I know it's not pretty, but it works.
True :-) However, I think my basic point still stands. Perl's current implementation of lvalue subs breaks implementation encapsulation - and can therefore make life considerably more complex if you have to change that implementation.
The classic case of this is when you have an attribute that you later decide can be calculated/set by a runtime method (trading space for speed).
Perl's lvalue subroutine implementation does not allow you to do this without resorting to tie (or possibly a source filter).
In reply to Re^2: $foo->bar = 14;
by adrianh
in thread $foo->bar = 14;
by Juerd
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |