October 2007 - Wine-Devel - list.winehq.org

Re: user32: Add more listbox message tests, make them pass under Wine
by Alexandre Julliard Oct. 16, 2007

Oct. 16, 2007

Dmitry Timoshkov <dmitry(a)codeweavers.com> writes: > @@ -533,8 +538,11 @@ static INT LISTBOX_GetItemFromPoint( LB_DESCR *descr, INT x, INT y ) > static void LISTBOX_PaintItem( LB_DESCR *descr, HDC hdc, const RECT *rect, > INT index, UINT action, BOOL ignoreFocus ) > { > - LB_ITEMDATA *item = NULL; > - if (index < descr->nb_items) item = &descr->items[index]; > + LB_ITEMDATA *item; > + > + if (index >= descr->nb_items) return; > + > + item = &descr->items[index]; This will break the painting of a focus rect for an empy listbox. We are supposed to show a focus rect on the first item when the listbox has focus, even if there are no items. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [SPAM] Re: USER32: fix EM_SETLIMITTEXT with zeroed args (with tests)
by Alexandre Julliard Oct. 16, 2007

Oct. 16, 2007

Alex Villacís Lasso <a_villacis(a)palosanto.com> writes: > * Implement OS version-dependent behavior for EM_SETTEXTLIMIT There's no reason to make this OS-dependent, unless you really have an app that checks the version and breaks if the limit is too high. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: wordpad: Refuse to load OLE compound storage files like Windows does
by Alexandre Julliard Oct. 16, 2007

Oct. 16, 2007

Dmitry Timoshkov <dmitry(a)codeweavers.com> writes: > + IStorage *stg; > + > + if (StgOpenStorage(szOpenFileName, NULL, STGM_SHARE_DENY_WRITE | STGM_READ, > + NULL, 0, &stg) == S_OK) > + { > + IStorage_Release(stg); > + MessageBoxW(hMainWnd, MAKEINTRESOURCEW(STRING_OLE_STORAGE_NOT_SUPPORTED), wszAppTitle, > + MB_OK | MB_ICONEXCLAMATION); > + return; > + } It would be more efficient to simply check the magic number, particularly since we already have support for doing that. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: ntdll: Fix NtAccessCheck for NULL, but present SACLs and DACLs.
by Juan Lang Oct. 15, 2007

Oct. 15, 2007

Hi Rob, - sd.sacl_len = (present ? sacl->AclSize : 0); + sd.sacl_len = ((present && dacl) ? sacl->AclSize : 0); Shouldn't that be ((present && sacl) ? sacl->AclSize : 0); ? --Juan

2 1

re: Make test drill, next steps, call for help with Winetest
by Dan Kegel Oct. 15, 2007

Oct. 15, 2007

>> There is a headless operation mode already, it detects >> if you run it as a service, and skips appropriate tests. > > Perhaps on Windows; I'm focused on Linux. How about wine net start winetest? <ducks> - Dan

1 0

Re: Have `make distclean` also remove .i and .s files
by Alexandre Julliard Oct. 15, 2007

Oct. 15, 2007

Gerald Pfeifer <gerald(a)pfeifer.com> writes: > After using GCC's -save-temps option to debug something, I noticed > that `make distclean` fails to actually remove .i and .s files. > > This patch fixes it. I hope it's the right place to make this change? It would be, but I don't think we want make clean to remove all possible output files from all possible tools. The current list is probably too much already, it should really be limited to files that are generated as part of the normal build process. -- Alexandre Julliard julliard(a)winehq.org

3 3

Re: First batch of Swedish translation patches
by James Hawkins Oct. 15, 2007

Oct. 15, 2007

On 10/15/07, Daniel Nylander <po(a)danielnylander.se> wrote: > > Here is the first batch of Swedish translation patches. > One patch per email. -- James Hawkins

1 0

Re: Patch for adding Swedish localization
by Robert Shearman Oct. 15, 2007

Oct. 15, 2007

Daniel Nylander wrote: > Here is the patch to add Swedish localization files to Wine. > Sorry, this is my first patch using git and I'm really unsure if I did > it correctly. > You need to use git-add to include the new files in the commit/patch you sent. -- Rob Shearman

