- Wine-Devel - list.winehq.org

Re: seh_try_macros2-00
by Shaheed R. Haque May 17, 2003

May 17, 2003

Hi Greg, Not sure if this is off-base, but is it correct to assume that __wine_pop_frame is based on setjmp/longjmp? If not, ignore the rest of this note. If it is, then one issue to think about is that all the user variables in the handler blocks which are also used in the try block must be declared volatile. For example, int x; try { ... x = ... } handler { ... y = x; } won't work right unless x is volatile (because the compile's flow analysis and register usage will get confused). Since the original code is unlikely to be so marked, you have a problem. In general, AFAICS, any solution implemented at the language level has this problem...without compiler support you are stuffed. I am in the middle of solving a similar problem at work (i.e. forward porting and old C codebase using both exceptions, and a homebrew ORB which also has exceptions handled in a unified way). The ONLY solution I could come up with was indeed to use C++'s native exceptions, but wrapped in macros in such a way that the legacy code works without change. In fact, it works better than the legacy code because the legacy coders sometimes forgot to mark variables as volatile :-). If this is any use to you, contact off-list (I am not subscribed), and I can assist with the conceptual version of the macros - I've long forgotten the MSVC exception details. Thanks, Shaheed

1 0

Re: Lotus Notes and wine and RedHat 9
by Joe Squire May 17, 2003

May 17, 2003

Thanks for your responses. Sorry I didn't reply earlier - been travelling. I noticed the latest wine newsletter that this may be fixed already in the CVS builds ( http://www.winehq.org/index.php?issue=170#Lotus%20Notes%20Breakage ) I will try installing that on the weekend and let you know how I get on. Cheers Joe >From: Mike Hearn <m.hearn(a)signal.qinetiq.com> >To: Uwe Bonnes <bon(a)elektron.ikp.physik.tu-darmstadt.de> >CC: Joe Squire <wiresquire(a)hotmail.com>, wine-devel(a)winehq.com, >Andreas Mohr <andi(a)rhlx01.fht-esslingen.de> >Subject: Re: Lotus Notes and wine and RedHat 9 >Date: 12 May 2003 10:02:10 +0100 > > > Hallo, > > > > it boiled down that Andi added a lot of functions in the user32 spec >file, > > with no implementation provided. These functions are mostly only >implemented > > in recent windows version, and Notes has to cope with these functions >not > > available. But as we added them in the spec file, this request will be > > granted, but later fails when this function should be executed. > > > > I notified Andi, but got no feedback so long. > >Perhaps a patch should be sent that comments out these functions in the >spec file until there are implementations of them? > >-- >Mike Hearn <m.hearn(a)signal.qinetiq.com> >QinetiQ - Malvern Technology Center > _________________________________________________________________ ninemsn Extra Storage is now available. No account expiration - no need to worry about losing your Hotmail account. Go to http://join.msn.com/?page=dept/home&pgmarket=en-au

1 0

xmms-winamp 0.4 vs. winex
by Tom Hibbert May 17, 2003

May 17, 2003

Hi all, I am wanting to pdate xmms-winamp (http://www.emulinks.de/xmms-winamp/) to run with WineX CVS and gcc 3.2. The original developer (mailto:uli(a)emulinks.de) tells me he made the original source just to see if it could be done and has no interest in keeping the project alive. I personally would like to see xmms-winamp updated to work with WineX CVS because we now have vertex shader support which gives a very good chance of running MilkDrop. So i set about firstly trying to make the current source compile on my system. The configure script is very brain-dead so I had to force a lot of valies. But, it still has trouble detecting the wine libraries. --- snip from configure for i in "" ".dll" do for j in "" -L/usr/local/lib -L/usr/lib -L/usr/X11R6/lib do if gcc -o $TMPFILE $TMPFILE.c $j -lkernel32${i} 2>/dev/null >/dev/null then echo -n found LIBEXT="$i" LIBPATH="$j" found=1 test -n "$LIBPATH" && echo -n " at $LIBPATH" test -n "$LIBEXT" && echo ", extension \"$LIBEXT\"" || echo ", no extension" break 2 fi done done --- end snip In this case the script is testing that the libraries are there and linkable by compiling a sample one liner program int main(int argc, char** argv){} and linking it against libkernel32.so. When I try and do this myself, I receive the following error messages: $ gcc -o test /tmp/xmmswinampconf.816.c -L/usr/lib/transgaming/WineX/lib -lkernel32 /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `casemap_lower' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `utf8_wcstombs' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `cp_enum_table' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `wine_errno_location' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `wine_gettid' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `casemap_upper' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `utf8_mbstowcs' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `cp_mbstowcs' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `wctype_table' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `wine_h_errno_location' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `cp_wcstombs' /usr/lib/transgaming/WineX/lib//libntdll.so: undefined reference to `cp_get_table' collect2: ld returned 1 exit status My thoughts are that kernel32 can no longer be linked in this method, can someone confirm? Thanks, Tom Hibbert Work: someday Home: +649 815 1838 Obtain my GPG Key from: http://cryptocracy.hn.org/pubkey.txt

1 0

Re: Fix ListView Bugs
by Dimitrie O. Paun May 17, 2003

May 17, 2003

On Fri, 16 May 2003, Robert Shearman wrote: > Fixes two bugs. > 1. Regression. The colour used for the background of text that wasn't > selected wasn't being set properly any more, as it had been moved into > notify_prepaint. This patch moves it back to where it was before and the > text background appears correctly. Maxim and I are working on an alternative patch, which also does a better job of handling subitem notifications. Alexandre, please don't apply. > 2. Don't clear the item header when column 0 is inserted - it would clear > anything already set in the header, such as information set by InsertItem Well, the header contains the text and the icon. When I insert a new column, I don't expect the old column text to be preserved. Do you have tests that show otherwise? -- Dimi.

