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

unixfs registered by default now in cvs

by Michael Jung

Hi, With the current CVS version, the unixfs shell namespace extension is now registered by default at the desktop. So if you do a 'regsvr32 shell32' you will see the unix filesystem in the file dialogs. It's probably still quite buggy though. It would be cool if we could get the biggest problems sorted out before the next release. So if you have an application, which doesn't work due to unixfs, please report to wine-devel. If you don't want to use unixfs, because you don't like it or because it totally breaks your setup, you can delete the following registry-key and you should be back to the old behaviour: HKLM\Software\Microsoft\Windows\CurrentVersion\Explorer\Desktop\Namespace\ {9D20AAE8-0625-44B0-9CA7-71889C2254D9} This will also help you to figure out if a problem is due to unixfs. If you do 'regsvr32 shell32' again, the key will be re-created. Bye, -- Michael Jung mjung(a)iss.tu-darmstadt.de

June 30, 2005

Re: Delays in FileOpen

by Uwe Bonnes

>>>>> "Michael" == Michael Jung <mjung(a)iss.tu-darmstadt.de> writes: Michael> On Wednesday 29 June 2005 23:36, Uwe Bonnes wrote: >> File Open Dialogs for me have some noticable initial delay when >> starting up. For example, try in internet explorer "Save under". I >> don't remember seeing that delay some time before. Dif recent >> shfolder changes cause that delay? Michael> There was a problem in SHGetPathFromIDList, which caused a Michael> delay in the file dialogs. I hope it was fixed with this patch, Michael> applied at the 2005/06/25. Michael> http://cvs.winehq.org/patch.py?id=18448 Michael> Does your local tree already contain this patch? Yes, my local tree contains the patch. But my first impression was incorrect, the delay is _not_ new. The test with ie6 where done with a wine-200410xx setup, and "save as" shows the delay too. But also other programs run against recent CVS wine show that initial delay until the file selection box starts up. Do other notice that delay too? E.g. the file -> open box from wordview97 takes several seconds on an Athlon2500. (about 500 file/directories including dot files/directory in $HOME where the file open box starts up). Bye -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

June 30, 2005

Re: [user/tests]: Re: Fix bug that left mouse buttons swapped after tests

by jakov＠vmlinux.org

> Saulius Krasuckas wrote: >>>ChangeLog: Fix bug that left mouse buttons swapped after tests >>> >>>Ivan. >> >> >> I think your patch causes test to fail on Wine, Ivan. My try goes next. > > Maybe that isn't a bad thing, the results from the latest build of > winetest shows this test fails > on windows 2000, windows nt, windows me, and windows 98se, so I somewhat > doubt the wine behaviour is > correct. > > Ivan. And this is exactly why no part of winetest should be allowed to fail. //Jakob

June 30, 2005

Fwd: Anybody looking for a pet Wine project?