3 6

make crosstest - a catastrophic saga
by Pedro Araújo Oct. 15, 2007

Oct. 15, 2007

Hi all, I've been consistently failing to 'make crosstest'... I've already tried my distro's repositories, Hans Leidekker's patches, all to no avail. I wish I could use the test suite, but I need some custom local tests and therefore I need to be able to build win32 executables. However, I'm stuck getting this: > security.cross.o: In function `test_ConvertSecurityDescriptorToString':/home/foo/Wine/dlls/advapi32/tests/security.c:1996: undefined reference to `_CreateWellKnownSid(a)16' > :/home/foo/Wine/dlls/advapi32/tests/security.c:2006: undefined reference to `_CreateWellKnownSid(a)16' > collect2: ld returned 1 exit status > make[2]: *** [advapi32_crosstest.exe] Error 1 > make[2]: Leaving directory `/home/fooer/Wine/dlls/advapi32/tests' > make[1]: *** [advapi32/tests/__crosstest__] Error 2 > make[1]: Leaving directory `/home/fooer/Wine/dlls' > make: *** [dlls/__crosstest__] Error 2 or this, if I 'make crosstest -C dlls/gdi32/tests/': > bitmap.cross.o: In function `test_GdiAlphaBlend':/home/foo/Wine/dlls/gdi32/tests/bitmap.c:1870: undefined reference to `_GdiAlphaBlend(a)44' > :/home/foo/Wine/dlls/gdi32/tests/bitmap.c:1872: undefined reference to `_GdiAlphaBlend(a)44' > :/home/foo/Wine/dlls/gdi32/tests/bitmap.c:1874: undefined reference to `_GdiAlphaBlend(a)44' > :/home/foo/Wine/dlls/gdi32/tests/bitmap.c:1875: undefined reference to `_GdiAlphaBlend(a)44' > :/home/foo/Wine/dlls/gdi32/tests/bitmap.c:1876: undefined reference to `_GdiAlphaBlend(a)44' > bitmap.cross.o:/home/foo/Wine/dlls/gdi32/tests/bitmap.c:1877: more undefined references to `_GdiAlphaBlend(a)44' follow > collect2: ld returned 1 exit status > make: *** [gdi32_crosstest.exe] Error 1 > make: Leaving directory `/home/fooer/Wine/dlls/gdi32/tests' What am I doing wrong? - Pedro. -- "Compatibility means deliberately repeating other people's mistakes." (David J. Wheeler)

2 1

Re: RICHED20: EM_SETTEXTEX must support RTF strings too
by Daniel Remenak Oct. 15, 2007

Oct. 15, 2007

When you fix bugs that are documented in comments, please fix the comments also. At the top of the file: EM_SETTEXTEX 3.0 (no rich text insertion handling, proper style?) And in the case itself /* FIXME: this should support RTF strings too, according to MSDN */ Thanks, Daniel Remenak On 10/13/07, a_villacis(a)palosanto.com <a_villacis(a)palosanto.com> wrote: > Another fix for Zoiper (www.zoiper.com) This app can display an About dialog > with a richedit control that displays program credits. The text is a RTF > string pasted with EM_SETTEXTEX, which should result in formatted text being > displayed, but instead results in raw display of RTF codes. The attached > patch fixes this. > > Changelog: > * EM_SETTEXTEX should also support RTF strings, just like WM_SETTEXT > > > >

1 0

re: Wine at Systems in Munich
by Dan Kegel Oct. 15, 2007

Oct. 15, 2007

It would be great to have a booth there, hope somebody can go. It's kind of hard to do a good booth on short notice, though. Let's remember that this is coming up next year, and plan for it. Can you send me contact info for the organizers? Thanks, Dan

5 4

Re: Fix casts in dlls/crypt32/decode.c
by Robert Shearman Oct. 15, 2007

Oct. 15, 2007

Gerald Pfeifer wrote: > } > if (!HIWORD(lpszStructType)) > { > - switch (LOWORD(lpszStructType)) > + switch ((DWORD)lpszStructType) > You're just passing the problem on to people doing 64-bit builds. You should use DWORD_PTR or use LOWORD everywhere, like you suggested. -- Rob Shearman

