> > I'm not yet convinced that we should generally move away from using
> > APR instead of moving APR closer to what we want.
>
> That would be a huge task and I see no sign of effort being applied to
> that problem.  I'd like to be wrong though?

Hmm, it might be easier to advance Harmony portlib to separate
(sub)project as alternative to APR.

>
> The portlib API is a good deal more flexible.  (Nevertheless, it might
> be an interesting exercise to implement portlib on APR.)

There is one more dependency on APR existing in DRLVM, it is log4cxx.
Anyway, I doubt we can afford moving from APR in near future, benefits
are not so evident to force this task.

--
Alexey

Reply via email to