Re: [bug-gnulib] same, utimecmp should depend on minmax

[Paul Eggert]

Derek Price <address@hidden> writes:

> Maybe, when I'm looking at only one file, but I think it takes a little
> extra brain power worrying when to and not to include minmax.h.

Sorry, I wasn't clear enough.  The problem I'm worried about is older
hosts where <sys/param.h> defines MIN, and where <sys/param.h> also
collides with other random include files.  If minmax.h has the
<sys/param.h> stuff in it, it will be more likely to cause problems in
files that include it but not <sys/param.h>.

(Though perhaps this is an argument against the <sys/param.h> stuff
entirely -- I don't know.)