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

Re: [PATCH 1/5] [WineHelp]: added basic infrastructure for a new wineconsole mode: line
by Frédéric Delanoy Dec. 4, 2012

Dec. 4, 2012

On Mon, Dec 3, 2012 at 10:12 PM, Eric Pouech <eric.pouech(a)orange.fr> wrote: > diff --git a/programs/wineconsole/line.c b/programs/wineconsole/line.c > new file mode 100644 > index 0000000..4b7fd29 > --- /dev/null > +++ b/programs/wineconsole/line.c > @@ -0,0 +1,334 @@ > +/* > + * a GUI application for displaying a console > + * Line back end > + * > + * Copyright 2002 Eric Pouech Guess that should be 2012 ;) Frédéric

1 0

Re: wshom: Basic support for REG_SZ values in RegRead() (try2)
by Jacek Caban Dec. 4, 2012

Dec. 4, 2012

Hi Nikolay, On 12/04/12 10:33, Nikolay Sivov wrote: > + if (name[len] != '\\') > + return HRESULT_FROM_WIN32(ERROR_PATH_NOT_FOUND); > + /* skip '\' after root name */ > + len++; > + > + /* at this point we have at least one '\' in path */ > + val = strrchrW(name, '\\'); > + keyname = SysAllocStringLen(&name[len], val-&name[len]); It's better, but now if there is only one backslash, you will have val-&name[len] == -1. Thanks, Jacek

1 0

Re: wshom: Basic support for REG_SZ values in RegRead()
by Jacek Caban Dec. 4, 2012

Dec. 4, 2012

Hi Nikolay, On 12/03/12 12:29, Nikolay Sivov wrote: > + /* skip '\' after root name */ > + if (name[len] == '\\') len++; It looks like if name[len] != '\\', then name should be considered invalid. > + > + val = strlenW(name); > + while (name[--val] != '\\' && val) > + ; > + keyname = SysAllocStringLen(&name[len], val-len); This won't work if name has no backslashes. > + /* skip leading '\' */ > + val++; > + > + type = REG_NONE; > + datalen = 0; > + ret = RegGetValueW(root, keyname, &name[val], RRF_RT_ANY, &type, NULL, &datalen); > + if (ret == ERROR_SUCCESS) > + { > + if (type == REG_SZ) > + { > + V_VT(value) = VT_BSTR; > + V_BSTR(value) = SysAllocStringLen(NULL, datalen-1); > + ret = RegGetValueW(root, keyname, &name[val], RRF_RT_ANY, &type, V_BSTR(value), &datalen); > + } > + else > + FIXME("type %d not handled\n", type); This should return an error. Jacek

2 1

Re: [PATCH] ddraw: Create a dynamic buffer if DDLOCK_DISCARDCONTENTS is used
by Rico Schüller Dec. 4, 2012

Dec. 4, 2012

On 03.12.2012 23:11, Stefan Dösinger wrote: > + DWORD pool = buffer->Caps & D3DVBCAPS_SYSTEMMEMORY ? WINED3D_POOL_SYSTEM_MEM : WINED3D_POOL_DEFAULT; ... > + return wined3d_buffer_create_vb(buffer->ddraw->wined3d_device, > + buffer->size, usage, pool, buffer, &ddraw_null_wined3d_parent_ops, > + wined3d_buffer); Is there a reason why you don't use "enum wined3d_pool" for pool? Cheers Rico

1 0

Re: [3/3] gdiplus: Bitmaps created from a 32-bit dib should be PARGB.
by Dmitry Timoshkov Dec. 4, 2012

Dec. 4, 2012

Vincent Povirk <madewokherd(a)gmail.com> wrote: > case 32: > - format = PixelFormat32bppRGB; > + if (bm.bmBits) > + format = PixelFormat32bppPARGB; > + else > + format = PixelFormat32bppRGB; Besides the test failure, to my eyes format = bm.bmBits ? PixelFormat32bppPARGB : PixelFormat32bppRGB; looks shorter and clearer. -- Dmitry.

