[11452] in Athena Bugs
Re: sun4 7.6P: codecenter
daemon@ATHENA.MIT.EDU (Calvin Clark)
Tue Nov 30 15:59:14 1993
Date: Tue, 30 Nov 93 15:58:56 -0500
From: Calvin Clark <ckclark@MIT.EDU>
To: alexp@MIT.EDU
Cc: bugs@MIT.EDU
In-Reply-To: Your message of Tue, 30 Nov 93 15:12:31 -0500.
> Please let me know if this solves your problem.
Since I was debugging a program which allocates pseudo-ttys on that
machine earlier, it is possible that some of them were left in a weird
state and this is what was causing the problem. The code given in the
is the program you sent me is the correct way to allocate ptys on a
Solaris machine (at least according to the documentation), so it's
probably not the fault of CodeCenter.
I think you might as well resolve the bug. If I see it again under
normal circumstances, I'll send in another report.
> Please also let me know if your other bug ("can't recognize 'const'") has been
> resolved so I can close it out. Thanks,
I think that the CodeCenter should recognize "const" even if the "ansi"
option is not set, just as gcc will recognize it even if you don't
specify the -ansi option. K&R C is not a standard. If it is the
intention that the default language interpereted by CodeCenter should
allow certain constructs used by "traditional" C compilers, that is
fine, but I don't see why it needs to lobotomize itself. It should
allow, but not require, ANSI C in the default mode. In my opinion, to
do otherwise is not necessarily a bug, but a poor design decision. If
in the default mode CodeCenter is trying to emulate some old C compiler,
which one is it trying to emualate? I know of quite a few, and they
are all different. Is there some great market need to support the
syntax of broken C compilers?
You may resolve the "const" report, too, but you might forward
my comments to the CodeCenter support list.
-Calvin
