Re: X_OK redefinition protection.

[Christopher Faylor <cgf at redhat dot com>]

On Fri, Apr 20, 2001 at 01:14:38PM -0400, Earnie Boyd wrote:
>Christopher Faylor wrote:
>> 
>> On Thu, Apr 19, 2001 at 11:02:07AM -0400, Earnie Boyd wrote:
>> >I've also sent the sys-unistd file to newlib.
>> >
>> >Earnie.
>> >2001-04-19  Earnie Boyd  <earnie@users.sourceforge.net>
>> >
>> >       * include/sys/file.h (X_OK): Remove redefinition warnings when
>> >       including both sys/unistd.h and sys/file.h.  Make the definition
>> >       consistent with sys/unistd.h.
>> 
>> I've checked in an alternate patch for this.
>> 
>> I don't think there is any reason to protect F_OK, W_OK, and R_OK.
>> Those definitions haven't changed for years.  The real problem was that
>> I somehow got the #ifdef wrong.
>> 
>> Are you actually seeing problems with F_OK redefinition?
>> 
>
>No, just with X_OK.  I just chose to protect the group so that if Cygwin
>decides to do something similar to any of the others it's already
>covered.  Why are they defined in both anyway?

No idea.  I see that other systems (not linux?) do something similar so I
suppose that the reason is lost in the mists of time.

It's not likely that I'll want to do anything similar with the other
variables so we should be ok.

cgf