by Vijay Kiran Kamuju

---------- Forwarded message ---------- From: Vijay Kiran Kamuju <infyquest(a)gmail.com> Date: Jun 30, 2005 3:42 PM Subject: Re: Anybody looking for a pet Wine project? To: Mike McCormack <mike(a)codeweavers.com> typelib info of hhctrl.ocx of winxp (used pe explorer) -------------------------------------------------------------------------- //HHCtrl 4.0 Type Library //Version: 4.0 HHCTRLLib; GUID = {ADB880A2-D8FF-11CF-9377-00AA003B7A11}; //Event interface for HHCtrl Dispatch _HHCtrlEvents; GUID = {ADB880A3-D8FF-11CF-9377-00AA003B7A11}; function Click(out ParamString: BSTR); stdcall; //HHCtrl Class CoClass HHCtrl; GUID = {52A2AAAE-085D-4187-97EA-8C30DB990436}; //IHHCtrl Interface Dispatch IHHCtrl; GUID = {ADB880A1-D8FF-11CF-9377-00AA003B7A11}; function QueryInterface(riid: ^GUID; out ppvObj: ^^VOID); stdcall; function AddRef: UI4; stdcall; function Release: UI4; stdcall; function GetTypeInfoCount(out pctinfo: ^UINT); stdcall; function GetTypeInfo(itinfo: UINT; lcid: UI4; out pptinfo: ^^VOID); stdcall; function GetIDsOfNames(riid: ^GUID; rgszNames: ^^I1; cNames: UINT; lcid: UI4; out rgdispid: ^I4); stdcall; function Invoke(dispidMember: I4; riid: ^GUID; lcid: UI4; wFlags: UI2; pdispparams: ^DISPPARAMS; out pvarResult: ^Variant; out pexcepinfo: ^EXCEPINFO; out puArgErr: ^UINT); stdcall; property-put Image(: BSTR); stdcall; property-get Image: BSTR; stdcall; //Click method function Click; stdcall; //Click method function HHClick; stdcall; //Print method function Print; stdcall; //syncURL method function syncURL(out pszUrl: BSTR); stdcall; //TCard method function TCard(out wParam: UINT_PTR; out lParam: LONG_PTR); stdcall; //Text Popup method function TextPopup(out pszText: BSTR; out pszFont: BSTR; out horzMargins: INT; out vertMargins: INT; out clrForeground: UI4; out clrBackground: UI4); stdcall; Alias UINT_PTR; UI4 Alias LONG_PTR; I4 //Deprecated HHCtrl Class 2 CoClass OldHHCtrl2; GUID = {41B23C28-488E-4E5C-ACE2-BB0BBABE99E8}; //Deprecated HHCtrl Class 1 CoClass OldHHCtrl1; GUID = {ADB880A6-D8FF-11CF-9377-00AA003B7A11}; On 6/30/05, Mike McCormack <mike(a)codeweavers.com> wrote: > > Hi All, > > If anybody is looking for a small project, I've got just the thing for you! > > hh.exe and hhctrl.ocx are two components of the Microsoft HTML help > engine. hh.exe is a small wrapper around hhctrl.ocx, which is the HTML > help viewer. > > hhctrl.ocx embeds IE and feeds it HTML documents decoded from .chm files > with itss.dll (the Infotech storage system). Implementing hhctrl.ocx > would make an excellent test case for IE embedding that Jacek is working > on, and for itss.dll. > > So most of the components are already in place (or will be soon) to make > HTML help work... any takers to fill in the dots? > > The reward is, like most of Wine code, to say "I made that work" :) > > Let me know if you're interested. > > Mike > > >

June 30, 2005

Re: [user/tests]: Re: Fix bug that left mouse buttons swapped after tests

by Ivan Leo Puoti

Saulius Krasuckas wrote: >>ChangeLog: Fix bug that left mouse buttons swapped after tests >> >>Ivan. > > > I think your patch causes test to fail on Wine, Ivan. My try goes next. Maybe that isn't a bad thing, the results from the latest build of winetest shows this test fails on windows 2000, windows nt, windows me, and windows 98se, so I somewhat doubt the wine behaviour is correct. Ivan.

June 30, 2005

Re: [user/tests]: Re: Fix bug that left mouse buttons swapped after tests

by Paul Vriens

>> ChangeLog: Fix bug that left mouse buttons swapped after tests >> >> Ivan. > > I think your patch causes test to fail on Wine, Ivan. My try goes next. > > > ChangeLog: > Saulius Krasuckas <saulius.krasuckas(a)ieee.org> > - Break from the loop to restore SM_SWAPBUTTON metrics. > - SetLastError() to see if restoration changes it. > > > Index: dlls/user/tests/sysparams.c > =================================================================== > RCS file: /home/wine/wine/dlls/user/tests/sysparams.c,v > retrieving revision 1.38 > diff -p -u -r1.38 sysparams.c > --- dlls/user/tests/sysparams.c 20 Jun 2005 15:36:17 -0000 1.38 > +++ dlls/user/tests/sysparams.c 30 Jun 2005 07:36:40 -0000 > @@ -1011,7 +1011,7 @@ static void test_SPI_SETMOUSEBUTTONSWAP( > rc=SystemParametersInfoA( SPI_SETMOUSEBUTTONSWAP, vals[i], 0, > SPIF_UPDATEINIFILE | SPIF_SENDCHANGE ); > if (!test_error_msg(rc!=vals[i],"SPI_{GET,SET}MOUSEBUTTONSWAP")) > - return; > + break; It's nice to break but the the end-result will be that the test fails on all windows versions I could my hands on and that the previous setting is restored. Not very useful it seams. Cheers, Paul.

