September 2002 - Wine-Devel

Re: winedbg: enhancing symbol lookup
by Alexandre Julliard Sept. 10, 2002

Sept. 10, 2002

Eric Pouech <eric.pouech(a)wanadoo.fr> writes: > --- debugger/break.c 30 Jul 2002 00:06:34 -0000 1.35 > +++ debugger/break.c 2 Sep 2002 19:39:59 -0000 > @@ -421,10 +421,15 @@ > DBG_VALUE value; > int i; > > - if (DEBUG_GetSymbolValue(name, lineno, &value, TRUE)) > + switch (DEBUG_GetSymbolValue(name, lineno, &value, TRUE)) > { > + case TRUE: > DEBUG_AddBreakpoint(&value, NULL, TRUE); > return; > + case FALSE: > + break; > + case ABORT: /* user aborted symbol lookup */ > + return; > } Please don't invent a third boolean state. A boolean is either TRUE or FALSE; if you need more than that you should define a new enum type with meaningful names. Reusing TRUE and FALSE will cause major confusion. -- Alexandre Julliard julliard(a)winehq.com

1 0

Re: winemaker: wineapploader generation
by Alexandre Julliard Sept. 10, 2002

Sept. 10, 2002

Francois Gouget <fgouget(a)codeweavers.com> writes: > * tools/winemaker > Generate wineapploader using autoconf rather than from the Makefile Actually it's better to generate this kind of file from the Makefile; this way you can change things without having to re-run configure (by passing the values directly to make). For that reason the autoconf guidelines state that configure should only generate makefiles, not other files. -- Alexandre Julliard julliard(a)winehq.com

1 0

RE: patches policy
by Patrik Stridvall Sept. 9, 2002

Sept. 9, 2002

> There is something concerning submitting patches that bothers me > to no end: inlining vs. attaching them. Yes!!! Finally. Another flamewar. :-) I will mainly comment how I submit my patches. I don't personally care that much what others do. However if Alexandre finds some way more convenient I can certain change... > I don't know about others, but for me 99/1 rule applies: I at least > skim over 99% of the patches inlines in the message, whereas > I bother to read at most 1% of the ones attached/tar.gzed/ziped. > Maybe I'm an extreme case, but there's got to be more to it than > my personal quirks. > > If a patch is sent to wine-patches, it's sent there for peer review. > If you don't want the review, send it to Alexandre directly, even > though I suggest this is avoided as much as possible. However, > it you do send it to wine-patches, please, *please* inline it! > > What about a nice patch submission policy: > -- unified diff only (required) Agreed (if the file already exists). > -- have a decent subject (recommended) Agreed. > -- a long description (optional, if the change warrants it) Agreed. > -- a meaningful ChangeLog entry (required) Agreed. > -- new files, if any, included in patch, diffed against > /dev/null (required) I currently use .tar files since it handles the case that Uwe was talking about that is new directories. But then, I use a script to generate my patches, so I can make the normal case (no new directory) as a diff against /dev/null. > -- patch inlined at the end of the message (required) If it is inlined at all it should certainly be at the end, agreed. > -- one changeset per message What about dependent changesets? Or changeset that are otherwise logically dependent (even if not syntactically or semantically)? I guess they could be perhaps is multiple messages like: Subject: Fix various bugs (part 1 of 3) Subject: Fix various bugs (part 2 of 3) Subject: Fix various bugs (part 3 of 3) I currently mail all my patches from Windows (Outlook) just out of habit I guess since it is where I send my other mail so that would require manual work. However I guess I could mail them directly from my Linux box using another script...

