wine-devel

wine-devel@winehq.org

2 participants
84469 discussions

Re: Patch : dlls/ntdll, dlls/kernel, dlls/user
by Dmitry Timoshkov 24 Oct '04

24 Oct '04

"Peter Quiring" <pquiring(a)hotmail.com> wrote: > I've added some stub's for 39 functions that I've come accross > while using Wine. I often try to use native DLLs as much as > possible but can't always unless these 39 stubs are added. > I've also attached a txt file listing the functions I've added. That's usually considered a bad idea to add stubs to .spec files if there is no real need for that. And that's really the case, since: 1. there is no real difference between a missing API and a stubbed in the .spec file one 2. using native dlls "as much as possible" is not encouraged in Wine at all, and especially for Wine *development* 3. you can't use native versions of ntdll, kernel32 and user32 in any case 4. Adding stubs for NT/win2k/win2k3 only APIs will make the things much worse for some apps, because they could start think that they are running under those platforms by checking presence of that APIs -- Dmitry.

1 0

OpenGL Extension Viewer
by Devils Cry 24 Oct '04

24 Oct '04

Hi there everybody! While browsing the web i found a little tool which i expect to be quite helpfull: the opengl extension viewer. Though its written for windows, it runs perfectly on wine and allows people working on d3d and so on to have a quick overview of the functions. With kind regards Nikolas

2 2

DLL search order
by Kenneth Porter 24 Oct '04

24 Oct '04

I know this has to be a FAQ, but I'm not finding it. I want to test a custom version of kernel32.dll without bothering other wine users on the box. What do I need to do to make an app search the local directory for it instead of using the one in /usr/lib/wine?

2 1

Re: Bug in recent shellpath.c change, with fix?
by Juan Lang 24 Oct '04

24 Oct '04

This certainly is a bug. Your patch fixes it, but I think it might be better (there's an unnecessary copy left, and I wonder whether _SHExpandEnvironmentStrings should be called at all). I'll submit what I think is a better patch shortly. Thanks, --Juan --- Dan Kegel <dank(a)kegel.com> wrote: > Hey, I think I found a bug in your recent patch. > Could you have a look and maybe approve my > patch or submit the real fix? > The app I was tryiing to install was Google Desktop > Search, > which is a free download from > http://desktop.google.com > > Thanks, > Dan > > -------- Original Message -------- > Subject: _SHExpandEnvironmentStrings() should return > input string if no % found to expand? __________________________________ Do you Yahoo!? Yahoo! Mail Address AutoComplete - You start. We finish. http://promotions.yahoo.com/new_mail

1 0

bug 2170: can't install Office '97
by Dan Kegel 24 Oct '04

24 Oct '04

http://bugs.winehq.org/show_bug.cgi?id=2170 This bug is marked 'closed', but it's still happening for me with today's CVS. Anyone know what's up with this? Thanks, Dan -- Trying to get a job as a c++ developer? See http://kegel.com/academy/getting-hired.html

2 2

RE: The wine user guide: an idea
by Scott Ritchie 24 Oct '04

24 Oct '04

Thanks for you comment. I've taken up the task of rewriting alot of the User Guide at the moment, and revising the Using Wine section may be the most important next step after the rewrite of the Intro. > What I would like is a section with common use cases of wine and > step-by-step instructions for each of them. It shouldn't be long - it > should be mostly links to other parts of the user guide. I think of > something like that... Not a bad idea, I'll certainly keep it in mind as I go through the other chapters. -Scott Ritchie

2 1

Re: Fix regression in CreateProcess
by Mike McCormack 24 Oct '04

24 Oct '04

Eric Pouech wrote: > As Mike reported, there was a bug in CreateProcess introduced with the > runtime info handling. > This patch should take care of it. > A+ Hi Eric, This patch fixes the problem! thanks for your help! Mike

1 0

The wine user guide: an idea
by Noam Raphael 23 Oct '04

23 Oct '04

Hello, First, let me apologize: I am not familiar with wine, and I didn't make a decent research before I posted this message. I just read through the wine user guide, and though of something that could be nice for a user as lazy as I am. Since the introduction asked for comments, I comment. What I would like is a section with common use cases of wine and step-by-step instructions for each of them. It shouldn't be long - it should be mostly links to other parts of the user guide. I think of something like that: CASE: I have a working windows machine, and I would like to use the installed programs from linux. (which happens to be my situation...) WHAT TO DO: 1. Install wine (see ...) 2. Copy all files into your home dir, for example, ~/wine/C 3. Run winecfg (see ...) ... This would give the users (or at least me) a nice warm feeling that what I want to do was already done and isn't so complicated (I hope - I haven't tried it yet). Have a good week, Noam Raphael P.S. if you reply, can you please CC me? I'm not a member of wine-devel.

