August 2009 - Wine-devel - list.winehq.org

Re: [trans] update Chinese translation for winecfg
by Henri Verbeet 12 Aug '09

12 Aug '09

2009/8/12 xiaq <xiaqqaix(a)gmail.com>: > The patch has been reviewed by the original translator, Hongbo Ni. I > have also included my name in the file. > You need to include your full name in patches sent to Wine.

1 0

Review: winex11.drv: Add workaround for 'phantom key' presence in X
by Andrew Eikum 12 Aug '09

12 Aug '09

A bug made itself present in VkKeyScan's handling of the less-than character on US (and other) keyboard layouts. VkKeyScan('<') in Windows returns 0x1bc, which is VK_OEM_COMMA with the shift key flag set. In Wine, VkKeyScan('<') returns 0xe2, which is VK_OEM_102, or the VK for the "infamous phantom key". The root cause of this is X's XKeysymToKeycode returning the keycode for the phantom key instead of the keycode for the comma key when passed the keysym for '<' (XK_less). This isn't incorrect on X's part, but inconvenient for Wine as most US keyboards don't have the <> key, but rather use the shifted comma key. The attached patch adds a new field in the main_key_tab structure. The new field indicates which keysym, if any, should be passed to XKeysymToKeycode instead of '<'. I've filled this field in for every keyboard type already defined in keyboard.c. Since winex11.drv is a pretty touchy part of Wine, I'm sending this patch to wine-devel for review. I don't have the hardware or software setup to test other language inputs, and LoadKeyboardLayout is not implemented so I can't just test with VkKeyScanEx. If you're using a foreign keyboard, especially one with a <> key, please test VkKeyScan's handling of the '<' character with this patch applied. Thanks Andrew

3 5

Re: ws2_32(2/8): Add af_irda.h (try 3)
by Alexandre Julliard 12 Aug '09

12 Aug '09

Juan Lang <juan.lang(a)gmail.com> writes: > +typedef struct _IAS_QUERY > +{ > + UCHAR irdaDeviceID[4]; > +#ifdef USE_WS_PREFIX > + char irdaClassName[WS_IAS_MAX_CLASSNAME]; > + char irdaAttribName[WS_IAS_MAX_ATTRIBNAME]; > +#else > + char irdaClassName[IAS_MAX_CLASSNAME]; > + char irdaAttribName[IAS_MAX_ATTRIBNAME]; > +#endif > + USHORT irdaAttribType; > + union { > + int irdaAttribInt; > + struct { > + ULONG Len; > + UCHAR OctetSeq[IAS_MAX_OCTET_STRING]; > + } irdaAttribOctetSeq; > + struct { > + ULONG Len; > + UCHAR CharSet; > + UCHAR UsrStr[IAS_MAX_USER_STRING]; > + } irdaAttribUsrStr; > + } irdaAttribute; > +} IAS_QUERY, *PIAS_QUERY; > + > +typedef struct _IAS_SET > +{ > +#ifdef USE_WS_PREFIX > + char irdaClassName[WS_IAS_MAX_CLASSNAME]; > + char irdaAttribName[WS_IAS_MAX_ATTRIBNAME]; > +#else > + char irdaClassName[IAS_MAX_CLASSNAME]; > + char irdaAttribName[IAS_MAX_ATTRIBNAME]; > +#endif > + union { > + int irdaAttribInt; > + struct { > + ULONG Len; > + UCHAR OctetSeq[IAS_MAX_OCTET_STRING]; > + } irdaAttribOctetSeq; > + struct { > + ULONG Len; > + UCHAR CharSet; > + UCHAR UsrStr[IAS_MAX_USER_STRING]; > + } irdaAttribUsrStr; > + } irdaAttribute; > +} IAS_SET, *PIAS_SET; > + > +#endif /* AF_IRDA_H */ These don't seem to match the PSDK headers or MSDN (which don't match each other either...) Which are the correct ones? -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: [7/13] windowscodecs: implement SetPixelFormat for BMP frame encoder
by Henri Verbeet 12 Aug '09

12 Aug '09

2009/8/12 Vincent Povirk <vincent(a)codeweavers.com>: > + if (memcmp(formats[i].guid, pPixelFormat, sizeof(GUID)) == 0) This works of course, but isn't IsEqualGUID() the preferred way to do that?

2 1

Re: [2/2] comctl32/listview: Block redrawing entirely after WM_SETREDRAW wParam=FALSE
by Alexandre Julliard 12 Aug '09

12 Aug '09

Nikolay Sivov <bunglehead(a)gmail.com> writes: > diff --git a/dlls/comctl32/listview.c b/dlls/comctl32/listview.c > index a65d832..9bca376 100644 > --- a/dlls/comctl32/listview.c > +++ b/dlls/comctl32/listview.c > @@ -9740,6 +9740,8 @@ static LRESULT LISTVIEW_Paint(LISTVIEW_INFO *infoPtr, HDC hdc) > { > TRACE("(hdc=%p)\n", hdc); > > + if (!is_redrawing(infoPtr)) return 0; > + > if (infoPtr->bNoItemMetrics && infoPtr->nItemCount) > { > infoPtr->bNoItemMetrics = FALSE; > @@ -9790,7 +9792,13 @@ static LRESULT LISTVIEW_PrintClient(LISTVIEW_INFO *infoPtr, HDC hdc, DWORD optio > LISTVIEW_EraseBkgnd(infoPtr, hdc); > > if (options & PRF_CLIENT) > + { > + BOOL redraw_old = infoPtr->bRedraw; > + > + infoPtr->bRedraw = TRUE; > LISTVIEW_Paint(infoPtr, hdc); > + infoPtr->bRedraw = redraw_old; > + } That's not very nice. You should move the checks so that you don't have to change the flag to paint. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: [1/6] comctl32/tests: Move v6 test initialisation routines into separate file
by Henri Verbeet 12 Aug '09

12 Aug '09

2009/8/10 Owen Rudge <orudge(a)codeweavers.com>: > This is the first of a series of patches to implement the IImageList > interface into comctl32, introduced in comctl32 v6. > > This first patch splits the routines for loading v6 of comctl32 > (requiring a side-by-side manifest) out of the listview tests and into a > general file, so that it can be used by the imagelist tests too. > I don't know if there's any reason these tests should be an exception, but the general rule is that tests are supposed to be contained within their own file, to allow for easy compilation.

6 9

Re: [2/2] msxml3: IXMLDOMElement doesnt support IObjectIdentity
by Jacek Caban 12 Aug '09

12 Aug '09

Alistair Leslie-Hughes wrote: > Hi, > This is just to silence a noisy fixme. > > Changelog: > msxml3: IXMLDOMElement doesnt support IObjectIdentity It's an interface used by jscript engine, so it would be better to silence it for all object that implements IDispatch. I'd suggest to move it to dispex_query_interface. Jacek

1 0

Re: [1/2] msxml3: Add IDispatchEx support to IXMLDOMElement
by Jacek Caban 12 Aug '09

12 Aug '09

Alistair Leslie-Hughes wrote: > Hi, > > > Changelog: > msxml3: Add IDispatchEx support to IXMLDOMElement @@ -46,6 +47,9 @@ typedef struct _domelem LONG ref; IUnknown *node_unk; IXMLDOMNode *node; + + /* IDispatchEx */ + DispatchEx dispex; } domelem; It would be probably better to add DispatchEx to node object. Then other node types will reuse it (initializing with different data). See how it's done in mshtml. Jacek

1 0

Re: shdocvw: Implement OleInPlaceObject InPlaceDeactivate
by Jacek Caban 12 Aug '09

12 Aug '09

Alistair Leslie-Hughes wrote: > Hi, > This fixes bug http://bugs.winehq.org/show_bug.cgi?id=18531 > > > Changelog: > shdocvw: Implement OleInPlaceObject InPlaceDeactivate It's not an implementation, but a bit better stub. Please leave the FIXME message. Jacek

1 0

Re: [3/5] comctl32/treeview: Make tests not depend from each other, replace some magics with macros
by Paul Vriens 12 Aug '09

12 Aug '09

Nikolay Sivov wrote: > Currently tests are dependent which makes it impossible to change only one, > could be possible actually, but not always. Anyway separate tests are better. > > Changelog: > - Make test not depend from each other, replace some magics with macros > Hi Nikolay, Looks like this one introduced 1 failure on several boxes (XP, W2K3, Vista and W2K8) and 1 more on Vista boxes: http://test.winehq.org/data/tests/comctl32:treeview.html Could you have a look? -- Cheers, Paul.

2 2