Re^10: Storing/parsing perl data structure in/from a file

[Wilderness]

Sorry to make it confusing but my intention is to be able to `do` a file with the information below and flag that key alpha->beta and later alpha->beta->gamma is defined twice. Duplicate keys can be at any level of the multi-level hash.

{
  alpha => {
    beta => {
      gamma => theta,
      delta => lambda,
    },
    beta => {
      gamma => zeta,
    },
  },
},

[LanX]

No this is your approach, you should explain your intention to use such a weird approach.

I'm sure we can help you, if we just knew why ... this smell like an XY Problem! =)

Your dump isn't even valid Perl code (barewords as values)!

[Wilderness]

Sorry to have completed pissed you off. It's ironical that I know it is an XY problem, yet I do not know what X and Y are...Anyways, after looking at all the responses, I wanted to know more about tying hashes and being able to override methods like STORE, FETCH, etc. to make them work for my multi-dimensional hash. But I haven't used OO-Perl before so I am struggling to write my own recursive methods to make it work for multi-level structures. I will go with YAML. It looks a lot more elegant than perl hashes in a file.

[LanX]

You didn't piss me off, it's just not easy to speculate about fictional use cases!

The tied hash doesn't work with nested literal hashes, they are cleaned at compile phase.

Though you can use nested literal arrays like demonstrated w/o information loss.

Assign them to a tied array, which STOREs them internally in a tied hash.

Or just a normal array which is parsed afterwards!

And yes YAML looks better! =)