DG/UX

[David Relson]

On Wed, 17 Dec 2003 11:48:27 +0100
Matthias Andree <matthias.andree at gmx.de> wrote:

> On Tue, 16 Dec 2003, Clint Adams wrote:
> 
> > > + #ifdef __DGUX__
> > 
> > Wouldn't #ifndef MAP_FAILED be more appropriate?
> 
> What if MAP_FAILED is an enum or const int? #ifndef won't catch that.
> POSIX doesn't state MAP_FAILED is a macro, just a "symbolic constant".
> 
> It would usually work though.
> 
> Of course we can have configure.ac check it, but I wanted to avoid
> polluting configure.ac with contrib/* checks and contrib/* with
> #include<config.h>.

Correct.  We've recently seen that a symbol that's #define'd before it's
enum'd causes an error message. I wasn't sure what would happen with a
#define after the enum, so I  tested it.  gcc-3.3.1 doesn't complain. 
It happily uses the most recent definition.

I'll put in the __DGUX__ version.  If/when we discover that something
more is needed, we can change it.

David
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.c
Type: application/octet-stream
Size: 547 bytes
Desc: not available
URL: <https://www.bogofilter.org/pipermail/bogofilter-dev/attachments/20031217/59d34609/attachment.obj>