Re: What's the point of byte-compile-define-keymap?

[Lars Ingebrigtsen]

Alan Mackenzie <acm@muc.de> writes:

> It looks to me that byte-compile-define-keymap (in
> lisp/emacs-lisp/bytecomp.el) saves its FORM argument in ORIG-FORM and
> always returns it unchanged.  This looks like an oversight.

It was rewritten from a byte-optimize function (but it doesn't optimise
anything any more), so it should probably be cleaned up a bit more.

> Also, would this function not better be named
> byte-compile-FILE-FORM-define-keymap, since it is the value of a
> byte-hunk-handler symbol property.
>
> Also[2], is this function really necessary?  Is define-keymap used
> sufficiently often to justify a special byte-hunk-handler for it?

It just used to issue warnings for invalid keymap syntaxes at compile
time.  And, yes, it will be used everywhere.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no