October 2003 - Wine-devel - list.winehq.org

Errors in user tests
by Sylvain Petreolle 25 Oct '03

25 Oct '03

Running the user tests, I get different results if run with 'make test' and manually in a terminal. If run with wine in a terminal: listbox gets 16 failures (no failure with make test) msg gets 24 failures (only 1 failure with make test.) If I launch the same runtest command that gets started from Makefile, I get no error. I dont understand what is happening here. With make: [syl(a)wine tests]$ make test ../../../tools/runtest -q -P wine -M user32.dll -T ../../.. -p user32_test.exe.so listbox.c && touch listbox.ok listbox.c: 48 tests executed, 16 marked as todo, 0 failures. Manual runtest, same command : [syl(a)wine tests]$ ../../../tools/runtest -q -P wine -M user32.dll -T ../../.. -p user32_test.exe.so listbox.c && touch listbox.ok listbox.c: 48 tests executed, 16 marked as todo, 0 failures. Manual runtest, flag -v : [syl(a)wine tests]$ ../../../tools/runtest -v -P wine -M user32.dll -T ../../.. -p user32_test.exe.so listbox.c && touch listbox.ok listbox.c:165: Testing single selection... listbox.c:167: ... with NOSEL listbox.c:169: Testing multiple selection... listbox.c:171: ... with NOSEL listbox.c: 48 tests executed, 16 marked as todo, 0 failures. Manual wine: [syl(a)wine tests]$ wine user32_test.exe.so listbox listbox.c:165: Testing single selection... listbox.c:129: Test failed: style 0, step click, field anchor: expected 1, got -1 listbox.c:134: Test failed: style 0, step step, field anchor: expected 2, got -1 listbox.c:167: ... with NOSEL listbox.c:129: Test failed: style 0x4000, step click, field selected: expected 1, got -1 listbox.c:129: Test failed: style 0x4000, step click, field anchor: expected 1, got -1 listbox.c:134: Test failed: style 0x4000, step step, field selected: expected 2, got -1 listbox.c:134: Test failed: style 0x4000, step step, field anchor: expected 2, got -1 listbox.c:134: Test failed: style 0x4000, step step, field caret: expected 2, got 1 listbox.c:169: Testing multiple selection... listbox.c:129: Test failed: style 0x8, step click, field anchor: expected 1, got -1 listbox.c:134: Test failed: style 0x8, step step, field anchor: expected 1, got -1 listbox.c:134: Test failed: style 0x8, step step, field selcount: expected 1, got 2 listbox.c:171: ... with NOSEL listbox.c:123: Test failed: style 0x4008, step init, field selected: expected -1, got 0 listbox.c:123: Test failed: style 0x4008, step init, field selcount: expected -1, got 0 listbox.c:129: Test failed: style 0x4008, step click, field anchor: expected 1, got -1 listbox.c:129: Test failed: style 0x4008, step click, field selcount: expected -1, got 0 listbox.c:134: Test failed: style 0x4008, step step, field anchor: expected 2, got -1 listbox.c:134: Test failed: style 0x4008, step step, field selcount: expected -1, got 0 listbox: 48 tests executed, 0 marked as todo, 16 failures. ===== Sylvain Petreolle (spetreolle_at_users_dot_sourceforge_dot_net) ICQ #170597259 Say NO to software patents Dites NON aux brevets logiciels "What if tomorrow the War could be over ?" Morpheus, in "Reloaded". ___________________________________________________________ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Mail : http://fr.mail.yahoo.com

1 0

Re: kernel32: Fix lstrcmp(i)A/W to correct return values and document this behaviour
by Dmitry Timoshkov 25 Oct '03

25 Oct '03

"Rolf Kalbermatter" <rolf.kalbermatter(a)citeng.com> wrote: > int WINAPI lstrcmpA(LPCSTR str1, LPCSTR str2) > { > - int ret = CompareStringA(GetThreadLocale(), 0, str1, -1, str2, -1); > - if (ret) ret -= 2; > - return ret; > + int ret = CompareStringA(GetThreadLocale(), 0, str1, -1, str2, -1); > + if (!ret) > + { > + ret = CompareStringA(GetSystemDefaultLCID(), 0, str1, -1, str2, -1); > + if (!ret) > + { > + if (!str1) > + return (str2 ? -1 : 0); > + else if (!str2) > + return 1; > + return strcasecmp(str1, str2); > + } > + } > + return ret - 2; > } Windows doesn't do anything of the above. A patch for CompareString I sent recently to wine-devel has lots of tests and fixes for all the differences we have found so far. It just needs a clean approach for unicode.org/windows collation table differences merge to be included in Wine. -- Dmitry.

1 0

Summary of translations
by Vincent Béron 24 Oct '03

24 Oct '03

This is the current summary of Wine translations. It lists languages for which at least one resource exists. It can be found here: http://mecano.gme.usherb.ca/~vberon/wine/translation All comments welcome. Vincent

3 4

Re: Translation support (wrc)
by Alexandre Julliard 24 Oct '03

24 Oct '03

Vincent Béron <vberon(a)mecano.gme.usherb.ca> writes: > Changelog: > Add support for comparing resources to wrc for translations. That's a clear case of preprocessor abuse... there has to be a better way to do that than with a 100-line macro. -- Alexandre Julliard julliard(a)winehq.com

2 1

Re: ClipCursor values init on start (resize missing update values)
by Carlos Lozano 24 Oct '03

24 Oct '03

El vie, 24 de oct de 2003, a las 23:08, Carlos Lozano escribio: > > - Windows init the clipcursor values with width and height of > the screen of the primary display monitor. > (Theme Hospital mouse is now fixed) Hmmm, it is even more complex what i had thought. When the resolution mode is changed, the values for clipcursor are reset to 0,0,X,Y. Now the question is: a) Should we modify the function SYSMETRICS_Set, and when the values SM_CXSCREEN or SM_CYSCREEN will get changed, call to the Clipcursor function b) or add a Clipcursor call after of every change resolution function?? Regards, Carlos. -- ___ _ \ | / Consulting | . |._ _ _| | ___ ___ ___ http://www.andago.com | || ' |/ . |<_> |/ . |/ . \__ GNU/Linux |_|_||_|_|\___|<___|\_. |\___/ _ \ __|\ \ / Carlos A. Lozano <___'/ | \ -_) __/\__ \ > < -_) [ carlos.lozano(a)andago.com ]\___|_| ____/ _/\_\___| [ calb(a)epsxe.com ] http://www.ePSXe.com

1 0

Re: FoldStringA (using W version)
by Jon Griffiths 24 Oct '03

24 Oct '03

Hi, > + goto FoldStringA_exit; > ^^^^^^^^^^^^^^^^^^^^^^ > No allocation has been done yet, just "return 0;" would be > enough! Indeed, although the above is functionally identical. Feel free to submit a patch if you feel strongly about it... > Interesting idea, but I wonder how you can distinguish between > equal 0 and FALSE without some real magic! > Same applies to lstrcmpi No magic required. Use GetLastError(), just like it says. Cheers, Jon ===== "Don't wait for the seas to part, or messiahs to come; Don't you sit around and waste this chance..." - Live jon_p_griffiths(a)yahoo.com __________________________________ Do you Yahoo!? The New Yahoo! Shopping - with improved product search http://shopping.yahoo.com

3 4

WineHQ: PATCH to fix IPersistFile::Load() and associated functions
by Subhobroto Sinha 24 Oct '03

