> Date: Sat, 09 Jul 2005 00:31:24 +0100
> From: "J. Grant" <[EMAIL PROTECTED]>
> CC: [email protected]
> 
> I have attached new  files, an updated version of the patch, and files
> files listing the lines removed (as un-used, and not in MinGW/MSYS
> generated config.h), and the lines added (to bring it in line with
> MinGW/MSYS generated config.h)
> [...]
> I have only moved defines around, to make it clearer to compare to MinGW
> or other generated config.h

Thanks for your work.

However, the last sentence I cited above seems to possibly contradict
the one above it.  In the former, you say that some lines were added
to bring the template in line with what MSYS generates when one runs
the configure script.  In the latter, you say that you only moved
lines around.  Which one is true, or what am I missing?

Adding lines based on MSYS output for the MinGW build bothers me a
bit: did you check that these lines are okay for the other W32 builds,
notably MSVC?  If you did, then I don't have any objections adding
those lines.

Likewise for the removed lines: the best way to make sure they can
indeed be removed is to verify that no source code uses them.  If some
source code uses them, we need to look at every code fragment that
does and decide whether some W32 build might need them after all.

> /* Use platform specific coding */
> #define WINDOWS32 1

This is strange: WINDOWS32 should be defined for _any_ W32 build.  Are
you saying that without this addition WINDOWS32 might not be defined
in some situation?  If so, then what is that situation?


_______________________________________________
Make-w32 mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/make-w32

Reply via email to