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

Re: uxtheme: also send WM_THEMECHANGED to child windows
by Dimi Paun July 3, 2005

July 3, 2005

On Sun, 2005-07-03 at 15:11 +0200, Frank Richter wrote: > +static void UXTHEME_broadcast_themechange (HWND hWnd) > +{ > + if (hWnd == NULL) > + { > + EnumWindows (UXTHEME_EnumChildProc, 0); > + } > + else > + { > + if (hWnd != NULL) PostMessageW(hWnd, WM_THEMECHANGED, 0, 0); > + EnumChildWindows (hWnd, UXTHEME_EnumChildProc, 0); > + } > +} This function would be more useful (and usage clearer) if it took the msg as an argument (also I think you can get rid of UXTHEME_EnumChildProc too, untested): +static BOOL UXTHEME_broadcast_msg (HWND hWnd, LPARAM msg) +{ + if (hWnd == NULL) + { + EnumWindows (UXTHEME_broadcast_msg, msg); + } + else + { + PostMessageW(hWnd, msg, 0, 0); + EnumChildWindows (hWnd, UXTHEME_broadcast_msg, msg); + } + return TRUE; +} Note: the 'if' for PostMessageW was superfluous. -- Dimi Paun <dimi(a)lattica.com> Lattica, Inc.

3 3

Possible regressions in IE6 SP1 install
by James Liggett July 3, 2005

July 3, 2005

Hello all, I've noticed some problems installing Internet Explorer 6 SP1 with any version of WINE newer than 20050310, and that includes CVS as of yesterday. Apparently, I'm not the only one; many other are also having issues with 20050419 and 20050524. With my current setup, I get this error repeatedly at 81% during installation: fixme:thunk:__regs_CommonUnimpStub generic stub: ? It keeps looping like that on and on forever, and the install never finishes, but the UI stays responsive. I've had many issues installing IE in the past, but nothing like this. Can anyone give me any leads as to what might cause this? If it might help, I've applied Oliver Stieber's latest DX9 patch, and have tried both Wine Tools and Sidenet. Thanks very much, James Liggett

7 9

Re: MSCMS: fix color directory handling
by Uwe Bonnes July 3, 2005

July 3, 2005

>>>>> "Hans" == Hans Leidekker <hans(a)it.vu.nl> writes: Hans> I found the culprit. There was a patch by Vincent Beron that Thanks, This patch lets "make test" in mscms/test succeed. B.t.w: Hans, I can't reach mirzam.it.vu.nl. Is it down? -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

2 1

Proposal: Let Oleaut32/tests/olepicture.c faile more gracefully with missing libgif
by Uwe Bonnes July 3, 2005

July 3, 2005

Hallo, while I found strong opposition with my proposal to fail more gracefully with a missing libgif, I still think the FIXME output is enough and the test shouldn't fail, blocking any further test in a "make test" run. To have a possible patch at least in the archive, please find it appended. Bye -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- Index: wine/dlls/oleaut32/tests/olepicture.c =================================================================== RCS file: /home/wine/wine/dlls/oleaut32/tests/olepicture.c,v retrieving revision 1.3 diff -u -r1.3 olepicture.c --- wine/dlls/oleaut32/tests/olepicture.c 27 Jun 2005 09:57:28 -0000 1.3 +++ wine/dlls/oleaut32/tests/olepicture.c 3 Jul 2005 14:43:38 -0000 @@ -165,6 +165,11 @@ handle = 0; hres = IPicture_get_Handle (pic, &handle); + if (hres == E_NOTIMPL) + { + todo_wine ok(hres == S_OK, "Probably no support for libgif/libungif, Aborting\n"); + return; + } ok(hres == S_OK,"IPicture_get_Handle does not return S_OK, but 0x%08lx\n", hres); ok(handle != 0, "IPicture_get_Handle returns a NULL handle, but it should be non NULL\n");

