Re: gir stability

[Ryan Lortie <desrt desrt ca>]

On Tue, 2012-01-31 at 15:37 -0500, Matthias Clasen wrote:
> But its a fair question to ask -
> at what point do we stop considering the introspected bindings a
> work-in-progress and treat annotations as equally part of the stable
> api ? Sounds like your take is that the answer has to be 'now'...

If we are to do this now then it means that we need to start talking
about what that implies.  Some ideas:

We'd have to start reviewing new API for proper bindings annotations as
we land it so that we get it right the first time...

It also means that we have to be less creative about some changes that
are C-compatible but not bindings-compatible in the future (like
changing a non-nullable parameter to be nullable or changing a gboolean
arg to a flags...).

If we wanted to really go crazy on this, we could do some sort of
introspection-API compatibility checking... store the last-known public
introspection API of the library and find out about changes as we make
them, hopefully ensuring that everything is strictly additive.

There may also be scope for helping us cope better on the introspection
side with things like renames: we could keep the old name in a
deprecated state, for example.

Cheers