June 2006 - Wine-devel - list.winehq.org

Re: msiexec: msiexec also supports options using a hyphen
by Alexandre Julliard 29 Jun '06

29 Jun '06

"James Hawkins" <truiken(a)gmail.com> writes: > Changelog: > * msiexec also supports options using a hyphen. Surely there's a better way to do this than duplicating all the string compares... -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: DDraw: Return DDERR_NOPALETTEATTCHED if there is no palette
by Stefan Dösinger 29 Jun '06

29 Jun '06

Am Donnerstag 29 Juni 2006 11:40 schrieb Stefan Dösinger: > According to the msdn this should happen, and native d3drm.dll depends on > this > > ChangeLog: > Stefan Dösinger: Return DDERR_NOPALETTEATTCHED if there is no palette Do not apply this patch, it is not completely correct. It breaks palettized textures in the dx7 sdk. I'll write a test case for this

1 0

winegcc atof() bug?
by Benjamin Fabricius 29 Jun '06

29 Jun '06

Hi, Im experiencing weird things when using atof() in certain places of my software and compiling with winegcc (wine20050725). It works fine when calling it out of the main function but as soon as i migrated some routines to another c file things became really strange. atoi works fine but as soon as i use atof with any "x.x" string i keep on getting some negative constant value as it seems. thought this might be HUGE_VAL of strtod but atof doesnt return any errors.. Anybody experience similar problems. unfortunately i am dependent on this wine version.. so changing wine versions won't help since current builds interfere with other parts of the software running in wine. Heres a really simple example code of what im trying to do: --- snip --- void print_set_parameters(char **setparams, int startofparams, int endofparams) { int param_num; double param_value; /* this works fine setparams has got the right * string values */ printf("test %d\n", startofparams); printf("test %s\n", setparams[3]); printf("test %s\n", setparams[4]); for(i = startofparams; i < endofparams; i++) { // works fine! param_num = atoi(setparams[i]); printf("string is %s, param number %d\n", setparams[i],param_num); i++; // heres where the trouble starts param_value = atof(setparams[i]); // tried this as an alternative same problem // param_value = strtod(".1", 0); // no errors reported // perror(__FUNCTION__); printf("param value %f\n", param_value); } } --- snip --- Any help would be great!! Cheers, ..Ben -- ------------------- Benjamin Fabricius German Student Lawo AG

1 0

msdn calling everything .net now?
by Dan Kegel 28 Jun '06

28 Jun '06

Lately, when I search for win32 API functions like CoGetObject, the top few hits I get at microsoft.com are the .net ones. I guess it's not a problem, but it feels weird.

2 1

automatic winetest builds stopped
by Ivan Leo Puoti 28 Jun '06

28 Jun '06

Does someone know why the tests at test.winehq.com/data aren't being built any more? Ivan.

3 3

Battlefront blending problem is fixed.
by Jesse Allen 28 Jun '06

28 Jun '06

Hi Henri, You latest round of wined3d patches fixed the blending problem in SW Battlefront including blocky text. Now I can properly aim at targets. :) However, some screen items seem to be a little more transparent than they should be. It doesn't fix the blocky text in BF1942. BF1942 is a d3d8 game and Battlefront is a d3d9 game. Thanks, Jesse

2 1

Freetype - Compile error when making a regression test
by Augusto Arcoverde da Rocha 28 Jun '06

28 Jun '06

Hi all, I'm here again with a little question about my dispersed efforts over Wine development. Now I'm trying to make some regression tests to identify when Lotus Notes started to show blank squares rather than icons for attachments. And by this way I get some errors compiling (see bellow) Wine versions 0.9.1 and 0.9.4, but with the version 0.9.9 this errors don't occurs. My guess is some version problem with freetype, but I didn't find any hints about this on the list history, though I have found some posts asking about the same problem. I tried to compile Wine on a Debian r3.1 with gcc versions 3.3 and 4.1, and with freetype 2.2.1, but didn't get any success. Any help is appreciated. Regards, --- 8< --- freetype.c:106: error: 'FT_Get_Module' undeclared here (not in a function) freetype.c:106: warning: type defaults to 'int' in declaration of 'pFT_Get_Module' freetype.c: In function 'WineEngGetFontData': freetype.c:3698: error: 'TT_Face' undeclared (first use in this function) freetype.c:3698: error: (Each undeclared identifier is reported only once freetype.c:3698: error: for each function it appears in.) freetype.c:3698: error: syntax error before 'tt_face' freetype.c:3699: error: 'SFNT_Interface' undeclared (first use in this function) freetype.c:3699: error: 'sfnt' undeclared (first use in this function) freetype.c:3703: error: syntax error before ')' token freetype.c:3703: error: 'tt_face' undeclared (first use in this function) freetype.c:3708: error: syntax error before ')' token freetype.c: In function 'is_hinting_enabled': freetype.c:3776: error: called object 'pFT_Get_Module' is not a function freetype.c:3777: warning: implicit declaration of function 'FT_DRIVER_HAS_HINTER' --- 8< ---

1 1

Wide Strings Query
by Andrew Talbot 28 Jun '06

28 Jun '06

I have written a patch to fix some of the write-strings warnings in regedit's listview.c file, in its OnGetDispInfo() fuction, but I need to know that my understanding is right, before I submit it. Am I correct in thinking that I can replace the _T() macros with TCHAR * variables, as exemplified with reg_szT, below? @@ -252,6 +252,19 @@ static void OnGetDispInfo(NMLVDISPINFO* plvdi) { static TCHAR buffer[200]; + static TCHAR reg_szT[] = {'R','E','G','_','S','Z',0}, plvdi->item.pszText = NULL; plvdi->item.cchTextMax = 0; @@ -263,43 +276,43 @@ case 1: switch (((LINE_INFO*)plvdi->item.lParam)->dwValType) { case REG_SZ: - plvdi->item.pszText = _T("REG_SZ"); + plvdi->item.pszText = reg_szT; break; Thanks, -- Andy.

1 0

re: msi: Download the MSI package if it is a remote URL
by Dan Kegel 28 Jun '06

28 Jun '06

Your patch ends in an empty sig, i.e. two dashes, which makes patch complain "patch ends in middle of a line" after finishing. I tried your patch to see if it fixed a similar problem in Ability Office (see http://ability.com/v4/download.php ) but sadly that still fails with fixme:urlmon:URLDownloadToCacheFileW ((nil) L"http://www.abilitydownload.com/download4/ability4_15/Ability.MSI" 0x7fd49f40 520 0 0x7fb9fae8) -- Wine for Windows ISVs: http://kegel.com/wine/isv

2 2

Re: WineD3D: Add one extra line to the dib section
by Saulius Krasuckas 28 Jun '06

28 Jun '06

* On Fri, 23 Jun 2006, Stefan [iso-8859-1] D�singer wrote: > Some DirectDraw apps access the surface memory via DWORDS and do not > take the necessary care at the end of the surface. The old DDraw code > allocated 4 extra bytes for the surface and used the unofficial > DIB_CreateDibSection function to create the DIB section. The normal > CreateDIBSection function doesn't use the requested size. To increase > the surface size to catch bad accesses one extra line is added to the > dib section. It's nice to hear you are working on this... > This patch addresses the regression in Diablo 1 and some odd crashes in > the DX7 sdk demos ...but the patch doesn't fix the Diablo 1 game on my box. I even made distclean here to no avail.

1 0