wine-devel

wine-devel@winehq.org

10 participants
84459 discussions

Theme configuration options
by Kevin Koltzau 08 Oct '03

08 Oct '03

I've come to a point in the development of uxtheme where configuration options will be needed, and am looking for feedback on how this should be done. I currently forsee the following options needed: EnableTheme ="Y|N" ThemeMethod = "XP|GTK|KDE|..." GTK, KDE, etc are conceivable, but wont be in initial implementation, will default to XP ThemeName="name of theme" for example "Luna" if ThemeMethod=="XP" and the default XP theme is installed These should probably go into the [Tweak.Layout] section (possibly only enabled with a "WineLook"="WinXP") and of course needs the ability for application specific settings The theme API also has a function EnableTheming(BOOL) which under XP turns theming on/off for all applications immediately (all applications are redrawn), and it is persistant across reboots. I havent checked up on if I can update the EnableTheme config option from code, or if I would need to possibly store this value elsewhere in the registry Another possible option would simply be to only apply that call to the current running app, and not make it persistant..but that would deviate from how windows handles it Thanks Kevin

5 9

Byteswapping
by Warren_Baird＠cimmetry.com 08 Oct '03

08 Oct '03

Hi all, There are some changes I'd like to submit as patches, but I'm not sure the best way to go about it. We need to add some byteswapping code in a few places to make things happier on big-endian machines. The problem is that the byteswap macros only exist inside wine/tools/wrc right now - we've added a new header containing the byteswap macros in wine/include, and so they are just duplicated in wine/include and wine/tools/wrc - but this is hardly the ideal solution. The best solution I can think of at the moment is to put the byteswap macros into wine/include/wine/port.h, which is one of the few main wine header that is included in the wrc code - but then we'll have to include config.h and port.h in at least one place they aren't currently included : dlls/ole32/storage32.c Does anyone have any better suggestions? Thanks, Warren

2 1

Problem with NONAMELESSUNION on MS_VC and Mingw
by Steven Edwards 08 Oct '03

08 Oct '03

Hello, I have been having problems compiling on Mingw and MS_VC + PSDK code that uses #define NONAMELESSUNION. This is not the first place I have ran in to this problem. When I try to build libwine I am getting the following error: loader.c: In function `fixup_resources': loader.c:193: error: union has no member named `s3' loader.c:194: error: union has no member named `s3' loader.c: At top level: loader.c:170: warning: `fixup_imports' defined but not used make: *** [loader.o] Error 1 I had the same problem on MS_VC with the PSDK but I cant test/submut another report on this right now as my broadband at home on my dev box with the PSDK is still out. Can anyone else with the PSDK check? Thanks Steven __________________________________ Do you Yahoo!? The New Yahoo! Shopping - with improved product search http://shopping.yahoo.com

1 0

Re: Abort glyph rendering if everything is clipped
by Warren_Baird＠cimmetry.com 08 Oct '03

08 Oct '03

> It would be better IMO to do that as part of the clip rectangle > calculation a few lines below. Well, it was intended as an optimization - if we move the check down to the "if (flags & (ETO_CLIPPED) ..." clause, we may end up doing a HeapAlloc, and calls to GetGlyphIndices and GetObject that we don't need to do. I'd much rather bail out as soon as we can and avoid doing all of that extra work... Warren

2 1

Build without optimization is broken again
by Pavel Roskin 08 Oct '03

08 Oct '03

Hello! Build without optimization is broken again in CVS: gcc -o wine -Wl,--section-start,.interp=0x3c000100 main.o -L../dlls -lntdll.dll -L../libs/wine -lwine -L../libs/unicode -lwine_unicode -L../libs/port -lwine_port ../dlls/libntdll.dll.so: undefined reference to `SetLastError' ../dlls/libntdll.dll.so: undefined reference to `GetCurrentProcessId' ../dlls/libntdll.dll.so: undefined reference to `GetCurrentThreadId' GetCurrentProcessId() and GetCurrentThreadId() are referenced by dlls/ntdll/virtual.c and dlls/ntdll/server.c respectively because they don't include ntdll_misc.h header that has replacement includes. SetLastError() is used by the set_status() wrapper in dlls/ntdll/heap.c. I believe NtCurrentTeb()->last_error is valid even for gcc x86 builds. At least it's used unconditionally in LdrInitializeThunk(), file dlls/ntdll/loader.c. ChangeLog: * dlls/ntdll/server.c, dlls/ntdll/virtual.c: Include ntdll_misc.h to suppress wrong dependencies when building Wine without optimization. * dlls/ntdll/heap.c: Define set_error unconditionally to eliminate dependency on SetLastError(). -- Regards, Pavel Roskin

