Re: [COMMIT] Support #'function-arglist with built-in special forms.
Ar an séú lá déag de mí Meitheamh, scríobh Stephen J. Turnbull:
Stephen J. Turnbull writes:
> Aidan Kehoe writes:
>
> > Unfortunately, actually, for my purposes, it’s not. If you look
> > in any DOC file and search for
> > dontusethis-set-symbol-value-handler, the argument list looks
> > like so:
> > arguments: (VARIABLE HANDLER-TYPE HANDLER &optional HARG \
> > KEEP-EXISTING))
>
> Ah, OK. I think this is probably my bug, then. make-docfile didn't
> work properly at all when Jerry first added the UNUSED stuff, so I
> hacked its parser to work with it (more or less). I'll take a look at
> make-docfile.c in a day or so and see if I can remember what was
> supposed to be happening there.
I've made some progress on this, I do remember what is going on.
However, I don't have any test cases except that one (ie, broken
docstrings). Aidan, can you tell me any other functions where you've
seen breakage?
I’ve only seen the broken docstrings. That should be enough for a test case,
though—e.g.:
(Assert (not (string-match "))$" (function-arglist #'event-to-character))))
The last argument of #'event-to-character will remain unused; you might want
to check that (function-max-args #'event-to-character) still gives four,
though. Or M-C-s [A-Z]))$ RET in DOC for other examples.
I wonder how much extra build time it would cost to put this in
Lisp.... Anyway, I think I'm going to refactor the C to make it
clearer what's going on.
--
¿Dónde estará ahora mi sobrino Yoghurtu Nghe, que tuvo que huir
precipitadamente de la aldea por culpa de la escasez de rinocerontes?
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches