Consider an input file with following enteries:
1 3 4 5 6
1 6 4 5 3
1 2 4
4 2 1
...
...
Each row is a set without duplicates.
THe file itself represents a set of rows.

I am currently using set::Scalar to represent each row. 
And Set:: Scalar to represent the set of rows.
The problem is that it contains duplicates values of sets.
i.e. the entries of set look like this
(3,4,5), (3,4,5) ,(1,4,5),....
if (3,4,5) is added twice.

How can one force the set to contain unique elemens.

thanks
sandeep

---

Clarification of Question by skg-ga on 04 Jul 2004 23:38 PDT

The data could occur  at an intermediate step of the program. I posed
the problem  this way for simplicity.

To rephrase the problem:
Given  a data structure, D that stores family of sets  and a query set
nset. Check if nset is already present in D.

Ofcourse i can go through each set in D and compare it with nset. But
i want to do this more efficently.

This is how i would have done in C++. Create a function comparison(Set
a, Set b) that provides <=> over sets and then use stl::map with this
comparison.

If the duplicates are in your data file, you could just pipe the data
file through a program such as uniq to filter out all of the
duplicates before processing.

Of course, the functionality of uniq can be implemented fairly simply
in Perl itself:

perl -ne 'print unless $s eq $_; $s = $_'

The data could occur  at an intermediate step of the program. I posed
the problem  this way for simplicity.

To rephrase the problem:
Given  a data structure, D that stores family of sets  and a query set
nset. Check if nset is already present in D.

Ofcourse i can go through each set in D and compare it with nset. But
i want to do this more efficently.

This is how i would have done in C++. Create a function comparison(Set
a, Set b) that provides <=> over sets and then use stl::map with this
comparison.

One cheap solution is to use Hash of arrays:
$str = join(":", @arr);
$HOA{$str} = 1;

To recover:
@arr = split(/:/, $str);