1 0

Re: Render strings longer than 128 characters as individual chars
by Warren_Baird＠cimmetry.com 08 Oct '03

08 Oct '03

Thanks for the suggestion, Huw. I'll try applying that xrender patch to our build of X and see if it fixes the problems we were having... > If we really have to have > a workaround in Wine then the string should at least be batched rather > than passed character by character. I agree that this would be better - but using existing code that we know works had a distinct appeal, and in our testing we only came across a *very* few places where this happened, so trying to optimize it seemed like it wasn't worth the effort. Warren

1 0

A link to a article on Wine.
by Hannu Valtonen 08 Oct '03

08 Oct '03

Hello, I noticed a link on LinuxToday to an article on Wine and thought people might be interested. http://insight.zdnet.co.uk/business/0,39020481,39116972,00.htm (Summary) Basically the writer tried to install Paintshop Pro. Couldn't get it to work (InstallShield problem) and was frustrated by that. On the other hand he does recommend CrossOver Office as a polished product for firms and Wine for those who wish to experiment on their own. - Hannu Valtonen Hannu.Valtonen(a)hut.fi

2 1

What does it take to run a third-party DLL under Wine?
by John Conneely 08 Oct '03

08 Oct '03

Hi! I need help determining the feasibility of writing a wrapper around a DLL so that I can call it from a Linux machine, perhaps as a SO. I'm looking for either a member of this list that could contract with us for support, or pointers to a more appropriate place where I can search for such a person. If you're interested in learning more, please read on - otherwise, thanks for your time! The DLL in question is one that interacts with call recording equipment in a telephone switch. Our customer has an incoming call center where they currently use windows machines on the desktops of the telephone agents. They're looking to replace these desktops with Linux machines, and are trying to convert to a 100% non-Microsoft shop. This DLL has five or six function calls that presumably communicates over the network with the switch. My initial thought was that we have three choices of how to architect a solution: 1) Port the code to Linux. I eliminated this as a choice because we don't have the source code for the DLL. 2) Use some kind of RPC mechanism (COBRA, SOAP, etc.) to have the DLL running under windows on a server. This would work in principle, and apparently it is the preferred solution by the maker of the DLL. However, it violates our client's no-Microsoft-code mandate, and our customer has rejected this as a solution. 3) Write a wrapper to use the DLL running under WINE. This is what I am pursuing now. I'm looking for someone who has done #3 for other DLLs, and understands the issues that might come up in getting a third party DLL to run under WINE without active support from the DLL's authors. I myself have used Linux quite extensively but have not had the opportunity to use WINE directly, so we will likely need some help in setting up our WINE environment. We've got funding to pursue this, so if you can show me that you've done something like this before, we will pay you to first do a feasibility analysis on the DLL in question, and then implement it. Thanks! John

3 3

Re: Move actual code out of assert statements
by Warren_Baird＠cimmetry.com 08 Oct '03

08 Oct '03

> The indentation is messed up. You should set your editor tab width to > 8 instead of 4. My apologies --- our internal coding standard has tabs set at 4 spaces - and I haven't figured out how to tell xemacs to use tabs of 8 spaces in some directories, and 4 spaces in others... I thought I had fixed any indentation issues in the patches I was submitting - guess I missed this one. I'll resubmit shortly... Warren

1 0

Wineconf survey - interim results - call for more votes
by Jeremy White 08 Oct '03

08 Oct '03

Hi folks, I've posted interim results at http://www.winehq.org/wineconf/midresults.html It looks like St. Paul and Munich are the leading contenders, but I'm concerned that we have a fairly modest turnout (although I and several others at CodeWeavers haven't voted yet). I haven't done any weighting yet, and I'm actually worried that my comments on weighting may have discouraged folks from responding. My plan was to only modestly increase the pull of major contributors; even votes from plebes like me will have value. So, again, if you would come to Wineconf, and haven't done so yet, please pop over to http://www.winehq.org/wineconf/ I'll call it final on Friday and tally results. Cheers, Jer

1 0

alsa, arts, etc
by dwillis＠stx.rr.com 08 Oct '03

08 Oct '03

how can i get alsa, arts, etc for audio instead of oss? i'm using wine for cooledit96 and the 3 second delay for oss emulation is quite annoying, but i'm still very happy to be using cooledit96 under linux. i'm using wine version 20030813. thanks, dave

