ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jose Alberto Fernandez" <j_a_fernan...@yahoo.com>
Subject Re: We need to stop the lies
Date Sat, 23 Feb 2002 10:33:16 GMT
From: "Peter Donald" <peter@apache.org>
> > For what I remembered of his description of
> > classloader hierarchy, I think it is too limiting.
> > If libA needs things from libB, his approach will
> > declare that "libA extends libB" which is fine unless
> > libA also needs things from libC. Now you cannot do it
> > because we have single inheritance (i.e., a
> > classloader onl has one parent classloader).
> 
> If the set of classes from each source is disjoint then there is no real need 
> for a strictly hierarchial approach. You can easily accomodate the 
> multiparent scenario. 
> 
> However what you describe does not need this and may be modeled as simple 
> aggregation of classloaders like most other container based APIs specify.
> 
So what is your point? Does mutant or myrmidom proposals do this right now?
>From the information given here, I do not think they do, but after all there are so many
checkins a day that I certaintly have no idea what is going on for sure.

What antlib's does is exactly classloader agregation. What I remember from the
ANT2 proposals (do not remember which) required one to give a global name
to the library in a descriptor, and then it required one to declare that the library
"extends" another. And that was implemented by parenting.

Unless you are planning to write some new proxy classloaders that allow for multiple
parents, I really do not understand yout comments above.

Jose Alberto



--
To unsubscribe, e-mail:   <mailto:ant-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:ant-dev-help@jakarta.apache.org>


Mime
View raw message