"Carsten Haerle" <
[hidden email]> wrote in message
news:bup6g1$qmu$01$
[hidden email]...
> Yes, of cause we didn't try to copy a process directly, but copied an
object
> which had a reference to a process. As there is already a method deepCopy,
I
> would prefer using this method instead of first serializing and
> deserializing. It should also be much faster.
#deepCopy is only really of use in strictly controlled circumstances where
you know the extent of what will be copied (and can you ever know that for
sure?). Frankly you are much better off providing a specific #copy
implementation on your classes that copies just those objects that need to
be copied. The #copy implementation added to Process itself by the
continuations package is a case in point - it creates a copy of the current
state of a process such that it can be resumed at a later date,
independently of the original.
Regards
Blair