August 2016 - Wine-Devel - list.winehq.org

Re: [PATCH] gdi32: Support special 'ttcf' tag in GetFontData(), fix individual collection item offset
by Huw Davies Aug. 12, 2016

Aug. 12, 2016

On Fri, Aug 12, 2016 at 01:47:39AM +0300, Nikolay Sivov wrote: > Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> > --- > dlls/gdi32/freetype.c | 53 ++++++++++++++++++++++++++++++++++++++++++--------- > 1 file changed, 44 insertions(+), 9 deletions(-) > > diff --git a/dlls/gdi32/freetype.c b/dlls/gdi32/freetype.c > index 5ddfc6a..f4aab0b 100644 > --- a/dlls/gdi32/freetype.c > +++ b/dlls/gdi32/freetype.c > @@ -394,6 +396,8 @@ struct tagGdiFont { > int codepage; > BOOL fake_italic; > BOOL fake_bold; > + BOOL is_ttc_item; > + ULONG ttc_item_offset; Since ttc_item_offset can never be zero in a ttc, you don't need the boolean. > BYTE underline; > BYTE strikeout; > INT orientation; > @@ -2189,7 +2193,7 @@ static FT_Face new_ft_face( const char *file, void *font_data_ptr, DWORD font_da > !pFT_Get_Sfnt_Table( ft_face, ft_sfnt_hhea ) || > !pFT_Get_Sfnt_Table( ft_face, ft_sfnt_head )) > { > - TRACE("Font %s/%p lacks either an OS2, HHEA or HEAD table.\n" > + TRACE("Font %s/%p lacks either an OS/2, hhea or head table.\n" > "Skipping this font.\n", debugstr_a(file), font_data_ptr); > goto fail; > } Unrelated. > @@ -4535,6 +4539,9 @@ static FT_Face OpenFontFace(GdiFont *font, Face *face, LONG width, LONG height) > font->ft_face = ft_face; > > if(FT_IS_SCALABLE(ft_face)) { > + FT_ULong len; > + DWORD header; > + > /* load the VDMX table if we have one */ > font->ppem = load_VDMX(font, height); > if(font->ppem == 0) > @@ -4543,6 +4550,22 @@ static FT_Face OpenFontFace(GdiFont *font, Face *face, LONG width, LONG height) > > if((err = pFT_Set_Pixel_Sizes(ft_face, 0, font->ppem)) != 0) > WARN("FT_Set_Pixel_Sizes %d, %d rets %x\n", 0, font->ppem, err); > + > + /* see if it's a TTC */ > + len = sizeof(header); > + if (!pFT_Load_Sfnt_Table(ft_face, 0, 0, (void*)&header, &len)) { > + const DWORD ttcf = 0x66637474; Please define and use the 'ttcf' tag. > @@ -4638,12 +4661,18 @@ static void free_font(GdiFont *font) > HeapFree(GetProcessHeap(), 0, font); > } > > +#define MS_MAKE_TAG( _x1, _x2, _x3, _x4 ) \ > + ( ( (FT_ULong)_x4 << 24 ) | \ > + ( (FT_ULong)_x3 << 16 ) | \ > + ( (FT_ULong)_x2 << 8 ) | \ > + (FT_ULong)_x1 ) We already have this define later on, we should probably move it to the top of the file (under the GET_BE_(D)WORD defines looks like a good place) and move all the various tag definitions up there too. This should be done as a separate, preliminary patch. > static DWORD get_font_data( GdiFont *font, DWORD table, DWORD offset, LPVOID buf, DWORD cbData) > { > FT_Face ft_face = font->ft_face; > FT_ULong len; > FT_Error err; > + DWORD tag; > > if (!FT_IS_SFNT(ft_face)) return GDI_ERROR; > > @@ -4652,7 +4681,17 @@ static DWORD get_font_data( GdiFont *font, DWORD table, DWORD offset, LPVOID buf > else > len = cbData; > > - table = RtlUlongByteSwap( table ); /* MS tags differ in endianness from FT ones */ > + table = RtlUlongByteSwap( tag = table ); /* MS tags differ in endianness from FT ones */ > + > + /* if font is a member of TTC, 'ttcf' tag allows reading from beginning of TTC file, > + 0 tag means to read from start of collection member data. */ > + if (font->is_ttc_item) > + { > + if (tag == MS_MAKE_TAG('t','t','c','f')) > + table = 0; > + else if (tag == 0) > + offset += font->ttc_item_offset; > + } I find the introduction of tag here confusiong. You should be able to do this without the second variable. > return GDI_ERROR; > } > return len; > @@ -8246,9 +8282,8 @@ static DWORD freetype_GetFontData( PHYSDEV dev, DWORD table, DWORD offset, LPVOI > return dev->funcs->pGetFontData( dev, table, offset, buf, cbData ); > } > > - TRACE("font=%p, table=%c%c%c%c, offset=0x%x, buf=%p, cbData=0x%x\n", > - physdev->font, LOBYTE(LOWORD(table)), HIBYTE(LOWORD(table)), > - LOBYTE(HIWORD(table)), HIBYTE(HIWORD(table)), offset, buf, cbData); > + TRACE("font=%p, table=%s, offset=0x%x, buf=%p, cbData=0x%x\n", > + physdev->font, debugstr_an((char*)&table, 4), offset, buf, cbData); Doesn't this print the tag backwards? Huw.