1 0

Palm Desktop Install
by Andrew Neil Ramage July 3, 2005

July 3, 2005

Reading that someone was able to install the Palm Desktop software, I decided to try myself. I downloaded the software from Palm, then tried installing it. This is the output. [andrew(a)zeus Windows]$ WINEDEBUG=+all [andrew(a)zeus Windows]$ wine Palm\ setup.exe fixme:thread:NtSetInformationThread Set ThreadImpersonationToken handle to (nil) fixme:thread:NtSetInformationThread Set ThreadImpersonationToken handle to 0x48 fixme:thread:NtSetInformationThread Set ThreadImpersonationToken handle to (nil) fixme:thread:NtSetInformationThread Set ThreadImpersonationToken handle to 0x50 fixme:powermgnt:SetThreadExecutionState (0x80000001): stub, harmless. fixme:powermgnt:SetThreadExecutionState (0x80000000): stub, harmless. Does anyone know what's wrong ? Below is a listing of my DOSDEVICES DIRECTORY. [andrew(a)zeus Windows]$ ls -l ~/.wine/dosdevices total 0 lrwxrwxrwx 1 andrew andrew 10 Jun 6 19:52 c: -> ../drive_c lrwxrwxrwx 1 andrew andrew 10 Jun 6 20:08 d: -> /mnt/cdrom lrwxrwxrwx 1 andrew andrew 11 Jun 6 19:53 e: -> /mnt/cdrom2 lrwxrwxrwx 1 andrew andrew 22 Jun 6 19:53 f: -> /home/andrew/winetools lrwxrwxrwx 1 andrew andrew 12 Jun 6 19:53 h: -> /home/andrew lrwxrwxrwx 1 andrew andrew 4 Jun 6 19:53 t: -> /tmp lrwxrwxrwx 1 andrew andrew 12 Jul 3 13:42 USB1 -> /dev/ttyUSB1 lrwxrwxrwx 1 andrew andrew 12 Jul 3 13:43 USB2 -> /dev/ttyUSB2 lrwxrwxrwx 1 andrew andrew 12 Jul 3 13:43 USB3 -> /dev/ttyUSB3 lrwxrwxrwx 1 andrew andrew 1 Jun 6 19:52 z: -> / -- Andrew You can be the captain I will draw the chart Sailing into destiny Closer to the heart Closer to the Heart by Rush (A Farewell to Kings, 1977) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

1 0

mscms/tests/profile.c
by Uwe Bonnes July 3, 2005

July 3, 2005

Hallo, make test in dlls/mscms/tests fails in profile.c. It seems that the test expects an installed ColorProfile.exe. However I think our police for the tests is to have them succeed without any other prerequisit. Hans, can you perhaps rethink your test strategy? Bye -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

3 5

Re: MSHTML: Added test of IOleObject::[SG]etClientSite
by Alexandre Julliard July 3, 2005

July 3, 2005

Jacek Caban <jack(a)itma.pwr.wroc.pl> writes: > Changelog: > - Added test of IOleObject::[SG]etClientSite > - Fix IOleObject_SetClientSite to pass tests The test fails here: htmldoc.c:232: Test failed: *pchKey=0x7fc7d2d9, expected NULL -- Alexandre Julliard julliard(a)winehq.org

1 0

Bug in dlls/comctl32/comctl32undoc.c
by Sascha Cunz July 3, 2005

July 3, 2005

Hi, playing around a bit with wine, i found a thing that is obviously a bug. In lines 1944 and 1977 of dlls/comctl32/comctl32undoc.c the value "0x7FFF" is used to check for the boundary of a 32 bit integer. However, there should be used MAX_INT instead. By the way: Looking at current MSDN, DPA and DSA looks rather documented than "undoc" :-) Regards Sascha P.S.: Please CC me on answers as i'm not subscribed here.

2 3

Get Firefox Button
by James Liggett July 2, 2005

July 2, 2005

Tom, I've been following your discussion about putting a Get Firefox button on the WineHQ front page. While I'm a big fan of Firefox, I think that we should advocate the more generic W3C standards instead, and use XHTML that is valid against those. That way, we can support pretty much any browser. What do you think? Regards, James Liggett

2 1

Re: [PATCH] FCI work for cabinet.dll [cabinet-fci-patch-03c.diff]
by Mike McCormack July 2, 2005

July 2, 2005

Hi Gerold Gerold J. Wucherpfennig wrote: > And now take this... Gulp. > + p_fci_internal->perf->erfOper = FCIERR_NONE; > + p_fci_internal->perf->erfType = ERROR_INVALID_DATA; > + p_fci_internal->perf->fError = TRUE; > + SetLastError(ERROR_INVALID_DATA); Your code seem to be doing this over and over again. Is there any reason not to make a macro so you can do something like: FCI_SET_ERROR(FCIERR_NONE, ERROR_INVALID_DATA ); maybe something like this: #define FCI_SET_ERROR( oper, type ) \ p_fci_internal->perf->erfOper = oper; \ p_fci_internal->perf->erfType = type; \ p_fci_internal->perf->fError = TRUE; \ SetLastError(type); Also it would be nice to split this into two patches, one changing all the error codes as per above, and one with all the other minor changes you made. A 75k patch will take a long time for Alexandre to digest... Mike

2 1