[Parseperl-discuss] Unique ID per PPI::Element

[Dan B. <mr....@gm...>]

I'm currently going through the motions of going from a very
development version of PPIx::Index to something useful in practice and
I've hit a stage where I feel having a unique ID for each PPI::Element
would be useful. My rationale behind this is that if someone has a
PPI::Document they are iterating over and want to access the indexed
information pertaining to a given node then they can just access that
data using its unique id. This may very well be a X/Y Problem and I'm
certainly open to suggestions but that's what my monkey brain has
managed to scratch in the dirt.

I haven't gotten around to implementing it yet but if I did it
probably use some PPIx::XPath magic (cos I'm lazy like that) and look
something like this:

use Digest::MD5 'md5_hex';
sub PPI::Element::id {
  my $el = shift;
  return md5hex( PPIx::XPath->new($el)->xpath . $el->location->[0] );
}

There are probably issues with that code, such as the fact that if you
have recurring Elements on the same line you'd get the same ID (but
that's fine for what I'm doing) and it probably needs the script path
to be unique too, but I guess that's what you get for knocking out
code off the cuff.

So any thoughts, suggestions, criticisms on this ID idea or anything
pertaining to indexing software would be very much appreciated.

Thanks,
Dan Brook aka broquaint