1 0

Re: [PATCH] ucrtbase: Add __stdio_common_vfwscanf
by Piotr Caban Aug. 12, 2016

Aug. 12, 2016

On 08/11/16 19:27, Daniel Lehman wrote: > + if (options & ~UCRTBASE_SCANF_MASK) > + FIXME("options %s not handled\n", wine_dbgstr_longlong(options)); > + if (options & UCRTBASE_SCANF_SECURECRT) It would be better to compare options with UCRTBASE_SCANF_SECURECRT when printing FIXME message. Your implementation doesn't support any other options. Thanks, Piotr

1 0

Re: [PATCH] msvcrt: Add _wcreate_locale
by Piotr Caban Aug. 12, 2016

Aug. 12, 2016

On 08/11/16 19:27, Daniel Lehman wrote: > + len = MSVCRT_wcstombs(NULL, locale, 0); > + if (!(str = MSVCRT_malloc(len))) return NULL; Please handle string conversion error. The allocated buffer is not big enough. Something like this should work: len = MSVCRT_wcstombs(NULL, locale, 0); if(len == -1) { ... } if(!(str = MSVCRT_malloc(len+1)) ... It may be even better to call wcstombs_s so invalid parameter handler is called (I didn't test if it should be called). Thanks, Piotr

1 0

Re: [v3] riched20: allow setting background text color
by Huw Davies Aug. 12, 2016

Aug. 12, 2016

On Thu, Aug 11, 2016 at 12:22:21PM -0600, Kenneth Haley wrote: > fixes https://bugs.winehq.org/show_bug.cgi?id=36625 > > Tested on Arch Linux. Now the background is correctly painted, but there's still the issue of the background disappearing when selecting a range within a run of text that has a background. It's easy to test this using Wine's wordpad. Huw.

1 0

Re: [PATCH 2/3] d3drm: Implement IDirect3DRMFrame*::GetScene. (v5)
by Józef Kucia Aug. 11, 2016

Aug. 11, 2016

On Fri, Aug 12, 2016 at 12:48 AM, Aaryaman Vasishta <jem456.vasishta(a)gmail.com> wrote: > v4: Shorter implementation. > v3: Directly traverse through parent frames instead of using GetParent. > > Signed-off-by: Aaryaman Vasishta <jem456.vasishta(a)gmail.com> > --- > dlls/d3drm/frame.c | 43 +++++++++++++++++++++++++++++++++++++------ > dlls/d3drm/tests/d3drm.c | 24 ++++++++++++++++++++++++ > 2 files changed, 61 insertions(+), 6 deletions(-) > > diff --git a/dlls/d3drm/frame.c b/dlls/d3drm/frame.c > index 2ff0241..4d7ab42 100644 > --- a/dlls/d3drm/frame.c > +++ b/dlls/d3drm/frame.c > @@ -1339,23 +1339,54 @@ static HRESULT WINAPI d3drm_frame1_GetRotation(IDirect3DRMFrame *iface, > > static HRESULT WINAPI d3drm_frame3_GetScene(IDirect3DRMFrame3 *iface, IDirect3DRMFrame3 **scene) > { > - FIXME("iface %p, scene %p stub!\n", iface, scene); > + struct d3drm_frame *frame = impl_from_IDirect3DRMFrame3(iface); > > - return E_NOTIMPL; > + TRACE("iface %p, scene %p.\n", iface, scene); > + > + if (!scene) > + return D3DRMERR_BADVALUE; > + > + *scene = &frame->IDirect3DRMFrame3_iface; Do you really need this assignment? > + > + while (frame->parent) > + frame = frame->parent; > + > + IDirect3DRMFrame3_AddRef(*scene = &frame->IDirect3DRMFrame3_iface); > + > + return D3DRM_OK; > }

