Can people explain why they are voting no to a feature that seems
really well put and thought?
There is no need to go from new extension with a new unproven API to directly into core as first try. It's much better to try things out in for example PECL and then to promote it to a core extension when all the kinks are worked out.
cheers,
Derick
There is no need to go from new extension with a new unproven API to directly into core as first try. It's much better to try things out in for example PECL and then to promote it to a core extension when all the kinks are worked out.
True, but in this case Dmitry clarified that the Api is mostly a mirror
of the Python or Lua FFI implementations, and both have been already
proven to work and satisfy most needs, so basically the wheel isn't been
reinvented here, it could be said that this is an adaptation of the
wheel which structure is already well known.
By the way, Merry Christmas to all!