February 2009 - Wine-Devel - list.winehq.org

Re: appdb issue: can't search for apps platinum on 1.0.x!

by Dan Kegel

On Wed, Feb 25, 2009 at 4:31 PM, Ben Klein <shacklein(a)gmail.com> wrote: > 2009/2/26 Dan Kegel <dank(a)kegel.com>: >> Our currently released version is 1.0, but the appdb's >> browse feature acts as if that version no longer exists. >> This will seriously confuse newcomers who are using >> the 1.0.1 version (e.g. anybody who installs a fresh >> copy of Ubuntu!). > > Someone mentioned on another thread (or possibly on IRC, I don't > recall) that 1.0-series is too old to be of concern to us. We don't > want test data for 1.0.x; we don't want bug reports for 1.0.x unless > they're still apparent in the development version. Development has > stopped on 1.0.x. That's a fine attitude from the developer's point of view, but that means that Wine *doesn't care* about Ubuntu users who expect to be able to use Wine by doing "add/remove" in the system menu. And I think we do care. Another way around this, as Scott Ritchie pointed out, is to arrange for what's in Ubuntu to be less stale. However, there are only two ways to do that: either do a stable release more often (which is difficult, and which Alexandre doesn't seem inclined to do), or get Ubuntu to accept an unstable snapshot into their stable repository (which I think they are not inclined to do). Yet another way to show that we care about Ubuntu users would be to make it drop-dead simple for the average user to add the Wine repository and get the latest wine. The current download instructions are really too complicated. We need instructions that are no more complicated than First: Click *here* to add WineHQ's repository Then: Do Applications / 'Add / Remove', and choose Wine That would compensate for the packages in Ubuntu's repo being stale. - Dan

Feb. 27, 2009

Re: search path redux - if office 2007 always uses a private riched20, why does wine interpose its own global one?

by Dan Kegel

On Tue, Feb 17, 2009 at 7:10 AM, Alexandre Julliard <julliard(a)winehq.org> wrote: > The version of the native dll, compared to the builtin. I could imagine > a heuristic where if the major version of native is higher than builtin > you default to native or something like that. I've updated http://bugs.winehq.org/show_bug.cgi?id=14980 with that hint. Say, have we considered making riched20 prefer native? That makes the app work, too. - Dan --- a/dlls/riched20/editor.c +++ b/dlls/riched20/editor.c @@ -2743,6 +2743,9 @@ BOOL WINAPI DllMain(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved TRACE("\n"); switch (fdwReason) { + case DLL_WINE_PREATTACH: + return FALSE; /* prefer native version */ +

