c99/c++ localised variable definition
Paul Richards
paul at originative.co.uk
Mon Jan 31 04:58:03 PST 2005
On Mon, Jan 31, 2005 at 11:26:09PM +1100, Andrew Reilly wrote:
> On Mon, Jan 31, 2005 at 10:26:30AM +0000, Paul Richards wrote:
> > 3) Usage in for loops may be more useful than other uses.
> [snip]
> > I think the loop usage though is one clear example where it is
> > clearer. I think there are others as well; where the usage of the
> > variable is clearly localised it is much easier to see a local
> > definition than to have to jump back and forth to find out what
> > variables are.
>
> I'd just like to raise a dissenting voice to this particular
> point. I find the for-loop initialization syntax a pernicious
> source of errors, mainly because of the non-intuitive scope of
> the definition. I.e., it looks like it's equivalent to "int i;
> for (i = 0;;)" but it isn't.
>
> If you carelessly c++-ify a loop like:
>
> for (int i = 0; i < N; i++)
> {
> if (some_condition(i)) break;
> }
> do_something_with(i); /* use finishing index */
>
> you can miss the fact that the value of i is used outside of the
> loop. The newly created scope for "i" shadows the presumably
> pre-existing definition of i at the top of the function, which
> is what do_something_with() gets to see.
Depends whether you find it non-intuitive or not. I find this usage
makes it clearer that the loop variable only exists within the scope
of the loop. If I needed to code the above example then I know that
my for construct is doing more than just looping, it's also returning
a value and therefore it needs to affect a variable outside of its
scope. I think use of this syntax makes it clearer what you intend
in each case.
The converse is true as well though, where a global is inadvertently
used as a loop variable.
--
Paul Richards
More information about the freebsd-arch
mailing list