Nested loops

robert44444uk has asked for the wisdom of the Perl Monks concerning the following question:

Hello Monks. I've managed to get the following modular exercise to work, which looks at a range of integers, and provides the total of those which are not divisible by 13 or 17.

#!/usr/bin/env perl
use warnings;
use strict;
use feature ':5.10';
use File::Slurp;

my @lines = (1..2310) ; 
my @s = (13,17);
my @list1 = (0..$s[0]-1);    
my @list2 = (0..$s[1]-1);

foreach $a (@list1)
{
    my @remain1 = grep { $_% $s[0] ne $a} @lines;
        
    foreach $b (@list2)
    {
        my @remain2 = grep { $_% $s[1] ne $b} @remain1;
        say $s[0], " ", $a, " ",$s[1]," ", $b, " ", scalar @remain2;
    } 
}
[download]

# but the text below which plans to add a further prime number to the integer sequence and puts in a further nested foreach loop, does not work. It gives messages that $c is not defined.

#!/usr/bin/env perl
use warnings;
use strict;
use feature ':5.10';
use File::Slurp;

my @lines = (1..2310); 
my @s = (13,17,19);
my @list1 = (0..$s[0]-1);
my @list2 = (0..$s[1]-1);
my @list3 = (0..$s[2]-1);

foreach $a (@list1)
{
    my @remain1 = grep { $_% $s[0] ne $a} @lines;
    foreach $b (@list2)
    {
        my @remain2 = grep { $_% $s[1] ne $b} @remain1;    
        foreach $c (@list3)
        {
            my @remain3 = grep { $_% $s[2] ne $c} @remain2;
            say $s[0], " ", $a, " ",$s[1]," ", $b, " ", $s[2]," ", $c,
+ " ", scalar @remain3;
        } 
    }
}
[download]

#I'm not sure what I am doing wrong here. BTW I am totally new to perl and to programming - this is my very first perl program!

Comment on Nested loops Select or Download Code

Replies are listed 'Best First'.
Re: Nested loops by nuance (Hermit) on Sep 13, 2015 at 13:42 UTC
$a and $b are special variables in perl, $c isn't. You have forgotten to put my statements in your foreach loops. The $a and $b that you're using are package variables that exist, so you don't get a waring. $c is not a pre-existing package variable so you get a warning. It's probably not a great idea to use $a and $b at all. If you were going to, the loops need to look like `foreach my $a (@list1)` *Nuance*	[reply] [d/l]
Re^2: Nested loops by robert44444uk (Acolyte) on Sep 13, 2015 at 13:45 UTC
Ahh, that's very kind, nuance. I keep forgetting to put "my" at the front of these things.	[reply]
Re^3: Nested loops by stevieb (Canon) on Sep 13, 2015 at 14:05 UTC
As nuance stated, `$a` and `$b` are special variables to perl (used in things like `sort`. They don't need to be declared with `my` because of this). It's highly advisable that you don't use these special variables except for where they are normally expected.	[reply] [d/l] [select]
Re^3: Nested loops by AnomalousMonk (Archbishop) on Sep 13, 2015 at 15:22 UTC
`$a` and `$b` are just the beginning of Perl's pre-defined special variables. Please see perlvar (or `perldoc perlvar` on your local machine) for the whole story. Give a man a fish: `<%-{-{-{-<`	[reply] [d/l] [select]
Re: Nested loops by Laurent_R (Canon) on Sep 13, 2015 at 15:03 UTC
You've been given the answer to your question about $c not being defined, as well as additional material on why this does not happen with $a and $b. I will not comment further on that. In my humble opinion, your algorithm is quite poor. The simple fact that, when given a third number to check, you need to add another nested loops, shows that this is not quite right. You need an algorithm that will still work without any code change even if we add one, two or more numbers to the list of divisors to check. You could use two different routes. One is to loop on your range of integers and, for each of them, check if it is a multiple of the numbers stored in @s, and to store it in a @result array (or to print it out) it it is not a multiple of any of the @s numbers. The other one would be to calculate all the multiples of the numbers in @s that are smaller than your upper bound, store them in a hash, and grep on that hash the full range. Update: A short one-liner test showing the second option above (a variation on the sieve of Eratosthenes algorithm for prime numbers): $ perl -e 'my @lines = (1..2310) ; my @s = (13, 17); my %multiples; > for my $div (@s) { > $multiples{$_ * $div} = 1 for 1..200; > } > my @result = grep {not exists $multiples{$_} } @lines; > print "@result"; > ' 1 2 3 4 5 6 7 8 9 10 11 12 14 15 16 18 19 20 21 22 23 24 25 27 28 29 3 +0 31 32 33 35 36 37 38 40 41 42 43 44 45 46 47 48 49 50 53 54 55 56 5 +7 58 59 60 61 62 63 64 66 67 69 70 71 72 73 74 75 76 77 79 80 81 82 8 +3 84 86 87 88 89 90 92 93 ... 2299 2300 2302 2303 2304 2305 2306 2307 2308 2309 2310 [download]	[reply] [d/l]
Re^2: Nested loops by robert44444uk (Acolyte) on Sep 14, 2015 at 21:42 UTC
Thank you all for your contributions and thank you Laurent_R for your improved code	[reply]
Re^3: Nested loops by shadowsong (Pilgrim) on Sep 14, 2015 at 22:26 UTC
Hi robert44444uk, Just to further add to what the Monks above have imparted - ever heard of recursion? #! perl -slw use strict; sub count_id (\@\@); # "count indivisible dividends" fnc prototype sub count_id (\@\@) { my ($divisors,$range) = (shift,shift); return 0 unless @$range && @$divisors; # check terminal condition # apply the $divisors to the dividend for this recursive call my ($aggregate,$dividend) = (0,pop @$range); for (@$divisors) { return count_id(@$divisors,@$range) if !($divid +end % $_); } return 1 + count_id(@$divisors,@$range); } my @divisors = (13,17,19); my @range = (1..2310); print "Number of indivisible items: ",count_id(@divisors,@range),"\n"; __END__ [download]	[reply] [d/l]
Re^4: Nested loops by robert44444uk (Acolyte) on Sep 20, 2015 at 18:58 UTC
Re: Nested loops by danaj (Friar) on Sep 16, 2015 at 02:58 UTC
`use ntheory ":all"; use feature "say"; say scalar(grep { gcd($_,131719)==1 } 1..2310);` [download] Seems much easier than all the looping or recursion. The gcd will return 1 if the two inputs have no common factors -- if it isn't divisible by 13, 17, or 19 in this example.	[reply] [d/l]
Re^2: Nested loops by robert44444uk (Acolyte) on Sep 20, 2015 at 18:56 UTC
Hi danaj, fancy seeing you here! Thank you for this, will give it a work through when I get some time. You know why I am trying to do this program from Mersenneforum and it will need to be pretty efficient as I want to add more primes to get to an under-researched prime gap formula. Hard to beat your 1 liner, I think.	[reply]