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

Hi,

I'm trying to split an AoA into different entries if individual scalar values are multiline values. (I thought this was called "pivoting", but looking at Data::Pivot suggests I'm wrong). Super Searching on "split multiline array" doesn't seem to reveal much.

What I've tried already doesn't work as intended (it should be pretty obvious what end result I'm looking for) and I don't think I need to know the width of the arrays.

Any help gratefully received. Does anyone know what the technique should be called for a start? 

use warnings;
use strict;
use Data::Dumper;

my @AoA = ( [qw/single cell values/],
            [qw/are really easy/],
            ["but
multiline
stilton",
"these
cells
is",
"aren't
suck
great"],
           [qw/back to life/],
           [qw/back to reality with more cells/]);

#This is where the garbage begins...
my @new_rows;
my $width = 3;
foreach my $row (@AoA) {
        my @cells = @$row;
        my @foo = map { split /\n/ } @cells;

        my $new_row     = 0;
        my $new_row_idx = 0;
        for(my $idx = 0; $idx<$#foo; $idx++) {
                if(($idx % $width) == 0) {
                        $new_row++;
                        $new_row_idx = 0;
                }
                $new_rows[$new_row][$new_row_idx] = $foo[$idx];
                $new_row_idx++;
        }
}

print Dumper(\@new_rows);

[download]
Cheers
Update: Forgot to say: here I'm only printing out the attempted newly-created array. I'd prefer to re-create the original array with the split data, with the rows in the correct order

davis
Kids, you tried your hardest, and you failed miserably. The lesson is: Never try.

Replies are listed 'Best First'.
Re: Splitting multiline scalars into different array entries
by Tanalis (Curate) on Aug 05, 2005 at 10:36 UTC
    I'm not 100% sure what you need, but assuming you need what I think you need .. :) .., replacing your loop with this works: 
    foreach my $row (@AoA) {
  my @foo = map { split /\n/ } @$row;

  for( my $i = 0; $i < $#foo; $i += $width ) {
    push @new_rows, [ @foo[ $i .. -1 + $i + $width ] ];
  }
}

    [download]
    This outputs:

    Read more... (1182 Bytes)

    Any use to you?

    -- Foxcub
    #include www.liquidfusion.org.uk

[reply]
[d/l]
[select]
      Sorry, should have provided wanted output. That's what I ended up with on my first try :-) wanted output:
      Read more... (1150 Bytes)
      Cheers anyway!

      davis
      Kids, you tried your hardest, and you failed miserably. The lesson is: Never try.
[reply]
[d/l]
        So you need to transpose - this thread should help.

        Flavio
        perl -ple'$_=reverse' <<<ti.xittelop@oivalf

        Don't fool yourself.
[reply]
        I'm certain that this isn't the best way, but it works: 
        foreach my $row (@AoA) {
  my @foo = break_up( $row );

  if( $foo[0] =~ /ARRAY/ ) {
    push @new_rows, @foo;
  } 
  else {
    push @new_rows, [ @foo ];
  }
}

print Dumper(\@new_rows);


sub break_up {
  my $row = shift;
  
  my $uses_newlines = 0;
  $uses_newlines++ if join( "", @$row ) =~ /\n/m;
  
  return @$row unless $uses_newlines;
  
  my @tmp;
  map { push @tmp, [ split /\n/ ] } @$row;
  
  my @row;
  for( my $i = 0; $i <= $#tmp; $i++ ) { 
    for( my $j = 0; $j <= $#tmp; $j++ ) {
      $row[$i][$j] = shift @{$tmp[$j]};
    }
  }
  
  return @row;
}

        [download]
        giving an output of
        Read more... (1150 Bytes)

        Interesting problem :) What's this for, if I can ask?

        -- Foxcub
        #include www.liquidfusion.org.uk

[reply]
[d/l]
[select]
Re: Splitting multiline scalars into different array entries
by broquaint (Abbot) on Aug 05, 2005 at 11:12 UTC 
    use Data::Dumper;

## your @AoA here ...

my @results = map {
  grep(/\n/, @$_)
    ? map([split "\n"], @$_)
    : $_
} @AoA;

print Dumper( \@results );

    [download]
    If you want to update the original data just change the my @results = to @AoA =. Also you could use Scalar::Util::first there instead of grep if you're looking for an optimal solution.
    HTH

    _________
    broquaint

[reply]
[d/l]
      Hmmm... that was my first result. See my desired output. Thanks anyway!

      davis
      Kids, you tried your hardest, and you failed miserably. The lesson is: Never try.
[reply]
        Ah yes, it seems I missed the transposition. 
        use Algorithm::Loops 'MapCar';

my @results = map {
  grep(/\n/,@$_)
    ? MapCar {[@_]} map([split "\n"], @$_)
    : $_
} @AoA;

        [download]
        And to be sure:
        Read more... (1342 Bytes)
        HTH

        _________
        broquaint

[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom