I wrote a brief description on how to do this problem here. The error you are getting is probably because some of your bins are not initialized. Put this in your code before the print statement:

$binarray[ $i ][ $j ] ||= 0;
[download]

My first guess is that your printing-loop is incorrect.

for(my $i = 0; $i < @binarray; $i++)
{
    for(my $j = 0; $j < @binarray; $j++)
    {
    print "$binarray[$i][$j]\n";
    }
}
[download]

If it isn't, then you need something like: for (my $j = 0; $j < @{ $binarray[$i] }; ...

Update: Also you are not getting an error, you are getting a warning. And if you want some more information about the erros/warnings you get, you can look in `perldoc perldiag` (the perldiag pod), and/or add 'use diagnostics' to your code. (note the last one should not be used in production code (that is to say, code that is finished))

   # Add a complete round angle, i.e. 360 degrees
   $angle += 360 if ($angle < 0);
[download]

In this case adding 180 amounts to the first part of the standard binning transformation of subtracting the minimum value, and dividing by the bin size. In fact the OP's two routines ConvertToFromZero and BinAngles just spread this transformation over two subroutines.

the lowliest monk

Oh, I now see they're angles indeed. What I want to point out is that the OP is doing a 180 degrees translation, while he could mean to evaluate the Principal Argument, in which case should use the transformation I propose above.

Of course, if the problem is only that of binning, for example to produce some graphics, it all reduces to choosing the correct x-axis; OTOH, if the OP wants to use these angles for some evaluations (e.g. sin, cos, stability of a system, whatever) he'd better use the correct angle wrapping function.

Flavio (perl -e 'print(scalar(reverse("\nti.xittelop\@oivalf")))')

Don't fool yourself.

#!/usr/bin/perl

use strict;
use warnings;

use Data::Dumper;

my @binarray;

while (<DATA>)
{
    my ($x,$y)=split / +/;
    
    my $ax = [ (0) x 72 ];
    my $ay = [ (0) x 72 ];
    
    $ax->[int ($x+180) / 5] = 1;
    $ay->[int ($y+180) / 5] = 1;
    
    push @binarray, [$ax, $ay];
}

print Dumper (\@binarray);


__DATA__
-70.6     -27.5 
-60.8      52.9
-123.0    -132.9 
110.1     119.9 
-62.4     -45.6
[download]



holli, /regexed monk/

I think your code is fine, though it could be made more succinct without sacrificing clarity, and it has a strong "C accent". :-). Here's how I would "perlify" it:

use strict;
use warnings;
my $input    = shift;
my $min      = -180;
my $range    = 360;
my $bin_size = 5;
my $n_bins   = $range/$bin_size;
my @binarray = map [ ( 0 ) x $n_bins ], 1..$n_bins;
  
open my $in, $input or die "Oops! Can't open $input: $!\n";
while ( <$in> ) {
  my @indices = map to_index( $_ ), split;
  $binarray[$indices[ 0 ]][$indices[ 1 ]]++;
}
close $in or die "Failed to close $input: $!\n";

for my $row ( @binarray ) {
  print "$_\n" for @$row;
}

sub to_index {
  return ( $_[ 0 ] - $min )/$bin_size;
}
[download]

the lowliest monk

Thanks. Yeah, well my strongest programming language is C, and I have been using it for a lot longer than I have been using perl (although i would not claim to be an expert at C even and am certainly a novice at perl!) so I guess its not surprising my code is a bit 'C' like in nature :)

Thanks for the tips. I'll make a perl programmer yet (in about a decade or so maybe ... lol.)

... but is the approach I used a reasonably sensible one in your opinion?

You have only described the manipulations you are doing to create your graph. Without some idea about what the raw data represents and what you are hoping to discover (or just show) with your graph, it is impossible to say whether your manipulations are good or not.

---

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.

Lingua non convalesco, consenesco et abolesco. -- Rule 1 has a caveat! -- Who broke the cabal?

"Science is about question the status quo. Questioning authority".

The "good enough" maybe good enough for the now, and perfection maybe unobtainable, but that should not preclude us from striving for perfection, when time, circumstance or desire allow.