3 2

re: RFC: 'test' bugzilla keyword
by Dan Kegel Sept. 9, 2002

Sept. 9, 2002

Francois Gouget wrote: > if we make it easy to track > conformance test issues in Bugzilla, we may get new contributers (sure I > realize it's not a magic bullet either). Thus I propose to add a 'test' > keyword to make it easy to find such issues in bugzilla for people > specifically interested in conformance tests. Excellent idea. The keyword 'test' might be a bit confusing -- some bug tracking systems use that word as a state. How about something short that has very few hits on Google? Like, um, "apifail" or "unitfail"? - Dan

4 6

[PATCH] winemaker fixes
by Martin Wilck Sept. 9, 2002

Sept. 9, 2002

Hi, the patch below fixes a few things in winemaker. It seems to work with my app, but needs more testing, especially for MFC apps. Modified files: tools: winemaker documentation: winemaker.man Changelog: - adapt winemaker to the new way winebuild handles imports. - fix crashes due to winemaker-generated wrapper app closing libraries (is this one correct??) - Fix some minor bugs in the wrapper.c template. - change semantics of --nomfc option such that it overrides automatic setting of --mfc if stdafx.h or stdafx.cpp are found. (this may cause problems to some people. For me the old behaviour was a pain in the neck, because VC++ generates stdafx.h always, even for pure C projects). Feedback, please ! - Add --nodlls option to suppress auto-generated import list for simple apps. - Adapt man page. -- Martin Wilck Phone: +49 5251 8 15113 Fujitsu Siemens Computers Fax: +49 5251 8 20409 Heinz-Nixdorf-Ring 1 mailto:Martin.Wilck(a)Fujitsu-Siemens.com D-33106 Paderborn http://www.fujitsu-siemens.com/primergy Index: tools/winemaker =================================================================== RCS file: /home/wine/wine/tools/winemaker,v retrieving revision 1.39 diff -u -r1.39 winemaker --- tools/winemaker 14 Apr 2002 19:31:23 -0000 1.39 +++ tools/winemaker 22 May 2002 16:49:18 -0000 @@ -233,6 +233,14 @@ my $TF_MFC=4; ## +# User has specified --nomfc option for this target or globally +my $TF_NOMFC=8; + +## +# --nodlls option: Do not use standard DLL set +my $TF_NODLLS=16; + +## # Initialize a target: # - set the target type to TT_SETTINGS, i.e. no real target will # be generated. @@ -435,23 +443,37 @@ push @{@$target[$T_LIBRARY_PATH]},$option; } elsif ($option =~ /^-l/) { push @{@$target[$T_LIBRARIES]},"$'"; - } elsif (@$target[$T_TYPE] != $TT_DLL and $option =~ /^--wrap/) { - @$target[$T_FLAGS]|=$TF_WRAP; - } elsif (@$target[$T_TYPE] != $TT_DLL and $option =~ /^--nowrap/) { - @$target[$T_FLAGS]&=~$TF_WRAP; - } elsif ($option =~ /^--mfc/) { - @$target[$T_FLAGS]|=$TF_MFC; + } elsif ($option =~ /^--wrap/) { if (@$target[$T_TYPE] != $TT_DLL) { @$target[$T_FLAGS]|=$TF_WRAP; + } else { + print STDERR "warning: option --wrap is illegal for DLLs - ignoring"; + }; + } elsif ($option =~ /^--nowrap/) { + if (@$target[$T_TYPE] != $TT_DLL) { + @$target[$T_FLAGS]&=~$TF_WRAP; + } else { + print STDERR "warning: option --nowrap is illegal for DLLs - ignoring"; } + } elsif ($option =~ /^--mfc/) { + @$target[$T_FLAGS]|=$TF_MFC; + @$target[$T_FLAGS]&=~$TF_NOMFC; } elsif ($option =~ /^--nomfc/) { @$target[$T_FLAGS]&=~$TF_MFC; - @$target[$T_FLAGS]&=~($TF_MFC|$TF_WRAP); + @$target[$T_FLAGS]|=$TF_NOMFC; + } elsif ($option =~ /^--nodlls/) { + @$target[$T_FLAGS]|=$TF_NODLLS; } else { print STDERR "error: unknown option \"$option\"\n"; return 0; } } + if (@$target[$T_TYPE] != $TT_DLL && + @$target[$T_FLAGS] & $TF_MFC && + !(@$target[$T_FLAGS] & $TF_WRAP)) { + print STDERR "info: option --mfc requires --wrap"; + @$target[$T_FLAGS]|=$TF_WRAP; + } return 1; } @@ -541,7 +563,7 @@ } elsif ($dentry =~ /\.c$/i and $dentry !~ /\.spec\.c$/) { push @sources_c,"$dentry"; } elsif ($dentry =~ /\.(cpp|cxx)$/i) { - if ($dentry =~ /^stdafx.cpp$/i) { + if ($dentry =~ /^stdafx.cpp$/i && !(@$project_settings[$T_FLAGS] & $TF_NOMFC)) { push @sources_misc,"$dentry"; @$project_settings[$T_FLAGS]|=$TF_MFC; } else { @@ -551,7 +573,7 @@ push @sources_rc,"$dentry"; } elsif ($dentry =~ /\.(h|hxx|hpp|inl|rc2|dlg)$/i) { push @sources_misc,"$dentry"; - if ($dentry =~ /^stdafx.h$/i) { + if ($dentry =~ /^stdafx.h$/i && !(@$project_settings[$T_FLAGS] & $TF_NOMFC)) { @$project_settings[$T_FLAGS]|=$TF_MFC; } } elsif ($dentry =~ /\.dsp$/i) { @@ -774,10 +796,13 @@ push @exe_list,$target; } # This is the default link list of Visual Studio, except odbccp32 - # which we don't have in Wine. Also I add ntdll which seems - # necessary for Winelib. - my @std_dlls=qw(advapi32.dll comdlg32.dll gdi32.dll kernel32.dll ntdll.dll odbc32.dll ole32.dll oleaut32.dll shell32.dll user32.dll winspool.drv); - @$target[$T_DLLS]=\@std_dlls; + # which we don't have in Wine. + my @std_dlls=qw(advapi32.dll comdlg32.dll gdi32.dll kernel32.dll odbc32.dll ole32.dll oleaut32.dll shell32.dll user32.dll winspool.drv); + if (@$target[$T_FLAGS] & $TF_NODLLS == 0) { + @$target[$T_DLLS]=\@std_dlls; + } else { + $target[$T_DLLS]=[]; + } push @{@$project[$P_TARGETS]},$target; # Ask for target-specific options @@ -982,7 +1007,8 @@ @$wrapper[$T_TYPE]=@$target[$T_TYPE]; @$wrapper[$T_INIT]=get_default_init(@$target[$T_TYPE]); @$wrapper[$T_FLAGS]=$TF_WRAPPER | (@$target[$T_FLAGS] & $TF_MFC); - @$wrapper[$T_DLLS]=[ "kernel32.dll", "ntdll.dll", "user32.dll" ]; + @$wrapper[$T_DLLS]=[ "kernel32.dll", "user32.dll" ]; + push @{@$wrapper[$T_LIBRARIES]}, "dl"; push @{@$wrapper[$T_SOURCES_C]},"@$wrapper[$T_NAME]_wrapper.c"; my $index=bsearch(@$target[$T_SOURCES_C],"@$wrapper[$T_NAME]_wrapper.c"); @@ -1634,32 +1660,10 @@ $rcname =~ s+([^/\w])+\\$1+g; print FILEO "rsrc $rcname.res\n"; } - print FILEO "\n"; - my %dlls; - foreach $dll (@{$global_settings[$T_DLLS]}) { - if (!defined $dlls{$dll}) { - print FILEO "import $dll\n"; - $dlls{$dll}=1; - } - } - if (defined $project_settings) { - foreach $dll (@{@$project_settings[$T_DLLS]}) { - if (!defined $dlls{$dll}) { - print FILEO "import $dll\n"; - $dlls{$dll}=1; - } - } - } - foreach $dll (@{@$target[$T_DLLS]}) { - if (!defined $dlls{$dll}) { - print FILEO "import $dll\n"; - $dlls{$dll}=1; - } - } # Don't forget to export the 'Main' function for wrapped executables, # except for MFC ones! - if (@$target[$T_FLAGS] == $TF_WRAP) { + if ((@$target[$T_FLAGS]&($TF_WRAP|$TF_WRAPPER|$TF_MFC)) == $TF_WRAP) { if (@$target[$T_TYPE] == $TT_GUIEXE) { print FILEO "\n@ stdcall @$target[$T_INIT](long long ptr long) @$target[$T_INIT]\n"; } elsif (@$target[$T_TYPE] == $TT_CUIEXE) { @@ -1867,6 +1871,7 @@ generate_list("${canon}_LIBRARY_PATH",1,@$target[$T_LIBRARY_PATH]); generate_list("${canon}_LIBRARIES",1,@$target[$T_LIBRARIES]); generate_list("${canon}_DEPENDS",1,@$target[$T_DEPENDS]); + generate_list("${canon}_IMPORTS",1,@$target[$T_DLLS]); print FILEO "\n"; generate_list("${canon}_OBJS",1,["\$(${canon}_C_SRCS:.c=.o)","\$(${canon}_CXX_SRCS:.cpp=.o)","\$(EXTRA_OBJS)"]); print FILEO "\n\n\n"; @@ -1952,7 +1957,7 @@ print FILEO "\t-\$(STRIP) \$(STRIPFLAGS) \$\@\n"; print FILEO "\n"; print FILEO "\$(${canon}_SPEC_SRCS:.spec=.spec.c): \$(${canon}_SPEC_SRCS) \$(${canon}_SPEC_SRCS:.spec=.tmp.o) \$(${canon}_RC_SRCS:.rc=.res)\n"; - print FILEO "\t\$(LD_PATH) \$(WINEBUILD) -fPIC \$(${canon}_DLL_PATH) \$(WINE_DLL_PATH) -sym \$(${canon}_SPEC_SRCS:.spec=.tmp.o) -o \$\@ -spec \$(SRCDIR)/\$(${canon}_SPEC_SRCS)\n"; + print FILEO "\t\$(LD_PATH) \$(WINEBUILD) -fPIC \$(${canon}_DLL_PATH) \$(WINE_DLL_PATH) -sym \$(${canon}_SPEC_SRCS:.spec=.tmp.o) -o \$\@ -spec \$(SRCDIR)/\$(${canon}_SPEC_SRCS) \$(${canon}_IMPORTS:%=-l%)\n"; print FILEO "\n"; my $t_name=@$target[$T_NAME]; if (@$target[$T_TYPE]!=$TT_DLL) { @@ -2168,7 +2173,7 @@ print STDERR "Usage: winemaker [--nobanner] [--backup|--nobackup] [--nosource-fix]\n"; print STDERR " [--lower-none|--lower-all|--lower-uppercase]\n"; print STDERR " [--lower-include|--nolower-include]\n"; - print STDERR " [--guiexe|--windows|--cuiexe|--console|--dll]\n"; + print STDERR " [--guiexe|--windows|--cuiexe|--console|--dll|--nodlls]\n"; print STDERR " [--wrap|--nowrap] [--mfc|--nomfc]\n"; print STDERR " [-Dmacro[=defn]] [-Idir] [-Pdir] [-idll] [-Ldir] [-llibrary]\n"; print STDERR " [--interactive] [--single-target name]\n"; @@ -2181,9 +2186,6 @@ exit (2); } - -project_init(\@main_project,""); - while (@ARGV>0) { my $arg=shift @ARGV; # General options @@ -2244,11 +2246,13 @@ $opt_flags&=~$TF_WRAP; } elsif ($arg eq "--mfc") { $opt_flags|=$TF_MFC; - $opt_flags|=$TF_MFC|$TF_WRAP; $needs_mfc=1; } elsif ($arg eq "--nomfc") { - $opt_flags&=~($TF_MFC|$TF_WRAP); + $opt_flags&=~$TF_MFC; + $opt_flags|=$TF_NOMFC; $needs_mfc=0; + } elsif ($arg eq "--nodlls") { + $opt_flags|=$TF_NODLLS; # Catch errors } else { @@ -2263,6 +2267,11 @@ usage(); } } + + if ($opt_flags & $TF_MFC && $opt_target_type != $TT_DLL) { + print STDERR "info: option --mfc requires --wrap\n"; + $opt_flags |= $TF_WRAP; + }; } if (!defined $opt_work_dir) { @@ -2278,6 +2287,8 @@ print_banner(); } +project_init(\@main_project,""); + # Fix the file and directory names fix_file_and_directory_names("."); @@ -3094,6 +3105,8 @@ * This is either CUIEXE for a console