4 12

Bugzilla Upgraded
by Jeremy Newman May 17, 2003

May 17, 2003

I have upgraded our installation of Bugzilla from 2.14 to 2.17.4. This solves some issues we were having. - My Bugs link was not working - Queries where busted. >From the looks of it, quite a bit has changed in this new version. So poke around and let me know an issues that crop up. Yes, I plan to convert the bugs site over to the new design soon. Today I mainly wanted just to get it up and running. -- Jeremy Newman <jnewman(a)codeweavers.com> CodeWeavers, Inc.

2 2

Re: ddraw: load OpenGL at runtime
by Brian Vincent (C) May 16, 2003

May 16, 2003

> IMHO we should have our monthly "architectural meeting" and have some > volunteer go through the log and add the relevant content to the Wine > architecture page -- wait, which page again?? ;) If someone wants to collect a log of such a meeting I can summarize it and add it to WWN. From there, we could have the existing status pages just point to the WWN issue where it's covered. The next question is when a good time would be to hold something like this. Based on the list traffic, it seems like most people are around Monday - Friday. I'm -6 GMT, so something like 1pm GMT would work well for me (anyone in California or west coast? that might be too early.) Or, if work interferes (it doesn't seem to) maybe Saturday or Sunday would be better? --------------- Brian Vincent Copper Mountain Telecom vincentb(a)coppercolorado.com

1 0

Re: Wine: gethostbyname implementation in WS2_32 is buggy
by Andreas Mohr May 16, 2003

May 16, 2003

On Fri, May 16, 2003 at 04:59:18PM +0100, wrote: > Hi, > > I am not confident how this patch-submission bussiness > works, so I will just quickly describe what's wrong > and how to fix it. > > The current implementation of gethostbyname in WS2_32 > (located in winsock/socket.c) doesn't handle a null > string argument correctly. > > According to MS definition of gethostbyname, when it > receives a null argument, it should return the same > result as if a call to gethostname was made. Currently > Wine doesn't do this: it simply crashes. > > Fix is simple: add something like: > if (name == NULL) return __ws_gethostbyname( > "localhost", WS_DUP_LINEAR ); Is this the way it's supposed to be handled? I interpreted this (in my original patch I sent very recently) as something like: --- dlls/winsock/socket.c 13 May 2003 04:43:44 -0000 1.126 +++ dlls/winsock/socket.c 16 May 2003 20:00:28 -0000 @@ -2893,6 +2893,17 @@ int ebufsize=1024; struct hostent hostentry; int locerr = ENOBUFS; + char local_name[256]; + + /* name == NULL means use the local host name */ + if (name == NULL) + { + if (WS_gethostname(local_name, sizeof(local_name)) != 0) + return NULL; +··············· + name = local_name; + } +···· host = NULL; Which way is the right one? First? Second? Both? :) -- Help prevent Information Technology Fascism! - before it's too late... http://www.againsttcpa.com

1 0

Re: Fix ListView Bugs
by Robert Shearman May 16, 2003

May 16, 2003

> Subject: Re: Fix ListView Bugs > From: Maxime =?ISO-8859-1?Q?Belleng=E9?= <maxime.bellenge(a)laposte.net> > To: wine-devel(a)winehq.org > Cc: Wine Patches <wine-patches(a)winehq.com> > Organization: > Date: 16 May 2003 19:07:38 +0200 > > Robert can you try the attached patch to see if it solves the point 1. > in your case ? > If it doesn't I think we should make a merge of our patches. > Unfortunately it doesn't. Rob

1 0

Re: Fix ListView Bugs
by Maxime Bellengé May 16, 2003

May 16, 2003

I think that patch is not correct because it does not correctly handle the subitem customdraw notification. Dimitrie O. Paun and I are working on a patch to solve that. Robert can you try the attached patch to see if it solves the point 1. in your case ? If it doesn't I think we should make a merge of our patches. Max On Fri, 2003-05-16 at 13:02, Robert Shearman wrote: > Fixes two bugs. > 1. Regression. The colour used for the background of text that wasn't > selected wasn't being set properly any more, as it had been moved into > notify_prepaint. This patch moves it back to where it was before and the > text background appears correctly. > 2. Don't clear the item header when column 0 is inserted - it would clear > anything already set in the header, such as information set by InsertItem > > ChangeLog: > - Make selected text appear correctly now > - Don't clear header for subitem 0 when column 0 is inserted > > Rob -- Maxime Bellengé <maxime.bellenge(a)laposte.net>

1 0

RFC: Store apartment model in TEB
by Robert Shearman May 16, 2003

May 16, 2003

Hi, I've sent it to wine-devel so I can get some comments on it. This patch will basically store the apartment model in a structure in the TEB like Windows. This will allow CoInitializeEx to return the right value, which in itself has got some apps working better on my machine. If this patch is ok, then the next question must be: what functions/interfaces depend on the apartment model? ChangeLog: - CoInitializeEx should store apartment model separately for each thread Rob

1 0