August 2013 - Wine-devel - list.winehq.org

Re: gdi32: Avoid a crash by loading AppKit before using Core Text if might use the Mac driver.
by Alexandre Julliard 02 Aug '13

02 Aug '13

Ken Thomases <ken(a)codeweavers.com> writes: > + /* If the Mac driver might be used, then load AppKit now before using the Core Text API. > + Otherwise, AppKit crashes on Mac OS X 10.7+ because CTFontDescriptor isn't toll-free > + bridged to NSCTFontDescriptor. That bridging is only set up if AppKit has been loaded > + when CTFontDescriptor is first used. */ > + /* @@ Wine registry key: HKCU\Software\Wine\Drivers */ > + if (!RegOpenKeyA(HKEY_CURRENT_USER, "Software\\Wine\\Drivers", &hkey)) > + { > + char buffer[MAX_PATH]; > + DWORD type, count = sizeof(buffer); > + > + if (!RegQueryValueExA(hkey, "Graphics", 0, &type, (LPBYTE) buffer, &count)) > + { > + char *name = buffer; > + > + usingMacDriver = FALSE; > + while (name) > + { > + char *next = strchr(name, ','); > + if (next) *next++ = 0; > + > + if (!strcmp(name, "mac")) > + { > + usingMacDriver = TRUE; > + break; > + } > + > + name = next; > + } > + } > + > + RegCloseKey(hkey); > + } That's ugly. freetype.c has no business knowing about the details of the graphics driver. -- Alexandre Julliard julliard(a)winehq.org

2 5

Re: riched20: Make ITextDocument stub reusable for ITextServices (try 4)
by Alexandre Julliard 02 Aug '13

02 Aug '13

Caibin Chen <tigersoldi(a)gmail.com> writes: > diff --git a/dlls/riched20/tomimpl.h b/dlls/riched20/tomimpl.h > new file mode 100644 > index 0000000..886c3a8 > --- /dev/null > +++ b/dlls/riched20/tomimpl.h > @@ -0,0 +1,59 @@ > +/* > + * RichEdit - TOM interfaces implementations > + * > + * Copyright 2013 by Caibin Chen > + * > + * This library is free software; you can redistribute it and/or > + * modify it under the terms of the GNU Lesser General Public > + * License as published by the Free Software Foundation; either > + * version 2.1 of the License, or (at your option) any later version. > + * > + * This library is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > + * Lesser General Public License for more details. > + * > + * You should have received a copy of the GNU Lesser General Public > + * License along with this library; if not, write to the Free Software > + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA > + */ > + > +#ifndef __RICHOLE_H > +#define __RICHOLE_H > + > +#include <tom.h> > + > +#include "editstr.h" > + > +typedef struct ITextDocumentImpl ITextDocumentImpl; > +typedef struct ITextSelectionImpl ITextSelectionImpl; > + > +/* ITextDocument */ > + > +struct ITextDocumentImpl { > + ITextDocument ITextDocument_iface; > + IUnknown *outer_unk; > + ME_TextEditor *editor; > + ITextSelectionImpl *txtSel; > +}; You shouldn't need a new header either. Particularly since implementation details shouldn't be global. > +/** > + * Create an ITextDocumentImpl object, which acts as an inner object of COM aggression. I'm sure writing COM code can make you feel aggressive, but that's not the right term ;-) > + * > + * The ITextDocumentImpl object will delegate all of its IUnknown calls to the specified > + * {@code outer_unk}. > + * > + * @param outer_unk The outer objects that creates and delegates ITextDocument > + * method calls to the created ITextDocumentImpl. > + * @param editor The editor implementation. > + * > + * @return An ITextDocumentImpl object that should be destroyed with > + * {@code ITextDocumentImpl_destroy} > + */ Please don't use that sort of documentation markup. Besides, things like parameters and return should be obvious from the code, you shouldn't have to add documentation for this, especially not for internal functions. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: xmllite: Improve attribute value parsing
by Alexandre Julliard 02 Aug '13

