Re: [boost] Win32 undefs/redefs [was File operations: Win32 question [was Re: the first version of abstract path...]]

From: Ross Smith (r-smith_at_[hidden])
Date: 2002-02-25 11:06:44

Alan Bellingham wrote:
> 
> Ross Smith <r-smith_at_[hidden]>:
> 
> >It occurs to me that, if you're going to write all that code for every
> >API function (pair) that you use anyway, it may be simpler just to leave
> >the Windows headers out altogether and simply write your own headers
> >with just the extern declarations for FooA() and FooW() and your
> >overloaded Foo(), losing the macros completely. (Also that it probably
> >wouldn't be terribly difficult to automate the header conversion
> >process.)
> 
> That _is_ the other possibility, yes.
> 
> However, I'm a bit leery of actually doing that, for a few reasons:
> 
> 1) Copyright.
> 2) The possibility of making mistakes.
> 3) It's difficult to avoid the real headers getting included anyway.
> 
> But automating it more is definitely the way forward.

Automation would solve (2), and I don't understand (3). As for (1), you
could check the Cygwin/Mingw headers instead of the Microsoft ones,
and/or write a spider script that extracts the declarations from the
MSDN web site (I've already got teh beginnings of that around here
somewhere).

-- 
Ross Smith ...................................... Auckland, New Zealand
r-smith_at_[hidden] ......................... http://storm.net.nz/~ross/
  "We need a new cosmology. New gods. New sacraments. Another drink."
                                                       -- Patti Smith