1 0

Re: wine devel documentation reorganization
by Mike Hearn 23 Oct '04

23 Oct '04

Spiffy. As part of my ongoing quest to make Wine hacking less scary, I've written some docs on SEH. I'll submit them to the kernel modules section when I get net access back. Just a heads up in case anybody else was thinking of covering that topic. Eric Pouech wrote: > As posted on wine-devel, here's a patch to reorganize the Wine > Developer's Guide. > It's gzip:ed because of the size (160KB). > Here's the changelog for those who don't want to open the gzip: > > A+ > Reorganizing wine-devel: > - killing the advanced part > (and moving its chapters in both developing Wine and Wine > archi parts) > - creating a new book on coding practices > from i18n.sgml, patches.sgml and porting.sgml (developing > Wine) > - creating a new book on some debugging strategies > from the old advanced book and cvs-regression.sgml > (developing Wine) > - creating a new book on the kernel modules (NTDLL & KERNEL32) > from architure.sgml / related DLLs and > address-space.sgml, console.sgml, threading.sgml > - creating a new book on the windowing > from architure.sgml / USER32 and related > - creating a new book on the graphical parts > from architure.sgml / GDI32 and related > Other changes: > - removed list of DLLs and their role (from the > modules overview) > - removed in X11 keyboard mapping section the part related > to submit a patch

2 1

Re: Fully Revised Wine User Guide Introduction
by Francois Gouget 23 Oct '04

23 Oct '04

On Fri, 22 Oct 2004, Scott Ritchie wrote: > The version 0.9 todo list (bug 85) specifies the need for a rewritten > intro to the Wine User Guide. > > This patch is that rewritten intro. It rewords things to make Wine > easier to understand for the new user, gives it a more professional > look, and removes some outdated information. Unlike my previous patch, > this one should be properly formatted, It still has awfully long lines. Are you sure you sent the right patch? -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ Linux, WinNT, MS-DOS - also known as the Good, the Bad and the Ugly.

1 0

Program To Check System Metrics
by William Poetra Yoga H 23 Oct '04

23 Oct '04

Hi guys (& gals), if you remember I was working on the caption buttons. I haven't submitted the patch yet (it's still in progress) and I wish to know the relations between system metrics, because I think they are very closely related to drawing on the screen, like borders, sizes, etc. I'm writing an app here, called metrics, but since this is my first Win32 program, I've encountered many problems (and solved only a few :-P). I've read bits of theForge's Win32 tutorial at www.winprog.org (I got the link from Wine), but I just can't figure out how to handle the "Refresh" and "Save" buttons properly. "Refresh" is for refreshing (regetting) the metric values, and save is for saving the values to a file. Please help me with this, and be informed that this is my first Win32 app, so bear with me... File is attached. _______________________________ Do you Yahoo!? Declare Yourself - Register online to vote today! http://vote.yahoo.com

1 0

IE6 install Regression
by Mike McCormack 23 Oct '04

23 Oct '04

Hi Eric, The following patch causes the IE6 installer to crash. Thanks to Rob for figuring out which patch caused the problem very quickly :) Mike Log message: Eric Pouech <pouech-eric(a)wanadoo.fr> - msvcrt: the file descriptors are now inherited between parent/child processes - kernel32.CreateProcess: make use of the *reserved2 fields for C-RunTime data inheritance http://www.winehq.org/hypermail/wine-cvs/2004/09/0039.html ... 001f:Call kernel32.GetStartupInfoA(77adfca4) ret=500b34f6 001f:Ret kernel32.GetStartupInfoA() retval=77adfca4 ret=500b34f6 trace:seh:EXC_RtlRaiseException code=c0000005 flags=0 addr=0x500b354f trace:seh:EXC_RtlRaiseException info[0]=00000000 trace:seh:EXC_RtlRaiseException info[1]=780704e4 trace:seh:EXC_RtlRaiseException eax=00570000 ebx=77efd6e8 ecx=00000040 edx=77b004e0 esi=780704e4 edi=500c7ab8 trace:seh:EXC_RtlRaiseException ebp=77adfcec esp=77adfc98 cs=0073 ds=007b es=007b fs=0033 gs=0000 flags=00210 202 trace:seh:EXC_CallHandler calling handler at 0x77c0e240 code=c0000005 flags=0