02 Aug '13

Nikolay Sivov <nsivov(a)codeweavers.com> writes: > +static void reader_normalize_space(xmlreader *reader, WCHAR *ptr) > +{ > + encoded_buffer *buffer = &reader->input->buffer->utf16; > + > + if (!is_wchar_space(*ptr)) return; > + > + if (*ptr == '\r' && *ptr == '\n') That's not going to happen very often... -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] oleaut32: ERR on 32bit typelibs on 64bit
by Alexandre Julliard 02 Aug '13

02 Aug '13

Marcus Meissner <meissner(a)suse.de> writes: > Triggered by the Novell Groupwise installer currently, > from the ISBE64W.exe Installshield Helper program. > > The typelib is contained directly in ISBE64W.exe itself :/ > > Basically if we encounter this case the program will crash > anyway (but with random backtrace), so output an ERR(). That looks like it's just hiding the problem. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: d3dx9 [patch 1/2]: Implement D3DXCreatePolygon
by Matteo Bruni 02 Aug '13

02 Aug '13

2013/8/1 Nozomi Kodama <nozomi.kodama(a)yahoo.com>: > + vertices = HeapAlloc(GetProcessHeap(), 0, 2 * (sides + 1) * sizeof(D3DXVECTOR3)); + if (!vertices) + { + TRACE("Not memory enough for vertex buffer\n"); + polygon->lpVtbl->Release(polygon); + return E_OUTOFMEMORY; + } + + hr = polygon->lpVtbl->LockVertexBuffer(polygon, D3DLOCK_DISCARD, (VOID **)&vertices); Why are you allocating memory and immediately dropping track of it? Recheck how ID3DXMesh::LockVertexBuffer works. Also I'm having a deja-vu here, this patch looks suspiciously similar (up to having this same issue) to one I've already reviewed in the past.

2 1

appwiz: Correct Wine Mono installer title in several languages
by Ken Sharp 02 Aug '13

02 Aug '13

Hi Alexandre, http://source.winehq.org/patches/data/97460 should be correct but would I be right in assuming you would rather have the changes made along with the next string for each language?

2 4

Re: [PATCH 4/5] d3dx9/tests: Add ID3DXConstantTable struct test.
by Matteo Bruni 02 Aug '13

02 Aug '13

2013/7/24 Rico Schüller <kgbricola(a)web.de>: > --- > dlls/d3dx9_36/tests/shader.c | 308 > +++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 308 insertions(+) > This is okay, but as a followup can you add some tests with mixed-type structs? Something like: struct { float f; int i; bool b; }; If you have already written tests of this kind, I'd like to know what does the compiler do in this case :)

2 6

Re: [1/2] ntdll: Implement NtQueryEvent
by Alexandre Julliard 02 Aug '13

02 Aug '13

Andrew Cook <ariscop(a)gmail.com> writes: > diff --git a/dlls/ntdll/tests/Makefile.in b/dlls/ntdll/tests/Makefile.in > index 10d6674..a44b880 100644 > --- a/dlls/ntdll/tests/Makefile.in > +++ b/dlls/ntdll/tests/Makefile.in > @@ -7,6 +7,7 @@ C_SRCS = \ > directory.c \ > env.c \ > error.c \ > + event.c \ > exception.c \ > file.c \ > generated.c \ Please don't add a new file for just a couple of tests, merge this into an existing file. -- Alexandre Julliard julliard(a)winehq.org