24 Oct '03

Hi. I finally produced the "official patch" after verifying that everything is OK. Thanks to Dimi for putting up a temporary diff for me meanwhile ;) This patch has been inlined as well as attached as a ZIP file, so suit yourself. All reports regaring this patch is to be sent to <subhobrotosinha at yahoo.com> with a proper subject prefixed by "IPersistFile patch" in the subject header. For the uninitiated, here's what has to be done: Please make sure that you have the 20031016 source (latest tarball) (1)Copy this .diff to dlls/shell32/ (2)Run 'patch' on this diff file (3)It should patch "shelllink.c" (4)Type 'make' to rebuild 'shell32.dll.so' (5)'su' out and overwrite/symlink the current 'shell32.dll.so' over the installed one Now IPersistFile::Load() should work as intended. Try out McCormack's 'winemenubuilder' or my WineLib 'linkresolve' (NOT my C++ version) If the diff is not proper, etc, please alert me at the mentioned address BTW: If I should have submitted at 'wine-patches' rather at 'wine-devel', please let me know :=) I really want this BUG fixed you know ;-) Regards Subhobroto Sinha --- shelllink.c.original 2003-10-22 17:56:31.000000000 +0530 +++ shelllink.c 2003-10-24 13:09:59.000000000 +0530 @@ -18,7 +18,7 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA * * NOTES - * Nearly complete informations about the binary formats + * Nearly complete informations about the binary formats * of .lnk files avaiable at http://www.wotsit.org * */ @@ -401,9 +401,49 @@ return S_OK; } +static HRESULT wineStream_LoadString( IStream* stm, LPWSTR *pstr ) +{ + DWORD count=0; + USHORT len; + LPWSTR str=NULL; + HRESULT r; + /* + From Win98 upwards unicode==TRUE.(Atleast for shortcuts..) + However, the value of SCF_UNICODE does NOT seem to be 0x1000. + Until then, we assume it true. + Anybody differing mail to <subhobrotosinha at yahoo.com>. + */ + TRACE("%p\n", stm); + + r = IStream_Read(stm, &len, sizeof(len), &count); + if(FAILED(r)||(count != sizeof(len))) return E_FAIL; + + len *= sizeof(WCHAR); + + TRACE("reading %d\n", len); + str = HeapAlloc(GetProcessHeap(), 0, len+sizeof(WCHAR)); + if(!str) return E_OUTOFMEMORY; + + count = 0; + r = IStream_Read(stm, str, len, &count); + if( FAILED(r)||( count!=len) ) + { + HeapFree( GetProcessHeap(), 0, str ); + return E_FAIL; + } + + TRACE("read %s\n", debugstr_w(str)); + str[count/2]=0; + *pstr = str; + return S_OK; +} static HRESULT Stream_LoadString( IStream* stm, BOOL unicode, LPWSTR *pstr ) { +/*This is the original function short circuited to the above till we find out correct 'SCF_UNICODE'. +Now it's assumed that the data's unicode. (This assumption failes iff the OS was <=Win95*/ +return (wineStream_LoadString(stm,pstr)); +#if 0 DWORD count; USHORT len; LPVOID temp; @@ -453,6 +493,7 @@ *pstr = str; return S_OK; +#endif } static HRESULT Stream_LoadLocation( IStream* stm ) @@ -533,6 +574,27 @@ if( FAILED( r ) ) return r; } + /*Fill in the target path. This marked portion will be rewritten once GetPathFromIDList() is completed*/ + //LPITEMIDLIST pidltemp=_dbg_ILGetNext(This->pPidl); + LPITEMIDLIST pidltemp=ILGetNext(This->pPidl); + char szPasha[128][MAX_PATH]={{0},{0}},szTemp[MAX_PATH]={0}; + int i=0,j=1; + do + { + _ILSimpleGetText(pidltemp,szPasha[i++],MAX_PATH); + pidltemp=ILGetNext(pidltemp); + }while(pidltemp->mkid.cb); + strcpy(szTemp,szPasha[0]); + for(;j<i;j++) + { + strcat(szTemp,szPasha[j]); + if(j!=(i-1))strcat(szTemp,"\\"); + } + strcat(szTemp,szPasha[j]); + This->sPath=HeapAlloc( GetProcessHeap(), 0,(strlen(szTemp)+1)*sizeof(WCHAR)); + MultiByteToWideChar(CP_ACP,0,szTemp,(strlen(szTemp)+1),This->sPath,(strlen(szTemp)+1)); + TRACE("%s\n",szTemp); + /*End marking*/ This->wHotKey = hdr.wHotKey; This->iIcoNdx = hdr.nIcon; FileTimeToSystemTime (&hdr.Time1, &This->time1); __________________________________ Do you Yahoo!? The New Yahoo! Shopping - with improved product search http://shopping.yahoo.com

2 1

IPersistFile patch
by Ivan Leo Murray-Smith 24 Oct '03

24 Oct '03

>BTW: If I should have submitted at 'wine-patches' >rather at 'wine-devel', please let me know :=) >I really want this BUG fixed you know ;-) If you send your patch to wine-patches and it's OK for Alexandre, your patch will get into the CVS, and into the next release of wine.

1 0

Winesetuptk
by Ivan Leo Murray-Smith 24 Oct '03

24 Oct '03

Rc1 had a bug that caused winesetuptk to crash when it looked for a real windows drive and didn't find one. Also, with rc2 winver is set to 98, as a lot of apps refuse to install on 95. source http://www003.portalis.it/115/download/prerelease/winesetuptk0.7rc2.tar.bz2 binary http://www003.portalis.it/115/download/prerelease/winesetuptk0.7rc2.bz2

1 0

Re: Winesetuptk 0.7 rc1
by Ivan Leo Murray-Smith 24 Oct '03

24 Oct '03

>My ignorance may be obvious here, but is there any reason it is 5 meg? >(I don't remember the old one being that big.) You can get a 700kb compressed binary at www003.portalis.it/115/download/prerelease/winesetuptk0.7rc1.bz2 freedback is welcome.

1 0