVS. Stefan Dösinger

2 2

Where to map the unixfs extension
by Michael Jung 27 Jun '05

27 Jun '05

Hi, What if we map the unixfs extension on the desktop instead of on mycomputer? So if HKCR\Software\Microsoft\Windows\Current Version\Explorer\Desktop\Namespace\{UnixDosFolderCLSID} is present, the desktop folder would forward ParseDisplayName calls to UnixDosFolder instead of MyComputer. We could de-activate the special case of enumerating all child's of MyComputer in the file dialogs iff unixfs is rooted at the desktop. This way we wouldn't have to introduce a second "MyUnixComputer" and the user could expand MyComputer, if he forgot where there C: drive is mapped. But he would'nt see the drives until he does so. Bye, -- Michael Jung mjung(a)iss.tu-darmstadt.de

1 0

Re: dlls/commdlg/printdlg.c : Now works PageSetupDlgA function
by Michael Stefaniuc 27 Jun '05

27 Jun '05

Hi, you may want to recheck your patch as it contains lots of cvs update conflicts. tft(a)office.etersoft.ru wrote: > It's too big patch, but break it to many parts very hard. > > ChangeLog: > Complex PageSetupDlg patch > [Snip] > +<<<<<<< printdlg.c ^^^^^^^^^^^^^^^^^^^^^ cvs update conflict > +/*********************************************************************** > + * DefaultPaintHook > + Default hook paint procedure that receives WM_PSD_* messages from the dialog box > + whenever the sample page is redrawn. > +*/ > + [Snip] bye michael -- Michael Stefaniuc Tel.: +49-711-96437-199 Sr. Network Engineer Fax.: +49-711-96437-111 Red Hat GmbH Email: mstefani(a)redhat.com Hauptstaetterstr. 58 http://www.redhat.de/ D-70178 Stuttgart

1 0

Re: Move Notification Window (aka systray) To A Separate Process
by Robert Shearman 27 Jun '05

27 Jun '05

Robert Shearman wrote: > Changelog: > Move notification window (aka systray) to a separate process. Sorry, this changelog should be as follows instead: Mike Hearn <mh(a)codeweavers.com> Robert Shearman <rob(a)codeweavers.com> Move notification window (aka systray) to a separate process. -- Rob Shearman

1 0