Front page | perl.perl5.porters |
Postings from September 2015
There's no harm in being restrictive in the *current* semantics of ~~, leaving any case of literal string or number on RHS as compiletime failure. That still leaves the doorway open to adding semantics later on, if we manage to grow a new mechanism that makes it more sensible than we currently have. I'm verymuch in favour of the restricted set of "these things are valid, all else is an error". It's far easier to add new cases later on that used to be errors, than to redefine existing ones. So lets be very conservative in what we add today. -- Paul "LeoNerd" Evans leonerd@leonerd.org.uk http://www.leonerd.org.uk/ | https://metacpan.org/author/PEVANSThread Previous | Thread Next