getting a three dimensional array from flat text

[littleman]

Well I came up with this:

open(INF,"flat.txt");
 my(@wlines) = <INF>;
 close(INF);
 foreach $t(@wlines) {
  $d .= $t;
 };
 @each_line = split(/\n/,$d);
 $cut =0;
 foreach $word_group(@each_line){
  @{$words[$cut]} = split(/\¦/,$word_group);
  $cut++;
 }

I was needed to dice up a flat text file into a three dimensional array and this is the approach I came up with. It is made to work with a pipe delimited ('¦') file.
like this:

a1¦a2¦a3...
b1¦b2¦b3..
and so on...

so $words[0][1] would be 'a2' and $words[1][0] would be 'b1'. I couldn't find anything like this on the web to do this so I hacked this code together, so if anyone has a cleaner approach, please post it here.

[littleman]

I just wanted to float this back up now that this forum is jumping. If anyone has an improvement / variation on this please post away.

[sugarkane]

I did it this way:

open (FP, "flat.txt");
chomp(@lines=<FP>);
close (FP);
$tmp1=0;
$tmp2=0;
foreach $i (@lines) {
@temp=split(/\¦/, $i);
foreach $foo (@temp) {$words[$tmp1][$tmp2++]=$foo;}
$tmp2=0;
$tmp1++;
}

I'm not sure if it's an improvement - I've not benchmarked it - but it's another way.

Your @{$words[$cut]} is an interesting construct, I must look into that...

[littleman]

Thanks SK,

You thought me something,. It looks like my scriptlet could be weeded down.
This:
foreach $t(@wlines) {
$d .= $t;
};
@each_line = split(/\n/,$d);
$cut =0;
foreach $word_group(@each_line){

could be replaced with this:
$cut =0;
foreach $word_group(@wlines){

I didn't know that perl will automatically divide up an array along the newline character. That is cool.