June 2008 - Wine-devel - list.winehq.org

Re: [d3d8] Fix CreateImageSurface implementation (with testcase)
by H. Verbeet 20 Jun '08

20 Jun '08

2008/6/20 Tobias Jakobi <liquid.acid(a)gmx.net>: > Patch fixes a problem with the IDirect3DDevice8::CreateImageSurface > implementation (wrong pool type of the returned surface object). This fixes > savegame screenshot bugs (black images instead of game scenes) in both Max > Payne and Max Payne 2. > > Affected bugs: > http://bugs.winehq.org/show_bug.cgi?id=9775 > http://bugs.winehq.org/show_bug.cgi?id=7801 > > Test case is included and verifies behaviour on both Windows XP and Vista. > This is my first patch for wine and also my first post on the wine-patches > ml, so if I'm doing anything wrong here please tell me and I try to do > better next time :-) > > Greets, > Tobias Jakobi > The change itself looks ok to me, provided the test passes on XP. Don't mix tabs and spaces though: > + /* > + D3DPOOL_DEFAULT = 0 > + D3DPOOL_MANAGED = 1 > + D3DPOOL_SYSTEMMEM = 2 > + D3DPOOL_SCRATCH = 3 > + */ I'm not sure this adds much (not that the original comment did either): > - /*MSDN: D3DPOOL_SCRATCH will return a surface that has identical characteristics to a surface created by the Microsoft DirectX 8.x method CreateImageSurface. > + /*MSDN: D3DPOOL_SCRATCH will return a surface that has identical characteristics to a surface created by the Microsoft DirectX 8.x method CreateImageSurface (quote from the DirectX9 documentation). > + However this is wrong behaviour and conflicting with the explanation from the original DirectX8 documentation. According to DX8 docs it should be D3DPOOL_SYSTEMMEM and NOT D3DPOOL_SCRATCH. > + For a testcase of the DX8 CreateImageSurface method look for test_image_surface_pool in dlls/d3d8/tests/surface.c It would probably be better to just describe the behaviour of D3DPOOL_SCRATCH there.

1 0

re: ntdll: rtlstr.c: Implement checking for control characters in RtlIsTextUnicode [try 2]
by Dan Kegel 20 Jun '08

20 Jun '08

The array doesn't make much sense if you're not going to iterate over it. It might be more readable if you replaced the array and long if with a switch statement. It would be nice to break out of the loop early. The last two suggestions together kind of mean a goto, which is ugly, but not too uncommon in wine. (You might want to verify that the generated code isn't too slow, but since len is limited to 256, that probably doesn't matter.) e.g. if (flags & IS_TEXT_UNICODE_CONTROLS) for (i = 0; i < len; i++) switch (s[i]) { case '\t': case '\n': case 'r': case 0x20: out_flags |= IS_TEXT_UNICODE_CONTROLS; goto done; default: } } done:

4 5

win9x and setupapi (devinst)
by Paul Vriens 20 Jun '08

20 Jun '08

Hi, I'm planning to do some work on setupapi. Most of the tests we are having produce either rubbish or fail on win9x. How much effort should I make to have win9x tests? It appears that every single test I'm touching is totally different when we compare win9x to the rest (NT+). Making sure we are not flooded with errors means skipping a lot of tests now already. I'm thinking of creating a "check for win9x" at the start of the devinst tests and skip them all if we are on win9x. Ideas, comments, suggestions? -- Cheers, Paul.

1 0

Re: dbghelp: Implement SymEnumerateSymbols64
by Eric Pouech 20 Jun '08

20 Jun '08

Jon Griffiths a écrit : > Hi, > > Title says it all. > > Cheers > Jon > > > > > > ------------------------------------------------------------------------ > > -@ stub SymEnumerateSymbols64s > +@ stdcall SymEnumerateSymbols64(long long long ptr ptr) > should be: @ stdcall SymEnumerateSymbols64(long double ptr ptr) -- Eric Pouech "The problem with designing something completely foolproof is to underestimate the ingenuity of a complete idiot." (Douglas Adams)

2 1

Re: winemenubuilder, wineshelllink: move escaping code into wineshelllink
by Vincent Povirk 20 Jun '08

20 Jun '08

Ignore that. I used Git wrong and ended up with an old incorrect revision. Vincent Povirk On Thu, Jun 19, 2008 at 9:59 PM, Vincent Povirk <madewokherd+d41d(a)gmail.com> wrote: > This change moves the escape functionality (replacing \ with \\) out > of winemenubuilder and into wineshelllink. The .desktop files > generated by winemenubuilder have the same contents as before. > > Because escaping is a function of the .desktop file format, it makes > more sense for wineshelllink to do the escaping. > > This will make a simpler fix possible for bug 13303 (which requires \ > to be escaped as \\\\, but only in the Exec field of .desktop files), > and a similar approach will be needed in wineshelllink for bug 13963 > (since paths may or may not need escaping depending on whether xml or > a .directory file is being written). > > Vincent Povirk >

1 0

Re: oleview: Add Russian translation
by Kirill K. Smirnov 19 Jun '08

19 Jun '08

Hi, > + IDS_OPEN "Октрыть" Typo. > + IDM_EXPERT, "Переключится между режимом эксперт/новичок" ь is missing in word 'Переключится'. -- Kirill

1 0

Internet Explorer 6, Wine1
by Leslie Viljoen 19 Jun '08

19 Jun '08

I can follow this nice howto if someone thinks there's a chance of it succeeding: http://gentoo-wiki.com/HOWTO_Install_Internet_Explorer_6_SP1#Internet_Explo… ..then I can report back. Should I bother? I have already tried ies4linux and it failed to install a working ie6. Les

1 0

Re: ntdll: rtlstr.c: Implement checking for control characters in, RtlIsTextUnicode [try 3]
by Zac Brown 19 Jun '08

19 Jun '08

Zac Brown wrote: > Implement setting of IS_TEXT_UNICODE_CONTROLS and > IS_TEXT_UNICODE_REVERSE_CONTROLS when control characters are present in > RtlIsTextUnicode. > > -Zac Brown > > > ------------------------------------------------------------------------ > > --- > dlls/ntdll/rtlstr.c | 57 +++++++++++++++++++++++++++++++++++++++++++++ > dlls/ntdll/tests/rtlstr.c | 2 - > 2 files changed, 57 insertions(+), 2 deletions(-) > > diff --git a/dlls/ntdll/rtlstr.c b/dlls/ntdll/rtlstr.c > index deec931..ead8974 100644 > --- a/dlls/ntdll/rtlstr.c > +++ b/dlls/ntdll/rtlstr.c > @@ -1591,6 +1591,23 @@ NTSTATUS WINAPI RtlFindCharInUnicodeString( > */ > BOOLEAN WINAPI RtlIsTextUnicode( LPCVOID buf, INT len, INT *pf ) > { > + /* > + * Enum for possible control characters. > + * BR = byte reversed, for use in byte reversed strings. > + */ > + enum control_chars { > + TAB = '\t', > + BR_TAB = 0x0900, > + CRET = '\r', > + BR_CRET = 0x0d00, > + NEWLINE = '\n', > + BR_NEWLINE = 0x0a00, > + SPACE = ' ', > + BR_SPACE = 0x2000, > + CJKSPACE = 0x3000, > + BR_CJKSPACE = 0x0030 > + }; > + > const WCHAR *s = buf; > int i; > unsigned int flags = ~0U, out_flags = 0; > @@ -1650,6 +1667,46 @@ BOOLEAN WINAPI RtlIsTextUnicode( LPCVOID buf, INT len, INT *pf ) > } > } > > + /* Check for control characters in string. */ > + if (flags & IS_TEXT_UNICODE_CONTROLS) > + { > + for (i = 0; i < len; i++) > + { > + switch (s[i]) > + { > + case TAB: > + case CRET: > + case NEWLINE: > + case SPACE: > + case CJKSPACE: > + out_flags |= IS_TEXT_UNICODE_CONTROLS; > + break; > + default: > + continue; > + } > + } > + } > + > + /* Check for control characters in byte reversed string. */ > + if ((flags & IS_TEXT_UNICODE_REVERSE_CONTROLS) && (out_flags & IS_TEXT_UNICODE_REVERSE_SIGNATURE)) > + { > + for (i = 0; i < len; i++) > + { > + switch (s[i]) > + { > + case BR_TAB: > + case BR_CRET: > + case BR_NEWLINE: > + case BR_SPACE: > + case BR_CJKSPACE: > + out_flags |= IS_TEXT_UNICODE_REVERSE_CONTROLS; > + break; > + default: > + continue; > + } > + } > + } > + > if (pf) > { > out_flags &= *pf; > diff --git a/dlls/ntdll/tests/rtlstr.c b/dlls/ntdll/tests/rtlstr.c > index aa4a391..5d77f9a 100644 > --- a/dlls/ntdll/tests/rtlstr.c > +++ b/dlls/ntdll/tests/rtlstr.c > @@ -1689,7 +1689,6 @@ static void test_RtlIsTextUnicode(void) > > flags = IS_TEXT_UNICODE_UNICODE_MASK; > ok(pRtlIsTextUnicode(unicode, sizeof(unicode), &flags), "Text should not pass a Unicode\n"); > - todo_wine > ok(flags == (IS_TEXT_UNICODE_STATISTICS | IS_TEXT_UNICODE_CONTROLS), > "Expected flags 0x6, obtained %x\n", flags); > > @@ -1718,7 +1717,6 @@ static void test_RtlIsTextUnicode(void) > > flags = IS_TEXT_UNICODE_REVERSE_MASK; > ok(!pRtlIsTextUnicode(be_unicode, sizeof(unicode) + 2, &flags), "Reverse endian should be Unicode\n"); > - todo_wine > ok(flags == (IS_TEXT_UNICODE_REVERSE_CONTROLS | IS_TEXT_UNICODE_REVERSE_SIGNATURE), > "Expected flags 0xc0, obtained %x\n", flags); > HeapFree(GetProcessHeap(), 0, be_unicode); > > > ------------------------------------------------------------------------ > > One more time... ignore this patch. Will send update later. -Zac

