From: Neil Hodgson <nhodgson@...>
Date: 2001-10-24T08:47:33+09:00
Subject: [ruby-talk:23124] Re: Bruce Eckel's opinion of Ruby

Avdi B. Grimm:

> I have to disagree with this statement.  There *are* cases where the
> block paradigm is simply not appropriate, and destructors are.
> Sometimes it's possible to kludge in a solution using the block
> paradigm; but it's not pretty.  It's not a terribly rare occurance to
> have objects which (a) /must/ release resources before being
> garbage-collected, and (b) have non-determinate lifetimes.

   Deterministic finalization (DF) is a great feature but the Jython
implementation of Python on the JVM does not provide this as it is hard to
implement over the Java based semantics of the JVM. Even if Jython were to
implement this for objects it is totally responsible for it needs to
interoperate with Java based code by, for example, returning pointers (or
wrappers) to these objects to Java code.

   It is unlikely that Python for .NET will have deterministic finalization
for the same reason. The lack of DF in .NET has caused a lot of angst among
us old COM developers since COM is based on reference counting with DF. The
safe portable route is currently garbage collection as that is used in the
most widely used / soon-to-be-used VMs and can also be implemented for
self-contained implementations like Ruby and standard Python.

   The difficulty in providing DF on the JVM and .NET is an example of how
these environments require changes to the languages hosted on them. I'm sure
there will be Ruby features (possibly closures) that can not be implemented
reasonably on these VMs.

   Neil