2 3

Re: Fix for bug 824 - proper handling of REG_MULTI_SZ
by Alexandre Julliard 23 Oct '04

23 Oct '04

Randy Galbraith <randy.galbraith(a)cox.net> writes: > This proposed patch addresses the concern raised by bug 824. This > patch changes the handling of REG_MULTI_SZ from REG_SZ to REG_BINARY. Do you actually have an app that depends on this? I agree that in theory we should preserve the data exactly, but this means giving up on an easily editable registry format, so it should be done only if something really needs it. -- Alexandre Julliard julliard(a)winehq.org

3 5

Re: winapi_check and 16 bit -register functions
by Alexandre Julliard 22 Oct '04

22 Oct '04

Francois Gouget <fgouget(a)free.fr> writes: > Yesterday's winapi_check path warns about 32 bit -register functions > that are implemented by stdcall functions, but not about 16 bit > -register functions implemented by stdcall functions. The reason is > that for the latter the calling convention is set to 'pascal' or > 'pascal16' in winapi_check. However I suspect these too should be > implemented as cdecl. They could, but that will require some changes to the .spec.c code, right now the calling convention is based on the type specified in the spec file, so pascal register functions have to be WINAPI. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: [dx9-12] Major change to desgin
by Alexandre Julliard 22 Oct '04

22 Oct '04

"Ann and Jason Edmeades" <us(a)the-edmeades.demon.co.uk> writes: > I have changed the design so only the Direct3D<interface><8/9> ones extend > IUnknown, and so there is no reference counting of the IWineD3D > objects. I don't think that's a good idea. If you are using COM interfaces then you need to use them the COM way, this means IWineD3D should derive from IUnknown and freeing it should be done through the Release method. If you don't want to use standard COM then don't declare COM interfaces at all, create your own mechanism. > Instead, when any Direct3D<interface><8/9> is Released to zero, then it > calls a Free method in its WineD3D equivalent interface. I also changed the > IWineD3D interface to contain, as a void *, a pointer to its 'parent' (by > which I mean the Direct3D8/9 interface) plus a function (GetParent) which > returns that pointer. This is bad too, you shouldn't use void*, you should use proper types, at the very least IUnknown*, but preferably define an IWineD3DUser interface or something like that. -- Alexandre Julliard julliard(a)winehq.org

2 4

Wine on OpenBSD
by Sascha Hanse 22 Oct '04

22 Oct '04

> >> I just tried to install the recent Wine-tarball (20041019) on my >> OpenBSD 3.6. First I had to remove kthread.c:275: the RFTHREAD flag. >> Then the'make depend && make' worked properly but the make install >> first fails with: >> >> /usr/bin/install -c serialui.dll.so >> /usr/local/lib/wine/serialui.dll.so cd `dirname setupapi/__install__` >> && make install gcc -g -O2 -o install install.c > >Your make tries to build the 'make install' target from install.c, >that won't work of course. I guess we'll have to rename that file, >unless OpenBSD make has an equivalent of the .PHONY target. > >-- >Alexandre Julliard >julliard_at_winehq.org Hi Julliard, thank you for your response. Could you tell me when or how this can be done? -- Anyone who considers arithmetical methods of producing random digits is, of course, in a state of sin. --John von Neumann (1951)

