On Fri, Aug 9, 2013 at 12:34 PM, Antoine Pitrou <report@bugs.python.org>wrote:

>
> Antoine Pitrou added the comment:
>
> > Can we agree on discarding the current implementation for now and then
> > rewriting it based on a tree builder instead of a parser wrapper?
>
> Only if it actually brings something (feature-wise, performance-wise,
> maintenance-wise, whatever) that the current implementation doesn't
> have.
>
> Otherwise, I'd rather check in the simple inheritance-to-composition
> change.
>
> I'm sorry that you noticed this after the alpha. I would have had a
> lower threshold for changes if you had manifested when I proposed the
> feature.
>

Antoine, I'm not yet very well versed in the release process, but ISTM that
if:

1. We know that a better API can be created
2. The implementation is not overly difficult

It makes sense to do the change even though Alpha 1 was already released.
Alphas are alphas, after all :-) I suppose that we can consult with Larry
and others if this is the only doubt holding the change back.