April 2008 - Wine-Devel - list.winehq.org

Re: dinput: Limit number of supported axis and buttons
by Christoph Frick April 24, 2008

April 24, 2008

On Thu, Apr 24, 2008 at 08:06:07AM -0600, Vitaliy Margolen wrote: > + FIXME("Can't support %d axis. Clamping down to 16\n", newDevice->axes); FIXME("Can't support %d axes. Clamping down to 16\n", newDevice->axes); -- cu

2 1

Re: mshtml #5: Added more IDispatchEx tests.
by Alexandre Julliard April 24, 2008

April 24, 2008

Jacek Caban <jacek(a)codeweavers.com> writes: > dlls/mshtml/tests/dom.c | 56 > +++++++++++++++++++++++++++++++---------------- > 1 files changed, 37 insertions(+), 19 deletions(-) It doesn't work here: ../../../tools/runtest -q -P wine -M mshtml.dll -T ../../.. -p mshtml_test.exe.so dom.c && touch dom.ok dom.c:1381: Test failed: GetTypeInfo failed: 8002802b dom.c:1394: Test failed: GetTypeInfo failed: 8002802b make[2]: *** [dom.ok] Error 2 -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: [1/7] D3D9: Make the offscreen format test a bit more tolerant
by H. Verbeet April 24, 2008

April 24, 2008

2008/4/24 Stefan Dösinger <stefan(a)codeweavers.com>: > FWIW, for my texop test I wrote a function to do this in a generic way. I'll probably submit that in the next few days.

1 0

Re: [3/4] user32: Set the vkey value to VK_PROCESSKEY when IME process key.
by Aric Stewart April 24, 2008

April 24, 2008

Thank you for these investigations This is really nice. I support those patches. I just wanted to also communicate a bit of additional information from my investigations just in case someone asks. I placed hooks on both WH_KEYBOARD_LL and WH_KEYBOARD. Even when the IME is active both of those hooks showed the VKEY of the actual keystroke. This showed me that all of this processing happened after process_keyboard_message. great work and thanks, -aric ByeongSik Jeon wrote: > When the MS Windows set the vkey value(MSG.wParam) to VK_PROCESSKEY? > > I have tested following code: > ---------------------------- > while ( GetMessage(&msg, 0, 0, 0) ) { > if ( msg.message == WM_KEYDOWN ) > printf("OLD: %x %x %x\n", msg.message, msg.wParam, msg.lParam); > TranslateMessage( &msg ); > if ( msg.message == WM_KEYDOWN ) > printf("NEW: %x %x %x\n", msg.message, msg.wParam, msg.lParam); > DispatchMessage( &msg ); > } > ----------------------------- > The result is: > ----------------------------- > OLD: 100 e5 130001 > NEW: 100 e5 130001 > OLD: 100 e5 250001 > NEW: 100 e5 250001 > OLD: 100 e5 130001 > NEW: 100 e5 130001 > ----------------------------- > e5 is VK_PROCESSKEY. > Before GetMessage, PeekMessage return, msg.wPram ahs seted to VK_PROCESSKEY. > > --- > dlls/user32/message.c | 9 +++++++-- > 1 files changed, 7 insertions(+), 2 deletions(-) > > > ------------------------------------------------------------------------ > >

1 0

Re: imm32: If ImmNotifyIME fails, ImmSetOpenStatus should also fail.
by ByeongSik Jeon April 24, 2008

April 24, 2008

Kusanagi Kouichi wrote: > if (!fOpen != !data->IMC.fOpen) > { > + if (!ImmNotifyIME( hIMC, NI_CONTEXTUPDATED, 0, IMC_SETOPENSTATUS)) > + return FALSE; > data->IMC.fOpen = fOpen; > - ImmNotifyIME( hIMC, NI_CONTEXTUPDATED, 0, IMC_SETOPENSTATUS); > ImmInternalSendIMENotify(data, IMN_SETSENTENCEMODE, 0); Hi, MS Windows's native imm32.dll return TRUE although NotifyIME return FALSE. I have tested it with the DDK's Japanese FakeIME.

1 0

xfs getting case-insensitive filesystem support...
by Dan Kegel April 24, 2008

April 24, 2008

Hrm. This is interesting: http://lwn.net/Articles/278961/ I wonder if Wine could run faster if it could delegate the case-insensitvity to the filesystem...

4 3

1.0 idea - ntdll - add a messagebox for missing dlls
by Steven Edwards April 24, 2008

April 24, 2008

Hi, My hope that was for Wine 1.0 we could add error handling for missing dlls and functions that made sense to the user rather than just silent failed or spewed messages to the console. If you have an application that depends on a dll such as OUTLOOK.EXE depending on OUTLIB.DLL you will get a message like this: err:module:import_dll Library OUTLLIB.dll (which is needed by L"Z:\\home\\sedwards\\.wine\\drive_c\\Program Files\\Microsoft Office\\OFFICE11\\OUTLOOK.EXE") not found err:module:LdrInitializeThunk Main exe initialization for L"Z:\\home\\sedwards\\.wine\\drive_c\\Program Files\\Microsoft Office\\OFFICE11\\OUTLOOK.EXE" failed, status c0000135 So instead how about throwing a nice message box? I started working on this but don't quite know how to properly pass the information to the MessageBoxW or maybe fill in the MessageBoxIndirect structure to display the error in a human readable format. I just don't really know. Of course once its done this would need to be broken out in to a helper function, its just a proof of concept.... Anyone want to help? -- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

1 0

re: no test structure in place for wine programs
by Dan Kegel April 23, 2008

April 23, 2008

Darren wrote: > Should I try to take the dll test structure (mainly just the > dlls/Maketest.rules.in file) and port this into the programs area? > > or is that overboard for that area? It's something we need eventually, but for now it would probably be enough to just attach a standalone test that invoked msiexec.exe and printed out PASS or FAIL to http://bugs.winehq.org/show_bug.cgi?id=9628 It's not worth the effort to create the generic app test framework inside the wine tree at the moment.

2 1

Re: Fix "warning: cast from pointer to integer of different size"
by Dmitry Timoshkov April 23, 2008

April 23, 2008

"Erik de Castro Lopo" <mle+win(a)mega-nerd.com> wrote: > /* get pointer to object containing list element */ > #define LIST_ENTRY(elem, type, field) \ > - ((type *)((char *)(elem) - (unsigned int)(&((type *)0)->field))) > + ((type *)((char *)(elem) - (unsigned long)(&((type *)0)->field))) It should be replaced with FIELD_OFFSET instead. -- Dmitry.

4 6

Re: Add implemantation of property "Name" (IHTMLWindow2)
by Jacek Caban April 23, 2008

April 23, 2008

Ivan Sinitsin wrote: > This patch add implementation of HTMLWindow2_put_name, HTMLWindow2_get_name in > module htmlwindow.c (mshtml.dll) > > Changelog: > Add implementation of HTMLWindow2_put_name, HTMLWindow2_get_name > > > > @@ -289,15 +289,43 @@ static HRESULT WINAPI HTMLWindow2_get_na > static HRESULT WINAPI HTMLWindow2_put_name(IHTMLWindow2 *iface, BSTR v) > { > HTMLWindow *This = HTMLWINDOW2_THIS(iface); > - FIXME("(%p)->(%s)\n", This, debugstr_w(v)); > - return E_NOTIMPL; > + nsAString aName; > + nsresult nsres; > + > + TRACE("(%p)->(%s)\n", This, debugstr_w(v)); > + > + nsAString_Init(&aName, v); > + nsres = nsIDOMWindow_SetName(This->nswindow, &aName); > + if (NS_FAILED(nsres)) { > + TRACE("Failed nsIDOMWindow_SetName %08x \n",nsres); > + return nsres; You shouldn't mix nsresult with HRESULT. Also you leak aName here. SetName should never return failure, so printinf ERR and returning S_OK is fine here. > + } > + nsAString_Finish(&aName); > + return S_OK; > } > > static HRESULT WINAPI HTMLWindow2_get_name(IHTMLWindow2 *iface, BSTR *p) > { > HTMLWindow *This = HTMLWINDOW2_THIS(iface); > - FIXME("(%p)->(%p)\n", This, p); > - return E_NOTIMPL; > + nsAString aName; > + nsresult nsres; > + TRACE("(%p)->(%p)\n", This, p); > + > + nsAString_Init(&aName, NULL); > + nsres = nsIDOMWindow_GetName(This->nswindow, &aName); > + if (NS_FAILED(nsres)) { > + TRACE("Failed nsIDOMWindow_GetName %08x \n",nsres); > + return nsres; Same as above. > + } > + > + nsres = nsAString_GetData(&aName, p); You can't do this, p is BSTR and must be allocated with SysAllocString. Look at other functions how to implement it. > + if (NS_FAILED(nsres)) { > + TRACE("Failed nsAString_GetData %08x \n",nsres); > + return nsres; > + } > + nsAString_Finish(&aName); > + > + return S_OK; > } Jacek

1 0