June 2012 - Wine-Devel - list.winehq.org

Re: [1/5] d3dx9: Implement D3DXLoadVolumeFromMemory.
by Matteo Bruni June 29, 2012

June 29, 2012

2012/6/27 Józef Kucia <joseph.kucia(a)gmail.com>: > --- Hi Józef, > +void copy_simple_data(const BYTE *src, UINT srcpitch, SIZE src_size, const PixelFormatDesc *srcformat, > + BYTE *dest, UINT destpitch, SIZE dst_size, const PixelFormatDesc *destformat, D3DCOLOR colorkey) DECLSPEC_HIDDEN; > + You are not actually using this function in the patch. > + if (dst_box->Left > dst_box->Right || dst_box->Right > desc.Width) > + return D3DERR_INVALIDCALL; > + if (dst_box->Top > dst_box->Bottom || dst_box->Bottom > desc.Height) > + return D3DERR_INVALIDCALL; > + if (dst_box->Front > dst_box->Back || dst_box->Back > desc.Depth) > + return D3DERR_INVALIDCALL; > + > + dst_width = dst_box->Right - dst_box->Left; > + dst_height = dst_box->Bottom - dst_box->Top; > + dst_depth = dst_box->Back - dst_box->Front; > + if (!dst_width || !dst_height || !dst_depth) > + return D3DERR_INVALIDCALL; You can avoid the last check by comparing e.g. (dst_box->Left >= dst_box->Right) above, instead of using ">". > + if (src_box->Left & (src_format_desc->block_width - 1) > + || src_box->Top & (src_format_desc->block_height - 1) > + || (src_box->Right & (src_format_desc->block_width - 1) > + && src_width != desc.Width) > + || (src_box->Bottom & (src_format_desc->block_height - 1) > + && src_height != desc.Height)) > + { > + WARN("Source box (%u, %u, %u, %u) is misaligned\n", > + src_box->Left, src_box->Top, src_box->Right, src_box->Bottom); > + return D3DXERR_INVALIDDATA; > + } I'm not quite sure of the right and bottom misalignment checks. If the tests confirm that, of course that's okay. > + for (slice = 0; slice < src_depth; slice++) > + { > + src_addr = src_memory; > + src_addr += slice * src_slice_pitch; Shouldn't you take into account src_box->Front?

2 1

Re: user32: Add test for ComboBox repositioning behavior (try 2).
by Marvin June 29, 2012

June 29, 2012

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=19658 Your paranoid android. === W7PRO (32 bit msg) === msg.c:8219: Test failed: did not get zero count as expected (1).

2 1

Re: gdiplus: Add a test to show that GdipCloneImage is not supposed to increase refcount of the source image stream, but clone it instead.
by Vincent Povirk June 29, 2012

June 29, 2012

I think it's most likely not keeping the input stream in the new image at all. It's probably also worth testing SelectActiveFrame, GetPropertyItem, and GetImageRawFormat with clones of images. Maybe they don't work.

2 1

Re: wbemprox: Add a Win32_LogicalDisk class stub.
by Hans Leidekker June 28, 2012

June 28, 2012

On Thu, 2012-06-28 at 21:53 +0300, John Yani wrote: > +static void fill_logicaldisk( struct table *table ) > +{ > + static const WCHAR caption[] = > + {'C',':',0}; > + static const WCHAR description[] = > + {'L','o','c','a','l',' ','F','i','x','e','d',' ','D','i','s','k',0}; > + static const WCHAR device_id[] = > + {'C',':',0}; > + struct record_logicaldisk *rec; > + FIXME("returns a fake logical disks table\n"); > + if (!(table->data = heap_alloc( sizeof(*rec) ))) > + { > + return; > + } > + rec = (struct record_logicaldisk *)(table->data); > + rec->caption = caption; > + rec->description = description; > + rec->device_id = device_id; > + table->num_rows = 1; > + TRACE("created %u row(s)\n", table->num_rows); That's better. Does NFS actually query the Caption and Description properties?

2 4

Re: user32: Add a test for ComboBox WM_WINDOWPOSCHANGED handling.
by Alexandre Julliard June 28, 2012

June 28, 2012

Sergey Guralnik <serhio(a)etersoft.ru> writes: > @@ -5807,6 +5830,25 @@ static void test_combobox_messages(void) > log_all_parent_messages--; > ok_sequence(WmKeyDownComboSeq, "WM_KEYDOWN/VK_DOWN on a ComboBox", FALSE); > > + wp.hwnd = combo; > + wp.hwndInsertAfter = HWND_TOP; > + wp.cx = wp.cy = wp.x = wp.y = 50; > + wp.flags = 0; > + flush_sequence(); > + SendMessage(combo, WM_WINDOWPOSCHANGED, 0, (LPARAM)&wp); It doesn't make much sense to send this message directly. What are you trying to test? -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [2/2] gdiplus: Increase refcount of the source stream when loading a bitmap instead of cloning it.
by Vincent Povirk June 28, 2012

June 28, 2012

This looks good to me. (I didn't think it was necessary to comment, but this is an important fix and I was surprised it was still New.)

1 0

Unused msvcp60 functions
by Francois Gouget June 28, 2012

June 28, 2012

I get the following warnings when compiling msvcp60: main.c:108:51: warning: ‘std_BADOFF_func’ defined but not used [-Wunused-function] misc.c:66:40: warning: ‘mutex_mutex_lock’ defined but not used [-Wunused-function] misc.c:73:40: warning: ‘mutex_mutex_unlock’ defined but not used [-Wunused-function] misc.c:80:40: warning: ‘mutex_mutex_ctor’ defined but not used [-Wunused-function] misc.c:87:40: warning: ‘mutex_mutex_dtor’ defined but not used [-Wunused-function] misc.c:175:58: warning: ‘set_new_handler_reset’ defined but not used [-Wunused-function] Indeed msvcp60.spec contains no _BADOFF, _Mutex or set_new_handler function. I don't see them in my 'database' of exported Windows APIs for msvcp60 but maybe the Windows dlls I built it from where not entirely up to date. Still it looks like these were added in later revisions and thus can be removed from the above files. misc.c:127:50: warning: ‘wctype’ defined but not used [-Wunused-function] This one is actually forwarded to msvcp90 according to msvcp60.spec, thus making this implementation redundant. So it should either be removed or msvcp60.spec should be modified. -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ "Only wimps use tape backup: _real_ men just upload their important stuff on ftp, and let the rest of the world mirror it ;)" -- Linus Torvalds

2 3

Re: windowscodecs: Make sure that stream is not reused once the decoder is initialized.
by Vincent Povirk June 28, 2012

June 28, 2012

Looks good to me.

1 0

Re: kernel32/tests: Remove an unused variable in the sync tests.
by Marvin June 28, 2012

June 28, 2012

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=19638 Your paranoid android. === WINEBUILD (build) === Patch failed to apply

1 0

Re: [4/5] d3dx9: Implement D3DXCreateVolumeTextureFromFileInMemoryEx.
by Matteo Bruni June 28, 2012

June 28, 2012

2012/6/27 Józef Kucia <joseph.kucia(a)gmail.com>: > --- > + for (mip_level = 0; mip_level < mip_levels; mip_level++) > + { > + hr = calculate_dds_surface_size(src_info, width, height, &src_pitch, &mip_level_size); > + if (FAILED(hr)) return hr; ... > + hr = D3DXLoadVolumeFromMemory(volume, palette, NULL, pixels, src_info->Format, > + src_pitch, mip_level_size, NULL, &box, filter, color_key); mip_level_size would be the source slice pitch, right? I'd just call it like that.

1 0

Re: [2/2] msvcp90: Sync spec files
by Owen Rudge June 28, 2012

June 28, 2012

On 28/06/2012 11:33, Owen Rudge wrote: > --- > dlls/msvcp60/msvcp60.spec | 638 > ++++++++++++++++++++++---------------------- Sorry, I ran the make_specfiles script without realising that msvcp60.spec shouldn't have been updated. I've resent the patch without these changes. -- Owen Rudge http://www.owenrudge.net/

1 0

Re: [2/2] msvcp90: Sync spec files
by Piotr Caban June 28, 2012

June 28, 2012

First patch looks correct. On 06/28/12 12:33, Owen Rudge wrote: > dlls/msvcp60/msvcp60.spec | 638 > ++++++++++++++++++++++---------------------- You can't update msvcp60.spec file this way (most of functions there should not be forwarded to msvcp90).

