I'm working on a module for parsing and extracting data out of ELF (Executable and Linkable Format) files which I intend to put on CPAN shortly. An ELF file contains a couple of important tables whose entries describe parts of the file (segments and sections). I have an object which holds on to objects for each of the tables. The tables aren't very interesting, but the table entries are so I want to provide access to the individual entries. Current options look like this:

use warnings;
use strict;
use ELF::Reader;

my $elfPath = $ARGV[1];
my $elfFile = ELF::Reader->new(filePath => $elfPath);

# Using index on the segments object
my $segments = $elfFile->GetSegments();

for my $segIndex (0 .. $elfFile->SegmentCount() - 1) {
    next if !${$segments}[$segIndex]->FileSize();
    print ${$segments}[$segIndex]->Describe(head => 16, tail => 16, wi
+dth => 32)
};

print "\n";

# Using an iterator
my $nextSeg = $elfFile->GetSegmentIter();

while (my $segment = $nextSeg->()) {
    next if !$segment->FileSize();
    print $segment->Describe(head => 16, tail => 16, width => 32)
}
[download]

Prints:

Type: PT_LOAD
Virtual load address:  0x00000000
Memory image size:     0x000129f8
Segment alignment:     0x00000008
00000000: 20040000 00010019 00010069 00010081
000129e8: 000129c8 60200000 000129d0 20000000

Type: PT_LOAD
Virtual load address:  0x00000000
Memory image size:     0x000129f8
Segment alignment:     0x00000008
00000000: 20040000 00010019 00010069 00010081
000129e8: 000129c8 60200000 000129d0 20000000
[download]

The question is: should I provide both access techniques, or just one (which?) or something else? There will be a number of different classes that provide similar acessors.