July 2010 - Wine-Devel - list.winehq.org

Re: [PATCH 05/10] d3dx9: Add comment to delineate vertex buffer and index buffer tests. (try 2)
by Henri Verbeet July 19, 2010

July 19, 2010

On 17 July 2010 21:13, Misha Koshelev <misha680(a)gmail.com> wrote: > --- > dlls/d3dx9_36/tests/mesh.c | 2 ++ > 1 files changed, 2 insertions(+), 0 deletions(-) > > diff --git a/dlls/d3dx9_36/tests/mesh.c b/dlls/d3dx9_36/tests/mesh.c > index cac515e..401e61e 100644 > --- a/dlls/d3dx9_36/tests/mesh.c > +++ b/dlls/d3dx9_36/tests/mesh.c > @@ -546,6 +546,7 @@ static void D3DXCreateBoxTest(void) > > if (hr == D3D_OK && box) > { > + /* Vertex buffer */ > number_of_vertices = box->lpVtbl->GetNumVertices(box); > todo_wine ok( number_of_vertices == 24, "Got result %u, expected 24\n",number_of_vertices); > > @@ -704,6 +705,7 @@ static void D3DXCreateSphereTest(void) > > if (sphere) > { > + /* Vertex buffer */ > number_of_vertices = sphere->lpVtbl->GetNumVertices(sphere); > todo_wine ok( number_of_vertices == 4, "Got result %u, expected 4\n",number_of_vertices); > Why didn't you add these in the original patches?

1 0

Re: [PATCH 04/10] d3dx9: Add tests for D3DXCreateBox. (try 2)
by Henri Verbeet July 19, 2010

July 19, 2010

Same comments apply as for the previous patches. With this being a larger dataset it makes even more sense to calculate the vertices instead of hardcoding them.

1 0

Re: [PATCH 03/10] d3dx9: Test raw vertex data for D3DXCreateSphere. (try 2)
by Henri Verbeet July 19, 2010

July 19, 2010

On 17 July 2010 21:13, Misha Koshelev <misha680(a)gmail.com> wrote: > +#include <assert.h> You don't use that in this patch. > + D3DXVECTOR3 test_vertex_data[8]; I think it makes sense to use the same kind of structure you're using in your implementation of D3DXCreateSphere() here. > + if (vertex_data) > + { > + test_vertex_data[0].x = 0.0f; test_vertex_data[0].y = 0.0f; test_vertex_data[0].z = 1.0f; > + test_vertex_data[1].x = 0.0f; test_vertex_data[1].y = 0.0f; test_vertex_data[1].z = 1.0f; > + test_vertex_data[2].x = 0.0f; test_vertex_data[2].y = 1.0f; test_vertex_data[2].z = -4.37114e-008f; > + test_vertex_data[3].x = 0.0f; test_vertex_data[3].y = 1.0f; test_vertex_data[3].z = -4.37114e-008f; > + test_vertex_data[4].x = -8.74228e-008f; test_vertex_data[4].y = -1.0f; test_vertex_data[4].z = -4.37114e-008f; > + test_vertex_data[5].x = -8.74228e-008f; test_vertex_data[5].y = -1.0f; test_vertex_data[5].z = -4.37114e-008f; > + test_vertex_data[6].x = 0.0f; test_vertex_data[6].y = 0.0f; test_vertex_data[6].z = -1.0f; > + test_vertex_data[7].x = 0.0f; test_vertex_data[7].y = 0.0f; test_vertex_data[7].z = -1.0f; > + Note that you can simply initialize test_vertex_data with these values. On the other hand, you could also just calculate the expected values.

1 0

Re: [PATCH 02/10] d3dx9: Add tests for D3DXCreateSphere vertex buffer description. (try 2)
by Henri Verbeet July 19, 2010

July 19, 2010

On 17 July 2010 21:13, Misha Koshelev <misha680(a)gmail.com> wrote: > + if (hr == D3D_OK) > + { > + todo_wine ok( vertex_buffer_description.Format == D3DFMT_VERTEXDATA, "Got result %x, expected %x (D3DFMT_VERTEXDATA)\n",vertex_buffer_description.Format,D3DFMT_VERTEXDATA); > + todo_wine ok( vertex_buffer_description.Type == D3DRTYPE_VERTEXBUFFER, "Got result %x, expected %x (D3DRTYPE_VERTEXBUFFER)\n",vertex_buffer_description.Type,D3DRTYPE_VERTEXBUFFER); > + todo_wine ok( vertex_buffer_description.Usage == 0, "Got result %x, expected %x\n",vertex_buffer_description.Usage,0); > + todo_wine ok( vertex_buffer_description.Pool == D3DPOOL_MANAGED, "Got result %x, expected %x (D3DPOOL_DEFAULT)\n",vertex_buffer_description.Pool,D3DPOOL_DEFAULT); > + expected = number_of_vertices * sizeof(D3DXVECTOR3) * 2; > + todo_wine ok( vertex_buffer_description.Size == expected, "Got result %x, expected %x\n",vertex_buffer_description.Size,expected); > + todo_wine ok( vertex_buffer_description.FVF == (D3DFVF_XYZ | D3DFVF_NORMAL), "Got result %x, expected %x (D3DFVF_XYZ | D3DFVF_NORMAL)\n",vertex_buffer_description.FVF,D3DFVF_XYZ | D3DFVF_NORMAL); > + } > + Those shouldn't need to be todo_wine, the tests shouldn't even get run with current Wine. I guess it doesn't matter much though.