2 2

Re: Add list for monitoring Appdb
by Jeremy Newman 22 Oct '04

22 Oct '04

I'm holding off on this patch until I get more comments on why we should or shouldn't create yet another mailing list. My personal opinion is that we don't need a mailing list for this. The change emails can go straight to the app owners themselves, and possibly to the appdb admins. I know I wouldn't care seeing update emails for apps I don't maintain. Just my opinion. I will create this list if there is enough interest in it. On Wed, 2004-10-13 at 10:38 -0700, Tony Lambregts wrote: > I am working on a series of changes that will let changes to the appdb > be emailed to administrators, application owners and other users. > > This will be similar to how bugzilla does it. > > When someone changes an application an email listing the changes gets > sent to the following > > - Appdb administrators. > - Application owners. > - Aplication monitors. > > One of the things that I think is a repository the way wine-bugs(a)winehq.org > does for bugzilla > > I need Jeremy to set up the email account and the mailman account. > > Change log: Add link for monitoring changes to application database > > files changed: lostwages/templates/en/forums.template > > > Tony Lambregts > > Index: forums.template > =================================================================== > RCS file: /home/wine/lostwages/templates/en/forums.template,v > retrieving revision 1.16 > diff -u -r1.16 forums.template > --- forums.template 15 Sep 2004 20:05:31 -0000 1.16 > +++ forums.template 13 Oct 2004 15:46:24 -0000 > @@ -97,6 +97,13 @@ > <a href="http://bugs.winehq.org/">Bug Tracking Database</a>. > > <li> > + [<a href="http://www.winehq.org/mailman/listinfo/wine-appdb">(Un-)Subscribe</a>] > + [<a href="http://www.winehq.org/hypermail/wine-appdb/">Archive </a>] > + <a href="mailto:wine-appdb@winehq.org">wine-appdb(a)winehq.org</a> > + A low traffic (2/day) read-only list for monitoring submissions to the > + <a href="http://appdb.winehq.org/">Application Database</a>. > + > + <li> > [<a href="http://www.winehq.org/mailman/listinfo/wine-tests-results">(Un-)Subscribe</a>] > [<a href="http://www.winehq.org/hypermail/wine-tests-results/">Archive</a>] > <a href="mailto:wine-tests-results@winehq.org">wine-tests-results(a)winehq.org</a> > > >

3 2

Re: shellpath.c: Avoid copying invalid data on error
by Alexandre Julliard 22 Oct '04

22 Oct '04

Juan Lang <juan_lang(a)yahoo.com> writes: > The following is a better patch, I think, in that > - it doesn't add a break statement to the default case > (some compilers warn on this) I seriously doubt that, what compiler does this for you? It's much better practice to always add break to all cases, otherwise when someone adds some code at the end of the switch they will forget to add a break to the previous case. > - WCHAR szBuildPath[MAX_PATH], szTemp[MAX_PATH]; > + WCHAR szBuildPath[MAX_PATH], szTemp[MAX_PATH] = { 0 }; This fills the whole buffer with zeroes, that's not really needed. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: dlls/shlwapi PathIsValidChar
by Alexandre Julliard 22 Oct '04

22 Oct '04

Ge van Geldorp <gvg(a)reactos.com> writes: > Did this patch: http://www.winehq.org/hypermail/wine-patches/2004/09/0446.html > fall through the cracks or is it not acceptable? I think a lookup table would be more elegant (and faster...) -- Alexandre Julliard julliard(a)winehq.org

1 0

unaligned access bug in winspool on sparc
by Eric Frias 22 Oct '04

22 Oct '04

I just ran across an evil little bug in the WINSPOOL_GetPrinter_2 function. It looks like this type of bug could be hiding in other API functions too. It causes a segmentation fault because of an unaligned access on Solaris (sparc). This function packs a PRINTER_INFO_2 structure and all of its variable-length fields into one contiguous block of memory. The structure is first, then several strings, and then a DEVMODE structure. The problem is that because the strings can be any length, the DEVMODE structure itself could be placed at an unaligned (odd) address. So on our system, wine runs correctly if we only have printers with an even number of letters in their name in the printcap file, but it crashes if any have an odd number of letters in their name. I put in a quick hack that aligns it to a 4-byte boundary. I think a two-byte boundary would suffice, but I decided to err on the safe side since I wasn't sure. I inserted the following block near line 1822 in info.c: ------ { int bytesOfPadding = (4 - (*pcbNeeded % 4)) % 4; if (space && bytesOfPadding <= left) { ptr += bytesOfPadding; left -= bytesOfPadding; } else space = FALSE; *pcbNeeded += bytesOfPadding; } ------ right before the line: ------ if(WINSPOOL_GetDevModeFromReg(hkeyPrinter, Default_DevModeW, ptr, left, &size, unicode)) { ----- What is the correct way to align this structure? Are there any macros that specify the necessary alignment for each architecture? I'm also wondering if there are any other functions that are likely to contain the same bug. Thanks, Eric

