GNU bug report logs - #67182
29.1; read-kbd-macro always return a vector

Previous Next

Package: emacs;

Reported by: "Jakub T. Jankiewicz" <jcubic <at> jcubic.pl>

Date: Wed, 15 Nov 2023 01:17:01 UTC

Severity: normal

Found in version 29.1

Done: Eli Zaretskii <eliz <at> gnu.org>

Bug is archived. No further changes may be made.

Full log


Message #11 received at 67182 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Monnier <monnier <at> iro.umontreal.ca>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: "Jakub T. Jankiewicz" <jcubic <at> jcubic.pl>, 67182 <at> debbugs.gnu.org
Subject: Re: bug#67182: 29.1; read-kbd-macro always return a vector
Date: Wed, 15 Nov 2023 07:45:16 -0500
>> This is the code I was always using in Emacs to send raw keys
>> to modes like ansi-term:
>> 
>> (defun raw (str)
>>   (interactive "sSend Raw Key: ")
>>   (term-send-raw-string (read-kbd-macro str)))
>> 
>> But this doesn't work anymore it throw an error [3] when calling:
>> 
>> (raw "C-c")

The above code already threw the same error when you call

    (raw "C-<mouse-1>")
or
    (raw "M-a")
or
    (raw "é")

:-(

Maybe it should be something like

    (defun raw (str)
      (interactive "sSend Raw Key: ")
      (let ((keys (kbd str))
            (string (condition-case err
                        (concat "" keys)
                      (wrong-type-argument
                       (let ((elem (nth 2 err)))
                         (error "%s is not a character"
                                (if (numberp elem)
                                    (prin1-char elem)
                                  elem)))))))
        (term-send-raw-string string)))

?

>> According to documentation `C-h f read-kbd-macro`
>> "The result will be a string if possible" but it seems that in new
>> GNU Emacs it's always a vector.
> Stefan, this is due to your change in edmacro-parse-keys as part of
> commit 87789330, whose log message says nothing about
> edmacro-parse-keys.  Did you really intend to make that
> backward-incompatible change?  If so, why?

Yes.  I want to eliminate the use of strings that stand for a sequence
of events because it does nothing more than leave latent bugs and create
confusion (between the strings used as input to `read-kbd-macro` and the
strings that used to be output by `read-kbd-macro`), while increasing
the complexity of the rest of the code which has to handle both vectors
and strings.


        Stefan





This bug report was last modified 1 year and 189 days ago.

Previous Next


GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.