"Jenda Krynicky" <[EMAIL PROTECTED]> wrote in message
3E01E02F.17624.1ED52A1C@localhost">news:3E01E02F.17624.1ED52A1C@localhost...
> From: "Perl" <[EMAIL PROTECTED]>
> > This should work (but beware - it is untested :)
> >
> > Rob
> >
> > my @required = qw(head1 head6 head8);
> > my $line;
> >
> > for (<DATA>)
>
> It's much better to use
> while (<DATA>)
>
> This "for (<DATA>)" forces Perl to read the whole file into an array
> in memory. (Well ... unless the optimizer is bright enough to change
> the command to "while (<DATA>)". Not sure it is. You should not
> depend on the optimizer.)
I agree completely. Also the optimiser shouldn't translate this to a while()
as optimisers aren't supposed to change code semantics at all.
> > {
> > chomp;
> > my @field = split /,/;
>
> While this may be fine for column headers I would not recommend doing
> this with actual data. What if some of the fields are quoted and what
> if they contain commas? use Text::CSV_XS instead.
>
Also true, but the problem was vaguely defined and I was coding for the
simplest case.
[snip code]
>
> P.S.: Rob, could you change the comment part of your mail address?
> It's kinda strange to see a post comming from "Perl". :-)
Yes, my apologies for this. I used to receive and send the newsgroup posts
as emails and just today set up the perl.org NNTP server - wrongly! Things
should be correct now.
Cheers,
Rob
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
