R. Joseph Newton wrote:
>
> Rob Dixon wrote:
>
> > > foreach (@lines) {
>
> # too many lines here
>
> > > }
> > >
> > > While the purpose of the above may bve totally
> > > incomprehesible, there is no question about what $_ is. <;:-o)
> >
> > One proviso here. I always feel very uncomfortable about
> > explicitly assigning to $_. As I implied in my post to
> > Wiggins, $_ is very much the equivalent to 'it', and I would
> > no more use 'it' over more than a simple English sentence
> > than I would use $_ across nested loops, procedure calls or
> > whatever. In English your lines above say
>
> You're right. I actually got carried away. I was having so
> much fun coming up with the most pointless code possible. that
> I forgot the main point. The use of the idefault $_ did serve
> its purpose, though, perverse as it was. I should probably
> have done the print outside the loop to demonstrate both the
> purpose and danger of using "it":
>
> my @lines = (<DATA>);
>
> foreach (@lines) {
> chomp;
> s/dog/cat/g;
> my @words = split;
> $words[0] = lc $words[0];
> $_ = join ' ', reverse @words;
> }
>
> print"$_\n" for @lines;
> #same data, results as previous sample
>
> ...has the same effect as printing from inside the loop.
> Which means that my array contents
>
> have either been effectively modified, or totally hashed,
> depending on whether that was the desired effect.
..except that, to be picky, you need
print ucfirst "$_\n" for @lines
outside the loop. And that this is a warning for 'foreach'
in general, not specifically for $_. If your loop had been
foreach my $line (@lines) {
:
}
then the named scalar would similarly have been aliased
with the array elements.
Cheers,
Rob
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>