1 0

Re: msdn: d3drs_wrapn[0-7] ???
by Devils Cry 22 Oct '04

22 Oct '04

Well, i aplied the changes raphael suggested and although i dont understand what they do they work :D @David Welch Yeah, already found that out, but thx anyway, i think i understand it a bit better now :) /Me moving on for validateshaders... lets see what that gets :D Cheers Nikolas

1 0

Re: Add function prototypes to wineconf
by Chris Morgan 22 Oct '04

22 Oct '04

I think the idea is that we should move the functionality of wineconf into winecfg so winecfg should be able to autodetect and import the configurations from existing windows installs. Wineconf hasn't been well maintained lately so maybe now is a good time to encourage people to implement that functionality in winecfg by removing it ;-) Chris > > From: Francois Gouget <fgouget(a)free.fr> > Date: 2004/10/22 Fri PM 12:21:32 EDT > To: wine-patches(a)winehq.org > Subject: Add function prototypes to wineconf > > > This tool generates a Wine configuration containing [Drive] sections! > In other words it's pretty outdated. It was last updated almost 18 > months ago! > > Is it even worth keeping? Maybe the best thing to do is to do a cvs > delete on it? > > In case we want to keep it I added function prototypes. However I did > not change its behavior any further. On my machine it runs but perl > issues a couple of warnings... the same one it was issuing before I ever > touched it. > > > Changelog: > > * tools/wineconf > > Add function prototypes. > Change the way functions are called and their declaration order so > perl can check the prototypes. > > > -- > Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ > "Lotto: A tax on people who are bad at math." -- unknown > "Windows: Microsoft's tax on computer illiterates." -- WE7U >

1 0

modal message box fixme
by Ivan Leo Puoti 22 Oct '04

22 Oct '04

I've written an app that calls a modal message box, if the uType parameter passed to MessageBox is MB_TASKMODAL or MB_SYSTEMMODAL a fixme is printed, but if it's MB_APPLMODA wine doesn't print a fixme. Is the attached fix correct (It works for me)? Ivan. ____________________________________________________________ Libero ADSL: navighi gratis a 1.2 Mega, senza canone e costi di attivazione. Abbonati subito su http://www.libero.it

2 1

Using Microsoft C++ Toolkit with Wine
by Dan Kegel 22 Oct '04

22 Oct '04

I've posted about this before, but never really tied things up in a neat package, so: Here's a document that describes how to use Microsoft C++ Toolkit on Linux to build Windows programs: http://www.kegel.com/wine/cl-howto.html I hope that helps a few people. I certainly find it convenient when writing Wine conformance tests. To build a Wine conformance test standalone, I start the test off with the lines #include <stdarg.h> #include <windef.h> #include <winbase.h> #ifndef STANDALONE #include "wine/test.h" #define ok2 ok #else #include <assert.h> #include <stdio.h> #define START_TEST(name) main(int argc, char **argv) #define ok(condition, msg) \ do { \ if(!(condition)) \ { \ fprintf(stderr,"failed at %d, msg:" msg "\n",__LINE__); \ exit(0); \ } \ } while(0) #define ok2(condition, msg, arg) \ do { \ if(!(condition)) \ { \ fprintf(stderr,"failed at %d, msg:" msg "\n",__LINE__, arg); \ exit(0); \ } \ } while(0) #define todo_wine #endif #include <winerror.h> and use ok2 for those calls to ok that have one argument after the format string. That .c file should be fine for either normal or standalone use. (I haven't tested ok2 yet under the normal wine build, but I think #define ok2 ok should work.) Then I can compile with cl -DSTANDALONE -D_X86_ myfile.c and get a quickie Windows executable for just that test file. - Dan -- Trying to get a job as a c++ developer? See http://kegel.com/academy/getting-hired.html

1 0

Re: Cleanup config
by Alexandre Julliard 22 Oct '04

22 Oct '04

"Dimitrie O. Paun" <dpaun(a)rogers.com> writes: > [wine] > -"Windows" = "c:\\windows" > -"System" = "c:\\windows\\system" > -"Temp" = "e:\\" > -"Path" = "c:\\windows;c:\\windows\\system;e:\\;e:\\test;f:\\" > -"Profile" = "c:\\windows\\Profiles\\Administrator" This is OK except for the windows dir, that one is still being used by the registry code. -- Alexandre Julliard julliard(a)winehq.org

2 2

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-devel