3 2

Re: widl: Generate dlldata files. [resend]
by Robert Shearman Oct. 15, 2007

Oct. 15, 2007

Dan Hipschman wrote: > if (!parse_only && do_header) write_interface($$); > if (!parse_only && do_idfile) write_iid($$); > pointer_default = $1.old_pointer_default; > + if (!parse_only && !in_typelib && is_attr($$->attrs, ATTR_OBJECT)) > + have_object = TRUE; > } > /* MIDL is able to import the definition of a base class from inside the > * definition of a derived class, I'll try to support it with this rule */ > @@ -828,6 +830,8 @@ interfacedef: interfacehdr inherit > if (!parse_only && do_header) write_interface($$); > if (!parse_only && do_idfile) write_iid($$); > pointer_default = $1.old_pointer_default; > + if (!parse_only && !in_typelib && is_attr($$->attrs, ATTR_OBJECT)) > + have_object = TRUE; > } > | dispinterfacedef { $$ = $1; } > ; > This seems like a bit of a hack because you can find out whether the parsed IDL contains any definitions for objects by traversing the interface tree. -- Rob Shearman

1 0

Re: Swedish translations
by Robert Shearman Oct. 15, 2007

Oct. 15, 2007

Daniel Nylander wrote: > I have translated a lot of programs and dlls to Swedish. > Would be great if someone could commit this to git. > > http://home.danielnylander.se/translations/wine/wine-sv.tar.gz > > > Please also add the following to the ./tools/wine.desktop > > Name[sv]=Windows-emulatorn Wine > > You need to submit a patch to wine-patches yourself for numerous reasons. You can find out information for doing so at http://www.winehq.org/site/git and http://www.winehq.org/site/sending_patches. -- Rob Shearman

1 0

Re: Swedish translations
by James Hawkins Oct. 15, 2007

Oct. 15, 2007

On 10/15/07, Daniel Nylander <po(a)danielnylander.se> wrote: > > I have translated a lot of programs and dlls to Swedish. > Would be great if someone could commit this to git. > It's your responsibility to get your own patches accepted. > http://home.danielnylander.se/translations/wine/wine-sv.tar.gz > Please read [1] to figure out how to send patches for inclusion in Wine. [1] http://winehq.org/site/sending_patches -- James Hawkins

1 0

Re: Fix casts in dlls/crypt32/decode.c
by Dmitry Timoshkov Oct. 15, 2007

Oct. 15, 2007

"Gerald Pfeifer" <gerald(a)pfeifer.com> wrote: > The patch below fixes the following diagnostics (rightfully) issued by > GCC 4.3 and later. If you prefer, we could use LOWORD everywhere, also > for the case labels, but my current patch seems to be more natural. > > decode.c:4365: warning: cast from pointer to integer of different size > --- decode.c 12 Oct 2007 10:24:15 -0000 1.87 > +++ decode.c 14 Oct 2007 14:17:58 -0000 > @@ -4360,103 +4360,103 @@ static CryptDecodeObjectExFunc CRYPT_Get > } > if (!HIWORD(lpszStructType)) > { > - switch (LOWORD(lpszStructType)) > + switch ((DWORD)lpszStructType) Your fix still won't work for a 64-bit target. You would need to use ULONG_PTR cast for that. 'if (!HIWORD(lpszStructType))' line also doesn't look right in the 64-bit case. dlls/crypt32/chain.c,CertVerifyCertificateChainPolicy() has the same problem. -- Dmitry.

1 0

Wiki looks much better
by Juan Lang Oct. 15, 2007

Oct. 15, 2007

Hi Dimi, I just checked out the Wiki after your most recent checkin. It does, indeed, look much better. Thanks for taking the time to look at it. And, after reviewing the change, I have to agree: it would be hard to predict that the browsers would do what they did based on the old CSS. :-/ --Juan

3 3

CSS changes to the Wiki
by Kai Blin Oct. 15, 2007

Oct. 15, 2007

Hi folks, I seem to remember that someone posted a patch to the wiki's CSS that will stop the lists from looking ugly. I can't find that patch, though. Can someone point me at it? Cheers, Kai -- Kai Blin WorldForge developer http://www.worldforge.org/ Wine developer http://wiki.winehq.org/KaiBlin Samba team member http://www.samba.org/samba/team/ -- Will code for cotton.