Feb. 27, 2009

storing the registry key class name to make PDFXChange-Viewer happy

by Patrick Pfeifer

Dear Developpers of Wine I like wine and I like Tracker Software's PDFXChange-Viewer as well. http://www.docu-track.com/home/prod_user/PDF-XChange_Tools/pdfx_viewer So, today I was not getting any rest until i found out why these two won't work together nicely. I mean they DO, actually. But I'm kind of touchy. I was wondering all the time, why most of the settings I changed were not restored as I exited and restarted the app. I then discovered that if wineserver was preserved ("-p" option) beween restarts they WERE restored. Further investigations unveiled that the problem is triggered by the fact that wineserver is not storing the registry key class name attributes. Something M$ must have come up with in XP but that is not very wideley used. PDFXChange-Viewer however seems to rely on it when parsing it's preferences from the registry on startup. Wineservers unicode string handling gave me a though time in implementing a solution to this but i finally managed. You can see that I went quiet straightforward: adding the class name next to the key name in the {system,user}.reg files. There are two different versions of the fix. I would prefer the second one because it preserves backwards compatibility: In case the key class attribute is set, it is put in brackets between the key name and modification time. Otherwise nothing changes. It would be really nice if this could be somehow integrated into wine. (Not only because I worked on it for a whole day, but also because i really miss the feature because i don't want to reconfigure PDFXChange-Viewer every time i start it and also because I would like to make that functionality available to all of my friends and all linux users as well. Greets Patrick [Software\\Tracker Software\\PDFViewer\\FullScreen\\MainView\\Bars\\0011;ParamItem] 1235760987 "Break"=dword:00000001 "ID"=dword:000081c9 "Name"="Properties" [Software\\Tracker Software\\PDFViewer\\FullScreen\\MainView\\Bars\\0011] [ParamItem] 1235761189 "Break"=dword:00000001 "ID"=dword:000081c9 "Name"="Properties" --- registry.c 2009-02-27 19:44:14.000000000 +0100 +++ registry.PP1.c 2009-02-27 19:49:03.000000000 +0100 @@ -176,7 +176,7 @@ dump_path( key->parent, base, f ); fprintf( f, "\\\\" ); } - dump_strW( key->name, key->namelen / sizeof(WCHAR), f, "[]" ); + dump_strW( key->name, key->namelen / sizeof(WCHAR), f, "[;" ); } /* dump a value to a text file */ @@ -245,6 +245,8 @@ { fprintf( f, "\n[" ); if (key != base) dump_path( key, base, f ); + fprintf( f, ";" ); + dump_strW( key->class, key->classlen / sizeof(WCHAR), f, "[]" ); fprintf( f, "] %ld\n", (long)key->modif ); for (i = 0; i <= key->last_value; i++) dump_value( &key->values[i], f ); } @@ -1108,20 +1110,22 @@ int prefix_len, struct file_load_info *info, int default_modif ) { - WCHAR *p; - struct unicode_str name; - int res, modif; - data_size_t len; + WCHAR *p, cls[256]; + struct unicode_str name, class; + int res, res2, modif; + data_size_t len, len2 = 256; if (!get_file_tmp_space( info, strlen(buffer) * sizeof(WCHAR) )) return NULL; len = info->tmplen; - if ((res = parse_strW( info->tmp, &len, buffer, ']' )) == -1) + if ((res = parse_strW( info->tmp, &len, buffer, ';' )) == -1 || + (res2 = parse_strW( cls, &len2, buffer+res, ']' ) == -1)) { file_read_error( "Malformed key", info ); return NULL; } - if (sscanf( buffer + res, " %d", &modif ) != 1) modif = default_modif; + + if (sscanf( buffer + res + res2, " %d", &modif ) != 1) modif = default_modif; p = info->tmp; while (prefix_len && *p) { if (*p++ == '\\') prefix_len--; } @@ -1138,7 +1142,10 @@ } name.str = p; name.len = len - (p - info->tmp + 1) * sizeof(WCHAR); - return create_key( base, &name, NULL, flags, modif, &res ); + class.str = cls; + class.len = len2 - sizeof(WCHAR); + + return create_key( base, &name, &class, flags, modif, &res ); } /* parse a comma-separated list of hex digits */ --- registry.c 2009-02-27 19:44:14.000000000 +0100 +++ registry.PP2.c 2009-02-27 19:59:28.000000000 +0100 @@ -245,6 +245,10 @@ { fprintf( f, "\n[" ); if (key != base) dump_path( key, base, f ); + if (key->class != NULL) { + fprintf( f, "] [" ); + dump_strW( key->class, key->classlen / sizeof(WCHAR), f, "[]" ); + } fprintf( f, "] %ld\n", (long)key->modif ); for (i = 0; i <= key->last_value; i++) dump_value( &key->values[i], f ); } @@ -1108,10 +1112,10 @@ int prefix_len, struct file_load_info *info, int default_modif ) { - WCHAR *p; - struct unicode_str name; - int res, modif; - data_size_t len; + WCHAR *p, *c; + struct unicode_str name, class; + int res, c_res, modif; + data_size_t len, c_len; if (!get_file_tmp_space( info, strlen(buffer) * sizeof(WCHAR) )) return NULL; @@ -1121,6 +1125,23 @@ file_read_error( "Malformed key", info ); return NULL; } + + c_len = info->tmplen - len; + if ((c_res = parse_strW( info->tmp + len / sizeof(WCHAR), &c_len, buffer + res, ']' )) == -1) + { + class.len = 0; + } + else + { + res += c_res; + + c = info->tmp + len / sizeof(WCHAR); + while (*c) { if (*c++ == '[') break; } + + class.str = c; + class.len = c_len - (c - (info->tmp + len / sizeof(WCHAR)) + 1) * sizeof(WCHAR); + } + if (sscanf( buffer + res, " %d", &modif ) != 1) modif = default_modif; p = info->tmp; @@ -1138,6 +1159,9 @@ } name.str = p; name.len = len - (p - info->tmp + 1) * sizeof(WCHAR); + + if (class.len > 0) + return create_key( base, &name, &class, flags, modif, &res ); return create_key( base, &name, NULL, flags, modif, &res ); }

