February 2014 - Wine-Devel - list.winehq.org

Re: [PATCH 5/5] d3dx9: Handle pixel formats larger than 32 bits in surface loading / filtering functions.

by Matteo Bruni

2014-02-14 17:03 GMT+01:00 Henri Verbeet <hverbeet(a)gmail.com>: > On 14 February 2014 15:53, Matteo Bruni <mbruni(a)codeweavers.com> wrote: >> + if (format->bits[c]) >> + { >> + mask = format->bits[c] == 32 ? ~0U : (1 << format->bits[c]) - 1; > You could probably also do that as "mask = ~0u >> (32 - > format->bits[c]);", since you already need to check "format->bits[c]" > for 0 anyway. > Good point, I'll resend this patch with that change.

Feb. 14, 2014

Re: [PATCH 5/5] d3dx9: Handle pixel formats larger than 32 bits in surface loading / filtering functions.

by Henri Verbeet

On 14 February 2014 15:53, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > + if (format->bits[c]) > + { > + mask = format->bits[c] == 32 ? ~0U : (1 << format->bits[c]) - 1; You could probably also do that as "mask = ~0u >> (32 - format->bits[c]);", since you already need to check "format->bits[c]" for 0 anyway.

Feb. 14, 2014

Re: Converting source to unicode, bug 28189

by Akihiro Sagawa

On Fri, 14 Feb 2014 06:25:38 -0500, Hugh McMaster wrote: > >7. Silent handling: Many times, it refers Silent variable before > > calling output_write function. How about moving the check to the > > function? > Do you mean that I should add the Silent boolean value as a variable > in the function call? No. I meant: [Existing code] static int some_function(void) { /* ... */ if (Silent) { output_write(STRING_FOO); output_write(STRING_BAR); } /* ... */ } [My idea] static int some_function(void) { /* ... */ output_write(STRING_FOO); output_write(STRING_BAR); /* ... */ } static void __cdecl output_write(UINT id, ...) { /* variable declarations */ if (Silent) return; /* do nothing */ /* then, LoadString, FormatMessage, etc. */ } Does this work for you?

Feb. 14, 2014

Re: [PATCH 2/2] mshtml: Use IWebBrowserPriv2IE9 instead of IWebBrowserPriv2IE8 in tests on IE9+.

by Marvin

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 https://newtestbot.winehq.org/JobDetails.pl?Key=4905 Your paranoid android. === w8 (32 bit htmldoc) === htmldoc.c:2858: Test failed: unexpected command 74 htmldoc.c:2858: Test failed: unexpected command 74

Feb. 14, 2014

Re: xmllite: Don't fail the whole parsing process because of unsupported 'xmlns' attribute.

by Nikolay Sivov

On 2/14/2014 11:23, Dmitry Timoshkov wrote: > This patch allows to successfully parse pretty complex XML task definitions. > > Please apply after other xmllite fixes. I think it's fine to relax a check here as you did if it really helps. Same applies to other patches. My main problem with all of that is that whole reader code needs to be revised and partially rewritten at some point (better sooner than later), to properly support chunked reads and corresponding state updates. Looking at it now I don't actually like the way I did it, it's not easily extensible and not very clean. Do you plan to add more code to xmllite itself or what we have is enough for taskschd so far?

Feb. 14, 2014

Re: configure: Add import library for xmllite.

by Henri Verbeet

On 14 February 2014 08:23, Dmitry Timoshkov <dmitry(a)baikal.ru> wrote: > -WINE_CONFIG_DLL(xmllite) > +WINE_CONFIG_DLL(xmllite,,[implib]) That's auto-generated by make_makefiles from Makefile.in.

Feb. 14, 2014

Re: ws2_32: Fix some edge cases for getaddrinfo in BSD and OSX

by Bruno Jesus

On Fri, Feb 14, 2014 at 9:27 AM, Dmitry Timoshkov <dmitry(a)baikal.ru> wrote: > Bruno Jesus <00cpxxx(a)gmail.com> wrote: > >> + /* servname tweaks required by OSX and BSD kernels */ >> + if (!servname) >> + serv = NULL; >> + else if (!servname[0]) >> + serv = "0"; >> + else >> + serv = servname; > > Shouldn't 'serv = "0"' be 'serv = ""' ? No, that is the trick. Linux supports both "" and "0" but OSX/BSD does not support "". So turning the "" into "0" solves the issue. > -- > Dmitry. Bruno

Feb. 14, 2014

Re: ws2_32: Fix some edge cases for getaddrinfo in BSD and OSX

by Dmitry Timoshkov

Bruno Jesus <00cpxxx(a)gmail.com> wrote: > + /* servname tweaks required by OSX and BSD kernels */ > + if (!servname) > + serv = NULL; > + else if (!servname[0]) > + serv = "0"; > + else > + serv = servname; Shouldn't 'serv = "0"' be 'serv = ""' ? -- Dmitry.

Feb. 14, 2014

RE: Converting source to unicode, bug 28189

by Hugh McMaster

Hi Akihiro, Many thanks for your comments. On Friday, 14 February 2014 3:06 AM, Akihiro Sagawa wrote: >Regarding your patch, attachment #47519, I roughly read the code and post comments at the following points: >5. Double quote handling: command_line is advanced when the code > finds a double quote character. When it updates wsCommandLine? Good catch! :) >7. Silent handling: Many times, it refers Silent variable before > calling output_write function. How about moving the check to the > function? Do you mean that I should add the Silent boolean value as a variable in the function call?

Feb. 14, 2014

Re: [PATCH 5/5] tests: Allow broken results to cause failures on the "windows" platform as well.

by Henri Verbeet

On 14 February 2014 11:58, Alexandre Julliard <julliard(a)winehq.org> wrote: > Couldn't you simply set the platform to "windows_broken" or something > like that? > In practice that's probably good enough, although strictly speaking it would break win_skip(). I don't care too strongly, I mostly sent this in in case someone else found it useful as well.

Feb. 14, 2014