At 5:54 PM -0800 3/2/03, Dave Whipp wrote:
Dan Sugalski wrote:
Okay, here's another shot at the semantics for objects. If folks, especially non-perl folks, would look this over and chime in, I'd much appreciate it.

The thing that I noticed was the lack of semantics for creation and

Hence the next missive, "Class Spec" :)


This time I was trying to define just the behaviour of objects that already exist, rather than specifying their creation or how the classes that own the objects behave.

I was going to punt on a lot of the rest for now, but I expect I really shouldn't, so I'lll work up try 2.5 tonight. :)

destruction. Will there be well defined creation semantics, with separate steps for allocation of memory, binding to a class, and initialization of attributes? In what order are these done (inc multiple inheritance issues)? What is the state of the object if an exception is thrown before construction is complete; and how is it cleaned up? Even with GC, we must ask similar questions with respect to object destruction. Will there be finalizers; in what order are they called, and can they throw exceptions?


--
                                        Dan

--------------------------------------"it's like this"-------------------
Dan Sugalski                          even samurai
[EMAIL PROTECTED]                         have teddy bears and even
                                      teddy bears get drunk

Reply via email to