1 0

Re: Remove LATER and REMIND bug resolutions
by Maarten Lankhorst 19 Jun '08

19 Jun '08

Hi Lei, 2008/6/17 Lei Zhang <thestig(a)google.com>: > It's been a week and nobody has responded to Dmitry's bug report [1] > yet, so I'm going to bring it up here. I am all for it, does anyone > have objections? Mark it as remind, I'll look at it later. On a more serious note, i agree that they should be removed, we have milestones that we could use for this instead. Cheers, Maarten. PS: Not my day. First sent reply privately, then to wine-patches >:(

1 0

Re: [PATCH 1/2] shell32: SHFileOperation: test added [try 2]
by James Hawkins 19 Jun '08

19 Jun '08

2008/6/18 Vitaly Perov <vitperov(a)etersoft.ru>: > Changelog: > - shell32: test added. It check whether SHFileOperation could correctly work > with input files mask '*.*' > Couple problems: --- a/dlls/shell32/tests/shlfileop.c +++ b/dlls/shell32/tests/shlfileop.c @@ -120,6 +120,7 @@ static void init_shfo_tests(void) CreateDirectoryA("testdir2", NULL); CreateDirectoryA("testdir2\\nested", NULL); createTestFile("testdir2\\one.txt"); + createTestFile("testdir2\\one2.txt"); createTestFile("testdir2\\nested\\two.txt"); } @@ -136,11 +137,14 @@ static void clean_after_shfo_tests(void) DeleteFileA("test4.txt\\test3.txt"); RemoveDirectoryA("test4.txt"); DeleteFileA("testdir2\\one.txt"); + DeleteFileA("testdir2\\one2.txt"); DeleteFileA("testdir2\\test1.txt"); DeleteFileA("testdir2\\test2.txt"); DeleteFileA("testdir2\\test3.txt"); DeleteFileA("testdir2\\test4.txt\\test1.txt"); DeleteFileA("testdir2\\nested\\two.txt"); + DeleteFileA("testdir2\\nested\\one.txt"); + DeleteFileA("testdir2\\nested\\one2.txt"); RemoveDirectoryA("testdir2\\test4.txt"); RemoveDirectoryA("testdir2\\nested"); RemoveDirectoryA("testdir2"); Why are you adding another test file? There are already multiple *.* files created in the test directory. + /* try moving with MASK '*.*' */ + shfo.pFrom = "testdir2\\*.*\0"; + shfo.pTo = "testdir2\\nested\0"; + retval = SHFileOperationA(&shfo); + ok(retval == ERROR_SUCCESS, "Expected ERROR_SUCCESS, got %d\n", retval); + ok(!file_exists("testdir2\\one.txt"), "Expected one.txt to not exist\n"); + ok(!file_exists("testdir2\\one2.txt"), "Expected one2.txt to not exist\n"); + ok(file_exists("testdir2\\nested\\one.txt"), "Expected one.txt to exist\n"); + ok(file_exists("testdir2\\nested\\one2.txt"), "Expected one.txt to exist\n"); This test is misleading. We already test for moving multiple files to a destination directory using a mask: line 1102 of the git version of this file. The misleading part is that what we don't test is moving multiple files to a destination directory when the FOF_MULTIDESTFILES flag is set (it's set in a previous test and never unset in your patch). This would be a much better addition to the tests (and easier to understand because it follows from the previous test): /* move files using glob to a dest dir with FOF_MULTIDESTFILES */ shfo.fFlags |= FOF_MULTIDESTFILES; set_curr_dir_path(from, "test?.txt\0"); set_curr_dir_path(to, "testdir2\0"); ok(!file_exists("testdir2\\test2.txt"), "The file is not moved yet\n"); ok(!file_exists("testdir2\\test4.txt"), "The directory is not moved yet\n"); retval = SHFileOperationA(&shfo); ok(retval == ERROR_SUCCESS, "Files and directories are moved to directory\n"); ok(file_exists("testdir2\\test2.txt"), "The file is moved\n"); ok(file_exists("testdir2\\test4.txt"), "The directory is moved\n"); ok(file_exists("testdir2\\test4.txt\\test1.txt"), "The file in subdirectory is moved\n"); This should immediately follow the similar test on line 1102. -- James Hawkins

2 2