2 2

Re: Render strings longer than 128 characters as individual chars
by Huw D M Davies 08 Oct '03

08 Oct '03

On Tue, Oct 07, 2003 at 01:03:08PM -0400, Warren_Baird(a)cimmetry.com wrote: > > > > (See attached file: patch_xrender_long_string.txt) This is most likely due to a buggy version of libxrender which Keith fixed with the attached patch to the XFree86 cvs. I'd suggest upgrading to a newer version of the library. If we really have to have a workaround in Wine then the string should at least be batched rather than passed character by character. Huw. -- Huw Davies huw(a)codeweavers.com

1 0

Re: Protected mode interrupt problem
by Jukka Heinonen 08 Oct '03

08 Oct '03

Mike McCormack wrote: > Two 16bit programs launch, one after the other. Each program hooks INT > 0, however since our interrupt tables are global to all tasks, the > second task retrieves the interrupt vector that the first task set. > > The first task exits, and restores the original INT 0 vector. When the > second task exits, tries to restore the INT 0 vector set by the first > task, which now has an invalid selector, so it crashes. > > After a bit of discussion and writing a test program, we discovered that > it appears that some interrupt vectors are task local, and some are > global. The local ones appear to be stored in the TDB (task descriptor > block), in offsets 0x26 - 0x4a. > > The attached patch fixes this, though we're not sure that it is correct > for DOS programs. Well, I don't know much about the relation between Win16 tasks and DOS. If this is how interrupt vectors are handled under Win16 then I guess the basic idea in the patch is correct. However, I can't see why this would cause a crash because as far as I know, int 0 vector is never called by Wine. It doesn't really matter but I'm just curious... (Perhaps they handle int 0 as a call chain?) Anyway, this patch fails to initialize the vectors so any attempt to use them causes a crash. These vectors are not currently used by Wine (except 0x3e) so this is not yet a problem but I would really like to see this patch changed so that those vectors are properly initialized (possibly lazily) at program startup and when new task is created (are they copied from another task or reset to default values?). And, I don't think TASK_GetCurrent() should be called from winedos and I thought use of WINE_PACKED was unnecessary and deprecated. -- Jukka Heinonen <http://www.iki.fi/jhei/>

2 2

RE: question about copyright and code, is this legal?
by Patrik Stridvall 08 Oct '03

08 Oct '03

> Say I have a piece of code, x (which is copyright) which > always returns the > same output for a given input (e.g. a function that implements a math > formula or a function that retrieves some value from an > internal data table > stored in the dll or whatever) > Lets then say that I havent seen this code (either origonal C code or > disassembly) > Say I was to write a program that calls this function for all > valid inputs > and prints the inputs and outputs. > If I was to take the output of this program and, without > refering to the > origonal function or the dump program in any way (i.e. using just the > "truth table" for the function) generate a function that, for > any given > input, generates the same output as the original, have I > broken the law? Short answer: In depends on the nature of the output data and in which country you are in. Long answer: If the output data for each input data is not mearly a pure statement of some facts but an unique expression about something (be it some facts or otherwise) it is not permitted in any country. Note however that in some cases it might be possible to apply some another function and get rid of the unique expression and merely get the pure fact though... However in some countries it still depends on the nature of the pure facts though. In the US, it doesn't matter at all, see the "Feist" decision. In the Commonwealth countries and some other countries including my home country Sweden, each fact in and of itself is not protected either, however large collections of facts have some sort of protection as a whole and to some lesser degree to subsets of the whole collection. This is in order to protect the effort of collecting the facts and while case laws AFAIR at least in Sweden is a little unclear, it mostly seems to apply to collections of facts that have required a significant effort to collect... So functions the produce results that are purely algoritmic in nature are not protected. As far as Wine is concerned though I can't think of any Windows API that: 1. Returns pure facts (not having any unique expression) 2. Would fulfill the requirement of needed a significant effort to collect to collect the fact. So no I wouldn't worry about it as far as Wine is concerned.

