[2184] in Kerberos-V5-bugs
Re: non-sized types in kadm5
daemon@ATHENA.MIT.EDU (Barry Jaspan)
Sat Aug 24 11:02:06 1996
Date: Sat, 24 Aug 96 11:01:51 -0400
From: "Barry Jaspan" <bjaspan@MIT.EDU>
To: raeburn@cygnus.com
Cc: krb5-bugs@MIT.EDU
In-Reply-To: <tx1raoxgr1g.fsf@cygnus.com> (message from Ken Raeburn on 24 Aug
1996 03:49:47 -0400)
Disagree. There are times when types appropriate to the architecture
are better. You might as well try to dictate integer endianness.
I just can't really imagine a situation where it is better to use a
type whose valid range you don't really know, unless you are writing
platform specific code with a known compiler, etc., in which case you
can assume you know what "int" etc. means. So, although sometimes it
can be okay to use unsized ints, when it is really better not to? And
if it is always not worse to use sized types, and often better to do
so, then I conclude it is sensible always to use sized types.
At least we agree on what is actually relevent to krb5 development,
though. :-)
Barry