5 8

Re: wined3d: Allow WINED3DFMT_R32F render targets
by H. Verbeet Oct. 14, 2007

Oct. 14, 2007

On 14/10/2007, Chris Robinson <chris.kcat(a)gmail.com> wrote: > May need more work for this to be proper, but we currently allow R16F render > targets under the same circumstances. It's also enough to make UT3 (which > requires it) happy. > Unless the performance problems with R32F have been fixed, this will cause performance regressions in applications that actually try to use it. (C&C3 for example).

3 5

Re: tools/wineprefixcreate.in: create default w32x86 path under spool/drivers
by Detlef Riekenberg Oct. 14, 2007

Oct. 14, 2007

On Mi, 2007-10-10 at 13:10 +0200, Markus Gömmel wrote: > + "$CROOT/windows/system32/spool/drivers/w32x86" \ The above Path is created, when we add a Printer Driver in wine. Please update your wine. When you do not start with a clean ~/wine, you need to remove the wine printers below "HKLM\System\CurrentControlSet\Control\Print\Printers" and the wine drivers below "HKLM\System\CurrentControlSet\Control\Print\Environments\*". A "wine notepad" does the rest. When an application try to access this path, I expect, that they fail later, because we have no fake files for "w32x86/3/wineps.drv" and "win40/0/wineps16.drv" yet. -- By by ... Detlef

2 2

Translations
by Daniel Nylander Oct. 14, 2007

Oct. 14, 2007

I would like to contribute to the Wine project by submitting Swedish translations. I would really like to translate the graphical tools (such as winecfg) but as I can see it, it's not possible at the moment. Have you thought about implementing some sort of i18n/l10n framework? Regards, Daniel Nylander Stockholm, Sweden

2 4

Re: libs/wine/mmap.c: conditionally provide reserve_area()
by Vitaliy Margolen Oct. 14, 2007

Oct. 14, 2007

Gerald Pfeifer wrote: > Noticed on FreeBSD; patch tested on FreeBSD 5.4 as well as SUSE Linux > 10.0. > > Gerald > > ChangeLog: > Only provide reserve_area() when we actually use it. > > Index: mmap.c > =================================================================== > RCS file: /home/wine/wine/libs/wine/mmap.c,v > retrieving revision 1.20 > diff -u -3 -p -r1.20 mmap.c > --- mmap.c 26 Jan 2007 13:48:42 -0000 1.20 > +++ mmap.c 14 Oct 2007 11:09:28 -0000 In this and all of your patches, please diff from the top of the tree. The above line should say "+++libs/wine/mmap.c". And it would be even better if you can use GIT instead of cvs. Vitaliy.

2 1

Re: USER32: fix EM_SETLIMITTEXT with zeroed args (with tests)
by Detlef Riekenberg Oct. 14, 2007

Oct. 14, 2007

On Fr, 2007-10-12 at 19:42 -0500, a_villacis(a)palosanto.com wrote: You must use your real name to get a patch accepted. > + /* Maximum limits set by EM_SETLIMITTEXT vary by OS version */ > + osVersion = GetVersion(); > + if (osVersion < 0x80000000) { > + /* WinNT, Win2K, WinXP */ > + limitSingleLine = 2147483646UL; > + limitMultiLine = 4294967295UL; > + } else { > + /* Win9x, WinME */ > + limitSingleLine = 32766; > + limitMultiLine = 65535; > + } The tests should not depend on the OS-Version. Please handle both possible results in the ok() > + /* Test default limit for single-line control */ > + trace("EDIT: buffer limit for single-line\n"); > + hwEdit = create_editcontrol(ES_AUTOHSCROLL | ES_AUTOVSCROLL, 0); > + r = SendMessage(hwEdit, EM_GETLIMITTEXT, 0, 0); > + ok(r == 30000, "Incorrect default text limit, expected 30000 got > %u\n", r); > + SendMessage(hwEdit, EM_SETLIMITTEXT, 0, 0); > + r = SendMessage(hwEdit, EM_GETLIMITTEXT, 0, 0); /* Win9x+ME: 32766; WinNT: 2147483646UL */ ok( (r == 32766) || (r == 2147483646UL), "got limit %u (expected 32766 or 2147483646)\n", r); > + ok(r == limitSingleLine, "Incorrect default text limit, expected > %u got %u\n", limitSingleLine, r); Your looks like a copy&paste error ("default text limit" is wrong here). Change the multiline-test the same way. To allow an easy review, you can split your patch: one patch for the implementation and a second patch for the test. Thanks for helping wine -- By by ... Detlef

