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

In this node I got some advice on how to post and retrieve multiple rows of data. I am now getting an error using this: "Can't use string ("1") as an ARRAY ref while "strict refs" in use at...". I am using Mason, and one of the comments in the aforementioned node was trying to eliminate this problem, but I still get the same error (pointing to the "init" line).

The form being posted looks like this: 
<input type="hidden" name="edit_item" value="1">
<input type="text" name="edit_name_1" value="hello" size="20">

[download]
The code that processes this looks as follows... 
my @editlineitems = @{ $ARGS{ edit_item } };
foreach my $line (@editlineitems) {
  my $name = $ARGS{"edit_name_$line"};
}

[download]
Any help would be appreciated!


Michael Jensen

Replies are listed 'Best First'.
Re: can't use string as an array ref....
by monarch (Priest) on Jul 13, 2005 at 00:30 UTC
    It is possible, even highly likely, that $ARGS{edit_item} is a scalar containing the string "1".

    The complaint would be that you are attempting to dereference an array by sticking @{...} around the scalar item.

    You can confirm whether this is the case by commenting out the (potentially) offending code, and replacing it with:

      {
    use Data::Dumper;
    $m->out( '<pre>' . Dumper( \%ARGS ) . '</pre>' );
  }

    [download]
[reply]
[d/l]
[select]
Re: can't use string as an array ref....
by monarch (Priest) on Jul 13, 2005 at 00:35 UTC
    So, after looking at your code, I'm guessing what you want is to process a form that looks something like
      <input type="text" name="edit_name_1" value="hello" size="20">
  <input type="text" name="edit_name_2" value="world" size="20">
  <input type="text" name="edit_name_3" value="form" size="20">

    [download]
    ..where you have a number of edit_name_n fields.

    I suggest you do the following:

      foreach my $key ( keys %ARGS ) {
    # skip any arguments that aren't of interest
    next if ( $key !~ m/^edit_name_(\d+)$/ );

    my $num = $1; # captured from the regexp
    my $name = $ARGS{$key};

    $m->out( "<p>Edit field $num = $name</p>" );
  }

    [download]
[reply]
[d/l]
[select]
      monarch has a nice, clean and safe proposal.

      The only thing I'd suggest is a sort befor keys, so that you get the edits in order.

           "Income tax returns are the most imaginative fiction being written today." -- Herman Wouk

[reply]
      That works really great, thank you! I have more than one field (I simplified it for the sake of easy duplication), so besides edit_name_N I may have edit_type_N, what is the best way to work that in too? Another "next if" won't work, and when I just do an if it gives me everything posted.

      Thanks a ton! I like this way better than the other way from the other node I think.


      Michael Jensen
[reply]
        There are numerous ways of approaching this. Two suggestions are:
          foreach my $key ( keys %ARGS ) {
    if ( $key =~ m/^edit_name_(\d+)$/ ) {
      my $num = $1; # captured from the regexp
      my $name = $ARGS{$key};
      $m->out( "<p>Edit name $num = $name</p>" );
    } elsif ( $key =~ m/^edit_type_(\d+)$/ ) {
      my $num = $1; # captured from the regexp
      my $type = $ARGS{$key};
      $m->out( "<p>Edit type $num = $type</p>" );
    }
  }

        [download]

        The second method is to generate hashes to play with:

          my %edit_name = ();
  my %edit_type = ();
  foreach my $key ( keys %ARGS ) {
    if ( $key =~ m/^edit_name_(\d+)$/ ) {
      $edit_name{$1} = $ARGS{$key};
    } elsif ( $key =~ m/^edit_type_(\d+)$/ ) {
      $edit_type{$1} = $ARGS{$key};
    }
  }

  # play with values
  foreach my $num ( sort { $a <=> $b } keys %edit_name ) {
    $m->out( "<br />Name number $num is " . $edit_name{$num} );
  }
  foreach my $num ( sort { $a <=> $b } keys %edit_type ) {
    $m->out( "<br />Type number $num is " . $edit_type{$num} );
  }

        [download]

        Of course the name and type might be related! In which case you might consider the following:

          my %edit = ();
  foreach my $key ( keys %ARGS ) {
    if ( $key =~ m/^edit_([^_]+)_(\d+)$/ ) {
      $edit{$2}->{$1} = $ARGS{$key};
    }
  }

  # now %edit contains a hash of rows. In each row
  #   is the type and name fields. e.g.
  # %edit = ( 1 => { name => 'John', type => 'student' },
  #           2 => { name => 'Lisa', type => 'teacher' }
  #         );
  $m->out( "<table>" );
  foreach my $row ( sort { $a <=> $b } keys %edit ) {
    if ( ! ( exists( $edit{$row}->{name} &&
             exists( $edit{$row}->{type} ) ) {
      die( "Missing field for row $row" );
    }
    $m->out( "<tr><td>" . $edit{$row}->{name} . "</td>"
             . "<td> . $edit{$row}->{type} . "</td></tr>"
           );
  }
  $m->out( "</table>" );

        [download]

        updated: cleaned up code

[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom