March 2007 - Wine-devel - list.winehq.org

Question about OpenGL/D3D
by Mike Schaadt 09 Mar '07

09 Mar '07

Has anyone tested using standard window objects in a window that also had an opengl/d3d context? I believe Terragen 2 is attempting to do this. If this hasn't been tested, than it might be the cause of all of the static that appears non opengl area, making the program nearly unusable. No messages get printed to the console from wine(Terragen 2 prints out a fair bit of debug info there, but nothing from wine that would cause the problem, just a few warnings that shouldn't be related). Any thoughts?

4 4

Re: Shader instruction table whitespace cleanup
by Ivan Gyurdiev 09 Mar '07

09 Mar '07

Kevin Wallerbos wrote: > This formatting patch cleans up the spacing in the pixel/vertexshader > instruction table, making it somewhat nicer on the eyes. The lines > have become a bit longer than practical on standard resolutions, but > that shouldn't be a problem as most lines were already too long. A better thing to do is to remove the useless GLNAME_REQUIRE_GLSL, and "undefined" instruction names. In fact, it may be nice to break out the instruction names out of this table completely, since they are only necessary for some instructions, required by the map2gl function. This table is also organized poorly, you have to search through the whole thing, and then do version comparisons to find anything. If possible, it would be great to make it O(1) accessible. ===== Also, as you point out: Before patch : "most lines were already too long" [ I'd say some lines were too long ] After patch: "all lines are too long" Not sure what is gained... please don't force others to your resolution preferences. [ looking at this on a widescreen at 1680 x 1050 and it still wraps! ]

1 0

Re: user32: avoid NULL pointer access in DefWindowProcA WM_NCCREATE
by Felix Nawothnig 09 Mar '07

09 Mar '07

Jan Zerebecki wrote: > CREATESTRUCTA *cs = (CREATESTRUCTA *)lParam; > /* check for string, as static icons, bitmaps (SS_ICON, SS_BITMAP) > * may have child window IDs instead of window name */ > - if (HIWORD(cs->lpszName)) > + if (cs && HIWORD(cs->lpszName)) cs is never NULL at that point. Felix

2 3

Don't compare file handles to NULL
by Francois Gouget 09 Mar '07

09 Mar '07

Jacek just sent a patch fixing the following bug: file = CreateFileW(fileName, GENERIC_READ, 0, NULL, OPEN_EXISTING, FILE_ATTRIBUTE_READONLY,NULL); - if(file) { + if(file != INVALID_HANDLE_VALUE) { This looks like something that could be turned into a nice Smatch test. Hint, hint... -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ In theory, theory and practice are the same, but in practice they're different.

4 13

Re: comctl32: treeview: Message sequencing tests
by Marcelo Duarte 09 Mar '07

09 Mar '07

Chris Peterson escreveu: > This is another patch for my CS130 assignment. It adds message > sequencing tests. Other changes include: I updated one of my previous > tests which was not testing the implied message; I changed a > SendMessageW to SendMessageA call in DoFocusTest to fix a problem on > Windows 98 due to a lack of default unicode support. Any > comments/suggestions are welcome. > Thanks! > -Chris > Hi! > > @@ -166,7 +315,8 @@ static void DoFocusTest(void) > assert(hChild2); > > ShowWindow(hMainWnd,SW_SHOW); > - SendMessageW(hTree, TVM_SELECTITEM, TVGN_CARET, (LPARAM)hChild); > + /* Using SendMessageA since Win98 doesn't have default unicode support */ > + SendMessageA(hTree, TVM_SELECTITEM, TVGN_CARET, (LPARAM)hChild); > Use SendMessage() instead. > static void TestGetSetInsertMarkColor(void) > { > COLORREF crColor = RGB(0,0,0); > - SendMessage( hTree, TVM_SETBKCOLOR, 0, crColor ); > - crColor = (COLORREF)SendMessage( hTree, TVM_GETBKCOLOR, 0, 0 ); > + SendMessage( hTree, TVM_SETINSERTMARKCOLOR, 0, crColor ); > + crColor = (COLORREF)SendMessage( hTree, TVM_GETINSERTMARKCOLOR, 0, 0 ); > ok(crColor == RGB(0,0,0), "Insert mark color reported as 0x%.8x, expected 0x00000000\n", crColor); >

2 1

Re: gdi32: Add a GetTextMetrics test, make it pass under Wine
by Lei Zhang 09 Mar '07

09 Mar '07

I tested it out on a couple machines and found many failure cases. (output attached) As a sanity check, I copied the deja vu family of fonts to a windows box and they passed the tests there. On 3/9/07, Dmitry Timoshkov <dmitry(a)codeweavers.com> wrote: > Hello, > > this patch should partially resolve a problem reported in the bug 5783. > > In Wine GetTextMetricsW returns wrong tmFirstChar/tmLastChar, and > GetFontUnicodeRanges returns wrong ranges for fonts with SYMBOL_CHARSET. > > Changelog: > gdi32: Add a GetTextMetrics test, make it pass under Wine. > > --- > dlls/gdi32/font.c | 9 ++- > dlls/gdi32/freetype.c | 4 +- > dlls/gdi32/tests/font.c | 141 ++++++++++++++++++++++++++++++++++++++++++++++- > 3 files changed, 149 insertions(+), 5 deletions(-) > > diff --git a/dlls/gdi32/font.c b/dlls/gdi32/font.c > index 1b1d8b4..785a990 100644 > --- a/dlls/gdi32/font.c > +++ b/dlls/gdi32/font.c > @@ -287,10 +287,12 @@ static void FONT_TextMetricWToA(const TEXTMETRICW *ptmW, LPTEXTMETRICA ptmA ) > ptmA->tmOverhang = ptmW->tmOverhang; > ptmA->tmDigitizedAspectX = ptmW->tmDigitizedAspectX; > ptmA->tmDigitizedAspectY = ptmW->tmDigitizedAspectY; > - ptmA->tmFirstChar = ptmW->tmFirstChar > 255 ? 255 : ptmW->tmFirstChar; > + ptmA->tmBreakChar = ptmW->tmBreakChar > 255 ? 255 : ptmW->tmBreakChar; > + /* This appears to be what Windows does */ > + ptmA->tmFirstChar = ptmW->tmBreakChar; > + if (ptmA->tmFirstChar >= 2) ptmA->tmFirstChar -= 2; > ptmA->tmLastChar = ptmW->tmLastChar > 255 ? 255 : ptmW->tmLastChar; > ptmA->tmDefaultChar = ptmW->tmDefaultChar > 255 ? 255 : ptmW->tmDefaultChar; > - ptmA->tmBreakChar = ptmW->tmBreakChar > 255 ? 255 : ptmW->tmBreakChar; > ptmA->tmItalic = ptmW->tmItalic; > ptmA->tmUnderlined = ptmW->tmUnderlined; > ptmA->tmStruckOut = ptmW->tmStruckOut; > @@ -1366,6 +1368,9 @@ BOOL WINAPI GetTextMetricsW( HDC hdc, TEXTMETRICW *metrics ) > /* device layer returns values in device units > * therefore we have to convert them to logical */ > > + metrics->tmDigitizedAspectX = GetDeviceCaps(hdc, LOGPIXELSX); > + metrics->tmDigitizedAspectY = GetDeviceCaps(hdc, LOGPIXELSY); > + > #define WDPTOLP(x) ((x<0)? \ > (-abs(INTERNAL_XDSTOWS(dc, (x)))): \ > (abs(INTERNAL_XDSTOWS(dc, (x))))) > diff --git a/dlls/gdi32/freetype.c b/dlls/gdi32/freetype.c > index db990fa..03adfe5 100644 > --- a/dlls/gdi32/freetype.c > +++ b/dlls/gdi32/freetype.c > @@ -4480,7 +4480,7 @@ static DWORD get_font_unicode_ranges(FT_Face face, GLYPHSET *gs) > { > DWORD num_ranges = 0; > > - if (face->charmap->encoding == FT_ENCODING_UNICODE && pFT_Get_First_Char) > + if (pFT_Get_First_Char) > { > FT_UInt glyph_code; > FT_ULong char_code, char_code_prev; > @@ -4488,7 +4488,7 @@ static DWORD get_font_unicode_ranges(FT_Face face, GLYPHSET *gs) > glyph_code = 0; > char_code_prev = char_code = pFT_Get_First_Char(face, &glyph_code); > > - TRACE("face encoding FT_ENCODING_UNICODE, number of glyphs %ld, first glyph %u, first char %04lx\n", > + TRACE("number of glyphs %ld, first glyph %u, first char %04lx\n", > face->num_glyphs, glyph_code, char_code); > > if (!glyph_code) return 0; > diff --git a/dlls/gdi32/tests/font.c b/dlls/gdi32/tests/font.c > index 43e9c3a..a996ff0 100644 > --- a/dlls/gdi32/tests/font.c > +++ b/dlls/gdi32/tests/font.c > @@ -408,7 +408,7 @@ static void test_GdiGetCharDimensions(void) > > if (!pGdiGetCharDimensions) > { > - skip("GetFontUnicodeRanges not available on this platform\n"); > + skip("GdiGetCharDimensions not available on this platform\n"); > return; > } > > @@ -1149,6 +1149,7 @@ static void test_GetFontUnicodeRanges(void) > > size = pGetFontUnicodeRanges(hdc, gs); > ok(size, "GetFontUnicodeRanges failed\n"); > + ok(gs->cRanges != 0, "GetFontUnicodeRanges returned empty ranges\n"); > #if 0 > for (i = 0; i < gs->cRanges; i++) > trace("%03d wcLow %04x cGlyphs %u\n", i, gs->ranges[i].wcLow, gs->ranges[i].cGlyphs); > @@ -1397,6 +1398,143 @@ else > ReleaseDC(0, hdc); > } > > +static void test_text_metrics(const LOGFONTA *lf) > +{ > + HDC hdc; > + HFONT hfont, hfont_old; > + TEXTMETRICA tmA; > + TEXTMETRICW tmW; > + UINT first_unicode_char = 0, last_unicode_char = 0; > + INT ret, test_char; > + const char *font_name = lf->lfFaceName; > + > + trace("Testing font metrics for %s, charset %d\n", font_name, lf->lfCharSet); > + > + hdc = GetDC(0); > + > + SetLastError(0xdeadbeef); > + hfont = CreateFontIndirectA(lf); > + ok(hfont != 0, "CreateFontIndirect error %u\n", GetLastError()); > + > + hfont_old = SelectObject(hdc, hfont); > + > + if (pGetFontUnicodeRanges) > + { > + DWORD size; > + GLYPHSET *gs; > + > + size = pGetFontUnicodeRanges(hdc, NULL); > + ok(size, "GetFontUnicodeRanges failed unexpectedly\n"); > + > + gs = HeapAlloc(GetProcessHeap(), 0, size); > + > + size = pGetFontUnicodeRanges(hdc, gs); > + ok(size, "GetFontUnicodeRanges failed\n"); > + ok(gs->cRanges != 0, "GetFontUnicodeRanges returned empty ranges\n"); > + > + first_unicode_char = gs->ranges[0].wcLow; > + last_unicode_char = gs->ranges[gs->cRanges - 1].wcLow + gs->ranges[gs->cRanges - 1].cGlyphs - 1; > + trace("for %s first char %x, last char %x\n", font_name, > + first_unicode_char, last_unicode_char); > + > + HeapFree(GetProcessHeap(), 0, gs); > + } > + > + SetLastError(0xdeadbeef); > + ret = GetTextMetricsA(hdc, &tmA); > + ok(ret, "GetTextMetricsA error %u\n", GetLastError()); > + > + trace("A: first %x, last %x, default %x, break %x\n", > + tmA.tmFirstChar, tmA.tmLastChar, tmA.tmDefaultChar, tmA.tmBreakChar); > + > + SetLastError(0xdeadbeef); > + ret = GetTextMetricsW(hdc, &tmW); > + ok(ret, "GetTextMetricsA error %u\n", GetLastError()); > + > + trace("W: first %x, last %x, default %x, break %x\n", > + tmW.tmFirstChar, tmW.tmLastChar, tmW.tmDefaultChar, tmW.tmBreakChar); > + > + /* This appears to be what Windows does */ > + test_char = tmW.tmBreakChar; > + if (test_char >= 2) test_char -= 2; > + if (test_char > 255) test_char = 255; > + ok(tmA.tmFirstChar == test_char, "tmFirstChar for %s %02x != %02x\n", > + font_name, tmA.tmFirstChar, test_char); > + > + /* This test failes for Marlett, Wingdings 2 and Wingdings 3, but succeeds > + * for all other 327 truetype fonts on my system. > + */ > +#if 0 > + test_char = tmW.tmLastChar > 255 ? 255 : tmW.tmLastChar; > + ok(tmA.tmLastChar == test_char, "tmLastChar for %s %02x != %02x\n", > + font_name, tmA.tmLastChar, test_char); > +#endif > + if (pGetFontUnicodeRanges) > + { > + ok(tmW.tmFirstChar == first_unicode_char, "tmFirstChar for %s %02x != %02x\n", > + font_name, tmW.tmFirstChar, first_unicode_char); > +/* FIXME: remove completely once Wine is fixed > + * GetFontUnicodeRanges in Wine returns wrong data for fonts with > + * SYMBOL_CHARSET. > + */ > +if (tmW.tmLastChar != last_unicode_char) > +todo_wine ok(tmW.tmLastChar == last_unicode_char, "tmLastChar for %s %02x != %02x\n", > + font_name, tmW.tmLastChar, last_unicode_char); > +else > + ok(tmW.tmLastChar == last_unicode_char, "tmLastChar for %s %02x != %02x\n", > + font_name, tmW.tmLastChar, last_unicode_char); > + } > + > + ret = GetDeviceCaps(hdc, LOGPIXELSX); > + ok(tmW.tmDigitizedAspectX == ret, "tmDigitizedAspectX %u != %u\n", > + tmW.tmDigitizedAspectX, ret); > + ret = GetDeviceCaps(hdc, LOGPIXELSY); > + ok(tmW.tmDigitizedAspectX == ret, "tmDigitizedAspectY %u != %u\n", > + tmW.tmDigitizedAspectX, ret); > + > + SelectObject(hdc, hfont_old); > + DeleteObject(hfont); > + > + ReleaseDC(0, hdc); > +} > + > +static INT CALLBACK enum_truetype_font_proc(const LOGFONT *lf, const TEXTMETRIC *ntm, DWORD type, LPARAM lParam) > +{ > + INT *enumed = (INT *)lParam; > + > + if (type == TRUETYPE_FONTTYPE) > + { > + (*enumed)++; > + test_text_metrics(lf); > + } > + return 1; > +} > + > +static void test_GetTextMetrics(void) > +{ > + LOGFONTA lf; > + HDC hdc; > + INT enumed; > + > + SetLastError(0xdeadbeef); > + GetTextMetricsW(0, NULL); > + if (GetLastError() == ERROR_CALL_NOT_IMPLEMENTED) > + { > + skip("Skipping GetTextMetrics test on a Win9x platform\n"); > + return; > + } > + > + hdc = GetDC(0); > + > + memset(&lf, 0, sizeof(lf)); > + lf.lfCharSet = DEFAULT_CHARSET; > + enumed = 0; > + EnumFontFamiliesExA(hdc, &lf, enum_truetype_font_proc, (LPARAM)&enumed, 0); > + trace("Tested metrics of %d truetype fonts\n", enumed); > + > + ReleaseDC(0, hdc); > +} > + > START_TEST(font) > { > init(); > @@ -1427,4 +1565,5 @@ START_TEST(font) > } > else > skip("Arial Black or Symbol/Wingdings is not installed\n"); > + test_GetTextMetrics(); > } > -- > 1.5.0.2 > > > > > >