1 0

Re: po: add missing url tag to nl.po
by Francois Gouget June 28, 2012

June 28, 2012

On Wed, 27 Jun 2012, Austin English wrote: > http://bugs.winehq.org/show_bug.cgi?id=31038 "Noot: Het wordt aanbevolen om distributiepakketten te gebruiken. Zie <a " -"href=http://wiki.winehq.org/Gecko voor details." +"href=http://wiki.winehq.org/Gecko</a> voor details." Note that this is missing a lot more than just </a>. This should be: "Noot: Het wordt aanbevolen om distributiepakketten te gebruiken. Zie <a " "href=\"http://wiki.winehq.org/Gecko\">http://wiki.winehq.org/Gecko</a> voor details." Fortunately it looks like nl is the only such case. -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ A polar bear is a cartesian bear after a coordinate transform.

1 0

Re: [1/5] windowscodecs: Implement MetadataHandler_GetValue. Take 2. Resend.
by Marvin June 28, 2012

June 28, 2012

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=19631 Your paranoid android. === WVISTAADM (32 bit metadata) === Failure running script in VM: The specified guest user must be logged in interactively to perform this operation

1 0

Re: [4/5] gdiplus: Add some tests for GdipGetPropertyItemSize and GdipGetPropertyItem. Take 2.
by Marvin June 28, 2012

June 28, 2012

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=19625 Your paranoid android. === W7PROX64 (64 bit image) === image.c:2817: Test failed: 0: expected property size 17 or 36, got 44 image.c:2817: Test failed: 2: expected property size 144 or 0, got 152 image.c:2817: Test failed: 3: expected property size 20 or 0, got 28 === TEST64_W7SP1 (64 bit image) === image.c:2817: Test failed: 0: expected property size 17 or 36, got 44 image.c:2817: Test failed: 2: expected property size 144 or 0, got 152 image.c:2817: Test failed: 3: expected property size 20 or 0, got 28

2 1

Re: [5/5] gdiplus: Add a bunch of tests for TIFF image properties. Resend.
by Marvin June 28, 2012

June 28, 2012

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=19626 Your paranoid android. === W7PROX64 (64 bit image) === image.c:2818: Test failed: 0: expected property size 17 or 36, got 44 image.c:2818: Test failed: 2: expected property size 144 or 0, got 152 image.c:2818: Test failed: 3: expected property size 20 or 0, got 28 === TEST64_W7SP1 (64 bit image) === image.c:2818: Test failed: 0: expected property size 17 or 36, got 44 image.c:2818: Test failed: 2: expected property size 144 or 0, got 152 image.c:2818: Test failed: 3: expected property size 20 or 0, got 28

1 0

Re: [3/4] gdiplus: Add some tests for GdipGetPropertyItemSize and GdipGetPropertyItem.
by Marvin June 28, 2012

June 28, 2012

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=19594 Your paranoid android. === W7PROX64 (64 bit image) === image.c:2811: Test failed: 0: expected property size 17 or 36, got 44 image.c:2811: Test failed: 2: expected property size 144 or 0, got 152 image.c:2811: Test failed: 3: expected property size 20 or 0, got 28 === TEST64_W7SP1 (64 bit image) === image.c:2811: Test failed: 0: expected property size 17 or 36, got 44 image.c:2811: Test failed: 2: expected property size 144 or 0, got 152 image.c:2811: Test failed: 3: expected property size 20 or 0, got 28

3 4

Re: wbemprox: Add a Win32_LogicalDisk class stub.
by Hans Leidekker June 27, 2012

June 27, 2012

On Wed, 2012-06-27 at 14:41 +0300, John Yani wrote: > +static void fill_logicaldisk( struct table *table ) > +{ > + static const WCHAR device_id[] = {'0',0}; > + struct record_logicaldisk *rec; > + FIXME("returns a fake logical disks table\n"); > + if (!(table->data = heap_alloc( sizeof(*rec) ))) > + { > + return; > + } > + rec = (struct record_logicaldisk *)(table->data); > + rec->device_id = heap_strdupW( device_id ); > + table->num_rows = 1; > + TRACE("created %u rows\n", table->num_rows); > +} There's no need to create a copy of device_id. Just assign it to rec->device_id and remove COL_FLAG_DYNAMIC from the column definition.