1 0

jscript FTBFS in Cygwin 1.7.22
by Ken Sharp 02 Aug '13

02 Aug '13

Probably better to post this here rather than the forums: http://forum.winehq.org/viewtopic.php?f=2&t=19501 "Not sure if this is a Wine bug (as usual) so I'd rather put this here than to open a new bug. On Cygwin 1.7.22 the compilation stops at jscript, apparently a conflict in the declaration of dtoa. $ make ccache cc -c -I/home/ken/wine-git/dlls/jscript -I. -I/home/ken/wine-git/include -I../../include -D__WINESRC__ -D_REENTRANT -Wall -pipe -fno-strict-aliasing -Wdeclaration-after-statement -Wempty-body -Wignored-qualifiers -Wstrict-prototypes -Wtype-limits -Wunused-but-set-parameter -Wwrite-strings -Wpointer-arith -Wlogical-op -gdwarf-2 -gstrict-dwarf -fno-omit-frame-pointer -g -O0 -D_WIN32 -o jscript_main.o /home/ken/wine-git/dlls/jscript/jscript_main.c ccache cc -c -I/home/ken/wine-git/dlls/jscript -I. -I/home/ken/wine-git/include -I../../include -D__WINESRC__ -D_REENTRANT -Wall -pipe -fno-strict-aliasing -Wdeclaration-after-statement -Wempty-body -Wignored-qualifiers -Wstrict-prototypes -Wtype-limits -Wunused-but-set-parameter -Wwrite-strings -Wpointer-arith -Wlogical-op -gdwarf-2 -gstrict-dwarf -fno-omit-frame-pointer -g -O0 -D_WIN32 -o lex.o /home/ken/wine-git/dlls/jscript/lex.c ccache cc -c -I/home/ken/wine-git/dlls/jscript -I. -I/home/ken/wine-git/include -I../../include -D__WINESRC__ -D_REENTRANT -Wall -pipe -fno-strict-aliasing -Wdeclaration-after-statement -Wempty-body -Wignored-qualifiers -Wstrict-prototypes -Wtype-limits -Wunused-but-set-parameter -Wwrite-strings -Wpointer-arith -Wlogical-op -gdwarf-2 -gstrict-dwarf -fno-omit-frame-pointer -g -O0 -D_WIN32 -o number.o /home/ken/wine-git/dlls/jscript/number.c /home/ken/wine-git/dlls/jscript/number.c:57:20: error: conflicting types for ‘dtoa’ In file included from /home/ken/wine-git/include/wine/port.h:47:0, from /home/ken/wine-git/dlls/jscript/number.c:20: /usr/include/stdlib.h:161:35: note: previous declaration of ‘dtoa’ was here Makefile:199: recipe for target `number.o' failed make: *** [number.o] Error 1 dlls/jscript/number.c: static inline void dtoa(double d, WCHAR *buf, int size, int *dec_point) /usr/include/stdlib.h: char * _EXFUN(dtoa,(double, int, int, int *, int*, char**));"

1 0

Re: ntdll: Implement compatible FindActCtxSectionString() for window class section (try3)
by Alexandre Julliard 02 Aug '13

02 Aug '13

Nikolay Sivov <nsivov(a)codeweavers.com> writes: > @@ -1031,6 +1105,17 @@ static BOOL parse_window_class_elem(xmlbuf_t* xmlbuf, struct dll_redirect* dll) > > if (!(entity->u.class.name = xmlstrdupW(&content))) return FALSE; > > + /* each class entry needs index, data and string data */ > + acl->actctx->wndclass_section_len += sizeof(struct wndclass_index); > + acl->actctx->wndclass_section_len += sizeof(struct wndclass_redirect_data); > + /* original name is stored separately */ > + acl->actctx->wndclass_section_len += aligned_string_len((content.len+1)*sizeof(WCHAR)); > + /* versioned name and module name are stored one after another */ > + len = get_assembly_version(assembly, NULL) + content.len + 2 /* null terminator and '!' separator */; > + len += strlenW(dll->name) + 1; > + acl->actctx->wndclass_section_len += aligned_string_len(len*sizeof(WCHAR)); > + acl->actctx->wndclass_count++; > + It would be better to do this at the time you build the structure. The parser shouldn't need to worry about this. -- Alexandre Julliard julliard(a)winehq.org

2 2