2 1

Re: [3/3] gdiplus: Bitmaps created from a 32-bit dib should be PARGB.
by Marvin Dec. 3, 2012

Dec. 3, 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=23202 Your paranoid android. === WXPPROSP3 (32 bit image) === image.c:962: Test failed: Expected 925707, got 139273 image.c:1075: Test failed: Expected 925707, got 139273 === W2K3R2SESP2 (32 bit image) === image.c:962: Test failed: Expected 925707, got 139273 image.c:1075: Test failed: Expected 925707, got 139273 === WVISTAADM (32 bit image) === image.c:962: Test failed: Expected 925707, got 139273 image.c:1075: Test failed: Expected 925707, got 139273 === W2K8SE (32 bit image) === image.c:962: Test failed: Expected 925707, got 139273 image.c:1075: Test failed: Expected 925707, got 139273 === W7PRO (32 bit image) === image.c:962: Test failed: Expected 925707, got 139273 image.c:1075: Test failed: Expected 925707, got 139273 === W7PROX64 (32 bit image) === image.c:962: Test failed: Expected 925707, got 139273 image.c:1075: Test failed: Expected 925707, got 139273 === TEST64_W7SP1 (32 bit image) === image.c:962: Test failed: Expected 925707, got 139273 image.c:1075: Test failed: Expected 925707, got 139273 === W7PROX64 (64 bit image) === image.c:962: Test failed: Expected 925707, got 139273 image.c:1075: Test failed: Expected 925707, got 139273 === TEST64_W7SP1 (64 bit image) === image.c:962: Test failed: Expected 925707, got 139273 image.c:1075: Test failed: Expected 925707, got 139273

1 0

Re: kernel32: Implement SetFileValidData. (try 2)
by Alexandre Julliard Dec. 3, 2012

Dec. 3, 2012

Hans Leidekker <hans(a)codeweavers.com> writes: > QuickBooks 2008 calls this function. > --- > configure.ac | 1 + > dlls/kernel32/file.c | 12 +++- > dlls/kernel32/tests/file.c | 136 ++++++++++++++++++++++++++++++++++++++++++++ > dlls/ntdll/file.c | 28 +++++++++ > include/ddk/ntddk.h | 5 ++ > include/winbase.h | 1 + > 6 files changed, 181 insertions(+), 2 deletions(-) Still no luck: ../../../tools/runtest -q -P wine -M kernel32.dll -T ../../.. -p kernel32_test.exe.so file.c && touch file.ok file.c:940: Tests skipped: Either no authority to volume, or is todo_wine for C:\users\julliard\Temp\ err=5 should be 3 file.c:940: Tests skipped: Either no authority to volume, or is todo_wine for C:\users\julliard\Temp\removeme\ err=5 should be 3 file.c:940: Tests skipped: Either no authority to volume, or is todo_wine for C:\ err=5 should be 3 file.c:940: Tests skipped: Either no authority to volume, or is todo_wine for \\?\C:\ err=5 should be 3 file.c:3528: Test succeeded inside todo block: SetFileValidData succeeded file.c:3532: Test failed: SetFileValidData failed 50 file.c:3548: Test succeeded inside todo block: SetFileValidData succeeded file.c:3552: Test failed: SetFileValidData failed 50 file.c:3555: Test failed: SetFileValidData failed 50 file.c:3559: Test succeeded inside todo block: SetFileValidData succeeded file.c:3563: Test failed: SetFileValidData failed 50 make: *** [file.ok] Error 7 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 3/3] shell32/tests: Add tests for DoEnvironmentSubstA/W
by Alexandre Julliard Dec. 3, 2012

Dec. 3, 2012

Detlef Riekenberg <wine.dev(a)web.de> writes: > + /* Not all variables are supported before vista */ > + if (broken(*bufferA == '%')) > + trace("%d: %s not supported\n", i, bufferA); > + else > + { > + if (i < 3) todo_wine > + ok(HIWORD(res) && (LOWORD(res) == (lstrlenA(bufferA)+1)) && (*bufferA != '%'), > + "%d: got %d/%d and %s (len %d)\n", > + i, HIWORD(res), LOWORD(res), bufferA, lstrlenA(bufferA)); > + > + if (i < 3) todo_wine > + ok(HIWORD(res2) && (LOWORD(res2) == (lstrlenW(bufferW)+1)) && (*bufferW != '%'), > + "%d: got %d/%d and %s (len %d)\n", > + i, HIWORD(res2), LOWORD(res2), wine_dbgstr_w(bufferW), lstrlenW(bufferW)); > + } > + } This doesn't make sense, you are only testing the cases that don't work. You should pick a few variables that work everywhere, and test them properly, including the resulting buffer contents. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: wshom: Basic support for REG_SZ values in RegRead()
by Marvin Dec. 3, 2012

Dec. 3, 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=23189 Your paranoid android. === WINEBUILD (build) === Can't determine base name

2 1

Re: comdlg32: Resize Find dialog to fit translations
by Dmitry Timoshkov Dec. 3, 2012

Dec. 3, 2012

Aurimas Fišeras <aurimas(a)gmail.com> wrote: > Controls resized to fit translations for more languages. Tested with > resource editor. Did you see a comment in that file just before dialog templates? -- Dmitry.

2 4