10:41 p.m.
Hi all,
as I was trying to get rebox2.el to work, I noticed that a lot of
functions use `^' as code letter when calling #'interactive. This is an
Emacs23 feature related to shifted motion (see below):
http://www.gnu.org/s/emacs/manual/html_node/emacs/Shift-Selection.html
http://www.gnu.org/s/emacs/manual/html_node/elisp/Using-Interactive.html
Have there been any discussions as to whether it is worth porting that
behavior? For now, I suppose, filtering the code letter `^' from the
#'interactive argument string would be sufficient just to get things
going. I don't think it is possible to advice #'interactive, though. A
macro doesn't seem to work either. Any other clever idea other than
advising all interactive functions in rebox2 with new interactive
statements?
Thanks
Marcus
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
3:22 p.m.
Marcus Harnisch writes:
Hi all,
as I was trying to get rebox2.el to work, I noticed that a lot of
functions use `^' as code letter when calling #'interactive.
*sigh* I could see that ---- heading for the fan from a mile away.
Patches to 21.5 welcome. I don't know what implementation is
appropriate in XEmacs, though. Some thought about how to make this
forward-compatible to future GNU extensions might be a good idea
though.
You'd have to ask Vin if he thinks this is appropriate for 21.4.
going. I don't think it is possible to advice #'interactive,
though. A
macro doesn't seem to work either.
No. `interactive' is a declaration; the form is never executed.
Rather, the argument to interactive is stored with the function and
interpreted by the Lisp runtime rather than by the main Lisp
interpreter.
Any other clever idea other than advising all interactive functions
in rebox2 with new interactive statements?
It's not clever, but it's probably easier than writing advice or
rewriting the interactive statements: replace all rebox2 defuns with
rebox-defun, a macro which filters the interactive declaration (if
any).
Something like (untested)
(defmacro rebox-defun (func args &rest expr-list)
(flet ((fix-interactive (expr)
(if (and (eq (car expr) 'interactive)
(stringp (cadr expr)))
(list 'interactive
(replace-in-string (cadr expr) "\\^" "")
(cddr expr))
expr)))
`(defun ,func ,args ,＠(mapcar #'fix-interactive expr-list))))
There are probably more concise/efficient/Common-Lisp-ly-correct ways
to write that. The mapcar is used because it's certainly possible for
the interactive declaration to be the 3rd or 4th argument to #'defun,
but I don't know for sure about later (I think it never has any
effect, though). It will only work at top-level in the function for
sure, and when executed interactive simply returns nil, so editing the
form throughout expr-list won't affect what any function defined by
rebox-defun returns.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
4:14 p.m.
Stephen J. Turnbull wrote:
Marcus Harnisch writes:
> Hi all,
>
> as I was trying to get rebox2.el to work, I noticed that a lot of
> functions use `^' as code letter when calling #'interactive.
*sigh* I could see that ---- heading for the fan from a mile away.
In the same vein, I discovered recently that GNU Emacs has two bits of
regexp syntax that we currently don't: \_> and \_<, meaning empty string
at the the end/beginning of /symbols/.
--
Resistance is futile. You will be jazzimilated.
Scientific site: http://www.lrde.epita.fr/~didier
Music (Jazz) site: http://www.didierverna.com
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
6:14 p.m.
Didier Verna writes:
In the same vein, I discovered recently that GNU Emacs has two bits
of
regexp syntax that we currently don't: \_> and \_<, meaning empty string
at the the end/beginning of /symbols/.
Ugly, although it's probably not that hard to implement. But aren't
POSIX classes more important?
And isn't it time we admitted that Jamie was right, and just not get
invovled with problem #2 (unless it's actually a solution)?
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
11:48 a.m.
Ar an cúigiú lá de mí na Nollaig, scríobh Stephen J. Turnbull:
Something like (untested)
(defmacro rebox-defun (func args &rest expr-list)
(flet ((fix-interactive (expr)
(if (and (eq (car expr) 'interactive)
(stringp (cadr expr)))
(list 'interactive
(replace-in-string (cadr expr) "\\^" "")
(cddr expr))
expr)))
`(defun ,func ,args ,＠(mapcar #'fix-interactive expr-list))))
There are probably more concise/efficient/Common-Lisp-ly-correct ways
to write that.
Sort of relatedly; everyone, don’t use #'flet, use #'labels. It’s far
cheaper at run time in byte-compiled code--no need to unwind-protect, no
need to store and retrieve the old function binding, and it’s equivalent to
#'flet, except that the function binding is lexical, not dynamic, which is
usually what you want anyway. Mutually- and simply-recursive functions are
supported in XEmacs and endorsed by Common Lisp, which is not the case for
#'flet.
--
‘Iodine deficiency was endemic in parts of the UK until, through what has been
described as “an unplanned and accidental public health triumph”, iodine was
added to cattle feed to improve milk production in the 1930s.’
(EN Pearce, Lancet, June 2011)
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
4735
days inactive
4738
days old
4 comments
4 participants
participants (4)
-
Aidan Kehoe -
Didier Verna -
Marcus Harnisch -
Stephen J. Turnbull