Re^4: Accessing an array of anonymous hashes

What do you think about this structure?

my $transaction_record_fields = {
  'Batch_Agency'         =>  { COL => 0, POS => [  1 ..   3] },
  'Batch_Date'           =>  { COL => 1, POS => [  4 ..  11] },
  'Batch_Type'           =>  { COL => 2, POS => [ 12 ..  12] },
};
[download]

Eventually, I'll be inserting some data into each column, so it will end up looking like this:

my $transaction_record_fields = {
  'Batch_Agency'         =>  { COL => 0, POS => [  1 ..   3], VALUE =>
+ "AAA" },
  'Batch_Date'           =>  { COL => 1, POS => [  4 ..  11], VALUE =>
+ "BBBBBBBB"},
  'Batch_Type'           =>  { COL => 2, POS => [ 12 ..  12], VALUE =>
+ "C"},
};
[download]

I like the idea of being able to insert by fieldname rather than by a numeric index, and this structure supports that.

OTOH, that fact that I have to hard-code the column index seems a little funky to me.

Comment on Re^4: Accessing an array of anonymous hashes Select or Download Code

Replies are listed 'Best First'.
Re^5: Accessing an array of anonymous hashes by eric256 (Parson) on Mar 26, 2007 at 18:41 UTC
Generaly you would want your data definition, and your data separate. You would then apply that definition to your data. I would think you end up with an AoA for the definition and then an array of data (or AoA if you have multiple rows of data. The column of the data then corresponds directly to the column of the definition. use strict; use warnings; #definition AoA my $transaction_records = [ [ 'Batch_Agency' => [ 1 .. 3] ], [ 'Batch_Date' => [ 4 .. 11] ], [ 'Batch_Type' => [ 12 .. 12] ], ]; my $i = 0; #build a map of column name to column my $key_map = { map { $_->[0] => $i++ } @$transaction_records }; #then your rows could look like my $row1 = [ 'agc', '05-JAN-2007', 'type']; my $row2 = [ 'agc', '05-JAN-2007', 'type']; #and you could access data like $row1->[ $key_map->{Batch_Agency} ] = 'abc'; #multiple rows could look like my $rows = [ $row1, $row2]; for my $i ( 0,1,2) { print $transaction_records->[$i]->[0], " = ", $row1->[$i], "\n"; } [download] ___________ Eric Hodges	[reply] [d/l]

Replies are listed 'Best First'.

Re^5: Accessing an array of anonymous hashes
by eric256 (Parson) on Mar 26, 2007 at 18:41 UTC

Generaly you would want your data definition, and your data separate. You would then apply that definition to your data. I would think you end up with an AoA for the definition and then an array of data (or AoA if you have multiple rows of data. The column of the data then corresponds directly to the column of the definition.

use strict;
use warnings;

#definition AoA   
my $transaction_records = [
   [ 'Batch_Agency'         =>  [  1 ..   3] ],
   [ 'Batch_Date'           =>  [  4 ..  11] ],
   [ 'Batch_Type'           =>  [ 12 ..  12] ],
];
my $i = 0;

#build a map of column name to column
my $key_map = { map { $_->[0] => $i++ } @$transaction_records };

#then your rows could look like
my $row1  = [ 'agc', '05-JAN-2007', 'type'];
my $row2  = [ 'agc', '05-JAN-2007', 'type'];

#and you could access data like
$row1->[ $key_map->{Batch_Agency} ]  = 'abc';

#multiple rows could look like
my $rows = [  $row1, $row2];


for my $i ( 0,1,2) {
    print $transaction_records->[$i]->[0], " = ", $row1->[$i], "\n";
}
[download]

___________
Eric Hodges

[reply]
[d/l]