1 Jun
2007
1 Jun
'07
3 p.m.
> > For example, what would you expect from:
> > N = 1;
> > A[N++] = N++;
On the other
hand, I'd fire anyone who wrote something like the above.
Out of a cannon,
hopefully. Unfortunately, I've seen my share of
messes caused by tricks similar to this.
. . . followed by complaints that the
compiler didn't do what they
expected, and claims that there is a bug in the compiler, because it
behaved differently than their "baby duck" compiler.
In the first semester of C, I try to teach students to optimize for
readability, and to avoid compound steps. But, in the second semester, I
get some students who have learned bad habits, and think that they are
"optimizing" their code when they attempt to write Holub-style
"puzzle"
code. Even an erudite explanation, such as what Dave posted, gets a
response of "well, it works on xxx compiler!"
It is true, that scope of variables IS defined, and should function as
expected. But there were a LOT of variations in early compilers, some
legit, and some bad.
Complex compound lines, even when they follow all of the rules, such as:
while (*T++=*S++);
make finding and isolating problems difficult.
--
Grumpy Ol' Fred cisin at xenosoft.com