I'm getting kind of tired of doing: #ifndef __NetBSD__ around prototypes
of re_comp and re_exec.  I find this situation wierd for two reasons.
1) It's 4.2 compatibility, or so CVS-1.3 says.  You'd think we could compile
   4.2 BSD stuff w/o probs.
2) Normally, I've seen most calls converted to "const"ness, but the problematic
   prototypes are un-"const"ed in NetBSD.
3) Nobody else seems to have a problem with it since there's no conditional
   cpp structures around it.

So what's the clean way to deal with it?  I remember somewhere that someone
suggested all this #if(n)def My_OS_Here was ugly, but I forget the alternative
he suggested... maybe #ifdef CONST_RE_COMP and the same old #ifdef __NetBSD__
around a #define CONST_RE_COMP in a global header or something.
The issue, of course (I think!), is that OS's mutate and yet the source code
(near the re_comp protos) really shouldn't show changes just because there's
a new supported OS or your fav. OS has mutated beyond recognition.

Then again, I apparently can't count and reason may have followed suit.
-- 
VaX#n8 (vak-sa-nate) - n, CS senior++ and Unix junkie - vax@ccwf.cc.utexas.edu
Deal with evil through strength, yet encourage good through trust.    - PGP me