RE: PATCH: Fully Resumable VM (yield acrosspcall/callback/meta/iter)

["Andre Carregal" <carregal@...>]

> > It's all there: demand, discussion, docs, patches.
> > Waiting for 5.2 means waiting another 1-2 years. Sad.

I agree that this hiatus is sad news, it would be great to have the merge.
However I'd like to add one more piece to the puzzle:

The launching of Lua 5.1 involves a lot more than the source itself.

Most developers wait for Lua final releases to decide for a code base
migration and any documentation (including PIL) may have to be updated
before this release. In PIL's case that is no small feat and involves a lot
more than simply updating the book. Imagine negotiating a book reprint twice
in a year...

I don't know if it is or not possible to merge RVM in 5.1 since I'm not
qualified to estimate the effort, but assuming that 5.2 would follow shortly
after is somewhat naive.

Most of us here understand the value of Mike's work and the importance of
the features for Lua, but the question may really boil down to which path is
better:

1) Postpone 5.1 until the creators find out the impact of an eventual
merging.
AFAICS that poses no impacts on the final release other than deadlines
themselves.

2) Assume RVM will be an extension for 5.1 and eventually merged on 5.2.
I'm afraid that may pose a possibly considerable impact on 5.1 documentation
and PIL compared to the RVM version. Would PIL be "compatible" with RVM and
vice versa? Would the difference mean a devaluation of any of them?

Again, I'm not qualified for the analysis, I'm just trying to estimate
project impacts here.

Andre Carregal