Johann 'Myrkraverk' Oskarsson writes:
Will you forgive me for not knowing about these details in the 21.5
branch?
Sure, but then the whole conversation is basically moot.
I think it would be a very bad idea to make a change that fixes no bug
in the last release *ever* of 21.4. If Vin has a different opinion,
fine; I've said my opinion, and I'm sure he will give it appropriate
weight.
What I meant, and stand by, is that external emodules that implement
types are so rare they're practically nonexistent;
Only in 21.4.
particularly those that do properties.
In this context, what's part of xemacs core doesn't matter,
because
any changes to the ABI* will fix those too.
Not for third parties who have their own types with properties.
XEmacs has long been the preferred platform for private variations
because we worry about issues like that. This wart isn't worth
changing that.
For better or worse, this little wart is part of the public API (and I
do mean API) for primitive types, and has been for many years. That
means that changing it is a potentially big deal for third parties.
If we're going to do it, we should do it in a way that breaks the
build, not just the runtime. (It still potentially means breaking
only runtime for external modules if they're supplied as binaries, but
that can't be helped.)
Steve
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta