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

Are *any* games as popular as World of Warcraft?

by Dan Kegel

I spent some time noodling around looking at what games are really popular (I wouldn't know firsthand, I don't play much). There's some data at http://www.amazon.com/gp/bestsellers/videogames (click on PC Games). I tried to cross-check titles with Google Trends to see which ones seem enduringly popular, but I haven't found any that beat World of Warcraft yet: http://www.google.com/trends?q=world+of+warcraft%2C+guild+wars%2C+tomb+raid… Is Google search popularity a bad measure, or does it reflect the public's mood well? - Dan

Oct. 1, 2007

Re: Janitorial : Rename 16-Bit files

by Alexandre Julliard

Rémi Assailly <remi.assailly(a)free.fr> writes: > I started 16 Bit separation. > This first patch renames : > - files containing 16 Bit functions only > - some 16 Bit functions. Please don't do that, there's no reason to rename things that are already separated properly. -- Alexandre Julliard julliard(a)winehq.org

Oct. 1, 2007

Re: gdi32: Prefer a Microsoft cmap table over other platform ones

by Alexandre Julliard

Dmitry Timoshkov <dmitry(a)codeweavers.com> writes: > the bug 9840 has the ttf font attached that has incomplete default unicode > charmap which Freetype selects. Assuming that Microsoft and Macintosh cmaps > are more widely used/tested, let's use a priority to choose from available > character maps. Freetype developers think that it's a better choice to fix > this kind of problem. > > Changelog: > gdi32: Prefer a Microsoft cmap table over other platform ones. This fails make test here: ../../../tools/runtest -q -P wine -M gdi32.dll -T ../../.. -p gdi32_test.exe.so font.c && touch font.ok fixme:font:WineEngGetKerningPairs encoding 1634889070 not supported font.c:761: Test failed: got matches 0, expected 26 fixme:font:WineEngGetKerningPairs encoding 1634889070 not supported font.c:761: Test failed: got matches 0, expected 26 fixme:font:WineEngGetKerningPairs encoding 1634889070 not supported font.c:761: Test failed: got matches 0, expected 26 fixme:font:get_font_unicode_ranges encoding 1634889070 not supported fixme:font:get_font_unicode_ranges encoding 1634889070 not supported make[2]: *** [font.ok] Error 3 -- Alexandre Julliard julliard(a)winehq.org

Oct. 1, 2007

[PATCH] Fix Ugly borders around undecorated windows in Gnome

by Kirill K. Smirnov

>This is my first attempt at submitting a patch to WINE, so please excuse >me if I have broken any conventions or am going through the wrong >channel. As it currently stands, wine is setting the 'MWM_DECOR_BORDER' >hint on managed but undecorated windows, such as iTunes, or Winamp. >This does not cause problems in KDE, but in Gnome it puts a gray border >around such windows. For me this causes problem in KDE too. Not just border but an ugly caption too :-( See http://www.winehq.org/pipermail/wine-devel/2007-September/059366.html >Not setting this hint does not seem to cause >problems in KDE and fixes the problem in Gnome. A patch to do just that >is attached. Comments? >--- dlls/winex11.drv/window_orig.c 2007-09-30 18:14:56.281259713 -0400 >+++ dlls/winex11.drv/window.c 2007-09-30 18:18:44.724576520 -0400 >@@ -591,11 +591,6 @@ > if (style & WS_MINIMIZEBOX) mwm_hints.decorations |= MWM_DECOR_MINIMIZE; > if (style & WS_MAXIMIZEBOX) mwm_hints.decorations |= MWM_DECOR_MAXIMIZE; > } >- if (ex_style & WS_EX_DLGMODALFRAME) mwm_hints.decorations |= >MWM_DECOR_BORDER; >- else if (style & WS_THICKFRAME) mwm_hints.decorations |= >MWM_DECOR_BORDER | MWM_DECOR_RESIZEH; >- else if ((style & (WS_DLGFRAME|WS_BORDER)) == WS_DLGFRAME) mwm_hints.decorations |= MWM_DECOR_BORDER; >- else if (style & WS_BORDER) mwm_hints.decorations |= MWM_DECOR_BORDER; >- else if (!(style & (WS_CHILD|WS_POPUP))) mwm_hints.decorations |= MWM_DECOR_BORDER; > } > XChangeProperty( display, data->whole_window, x11drv_atom(_MOTIF_WM_HINTS), You overdid here a little: WS_THICKFRAME must have thick border. As I think, we should not mix WS_BORDER and MWM_DECOR_BORDER. They have DIFFERENT semantics. WS_BORDER is 1px border, but MWM_DECOR_BORDER is a thick border. What for me, I'd deleted just these lines: >- else if (style & WS_BORDER) mwm_hints.decorations |= MWM_DECOR_BORDER; >- else if (!(style & (WS_CHILD|WS_POPUP))) mwm_hints.decorations |= MWM_DECOR_BORDER; and draw thin 1px border 'manually' (I mean, in user32, do not left it to Window Manager) -- Kirill

