Gary Stainburn wrote: > > > Gary Stainburn wrote: > > > I'm working for the first time with object, my $self being a ref to an > > > anonymous hash, i.e. > [snip] > > > Now when I wish to DESTROY the hash, I need to free the memory used by > > > the hash, including the two arrays _Links and _Signals. > > > > > > Am I right in thinking that the arrays, along with the scalars will be > > > deleted by the garbage collector when the references in the hash are > > > deleted? > > > > > > Am I also right in thinking that this will happen if I lose the reference > > > in $self by simply running: > > > > > > $self=undef; > > > > Yes. But why do you think you need to write a DESTROY method? It's only > > necessary if the object's destruction must include more than just > > deallocating object data (for instance if a running instance count is > > being kept and needs decrementing). If the method is defined, Perl will > > call DESTROY when it is about to automatically release all of the object > > data anyway. > > The full DESTROY will also delete link-list references to/from other blocks of > track, as well as cascade deletes of fittings (signals, etc) associated with > that track block section.
OK, but Perl will also cascade the deletes for you unless you have cyclic references in your data. If you have a doubly linked list then this may be a problem, as you then have to arrange that object data is destroyed only if it is referenced from within an 'island' of objects which cannot be accessed from elsewhere. Rob -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] <http://learn.perl.org/> <http://learn.perl.org/first-response>
