You can do it one of two ways.

1. Check that the thing is valid before hand (as ref $coderef in my class below does)
2. Check when you call that the result is valid (as the trigger_event sub will do if you remove the test in the first point)

At the risk of getting a TL;DR, here's something that takes coderefs and then dispatches events with them, as I wrote this, I realised that it's surprisingly long for a simple example ;)

This is the class that takes care of dispatching the events (to a single handler for each event type)

This is the client code, that registeres event handlers (to be used as callbacks) and triggers the actual events on the object. Normally, the object would find it's own events, and handle them within a run() type method (Tl's MainLoop for example)

The class will reject 'antelope' because it's not a real coderef, although if you were to pass in 'main::antelope' as the name, and remove the check for ref $coderef, you should be able to use just names of subs. I don't see why you would want to do this, since it's the same number of characters to write \& as it is to enclose the sub's name quote likes

Justification: this does answer your question about checking if it's callable before hand (because it uses ref $subref to decide if it can be callled (long before the events actually turn up). This is a fairly painless way to check if something is callable, you can also use anonymous subs if you want

Hope this helps.

$ perl -le 'print ref (\&no_such_sub);' CODE Useless test.

Oh, my!

@_=qw; ask f00li5h to appear and remain for a moment of pretend better than a lifetime;;s;;@_[map hex,split'',B204316D8C2A4516DE];;y/05/os/&print;