Why is part of slice duplicated?

the_0ne has asked for the wisdom of the Perl Monks concerning the following question:

Hi monks, haven't been on in awhile and haven't posted in an even longer while. Having trouble with a hash slice. Somehow I'm duplicating a slice of data, but have no idea how I'm doing it. Here's the relevant code...

$to = "foo\@bar.com";
$subject = "New CDs now";
$file_name = "AAAAA.txt";
$from = "that\@person.com";
$last_dir = "Spam";

$spam->{$to} = $to;
$spam->{$to}->{$subject} = $subject;
$spam->{$to}->{$subject}->{$file_name} = $file_name;
$spam->{$to}->{$subject}->{$file_name}->{'path'} = $last_dir;
$spam->{$to}->{$subject}->{$file_name}->{'from'} = $from;

$to = "foo2\@bar.com";
$subject = "New CDs now";
$file_name = "BBBBB.txt";
$from = "that2\@person.com";
$last_dir = "Spam";

$spam->{$to} = $to;
$spam->{$to}->{$subject} = $subject;
$spam->{$to}->{$subject}->{$file_name} = $file_name;
$spam->{$to}->{$subject}->{$file_name}->{'path'} = $last_dir;
$spam->{$to}->{$subject}->{$file_name}->{'from'} = $from;

print "\n";
for my $email_slice (sort keys %{$spam}) {
  print "email_slice - $email_slice\n";

  for my $subject_slice (sort keys %{$email_slice}) {
    print "subject_slice - $subject_slice\n";

    for my $fn_slice (sort keys %{$subject_slice}) {
      print "fn_slice - $fn_slice\n";
    }
  }

  print "\n";
}
[download]

Sorry for the sloppy code, however, this comes from a very large script and just to figure this part out I wrote this code into a separate smaller script for debugging.
Most of you should be able to just paste this code into a vi session and run it, nothing special. Here's the output...

email_slice - foo2@bar.com
subject_slice - New CDs now
fn_slice - AAAAA.txt
fn_slice - BBBBB.txt

email_slice - foo@bar.com
subject_slice - New CDs now
fn_slice - AAAAA.txt
fn_slice - BBBBB.txt
[download]

Notice how the fn_slice is duplicated? What I'm trying to get is...

email_slice - foo2@bar.com
subject_slice - New CDs now
fn_slice - BBBBB.txt

email_slice - foo@bar.com
subject_slice - New CDs now
fn_slice - AAAAA.txt
[download]

Notice how the AAAAA.txt and BBBBB.txt are separated by their respective email addresses? I can not imagine how the fn_slice is getting duplicated? I think it has something to do with the $subject since it's the same, however, the email_slice should take care of that and separate it.

Thank you monks for your help, AGAIN! :)

Comment on Why is part of slice duplicated? Select or Download Code

Replies are listed 'Best First'.
Re: Why is part of slice duplicated? by Abigail-II (Bishop) on May 28, 2002 at 17:18 UTC
`$spam->{$to} = $to; $spam->{$to}->{$subject} = $subject; $spam->{$to}->{$subject}->{$file_name} = $file_name; $spam->{$to}->{$subject}->{$file_name}->{'path'} = $last_dir; $spam->{$to}->{$subject}->{$file_name}->{'from'} = $from;` [download] You are doing some very fishy stuff here, which `use strict;` would not have allowed you to. First you set `$spam->{$to}` to `$to`, which is a string. But then, in the next line, with `$spam->{$to}->{$subject}`, you are treating it as a hashreference! And later on, you are doing the same trick with `$subject` and `$file_name`. You should use `Data::Dumper` and display the context of `$spam`. You will be surprised what it contains (or rather, what it doesn't contain). Remember, unless you know what you are doing, always use strict;. Abigail	[reply] [d/l] [select]
Re: Re: Why is part of slice duplicated? by the_0ne (Pilgrim) on May 28, 2002 at 17:48 UTC
Ok, I see what you mean. I added 'use strict;' and I get... `Can't use string ("foo@bar.com") as a HASH ref while "strict refs" in +use at ./test.pl line 14.` [download] How can I fill the values of this structure then?... `# Data Structure. #$spam_hash = { # 'email@address.com' => # { # 'Subject1' => { # cache_name1 => { # path => 'path', # from => 'from' # }, # cache_name2 => { # path => 'path', # from => 'from' # } # }, # 'Subject2' => { # cache_name1 => { # path => 'path', # from => 'from' # }, # cache_name2 => { # path => 'path', # from => 'from' # } # } #};` [download] Until I came across a duplicated $subject, this script was working perfectly. (I guess I was getting lucky.) I need to dynamically fill the subject and cache_names to becomes slices of the email address. I'm confused on how to change my structure to one that will be functional.	[reply] [d/l] [select]
Re: Why is part of slice duplicated? by Abigail-II (Bishop) on May 28, 2002 at 18:01 UTC
Well, I could give an answer and that will work, until you run into the situation you describe: Until I came across a duplicated $subject, this script was working perfectly. What do you want to do if you hit a duplicate subject? Abigail	[reply]
Re: Re: Why is part of slice duplicated? by the_0ne (Pilgrim) on May 28, 2002 at 18:13 UTC
Re: Why is part of slice duplicated? by Abigail-II (Bishop) on May 28, 2002 at 18:37 UTC
Some notes below your chosen depth have not been shown here
Re: Why is part of slice duplicated? by danger (Priest) on May 28, 2002 at 17:16 UTC
You are accidentally using symbolic references and do not have the structure you think you have --- using strict would have told you something was wrong right away. You can see a similar problem and discussion here.	[reply]


go ahead... be a heretic
	PerlMonks