based application or * GUIEXE for a regular windows application. */ +#define GUIEXE 0 +#define CUIEXE 1 #define APP_TYPE ##WINEMAKER_APP_TYPE## /** @@ -3138,11 +3151,11 @@ int WINAPI WinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, PSTR szCmdLine, int iCmdShow) #else -int WINAPI Main(int argc, char** argv, char** envp) +int WINAPI main(int argc, char** argv, char** envp) #endif { void* appLibrary; - HINSTANCE hApp,hMFC,hMain; + HINSTANCE hApp = 0, hMFC = 0, hMain = 0; void* appMain; char* libName; int retcode; @@ -3212,7 +3225,7 @@ } /* Get the address of the application's entry point */ - appMain=(WinMainFunc*)GetProcAddress(hMain, appInit); + appMain=GetProcAddress(hMain, appInit); if (appMain==NULL) { char format[]="Could not get the address of %s (%d)"; char* msg; @@ -3232,11 +3245,6 @@ #endif /* Cleanup and done */ - FreeLibrary(hApp); - if (hMFC!=NULL) { - FreeLibrary(hMFC); - } - dlclose(appLibrary); free(libName); return retcode; Index: documentation/winemaker.man =================================================================== RCS file: /home/wine/wine/documentation/winemaker.man,v retrieving revision 1.5 diff -u -r1.5 winemaker.man --- documentation/winemaker.man 2 Oct 2001 17:47:34 -0000 1.5 +++ documentation/winemaker.man 22 May 2002 16:49:19 -0000 @@ -17,7 +17,7 @@ ] .br [ -.IR "--guiexe " "| " "--windows " "| " "--cuiexe " "| " "--console " "| " "--dll " +.IR "--guiexe " "| " "--windows " "| " "--cuiexe " "| " "--console " "| " "--dll " "| " "--nodlls " ] .br [ @@ -121,6 +121,14 @@ i.e. for which it does not know whether it is an executable or a library, it should assume it is a library. .TP +.I --nodlls +This option tells winemaker not to use the standard set of winelib libraries +for imports. That is, any DLL your code uses must be explicitly passed to +winemaker with -i options. +The standard set of libraries is: advapi32.dll, comdlg32.dll, gdi32.dll, +kernel32.dll, odbc32.dll, ole32.dll, oleaut32.dll, shell32.dll, user32.dll, +winspool.drv. +.TP .I --wrap Specifies that executable targets should be built as libraries and a small executable wrapper generated for them. This technique is sometimes required @@ -137,7 +145,9 @@ generates wrappers for these targets that are executables. .TP .I --nomfc -Specifies that targets are not MFC-based. This is the default. +Specifies that targets are not MFC-based. This option disables use of MFC libraries +even if winemaker encounters files "stdafx.cpp" or "stdafx.h" that would cause it +to enable MFC automatically if neither --nomfc nor --mfc was specified. .TP .I -Dmacro[=defn] Adds the specified macro definition to the global list of macro definitions. @@ -149,8 +159,7 @@ Appends the specified directory to the global library path. .TP .I -idll -Adds the Winelib library to the global list of Winelib libraries to import -in the spec file. +Adds the Winelib library to the global list of Winelib libraries to import. .TP .I -llibrary Adds the specified library to the global list of libraries to link with.

2 5

Recent change causes bad QuickBooks behaviour(?)
by Carl Sopchak Sept. 9, 2002

Sept. 9, 2002

I have been using Wine to run QuickBooks fairly successfully for several months now. However, I frequently do CVS updates because I'm also trying to run Quicken, and there are several issues that I'm hoping will get fixed this way. (I have fixed a few of them myself, and am working on one now, but my time [like everyone else's, I'm sure] is limited...) It's been a few weeks since I've run QuickBooks (8/19 was my last transaction date, so it'd be somewhere around there...). When I went to run it today (after several CVS updates on wine, including "moments ago"), after the company data file is opened, the main menu bar goes away. This was definately NOT a problem the last time I ran QuickBooks, so I surmise that it is the result of a change committed to CVS within the past two to three weeks that is causing this behaviour. It happens in both managed and unmanaged window mode. Basically, the main menu bar should be at the top of the window, but the configurable taskbar / buttonbar is displayed "over" it. The accellerator keys for the menu bar are no longer recognised (the program beeps if you try to use <alt>f [for example] to get to the File menu). I am more than willing to try to debug this. I'd really rather not go back to Windows to run QuickBooks, but I HAVE TO HAVE it (for running my company's finances!) But, before I spend days trying to find out what changed in the past few weeks, I was wondering (no, hoping, really) if anyone might recognize this problem as being a possible result of a recent CVS committed change. Well, in the mean time, I guess I'll look through the CVS commits from the past few weeks (arghhh!)... Any help would be greatly appreciated... Thanks, Carl

3 6

RE: Wine DLLs compiled under Visual C
by Patrik Stridvall Sept. 9, 2002

Sept. 9, 2002

> > Yes. I have submitted most of my changes of that kind to the > > CVS so the current status of that is quite good. > > What is the status of WINE under MS_VC? I ask because I dont > have a copy of MS_VC and I would like > to see if there is anything that builds right with it but not > with Mingw/Cygwin. What > dlls/programs can you build atm? The compile status in general is quite good. The link status however is not. A lot more work remains.

1 0

RE: Wine DLLs compiled under Visual C
by Patrik Stridvall Sept. 9, 2002

Sept. 9, 2002

> I want to get some comments about how others here try to get the DLLs > compiled > under Visual C. I went ahead and tried to do this for > shell32.dll and after > a > little work it went so far fine. However after creating a > config.h file for > Visual C with a few definitions such as for inline, I had to > make some minor > changes to some of the c source files such as adding an > #include "config.h" > and sometimes #include "wine/port.h" to the beginning. Yes. I have submitted most of my changes of that kind to the CVS so the current status of that is quite good. > Is this the recommended way of doing things or is there a reason that > "config.h" > shouldn't be included in some of the dll source files? I ask because I > noticed > that in a few of the shell32 files it is included but in the > several of them > it wasn't and in a few it doesn't (yet) need to be. Use msvcmaker. It generates Visual Studio 6.0 .dsw and .dsp files as well as an include/config.h file. cd wine ./tools/winapi/msvcmaker Then open wine.dsw in Visual Studio. PS. Note that msvcmaker only runs under Unix. I use Samba to export the directory to Windows.

2 1

Re: Trying to debug Paradise Poker client
by Per Nystrom Sept. 9, 2002

Sept. 9, 2002

Ok, these are the lines that concern me when I'm running it with the builtin advapi32 dll: 080995d0:Call advapi32.CryptAcquireContextA(41ee6570,00000000,00000000,00000001,f0000000) ret=40cc424b 080995d0:Ret advapi32.CryptAcquireContextA() retval=00000000 ret=40cc424b 080995d0:Call advapi32.CryptAcquireContextA(41ee6570,00000000,40d0e840 "Intel Hardware Cryptographic Service Provider",00000016,00000000) ret=40cc429c 080995d0:Ret advapi32.CryptAcquireContextA() retval=00000000 ret=40cc429c What can be done to fix it? Thanks, Per Nystrom

2 1

Wine DLLs compiled under Visual C
by Rolf Kalbermatter Sept. 9, 2002

Sept. 9, 2002

I want to get some comments about how others here try to get the DLLs compiled under Visual C. I went ahead and tried to do this for shell32.dll and after a little work it went so far fine. However after creating a config.h file for Visual C with a few definitions such as for inline, I had to make some minor changes to some of the c source files such as adding an #include "config.h" and sometimes #include "wine/port.h" to the beginning. Is this the recommended way of doing things or is there a reason that "config.h" shouldn't be included in some of the dll source files? I ask because I noticed that in a few of the shell32 files it is included but in the several of them it wasn't and in a few it doesn't (yet) need to be. Rolf Kalbermatter

1 0