Feb. 27, 2009

re: wine shirts

by Dan Kegel

That's a nice, simple design, but something's missing. Oddly, the original drunken penguin shirts, or ripoffs thereof, seem to still be available at http://www.ixsoft.de/software/products/CWTSHIRTDP-L.html Original artwork is at ftp://wine.codeweavers.com/pub/wine/logos/ I would kind of like a shirt that had the penguin opening the bottle on the front (text: Support the Wine Project), and the drunken penguin on the back (text: I did my part, or something like that) but I'd have to think a bit more on the text. Other old links: Martin's design: http://cross-lfs.org/~mlankhorst/t-shirt.png Ismael Barros put some effort into this before: http://www.winehq.org/pipermail/wine-devel/2008-December/070894.html but nobody could think up good text. - Dan

Feb. 27, 2009

Time to integrate win16 test suite?

by Dan Kegel

Now that we support building 16 bit executables, it seems like a good time to think about integrating the 16 bit test suite, currently hibernating at http://win16test.googlecode.com Any takers?

Feb. 27, 2009

Re: twain_32/tests: Link with twain_32.dll.

by Jeremy White

Francois Gouget wrote: > --- > > winetest can detect if twain_32.dll is there or not, and if it's missing > there's nothing to test anyway. Note that make_makefiles will need to be > run. This patch breaks make crosstest for me: [apevia:~/w/wine/dlls/twain_32/tests] make crosstest i586-mingw32msvc-gcc dsm.cross.o testlist.cross.o -o twain_32_crosstest.exe -L../../../dlls -L../../../dlls/twain_32 -L../../../dlls/user32 -L../../../dlls/gdi32 -L../../../dlls/kernel32 -ltwain_32 -luser32 -lgdi32 -lkernel32 /usr/lib/gcc/i586-mingw32msvc/4.2.1-sjlj/../../../../i586-mingw32msvc/bin/ld: cannot find -ltwain_32 collect2: ld returned 1 exit status make: *** [twain_32_crosstest.exe] Error 1 (it also breaks make test for me, but that may be operator error). But, what's more, I do not see the point. Twain_32 is not a Microsoft DLL; it is not present by default on most Windows systems. Further, the API specifications are quite clear: you're supposed to LoadLibrary Twain32, and all Twain applications I've tested do so. I don't see the benefit in making our test behave differently than the recommended behavior for this DLL. Cheers, Jeremy

Feb. 27, 2009

Re: gdi32: Revert an unrelated part of 028617b90ba586bdb30723c700eea888c159ada7.

by Byeongsik Jeon

And, my patch [1/2] is wrong. Dmitry's patch is right. Additionally, my patch [2/2] has to committed. Because of the other problem. It's also my mistake. http://bugs.winehq.org/show_bug.cgi?id=17218#c18 2009-02-27 (Fri), 21:50 +0900, Byeongsik Jeon wrote: > Yes!!! See the http://bugs.winehq.org/show_bug.cgi?id=17218 > > http://www.winehq.org/pipermail/wine-patches/2009-February/068752.html > > http://www.winehq.org/pipermail/wine-patches/2009-February/068753.html > > > > 2009-02-27 (Fri), 17:57 +0800, Dmitry Timoshkov wrote: > > It was not a very nice idea to break bitmap fonts which request a custom > > width (therefore a transformation) in a patch that pretended to do something > > unrelated. > > --- > > dlls/gdi32/freetype.c | 6 +++--- > > 1 files changed, 3 insertions(+), 3 deletions(-) > > > > diff --git a/dlls/gdi32/freetype.c b/dlls/gdi32/freetype.c > > index 6ca35c3..54bb999 100644 > > --- a/dlls/gdi32/freetype.c > > +++ b/dlls/gdi32/freetype.c > > @@ -4606,9 +4606,9 @@ DWORD WineEngGetGlyphOutline(GdiFont *incoming_font, UINT glyph, UINT format, > > } > > > > if(ft_face->glyph->format != ft_glyph_format_outline && > > - (needsTransform || format == GGO_NATIVE || format == GGO_BEZIER || > > - format == GGO_GRAY2_BITMAP || format == GGO_GRAY4_BITMAP || > > - format == GGO_GRAY8_BITMAP)) > > + (format == GGO_NATIVE || format == GGO_BEZIER || > > + format == GGO_GRAY2_BITMAP || format == GGO_GRAY4_BITMAP || > > + format == GGO_GRAY8_BITMAP)) > > { > > TRACE("loaded a bitmap\n"); > > LeaveCriticalSection( &freetype_cs ); > > > >