1 0

Re: user32: Make sure that main window state is consistent during the tests. (Resend)
by testbot＠testbot.winehq.org July 19, 2010

July 19, 2010

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 http://testbot.winehq.org/JobDetails.pl?Key=3509 Your paranoid android. === WVISTAADM (32 bit win) === win.c:2271: Test failed: 001401C8: expected prev 001001C6, got 003B01B4 win.c:2272: Test failed: 001001C6: expected next 001401C8, got 003B01B4 win.c:2198: Test failed: 005C01BE: expected next 000C01B6, got 001201C2 win.c:2271: Test failed: 003C01B4: expected prev 001601C8, got 001301C2 win.c:2272: Test failed: 001601C8: expected next 003C01B4, got 001301C2 === W2K8SE (32 bit win) === win.c:2271: Test failed: 00190116: expected prev 000F0144, got 002F0114 win.c:2272: Test failed: 000F0144: expected next 00190116, got 002F0114 win.c:2198: Test failed: 005D0118: expected next 000B0126, got 00270122 win.c:2271: Test failed: 00300114: expected prev 001B0116, got 00280122 win.c:2272: Test failed: 001B0116: expected next 00300114, got 00280122 === W7PRO (32 bit win) === win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 006E01A4 fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 007001A4 fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 000501AA fg 00030050 in hook HCBT_SETFOCUS win.c:2674: Test failed: GetActiveWindow() = 00000000 win.c:2674: Test failed: GetFocus() = 00000000 win.c:2700: Test failed: GetActiveWindow() = 00000000 win.c:2700: Test failed: GetFocus() = 00000000 win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 007101A4 fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 000701AA fg 00030050 in hook HCBT_SETFOCUS win.c:2759: Test failed: GetActiveWindow() = 00000000 win.c:2759: Test failed: GetFocus() = 00000000 win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 000501A0 fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 003001AA fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 003101AA fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 000A0136 fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 000E0136 fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 000F0136 fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 008301A4 fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 0007014E fg 00030050 in hook HCBT_SETFOCUS win.c:823: Test failed: wrong dwWindowStatus: 0000 != 0001 active 000501A0 fg 00030050 in hook HCBT_SETFOCUS

1 0

Re: [PATCH 1/2] include/shell32: Shell_MergeMenus should be declared in shlobj.h and exported by name.
by testbot＠testbot.winehq.org July 19, 2010

July 19, 2010

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 http://testbot.winehq.org/JobDetails.pl?Key=3414 Your paranoid android. === WNT4WSSP6 (32 bit shlfileop) === shlfileop.c:2197: Test failed: SHGetNewLinkInfoA succeeded

2 1

Re: kernel32: add a test for CreateFileA()
by Nikolay Sivov July 18, 2010

July 18, 2010

On 7/18/2010 10:17, Austin English wrote: > Passed all the vm's on wtb. Hi, Austin. This looks strange: > + > + hFile = CreateFileA("c:\\*.*", GENERIC_READ,FILE_SHARE_READ,NULL,OPEN_EXISTING,FILE_ATTRIBUTE_NORMAL,NULL ); > + ok(GetLastError() == ERROR_INVALID_NAME || broken(GetLastError() == ERROR_FILE_NOT_FOUND) || broken(GetLastError() == ERROR_PATH_NOT_FOUND), > + "CreateFileA should have returned ERROR_INVALID_NAME on %s, but got %u\n", filename, GetLastError()); > + CloseHandle( hFile ); If a call failed then you should test for null hFile I suppose and remove CloseHandle(). Also when testing for last error a common rule is to set it first before the call. Does anything depend on this having 3 possible return values?

3 2

Re: kernel32: add a test for CreateFileA()
by James McKenzie July 18, 2010

July 18, 2010

Austin English wrote: > Passed all the vm's on wtb. > > > ------------------------------------------------------------------------ > > + hFile = CreateFileA("c:\\*.*", GENERIC_READ,FILE_SHARE_READ,NULL,OPEN_EXISTING,FILE_ATTRIBUTE_NORMAL,NULL ); + ok(GetLastError() == ERROR_INVALID_NAME || broken(GetLastError() == ERROR_FILE_NOT_FOUND) || broken(GetLastError() == ERROR_PATH_NOT_FOUND), + "CreateFileA should have returned ERROR_INVALID_NAME on %s, but got %u\n", filename, GetLastError()); + CloseHandle( hFile ); Nice negative test, but how about one that should pass? James McKenzie

2 1

re: D3DXCreateTeapot and reverse engineering?
by Dan Kegel July 18, 2010

July 18, 2010

IMHO, our criterion is "we only need to implement what apps really need", and for the moment, that app probably could live with a stub for D3DXCreateTeapot. When we run into a complaint about our lousy teapot, we can create a better one. So relax, just do a stub, and move on to more important things. Also, as Roderick says, sucking tables out of Windows is not ok, for copyright reasons.

5 6

D3DXCreateTeapot and reverse engineering?
by Misha Koshelev July 18, 2010

July 18, 2010

Dear All: I just wanted to double check before I do anything with D3DXCreateTeapot. I have several options for implementing this: Option A -------- One, the slightly harder option, involves the original dataset: http://www.sjbaker.org/teapot/teaset.tgz Option B -------- Another, much simpler method, would be to: a) Call D3DXCreateTeapot b) D3DXCreateTeapot returns a vertex and index buffer c) I can easily use these to create a header file that _exactly_ reproduces the vertices and indices created by Windows. My question: is Option B allowed, and if not, why not? Thank you Misha p.s. This is for this bug: http://bugs.winehq.org/show_bug.cgi?id=22918

3 2