3 2

Regression in lstrcmpiA (occurred in late June, NLS related)
by Troy Rollo 08 Oct '03

08 Oct '03

When lstrcmpiA was moved from ole2nls.c to locale.c, (around 28th June) the results of comparisons in some cases became reversed. For example, the underscore now returns as greater than alphabetic characters, whereas it used to return as less than alphabetic characters. The older behaviour was consistent with Win2k. The output below is from the following source: ---begin test program--- #include <windows.h> #include <stdio.h> char *test_strings[] = { "_", "A", "a", "z", "Z", 0 }; void test_string(char *pch) { char **ppch = test_strings; while (*ppch) { printf("%s\t%s\t%d\n", pch, *ppch, lstrcmpiA(pch, *ppch)); ++ppch; } } int main(int argc, char **argv) { char **ppch = test_strings; while (*ppch) test_string(*ppch++); return 0; } ---end test program--- ---Wine output from immediately before the change--- _ _ 0sorts _ A -1 _ a -1 _ z -1 _ Z -1 A _ 1 A A 0 A a 0 A z -1 A Z -1 a _ 1 a A 0 a a 0 a z -1 a Z -1 z _ 1 z A 1 z a 1 z z 0 z Z 0 Z _ 1 Z A 1 Z a 1 Z z 0 Z Z 0 ---End--- ---Wine output from immediately after the change--- _ _ 0 _ A 1 _ a 1 _ z 1 _ Z 1 A _ -1 A A 0 A a 0 A z -1 A Z -1 a _ -1 a A 0 a a 0 a z -1 a Z -1 z _ -1 z A 1 z a 1 z z 0 z Z 0 Z _ -1 Z A 1 Z a 1 Z z 0 Z Z 0 ~---End---

8 26

Re: winebuild: small docu update
by Alexandre Julliard 08 Oct '03

08 Oct '03

"Dimitrie O. Paun" <dpaun(a)rogers.com> writes: > Since these are internal use only, let's not document them > for external people, it will just cause problems. I disagree, it's much better to have documentation stating that they are internal than not to have any documentation at all. -- Alexandre Julliard julliard(a)winehq.com

1 0

Re: Move actual code out of assert statements
by Alexandre Julliard 08 Oct '03

08 Oct '03

Warren_Baird(a)cimmetry.com writes: > (See attached file: patch_code_out_of_asserts.txt) The indentation is messed up. You should set your editor tab width to 8 instead of 4. -- Alexandre Julliard julliard(a)winehq.com

1 0

Re: Abort glyph rendering if everything is clipped
by Alexandre Julliard 08 Oct '03

08 Oct '03

Warren_Baird(a)cimmetry.com writes: > (See attached file: patch_xrender_abort_if_clipped.txt) It would be better IMO to do that as part of the clip rectangle calculation a few lines below. -- Alexandre Julliard julliard(a)winehq.com

1 0

GDI and DirectX test suite
by Steven Edwards 07 Oct '03

07 Oct '03

Hello, I have post a collection of tests with source code for the GDI and DirectDraw 7 interfaces at http://mail.gleneagle.net/sedwards/gdi/ The authors of the Windows2000 API Blackbook have aggreed to license the tests as public domain with a request for advertising. The ReactOS project has already started using these tests to help in our implementation and they may make a good test for WINE and Winelib. Thanks Steven __________________________________ Do you Yahoo!? The New Yahoo! Shopping - with improved product search http://shopping.yahoo.com

2 1

Re: wine/ windows/winpos.c server/window.c server/ ...
by Martin Fuchs 07 Oct '03

07 Oct '03

Hello Alexandre, I see, you implemented storage of the four global shell and takman windows. Btw.: great unified aproach, this single get/set server call. :) But you did not yet insert the altered focus handling in set_active_window() from my patch. Is this just comming next, or is there another particular reason for not doing this? Martin On 07.10.2003 05:40:24 Alexandre Julliard wrote: > ChangeSet ID: 9656 > CVSROOT: /home/winehq/opt/cvs-commit > Module name: wine > Changes by: julliard(a)wine.codeweavers.com 2003/10/06 22:40:24 > > Modified files: > windows : winpos.c > server : window.c trace.c request.h protocol.def > include/wine : server_protocol.h > dlls/user : focus.c > > Log message: > Store the global shell, progman and taskman windows in the server > (based on a patch by Martin Fuchs). > > Patch: > http://cvs.winehq.com/patch.py?root=/home/winehq/opt/cvs-commit&id=9656 > > Old revision New revision Changes Path > 1.147 1.148 +0 -81 wine/windows/winpos.c > 1.23 1.24 +61 -0 wine/server/window.c > 1.182 1.183 +20 -0 wine/server/trace.c > 1.88 1.89 +2 -0 wine/server/request.h > 1.84 1.85 +18 -0 wine/server/protocol.def > 1.85 1.86 +27 -1 wine/include/wine/server_protocol.h > 1.5 1.6 +131 -0 wine/dlls/user/focus.c

2 5

Re: Abort glyph rendering if everything is clipped
by Dimitrie O. Paun 07 Oct '03

07 Oct '03

On Tue, 7 Oct 2003 Warren_Baird(a)cimmetry.com wrote: Nit: why not use IsRectEmpty() ? -- Dimi.

