September 2013 - Wine-Devel

Re: [1/3] gdi32: Return GDI_ERROR when requested empty glyph bitmaps. (try 2)

by Marvin

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://newtestbot.winehq.org/JobDetails.pl?Key=2350 Your paranoid android. === w7u (32 bit font) === No test summary line found Failure running script in VM: network read timed out

Sept. 26, 2013

Re: [PATCH 1/2] gdi32: fix for rotated Arc, ArcTo, Chord and Pie drawing problem (try 2)

by Alexandre Julliard

Daniel Wendt <daniel.wendt(a)linux.com> writes: > @@ -308,6 +308,11 @@ static int get_arc_points( PHYSDEV dev, const RECT *rect, POINT start, POINT end > } > > memmove( points, points + count, (pos - count) * sizeof(POINT) ); > + > + if (GetGraphicsMode( dev->hdc ) == GM_ADVANCED) > + { > + LPtoDP( dev->hdc, points, pos - count ); > + } > return pos - count; > } > > @@ -322,7 +327,17 @@ static BOOL draw_arc( PHYSDEV dev, INT left, INT top, INT right, INT bottom, > BOOL ret = TRUE; > HRGN outline = 0, interior = 0; > > - if (!get_pen_device_rect( pdev, &rect, left, top, right, bottom )) return TRUE; > + if (GetGraphicsMode(dev->hdc ) != GM_ADVANCED) > + { > + if (!get_pen_device_rect( pdev, &rect, left, top, right, bottom )) return TRUE; > + } > + else > + { > + rect.left = left; > + rect.top = top; > + rect.right = right; > + rect.bottom = bottom; > + } I don't think that you want to draw arcs in logical space, that can be very inefficient. -- Alexandre Julliard julliard(a)winehq.org

Sept. 26, 2013

Re: [PATCH 3/3] msvcrt: Call MSVCRT_strtoi64_l in strtoul implementation (try2)

by Marvin

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://newtestbot.winehq.org/JobDetails.pl?Key=2349 Your paranoid android. === build (build) === Patch failed to apply

Sept. 26, 2013

Re: imm32: ImmGetDescriptionA return does not include NULL byte

by Alexandre Julliard

Aric Stewart <aric(a)codeweavers.com> writes: > @@ -1684,7 +1684,7 @@ UINT WINAPI ImmGetDescriptionA( > > HeapFree( GetProcessHeap(), 0, buf ); > > - return len; > + return min( len - 1, uBufLen ); > } This may do the right thing on failure, but in an ugly way. A few overflow tests would be a good idea. -- Alexandre Julliard julliard(a)winehq.org

Sept. 26, 2013

Re: [PATCH 1/6] riched20: Implement ITextDocument_fnOpen.

by Alexandre Julliard

Jactry <jactry92(a)gmail.com> writes: > + chBuffer = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, size+2); > + if(!chBuffer) > + { > + CloseHandle(hFile); > + return HRESULT_FROM_WIN32(GetLastError()); > + } It's not useful to zero the buffer since you are reading into it. Also GetLastError won't do anything useful here. -- Alexandre Julliard julliard(a)winehq.org

Sept. 26, 2013

Re: [2/3] gdi32: Return fake BBox when requested empty glyph metrics. (try 2)

by Alexandre Julliard

Akihiro Sagawa <sagawa.aki(a)gmail.com> writes: > @@ -6519,11 +6519,17 @@ static DWORD get_glyph_outline(GdiFont *incoming_font, UINT glyph, UINT format, > } > > lpgm->gmBlackBoxX = (right - left) >> 6; > + if (lpgm->gmBlackBoxX == 0) > + lpgm->gmBlackBoxX = 1; > lpgm->gmBlackBoxY = (top - bottom) >> 6; > + if (lpgm->gmBlackBoxY == 0) > + lpgm->gmBlackBoxY = 1; > lpgm->gmptGlyphOrigin.x = origin_x >> 6; > lpgm->gmptGlyphOrigin.y = origin_y >> 6; > abc->abcA = left >> 6; > abc->abcB = (right - left) >> 6; > + if (abc->abcB == 0) > + abc->abcB = 1; > abc->abcC = adv - abc->abcA - abc->abcB; It would be a good idea to add tests for the ABC metrics too. -- Alexandre Julliard julliard(a)winehq.org

Sept. 26, 2013

Re: (try 5)[1/4] imm32: Move thread data from TLSEntry to an internal list

by Alexandre Julliard

Aric Stewart <aric(a)codeweavers.com> writes: > @@ -530,12 +591,18 @@ static BOOL CALLBACK _ImmAssociateContextExEnumProc(HWND hwnd, LPARAM lParam) > */ > BOOL WINAPI ImmAssociateContextEx(HWND hWnd, HIMC hIMC, DWORD dwFlags) > { > + IMMThreadData* thread_data = NULL; > + > TRACE("(%p, %p, 0x%x):\n", hWnd, hIMC, dwFlags); > > - if (!IMM_GetThreadData()->defaultContext) > - IMM_GetThreadData()->defaultContext = ImmCreateContext(); > + thread_data = IMM_GetInitializedThreadData(); > + if (!thread_data) > + return FALSE; > > - if (!hWnd) return FALSE; > + LeaveCriticalSection(&threaddata_cs); > + > + if (!hWnd) > + return FALSE; > > switch (dwFlags) > { > @@ -543,7 +610,7 @@ BOOL WINAPI ImmAssociateContextEx(HWND hWnd, HIMC hIMC, DWORD dwFlags) > ImmAssociateContext(hWnd,hIMC); > return TRUE; > case IACE_DEFAULT: > - ImmAssociateContext(hWnd,IMM_GetThreadData()->defaultContext); > + ImmAssociateContext(hWnd,thread_data->defaultContext); > return TRUE; > case IACE_IGNORENOCONTEXT: > if (GetPropW(hWnd,szwWineIMCProperty)) You shouldn't be accessing thread data outside of the critical section. -- Alexandre Julliard julliard(a)winehq.org

Sept. 26, 2013

Re: [PATCH 2/2] msvcrt: Call MSVCRT_strtoi64_l in strtoul implementation

by Marvin

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://newtestbot.winehq.org/JobDetails.pl?Key=2346 Your paranoid android. === build (build) === Patch failed to apply

Sept. 26, 2013

Re: kernel32/tests: Add DuplicateHandle test to the file access tests.

by Marvin

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://newtestbot.winehq.org/JobDetails.pl?Key=2345 Your paranoid android. === build (build) === Patch failed to apply

Sept. 26, 2013

Re: dlls: Replace the remaining CONSTs.

by Marvin

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://newtestbot.winehq.org/JobDetails.pl?Key=2344 Your paranoid android. === build (build) === Patch failed to apply

Sept. 26, 2013