1 0

ODBC on 64bit system
by Daniel Jelinski June 27, 2012

June 27, 2012

Hello all, Is it possible to run a 32bit application using 64bit ODBC drivers? I already checked that standard 32bit wine compilation does not work - but will it work with WoW64? I am trying to get an application to connect to SQL Server via tdsodbc, but unfortunately tdsodbc:i386 in ubuntu 11.10 is broken... Regards, Daniel

1 0

Re: [1/5] windowscodecs: Implement IWICBitmapScaler::GetSize.
by Dmitry Timoshkov June 27, 2012

June 27, 2012

Vincent Povirk <madewokherd(a)gmail.com> wrote: > static HRESULT WINAPI BitmapScaler_GetSize(IWICBitmapScaler *iface, > UINT *puiWidth, UINT *puiHeight) > { > - FIXME("(%p,%p,%p): stub\n", iface, puiWidth, puiHeight); > + BitmapScaler *This = impl_from_IWICBitmapScaler(iface); > + TRACE("(%p,%p,%p)\n", iface, puiWidth, puiHeight); > > - return E_NOTIMPL; > + if (!This->source) > + return WINCODEC_ERR_WRONGSTATE; > + > + *puiWidth = This->width; > + *puiHeight = This->height; > + > + return S_OK; This patch and remaning patches in the sequence completely miss any parameter checking. -- Dmitry.

2 1

Re: d3dx9_36 [patch 2 of 3]: Implementation of D3DXSHScale
by Matteo Bruni June 27, 2012

June 27, 2012

2012/6/27 Nozomi Kodama <nozomi.kodama(a)yahoo.com>: > -@ stub D3DXSHScale(ptr long ptr ptr) +@ stdcall D3DXSHScale(ptr long ptr ptr) ... +FLOAT* WINAPI D3DXSHScale(FLOAT *out, UINT order, CONST FLOAT *a, CONST FLOAT scale) The last parameter doesn't match between the function prototype and the spec entry. Also, apparently MSDN agrees with your prototype, while the docs in the last DirectX SDK have a pointer there. I guess the tests demonstrate that the parameter is actually a plain float... + for (i = 0; i < 100; i++) + { + a[i] = (FLOAT)i; + b[i] = (FLOAT)i; + } No need to cast. + D3DXSHScale(b, order, a, 5.0f); Not important, but you could also verify that the returned pointer is == b. +/* D3DXSHScale does not modify the elements of the array after the order * order-th element */ Can you please indent those comments? + return; +} That's not needed at all. Also, while you're at it, you should probably merge the two for loops in the test, like Francois did in http://www.winehq.org/pipermail/wine-patches/2012-June/115464.html

1 0

Re: wbemprox: Add a Win32_LogicalDisk class stub.
by Hans Leidekker June 27, 2012

June 27, 2012

On Wed, 2012-06-27 at 11:34 +0300, John Yani wrote: > +static void fill_logicaldisk( struct table *table ) > +{ > + static const WCHAR fmtW[] = {'%','u',0}; > + WCHAR device_id[11]; > + struct record_logicaldisk *rec; > + UINT num_rows = 1; > + FIXME("returns a fake logical disks table\n"); > + if (!(table->data = heap_alloc( sizeof(*rec) * num_rows ))) > + { > + return; > + } > + rec = (struct record_networkadapter *)(table->data); This should be (struct record_logicaldisk *)(table->data); > + sprintfW( device_id, fmtW, 0 ); > + rec->device_id = heap_strdupW( device_id ); > + > + TRACE("created %u rows\n", num_rows); > + table->num_rows = num_rows; > +} You don't need the num_rows variable since it's a fixed value. You also don't need to build the device id, it can just be a static string.

1 0

Re: [4/4] gdiplus: Add a bunch of tests for TIFF image properties.
by Marvin June 27, 2012

June 27, 2012

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=19595 Your paranoid android. === W7PROX64 (64 bit image) === image.c:2812: Test failed: 0: expected property size 17 or 36, got 44 image.c:2812: Test failed: 2: expected property size 144 or 0, got 152 image.c:2812: Test failed: 3: expected property size 20 or 0, got 28 === TEST64_W7SP1 (64 bit image) === image.c:2812: Test failed: 0: expected property size 17 or 36, got 44 image.c:2812: Test failed: 2: expected property size 144 or 0, got 152 image.c:2812: Test failed: 3: expected property size 20 or 0, got 28

1 0

Re: d3dx9_36 [patch 3 of 3]: Implementation of D3DXSHMultiply2
by Marvin June 27, 2012

June 27, 2012

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=19586 Your paranoid android. === WINEBUILD (build) === Patch failed to apply

1 0

Re: d3dx9_36 [patch 2 of 3]: Implementation of D3DXSHScale
by Marvin June 27, 2012

June 27, 2012

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=19585 Your paranoid android. === WINEBUILD (build) === Patch failed to apply

1 0