1 0

Re: libs/unicode new conversions
by Alexandre Julliard 07 Oct '03

07 Oct '03

Jon Griffiths <jon_p_griffiths(a)yahoo.com> writes: > This patch adds two new conversions to libs/unicode, in preparation > for their > use by an upcoming patch adding FoldStringA/W. The first converts > from any > unicode number into its unicode 0-9 representation. The second maps > from > compatibility characters to unicode. > > The rationale for placing them in libs/unicode is that both these > conversions > may be required by calls other than FoldString (at least the numeric > mapping > will be used to parse native language digits in oleaut32). IMO you should export the FoldString functionality from libs/unicode, not the tables. Also I'm not sure how you built the tables, but they need to be generated from the Unicode standard data, and the code to do that should be added to cpmap.pl. -- Alexandre Julliard julliard(a)winehq.com

2 2

cdrom problem? winefile crash
by Moreno 07 Oct '03

07 Oct '03

Hi, fixme:cdrom:CDROM_GetInterfaceInfo CD-ROM device (11, 0) not supported what it mean? this is my cd is (11, 0) and then? brw-rw---- 1 moreno cdwriter 11, 0 gen 1 1970 /dev/scsi/host0/bus0/target0/lun0/cd i don't know if is for that but winefile crash!! fixme:cdrom:CDROM_GetInterfaceInfo CD-ROM device (11, 0) not supported err:dosfs:DRIVE_GetLabel Invalid or unreadable superblock on /dev/fd0 (A:). Invalid or unreadable superblock on /dev/fd0 (A:). Maybe not FAT? fixme:bitmap:CreateBitmap planes = 35 wine: Unhandled exception (thread 0009), starting debugger... fixme:cdrom:CDROM_GetInterfaceInfo CD-ROM device (11, 0) not supported WineDbg starting on pid 8 thx -- Moreno <morenoca(a)freemail.it>

2 1

Re: Oleaut32 3
by Dimitrie O. Paun 07 Oct '03

07 Oct '03

On Mon, 6 Oct 2003, Jon Griffiths wrote: > Hi, > > Part 3 of oleaut32 updates. Hi Jon, Cool stuff, but I don't understand why we have all these forwards: +HRESULT WINAPI VarDecFromUI1(BYTE bIn, DECIMAL* pDecOut) +{ + return _VarDecFromUI1(bIn, pDecOut); +} We don't seem to pass any additional info to _VarDecFromUI1, why not simply get rid of it, and implement stuff in the real VarDecFromUI1? -- Dimi.

3 4

Focus problems
by Pavel Roskin 06 Oct '03

06 Oct '03

Hello! My patch for fixing the focus of some programs (notepad, winecfg) on startup hasn't been applied, so I'm guessing what the reason might be. One reason may be that the testing could have found another bug that might seem to be related. When Wine uses non-managed windows without desktop, they don't get the keyboard focus initially. It affects even programs what show active captions, such as Winemine. When Winemine is started, the keyboard events go to the original window. The same happens after the "About" dialog is run. But when the "About" dialog is dismissed, the main window of Winemine gets keyboard focus. It seems to be another example of not doing something because the initial window is "already focused". This is a separate problem, it affects a wider class of programs and doesn't affect Wine in the desktop mode. If fixing this is considered precondition to fixing all other focus problems, I'd like to know that. Another possible reason I can think of is because SetFocus is supposed to affect widgets before the application window is shown. That's something that needs testing. I've made a patch that only disables activation of the parent window if it's invisible, but still focuses the window for which SetFocus() was called. The patch is attached. Before I start time-consuming verification, I'd like to know if I going in the right direction. There is another, third focus problem. Windows should not cover modal dialogs. Shareware Windows Commander 5.0 is useless in the desktop mode because the main window covers the "nag screen" on startup. The workaround is to enter "foo" in the command line and press Enter - there will be an error message, and after it's dismissed, the "nag screen" will pop up. Again, it's a separate problem that I'm not trying to solve at this point. I'll appreciate an advise in which order to proceed and what kind of verification is expected for the changes in the focus code. -- Regards, Pavel Roskin

2 1

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