[prev in list] [next in list] [prev in thread] [next in thread] 

List:       wine-devel
Subject:    Re: Testers on windows needed
From:       Stefan Leichter <Stefan.Leichter () camline ! com>
Date:       2009-07-28 20:51:27
Message-ID: 200907282251.28225.Stefan.Leichter () camline ! com
[Download RAW message or body]

Am Tuesday 28 July 2009 18:56:46 schrieb Hin-Tak Leung:
> --- On Tue, 28/7/09, Stefan Leichter <Stefan.Leichter@camline.com> wrote:
> > Hello,
> >
> > i'm looking for some good souls testing the attached patch
> > one windows.
> >
> > While looking for bug 7701 i found that the function
> > __vbaNew2 of the
> > MSVBVM60.dll does not like return values from
> > SHGetFileInfoA bigger than
> > 0x7fff
> >
> > Can you please report at least the results of the trace
> > statements and if any
> > modified test fails for you, the console output is fine for
> > me too.
> >
> > Thanks for your help
> > Stefan
>
> If I remember correctly, MSVBVM60.dll isn't one of core wine dll's, so it
> would need to be installed via winetricks or copied over from windows i.e.
> it has to be native, no built-ins... wouldn't be the same as running your
> test under wine?
>
> In a different thought, I think your question is less about whether
> MSVBVM60.dll like certain return values (presumably you already verified
> that at least one case it does not), but what versions of MSVBVM60.dll are
> out there, and do they behave all in the same way... i.e. you probably like
> people to report on the version of MSVBVM60.dll (from winedump, for
> example) also, rather than just test it and report success/failure?

The tests are not for the function __vbaNew2 of MSVBVM60.dll. The tests are 
for SHGetFileInfoA function in shell32.dll. The implementaion of the 
SHGetFileInfoA function in wine makes the __vbaNew2 function fail. Therefore 
the tests of the function SHGetFileInfoA were made stricter to satisfy the  
function __vbaNew2. If the modified test are succeeding on all windows 
versions, the implementaion of SHGetFileInfoA in wine can be changed to 
succeed too, and the problem is fixed.


[prev in list] [next in list] [prev in thread] [next in thread]