June 30, 2005

Re: Delays in FileOpen

by Michael Jung

On Wednesday 29 June 2005 23:36, Uwe Bonnes wrote: > File Open Dialogs for me have some noticable initial delay when starting > up. For example, try in internet explorer "Save under". I don't remember > seeing that delay some time before. Dif recent shfolder changes cause that > delay? There was a problem in SHGetPathFromIDList, which caused a delay in the file dialogs. I hope it was fixed with this patch, applied at the 2005/06/25. http://cvs.winehq.org/patch.py?id=18448 Does your local tree already contain this patch? Bye, -- Michael Jung mjung(a)iss.tu-darmstadt.de

June 30, 2005

Re: Using ReactOS Registry format

by Jürgen Schmied

Hi! Wine had support for reading win2000 registry files once. It was dropped some time ago. I had implemented it around 1999/2000. Have a look in the CVS. Bye Juergen > Hi Martin, > Thanks for this info. It looks like this might work. How do I contact > Eric Kohl? > > Thanks, > James > > On Tue, 2005-06-21 at 17:57 +0200, Martin Fuchs wrote: > > Hi James, > > > > > Last night Martin Fuchs suggested that we look into using ReactOS's > > > registry format in order to be compatible with Windows registry > > > databases. I have the latest release of ReactOS running on QEMU on my > > > box, so I checked it out. Basically, they're using the same regedit > > > program from Wine, missing find command and all (Which I too feel is a > > > pain in the neck). > > > > A bunch of dlls and applications are synced between Wine and ROS from > > time to time, also including regedit. What about implemting the > > missing find functionality at your own to bring both projects a little > > foreward? ;-) > > > > > > > I looked at the config stuff, and I found what looked > > > like some binary database files for each of the main registry sections. > > > Unfortunately, there's no documentation at all on any of this on their > > > website. > > > > I can forward you a mail from Steven Edwards to bring a bit light into > > this issue: > > > > S> I think ReactOS's registry is binary compatible with NT4. It and > > the windows 2000 format was > > S> documented/reversed for samba and the linux ntchpwd bootdisk > > projects. If I remeber right Eric > > S> Kohl offered to release some of his work to Wine for the binary > > format so you might want to ping > > S> him about the implementation details. > > > > So Eric Kohl would be the right man to ask about the internals of NT's > > registry format. > > > > > If we decide to go this route, we may be in for a hell of a lot > > > of work. But, I do agree with all of your points. I think the current > > > system could use some improvement, especially in the area of searching. > > > Let me know what you think of all this. > > > > > > Regards, > > > > Martin > > >

June 30, 2005

Re: Mouse-buttons swapped after winetest

by Saulius Krasuckas

* On Fri, 24 Jun 2005, Paul Vriens wrote: > * On Fri, 2005-06-24 at 17:44, Saulius Krasuckas wrote: > > > > Then would be nice to print "err" in following format "err=%0x%08lx" or > > such after a restoration. Any more ideas? Sorry, I meant 0x%08lx. > Why do you want to change that? I thought %08lx is pretty common in most > of the Wine code. (I remember seeing your email on wine-devel about that > though). I just see two versions of formatting in Wine and want to get rid of one of it. Not a large difference which. OTOH, which one case brings you more clearness while reading logs: 00000021 or 0x0000021 ? I prefer the latter one.

June 30, 2005

Anybody looking for a pet Wine project?

by Mike McCormack

Hi All, If anybody is looking for a small project, I've got just the thing for you! hh.exe and hhctrl.ocx are two components of the Microsoft HTML help engine. hh.exe is a small wrapper around hhctrl.ocx, which is the HTML help viewer. hhctrl.ocx embeds IE and feeds it HTML documents decoded from .chm files with itss.dll (the Infotech storage system). Implementing hhctrl.ocx would make an excellent test case for IE embedding that Jacek is working on, and for itss.dll. So most of the components are already in place (or will be soon) to make HTML help work... any takers to fill in the dots? The reward is, like most of Wine code, to say "I made that work" :) Let me know if you're interested. Mike

June 30, 2005