Feb. 27, 2009

Re: msi: Add reinstall tests.

by Detlef Riekenberg

On Do, 2009-02-26 at 10:50 +0100, Hans Leidekker wrote: > + state = 0xdeadbee; > + action = 0xdeadbee; > + r = MsiGetFeatureState(hpkg, "one", &state, &action); > + ok( r == ERROR_UNKNOWN_FEATURE, "Expected ERROR_UNKNOWN_FEATURE, > got %d\n", r ); > + ok( state == 0xdeadbee, "Expected 0xdeadbee, got %d\n", state); > + ok( action == 0xdeadbee, "Expected 0xdeadbee, got %d\n", action); > + > + state = 0xdeadbee; > + action = 0xdeadbee; > + r = MsiGetFeatureState(hpkg, "two", &state, &action); > + ok( r == ERROR_UNKNOWN_FEATURE, "Expected ERROR_UNKNOWN_FEATURE, > got %d\n", r ); > + ok( state == 0xdeadbee, "Expected 0xdeadbee, got %d\n", state); > + ok( action == 0xdeadbee, "Expected 0xdeadbee, got %d\n", action); > + > + state = 0xdeadbee; > + action = 0xdeadbee; > + r = MsiGetFeatureState(hpkg, "three", &state, &action); > + ok( r == ERROR_UNKNOWN_FEATURE, "Expected ERROR_UNKNOWN_FEATURE, > got %d\n", r ); > + ok( state == 0xdeadbee, "Expected 0xdeadbee, got %d\n", state); > + ok( action == 0xdeadbee, "Expected 0xdeadbee, got %d\n", action); This is huge amount of almost the same code in that test. Please use tables. -- By by ... Detlef

Feb. 27, 2009

Re: gdi32: Revert an unrelated part of 028617b90ba586bdb30723c700eea888c159ada7.

by Byeongsik Jeon

Yes!!! See the http://bugs.winehq.org/show_bug.cgi?id=17218 http://www.winehq.org/pipermail/wine-patches/2009-February/068752.html http://www.winehq.org/pipermail/wine-patches/2009-February/068753.html 2009-02-27 (Fri), 17:57 +0800, Dmitry Timoshkov wrote: > It was not a very nice idea to break bitmap fonts which request a custom > width (therefore a transformation) in a patch that pretended to do something > unrelated. > --- > dlls/gdi32/freetype.c | 6 +++--- > 1 files changed, 3 insertions(+), 3 deletions(-) > > diff --git a/dlls/gdi32/freetype.c b/dlls/gdi32/freetype.c > index 6ca35c3..54bb999 100644 > --- a/dlls/gdi32/freetype.c > +++ b/dlls/gdi32/freetype.c > @@ -4606,9 +4606,9 @@ DWORD WineEngGetGlyphOutline(GdiFont *incoming_font, UINT glyph, UINT format, > } > > if(ft_face->glyph->format != ft_glyph_format_outline && > - (needsTransform || format == GGO_NATIVE || format == GGO_BEZIER || > - format == GGO_GRAY2_BITMAP || format == GGO_GRAY4_BITMAP || > - format == GGO_GRAY8_BITMAP)) > + (format == GGO_NATIVE || format == GGO_BEZIER || > + format == GGO_GRAY2_BITMAP || format == GGO_GRAY4_BITMAP || > + format == GGO_GRAY8_BITMAP)) > { > TRACE("loaded a bitmap\n"); > LeaveCriticalSection( &freetype_cs );

Feb. 27, 2009

Re: gdi32: Revert an unrelated part of 028617b90ba586bdb30723c700eea888c159ada7.

by Michael Karcher

Am Freitag, den 27.02.2009, 17:57 +0800 schrieb Dmitry Timoshkov: > It was not a very nice idea to break bitmap fonts which request a custom > width (therefore a transformation) in a patch that pretended to do something > unrelated. It also "breaks" fake italic bitmap fonts, i.e. previously the fake italic transformation was just ignored, so roman characters were displayed, now WineEngGetGlyphOutline runs into this error case, breaking applications trying to use "MS Sans Serif" in italic. Regards, Michael Karcher

Feb. 27, 2009