1 0

Re: [PATCH] wininet: NULL terminate url string passed to InternetCrackUrlW from InternetCrackUrlA.txt (try 3)
by Francois Gouget Oct. 14, 2007

Oct. 14, 2007

On Sat, 13 Oct 2007, Nigel Liang wrote: > Hi, > > InternectCrackUrlW calls strchrW which assumes a NULL-terminated string. Will > crash without NULL termination if you upload/download for long period of time. The MSDN has this to say about InternetCrackUrlW(): dwUrlLength Size of the lpszUrl string, in TCHARs, or zero if lpszUrl is an ASCIIZ string. So if InternetCrackUrlW() assumes that lpwszUrl is zero-terminated then it is buggy. So the fix belong in InternetCrackUrlW(), not in InternetCrackUrlA(). -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ It really galls me that most of the computer power in the world is wasted on screen savers. Chris Caldwell from the GIMPS project http://www.mersenne.org/prime.htm

1 0

Wine crashes while trying to use Lexmark Network Scanback utility
by King InuYasha Oct. 14, 2007

Oct. 14, 2007

Ehh, well, I was trying to use the Lexmark Scanback Utility to operate my Lexmark OptraImage 322 and it crashed after clicking the button to setup a new profile.... Installed original MFC42.DLL in the Scanback program folder... Crash log: [InuTaisho(a)PIIMMX-GTWE4200 ScanBack]$ wine scanwiz.exe wine: Call from 0x40ee68 to unimplemented function MFC42.DLL.6905, aborting wine: Unimplemented function MFC42.DLL.6905 called at address 0x40ee68 (thread 0038), starting debugger... Unhandled exception: unimplemented function MFC42.DLL.6905 called in 32-bit code (0x602e4d4c). Register dump: CS:0073 SS:007b DS:007b ES:007b FS:0033 GS:003b EIP:602e4d4c ESP:0034dbd4 EBP:0034dc38 EFLAGS:00000206( - 00 - -IP1) EAX:00001af9 EBX:6032fa78 ECX:0034e938 EDX:00030068 ESI:0034dbe0 EDI:0003006c Stack dump: 0x0034dbd4: 00000038 00030068 0100101b 80000100 0x0034dbe4: 00000001 00000000 0040ee68 00000002 0x0034dbf4: 0041cca6 00001af9 0034dc3c 6062e98e 0x0034dc04: 5f422d22 00030068 00000000 00000038 0x0034dc14: 00030068 00001036 00000000 00000020 0x0034dc24: 00000000 00000000 00000000 0034e8d8 Backtrace: =>1 0x602e4d4c in ntdll (+0x24d4c) (0x0034dc38) 2 0x0040ee68 in scanwiz (+0xee68) (0x0034e938) 3 0x00000001 (0x00419bc8) 4 0x0040fed0 in scanwiz (+0xfed0) (0x0041524c) 5 0x25ff0041 (0x837c25ff) 6 0x00000000 (0x00000000) 0x602e4d4c: subl $4,%esp Modules: Module Address Debug info Name (67 modules) PE 400000- 420000 Export scanwiz PE 10000000-10035000 Deferred scanwizg ELF 41d2d000-41d4a000 Deferred ld-linux.so.2 ELF 41d4c000-41d60000 Deferred libresolv.so.2 ELF 426fe000-42852000 Deferred libc.so.6 ELF 42854000-4287d000 Deferred libm.so.6 ELF 4287f000-42884000 Deferred libdl.so.2 ELF 4289b000-428b3000 Deferred libpthread.so.0 ELF 