Split problem

[lmbylsma]

I have a data file that looks something like this:

1 ABABAB 123456 2223 0.626903553299492
1 ABABAB 123465 60 0.0169204737732657
2 ABABAB 123546 60 0.0169204737732657
1 ABABAB 123564 9 0.00253807106598985
7 AABBAB 123645 8 0.00225606316976875
1 ABABAB 123654 13 0.00366610265087422
10 AAABBB 124356 131 0.0369430344049633

What I want to do is put all this information into an array, splitting it on \t but then further splitting columns 2 and 3 (e. g. ABABAB, 123456) by character so that the letters and numbers will each be in a separate element of the array.

I know how I can use split to split up everything by where the tabs are @array=(/\t/, $line) or everything by character @array=(//, $line), but I can't figure out how to tell it to split differently depending on which column its on.

Is there a simple way to do this?

 

[raklet]

It is simple to do, but not in a single operation. I would do something like this:

my ($col1,$col2,$col3,$col4,$col5) = split/\s+/,$line;
push @array, $col1;
push @array, split//,$col2;
push @array, split//,$col3;
push @array, $col4;
push @array, $col5;
print "$_\n" foreach @array;

 

[lmbylsma]

That would work if I just wanted to print it out, but I want to be able to work with the individual elements later on. With that example it puts everything from a line into one element resulting in a single dimensional array, but I need to refer to specific elements like the "A" in the first line by referring to @array[0]->[1], etc.

 

[siberian]

Sure so just do this then

push $array[1], split//,$col2;
push $array[2], split//,$col3;

Assign the array of element to the array. An array of arrays.


You may have to twiddle that syntax a bit , its conceptual, I did not test it.

 

[mikevh]

How about splitting the lines in the usual way and then
splitting the relevant columns of that array later (or
even using substr) to access the individual characters?

Example:

#!perl
use strict;

open (FILE, "ibys.txt") || die qq(Can't open "ibys.tst" for input\n);

my @array;

while (defined($_=<FILE>)) {
	chomp;
	push @array, [ split /\s+/, $_ ];
}

for (@array) {
        #print chars in 2nd col., comma-separated
	print join(&quot;,&quot;, split(//, $_->[1])), &quot;\n&quot;;

        #print chars in 3rd col., blank-separated, a
        #different way
	for my $i (0..length($_->[2] - 1)) {
		print substr($_->[2], $i, 1), &quot; &quot;;
	}
	print &quot;\n&quot;;
}

Output using example code and data you posted:
A,B,A,B,A,B
1 2 3 4 5 6
A,B,A,B,A,B
1 2 3 4 6 5
A,B,A,B,A,B
1 2 3 5 4 6
A,B,A,B,A,B
1 2 3 5 6 4
A,A,B,B,A,B
1 2 3 6 4 5
A,B,A,B,A,B
1 2 3 6 5 4
A,A,A,B,B,B
1 2 4 3 5 6

 

[lmbylsma]

mikevh, Sure that works for printing but that's not what I want to do with it. I'm going to be doing some more complicated things for which I need everything to be in one array where I can access specific elements.

What Siberian suggested is too simple to work, but if I went along those lines I'd actually have to do something like this because each piece that is split off of $col2 and $col3 needs to go into a separate element:

while (<>) {

chomp;

my ($col1,$col2,$col3,$col4,$col5) = split/\t/,$_;
my @col2=split //, $col2;
my @col3=split //, $col3;

$array[$i]->[0]=$col1; #first column

$array[$i]->[1]=$col2[0]; #2nd column split into 6 elements
$array[$i]->[2]=$col2[1];
$array[$i]->[3]=$col2[2];
$array[$i]->[4]=$col2[3];
$array[$i]->[5]=$col2[4];
$array[$i]->[6]=$col2[5];

$array[$i]->[7]=$col3[0]; #3rd column split into 6 elements
$array[$i]->[8]=$col3[1];
$array[$i]->[9]=$col3[2];
$array[$i]->[10]=$col3[3];
$array[$i]->[11]=$col3[4];
$array[$i]->[12]=$col3[5];

$array[$i]->[13]=$col4; #4th column
$array[$i]->[14]=$col5; #5th column

$i++;

}


Which seems really inefficient. It works for what I want to do, but isn't there simpler more efficient way? If I had a lot more columns than this, this would be really hard to do in this manner.

 

[mikevh]

You may find this a bit neater.

#!perl
use strict;

my @array;
my $i=0;

while (<>) {
 
 chomp;
 
 my ($col1,$col2,$col3,$col4,$col5) = split/\s+/,$_;
 
 $array[$i]->[0]=$col1;  #first column
 
 $array[$i]->[1]=[ split //, $col2 ]; #2nd col split into 6 elems 
 
 $array[$i]->[2]=[ split //, $col3 ]; #3rd column split into 6 elements
 
 $array[$i]->[3]=$col4; #4th column
 $array[$i]->[4]=$col5; #5th column
 
 $i++;

}

for (@array) {
	#print contents of 2nd column
	for (@{$_->[1]}) {
		print qq($_,);
	}
	print &quot;\n&quot;;
	#print contents of 3rd column
	for (@{$_->[2]}) {
		print qq($_ );
	}
	print &quot;\n&quot;;
}

There's a nice module called Data:umper which I've recently discovered. If you have this on your machine,
you can say

use Data:umper;

near the top of your program, and

print Dumper(@array);

to see what @array looks like. It's a very helpful debugging tool when you're working with complicated data
structures.

 

[mikevh]

Hmm, even more concise and perhaps also clearer:

while (<>) {
 chomp; 
 my ($col1,$col2,$col3,$col4,$col5) = split/\s+/,$_;
 
 push @array, [ 
                $col1, 
                [ split //, $col2 ],
                [ split //, $col3 ],
                $col4,
                $col5,
               ]; 
}

or, with $i:

my $i=0;

while (<>) {
 chomp;
 my ($col1,$col2,$col3,$col4,$col5) = split/\s+/,$_;
 
 $array[$i++] = [ 
                  $col1, 
                  [ split //, $col2 ],
                  [ split //, $col3 ],
                  $col4,
                  $col5,
                 ]; 
}