1 0

re: Writing a winelib plugin
by Dan Kegel 09 Mar '07

09 Mar '07

Shachar wrote: >It used to not be possible to write a plugin, to be loaded from a >standard Linux ELF program, that will be itself a winelib shared object. >All sorts of issues regarding running wineserver and memory layout >initializations were problematic. > >I'm wondering whether there is any news on that front. Is it still not >practically possible to do this? Last time I asked about this, I got the impression that there was some hope but nobody was working on it. And I think it would require linking something statically into the program, too, so if you're trying to foist winelib shared objects onto an unsuspecting Big App, you're probably going to be out of luck even if this gets worked on. What application did you have in mind?

3 2

Re: [Bug 7503] Eve Online pasword box on login screen doesn't work
by Lei Zhang 09 Mar '07

09 Mar '07

Can we change wineprefixcreate to copy the truetype fonts (if any) from /usr/X11R6/lib/X11/fonts/TTF or /usr/share/X11/fonts/TTF or [insert distro specific X11/fonts/TTF directory] to $WINEPREFIX/drive_c/windows/fonts/ ? On 2/20/07, Wine Bugs <wine-bugs(a)winehq.org> wrote: > http://bugs.winehq.org/show_bug.cgi?id=7503 > > > > > > ------- Additional Comments From kelfe(a)gmx.de 2007-20-02 16:57 ------- > to fix the issue you need to copy the font file arial.ttf into > .wine/drive_c/windows/fonts folder (.wine is a hidden folder) > > the arial.ttf can be found in the mscorefonts packages or search engine/windows box > > I don't know if font files should be placed in the fonts folder by wine or by > the user. So close this bug or keep it open until basic fonts are installed by > wine when creating the .wine folder. > > -- > Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email > ------- You are receiving this mail because: ------- > You are the assignee for the bug, or are watching the assignee. > > >

3 3

New opengl thread context selection patches for testing
by Stefan Dösinger 09 Mar '07

09 Mar '07

Hi, Here is a new patchset for testing which implements duplicating gl contexts for new threads in wined3d. Again, it does not implement any synchronisation measures(except ENTER_GL and glFinish), so running multithreaded games is still kinda a lottery. This patchset also contains a patch for offscreen rendering, which may fix the problems introduced with TES: Oblivion before. This is just a format-patch of my git tree, the first 3 patches are unrelated. One of them is a clone of Henri's already sent patch, and the other 2 patches were sent by me already. Happy testing, Stefan

3 7

Re: [2/2] wined3d: IWineD3DDeviceImpl_UpdateSurface: return D3DERR_INVALIDCALL if source and dest format do not match
by Fabian Bieler 09 Mar '07

09 Mar '07

Please ignore this patch.

1 0