428b5000-428b8000 Deferred libxau.so.6 ELF 428ba000-428c0000 Deferred libxdmcp.so.6 ELF 428c2000-429c4000 Deferred libx11.so.6 ELF 429c6000-429d6000 Deferred libxext.so.6 ELF 429d8000-42a62000 Deferred libfreetype.so.6 ELF 42a64000-42a85000 Deferred libexpat.so.0 ELF 42aaf000-42ade000 Deferred libfontconfig.so.1 ELF 42ae0000-42ae9000 Deferred libxrender.so.1 ELF 42af5000-42aff000 Deferred libxcursor.so.1 ELF 42b01000-42b08000 Deferred libxrandr.so.2 ELF 42b0a000-42b0d000 Deferred libxinerama.so.1 ELF 42b0f000-42b14000 Deferred libxfixes.so.3 ELF 42bc5000-42bce000 Deferred libsm.so.6 ELF 42bd0000-42bea000 Deferred libice.so.6 ELF 4347a000-43485000 Deferred libdrm.so.2 ELF 43487000-434f7000 Deferred libgl.so.1 ELF 43fcb000-43fd0000 Deferred libxxf86vm.so.1 PE 5f400000-5f4ed000 Deferred mfc42 ELF 6001d000-60132000 Deferred libwine.so.1 ELF 602a3000-6034b000 Export ntdll<elf> \-PE 602c0000-6034b000 \ ntdll ELF 60374000-60380000 Deferred libnss_files.so.2 ELF 60380000-604b5000 Deferred kernel32<elf> \-PE 603a0000-604b5000 \ kernel32 ELF 604b5000-604cf000 Deferred version<elf> \-PE 604c0000-604cf000 \ version ELF 604cf000-604e3000 Deferred lz32<elf> \-PE 604e0000-604e3000 \ lz32 ELF 604e3000-6054f000 Deferred msvcrt<elf> \-PE 604f0000-6054f000 \ msvcrt ELF 6054f000-6059c000 Deferred advapi32<elf> \-PE 60560000-6059c000 \ advapi32 ELF 6059c000-606ea000 Deferred user32<elf> \-PE 605c0000-606ea000 \ user32 ELF 606ea000-607f6000 Deferred shell32<elf> \-PE 60700000-607f6000 \ shell32 ELF 607f6000-60853000 Deferred shlwapi<elf> \-PE 60800000-60853000 \ shlwapi ELF 60853000-6091a000 Deferred comctl32<elf> \-PE 60860000-6091a000 \ comctl32 ELF 6091a000-60978000 Deferred rpcrt4<elf> \-PE 60930000-60978000 \ rpcrt4 ELF 60978000-60997000 Deferred iphlpapi<elf> \-PE 60980000-60997000 \ iphlpapi ELF 609ab000-609d8000 Deferred ws2_32<elf> \-PE 609b0000-609d8000 \ ws2_32 ELF 60a28000-60abd000 Deferred winex11<elf> \-PE 60a40000-60abd000 \ winex11 ELF 60c8e000-60cac000 Deferred imm32<elf> \-PE 60ca0000-60cac000 \ imm32 ELF 60cbb000-60cef000 Deferred uxtheme<elf> \-PE 60cc0000-60cef000 \ uxtheme ELF 627d4000-62878000 Deferred gdi32<elf> \-PE 627f0000-62878000 \ gdi32 ELF 63167000-63213000 Deferred ole32<elf> \-PE 63180000-63213000 \ ole32 ELF 6875e000-68778000 Deferred wsock32<elf> \-PE 68760000-68778000 \ wsock32 ELF 7bf00000-7bf03000 Deferred <wine-loader> Threads: process tid prio (all id:s are in hex) 00000037 (D) C:\Program Files\Lexmark\ScanBack\scanwiz.exe 00000038 0 <== 00000013 00000015 0 00000014 0 0000000f 00000016 0 00000010 0 0000000a 0000000c 0 0000000b 0 wine: Call from 0x40ee68 to unimplemented function MFC42.DLL.6905, aborting wine: Call from 0x40ee68 to unimplemented function MFC42.DLL.6905, aborting

3 3