Oct. 1, 2007

Re: msinkaut.idl: Add stub IInkCollector interface (resend)

by Alexandre Julliard

"John Klehm" <xixsimplicityxix(a)gmail.com> writes: > typedef float single; > + > + [ > + odl, > + uuid(F0F060B5-8B1F-4A7C-89EC-880692588A4F), > + dual, > + oleautomation > + ] > + interface IInkCollector : IDispatch { > + }; You have to list all the functions, a partial interface definition can't work right. -- Alexandre Julliard julliard(a)winehq.org

Oct. 1, 2007

Re: ole32: Fix some memory leaks

by Alexandre Julliard

Andrew Talbot <andrew.talbot(a)talbotville.com> writes: > diff --git a/dlls/ole32/compositemoniker.c b/dlls/ole32/compositemoniker.c > index c49e9b8..88f141e 100644 > --- a/dlls/ole32/compositemoniker.c > +++ b/dlls/ole32/compositemoniker.c > @@ -1817,8 +1817,10 @@ CompositeMonikerImpl_Construct(IMoniker* > This->tabSize+=BLOCK_TAB_SIZE; > This->tabMoniker=HeapReAlloc(GetProcessHeap(),0,This->tabMoniker,This->tabSize*sizeof(IMoniker)); > > - if (This->tabMoniker==NULL) > + if (This->tabMoniker==NULL) { > + HeapFree(GetProcessHeap(), 0, This); > return E_OUTOFMEMORY; This->tabMoniker is leaked too in that case, it would be good to fix it while you are at it. -- Alexandre Julliard julliard(a)winehq.org

Oct. 1, 2007

Re: [PATCH 1/2] winex11.drv: Use ClipCursor to grab/ungrab the pointer.

by Alexandre Julliard

Vitaliy Margolen <wine-patches(a)kievinfo.com> writes: > +/*********************************************************************** > + * do_grab_pointer > + * > + * Helper to call grab_pointer from any thread. > + * > + * x11_lock have to be held when calling this function. It will be > + * released upon return. Don't do that, lock/unlock should be done at the same level, it makes it really hard to verify otherwise. -- Alexandre Julliard julliard(a)winehq.org

Oct. 1, 2007

Re: user32: Add tests for DefWindowProc processing the undocumented 0x3Bmessage.

by Dmitry Timoshkov

"Robert Shearman" <rob(a)codeweavers.com> wrote: > --- > dlls/user32/tests/msg.c | 71 > +++++++++++++++++++++++++++++++++++++++++++++++ > 1 files changed, 71 insertions(+), 0 deletions(-) > > Found from observations with Spy++ whilst testing log-off behaviour in > XP. I don't see any point in using this message type over the documented > messages though. http://wiki.winprog.org/wiki/Windows_messages names 0x003b as WM_CLIENTSHUTDOWN. That page also lists some other not documented messages. -- Dmitry.

Oct. 1, 2007

Re: Weekly Newsletter

by Scott Ritchie

EA Durbin wrote: > Who maintains the weekly newsletter? I noticed it hasn't been updated since May. > No one, which is why it hasn't been updated. Would you like to fill the job? Thanks, Scott Ritchie

Oct. 1, 2007