2 1

Re: [PATCH 2/3] d3drm: Implement IDirect3DRMFrame*::GetScene. (v4)
by Henri Verbeet Aug. 11, 2016

Aug. 11, 2016

On 10 August 2016 at 17:41, Aaryaman Vasishta <jem456.vasishta(a)gmail.com> wrote: > + *scene = &frame->IDirect3DRMFrame3_iface; > + > + while ((frame = frame->parent)) > + *scene = &frame->IDirect3DRMFrame3_iface; > + > + IDirect3DRMFrame3_AddRef(*scene); How about: while (frame->parent) frame = frame->parent; IDirect3DRMFrame3_AddRef(*scene = &frame->IDirect3DRMFrame3_iface);

2 1

Re: [PATCH] ntdll: Set Rip in for longjmp in RtlRestoreContext
by Marvin Aug. 11, 2016

Aug. 11, 2016

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=25000 Your paranoid android. === wvistau64 (32 bit exception) === exception.c:523: Test failed: 42: Unexpected eip 0x2d0000/0x2d0001 exception: unhandled exception 80000004 at 002D0000 exception.c:523: Test failed: 42: Unexpected eip 0x77baed97/0x2d0001 === w7u (32 bit exception) === exception.c:523: Test failed: 42: Unexpected eip 0x280000/0x280001 exception: unhandled exception 80000004 at 00280000 exception.c:523: Test failed: 42: Unexpected eip 0x77cde34f/0x280001 === w1064 (32 bit exception) === An error occurred while waiting for the test to complete: network read timed out (wait2/connect:AgentVersion.h:0/9) The test VM has crashed, rebooted or lost connectivity (or the TestAgent server died) === w1064 (64 bit exception) === An error occurred while waiting for the test to complete: network read timed out (wait2/connect:AgentVersion.h:0/9) The test VM has crashed, rebooted or lost connectivity (or the TestAgent server died)

1 0

Re: [PATCH 2/2] include: cpp_quote D3D10CreateDeviceAndSwapChain1.
by Henri Verbeet Aug. 11, 2016

Aug. 11, 2016

On 7 August 2016 at 21:22, Stefan Dösinger <stefandoesinger(a)gmx.at> wrote: > D3D10_DRIVER_TYPE is defined in d3d10misc.h, so MIDL complains that it > doesn't know about it. Would introducing d3d10misc.idl help?

2 3

xaudio2: More than 256 audio sources
by Fabian Maurer Aug. 11, 2016

Aug. 11, 2016

Some applications run out of sources when using xaudio2 so I'm working on a patch that allows more than the 256 sources provided by OpenAL. I've tried a few different methods, and the current should be the easiest and most compatible. Still, I'm not perfectly familiar with the implementation of xaudio2, and I ran into lots of hard to resolve issues, so I figured I'd ask for feedback before submitting the patch. Any feedback or mistakes regarding my implementation?

2 1

Re: [PATCH 3/5] programs/services: Break services into Loader groups
by Sebastian Lackner Aug. 11, 2016

Aug. 11, 2016

On 09.08.2016 18:40, Aric Stewart wrote: > > Sort the groups by the service's tag to start in order > > Signed-off-by: Aric Stewart <aric(a)codeweavers.com> > --- > programs/services/services.c | 101 ++++++++++++++++++++++++++++++++----------- > 1 file changed, 75 insertions(+), 26 deletions(-) > I believe this can be solved a bit easier. Why not just a sort function callback which sorts for both group and tag?

2 3