Cirollo has asked for the wisdom of the Perl Monks concerning the following question:

What is the best way to count iterations in a foreach loop? Usually, if I need to count iterations, I'll just increment a variable each time through - something like 
$i = 0;
foreach (@array) {
    # do stuff
    $i++;
}

[download]
Is there a better way to do this? Maybe through one of Perl's special variables? I thought about searching for the array index containing the current value (like, "what is the index of @array that contains $_"). Also, I'm not necessarily looking for a solution to give me a total end count of iterations, but more for telling me how many iterations I have gone through so I can say things like "print this line every 3rd time through the loop"

Replies are listed 'Best First'.
Re: Of foreach loops and counting
by eak (Monk) on Aug 09, 2000 at 19:42 UTC
    The easiest way I know of is to do something like. 
    foreach (0..$#array){
  # do stuff, $_ holds iteration num.
}

    [download]


    --eric
[reply]
[d/l]

      Or if you really want to use one of Perl's special variables, try this: 

      
for ($[..$#array) {
  ## Foo
}

      [download]
      since "0" is not necessarily the first element of the array. Of course, nobody should ever, ever change the $[ variable except as part of an obfuscated perl program. :)

[reply]
[d/l]
      Ok, and then I would just get my array values with $array[$_] instead of having the value be in $_. Sounds good to me. Thanks.
[reply]
(jeffa) Re: Of foreach loops and counting
by jeffa (Bishop) on Aug 09, 2000 at 19:47 UTC
    Hmm, if you need to keep track of iteration in a loop, then you should use for instead of foreach. 
    for (my $i=0; $i < $#array; $i++) {

    #print every third line
    print "$array[$i]\n" if (($i % 3)==0);
    
    #or even more consise
    print "$array[$i]\n" unless ($i%3);
}

    [download]
    I know that for is a synonym for foreach, but I simply prefer to use a for construct if I am going to need the current iteration. Guess it's the C inside me. Oh, thanks to KM - I'll be more idiomatic next time. ;)

    /me seeks Priest to exorcise evil C demons.

[reply]
[d/l]
      I just wanted to comment that this is a very C like way of doing this, and is more idiomatically written as (or something similar as far as the loop construct):

      for (0..$#array) {
    print $array[$_];
}

      [download]

      Cheers,
      KM

[reply]
[d/l]
      for is a synonym of foreach. you make the decision on which one to use based on readabilty, since both behave exactly the same.
      --eric
[reply]
Re: Of foreach loops and counting
by ferrency (Deacon) on Aug 09, 2000 at 20:18 UTC
    For testing "every third iteration," I do this sometimes:

    my $n = 0;
foreach (@array) {
    next if I_dont_care_about($_);
    do_something_with ($_);
    print "." if !($n++ % 3);
}
print "\n";

    [download]
    If you aren't necessarily processing every item in the list, but only want to count the number of elements you process, then using a temporary variable as you were doing is a useful alternative to looping over the array indices. It also works when you're doing things like looping over hash keys, where you can't count on having a numeric index implicitly; or processing lines of a file with a while (<>) loop.

    Update: KM, that is definitely true, and a more efficient and succinct way to count lines when you're interested in the total line count. The method I presented is still useful when you want to count the number of lines you processed, not the total line count so far.

    Alan

[reply]
[d/l]
[select]
      or processing lines of a file with a while (<>) loop.

      In that case, you can use $. (perldoc perlvar) to know what line you are in. For example:

      #!/usr/bin/perl

while (<DATA>) {
        print $. . ": $_";
}

__DATA__
a
b
c
d
e
f
g
h
i

# Would display when run...

$ ./test.pl
1: a
2: b
3: c
4: d
5: e
6: f
7: g
8: h
9: i

      [download]

      Cheers,
      KM

[reply]
[d/l]

Back to Seekers of Perl Wisdom