Re: 3.0.1 release? (Reports on mingw build)

[Tatsuro MATSUOKA]

Hello

> >  OK, I made this change.
> 
> I don't understand why this change is needed. At that point, windows.h
> should already
> have been included using the following sequence:
> 
> file-ops.h > str-vec.h > Array.h > lo-utils.h > windows.h (around line 69).
> 
> I think that the problem you have is a symptom of another problem:
> HAVE_LOADLIBRARY_API
> is not defined, that means that the configure check for LoadLibrary failed.
> 
> The patch above is not a problem for MSVC. But when something does not work
> under MinGW, the first thing you should ask yourself is "Damn, why does it 
> work
> under MSVC?". Because if it wouldn't work, then I would also have the problem,
> and I would probably fix it ASAP... :-)

What you said is right at this moment, I think.  But I ... could not aware that 
that sequence.
Under make of losystep.cc, I could not really comipile it.  Errors appread were 
typical 
to the lost of windows.h.  So I consider the patch at that moment.

>Damn, why does it work under MSVC?"  
I also think what you said is completly right now.  However, I thought that it 
was a 
careless mistake at the integrating the unixy and windowsish codes at that 
moment.
That was my careless mistake.

However the problem is why the detection at configure check did not work on 
MinGW.
Sorry I have no idea at this moment because it is midnight in Japan and the 
problem is beyond 
my head now.   Please wait a while to check what was happend in MinGW.

If it is not required at patch for msvc, it should be tempoaral patch for the 
MinGW build.
And modification will be deleteded at the moment for the 3.0.1 release.

We have to solve it until 3.0.2 release.

That's all What I can reply at the moment.

Regards

Tatsuro
  





--------------------------------------
GANBARE! NIPPON! Win your ticket to Olympic Games 2008.
http://pr.mail.yahoo.co.jp/ganbare-nippon/