bug#69893: 29.2; Valid key "<TAB>" not accepted by `keymap-global-set'

bug#69893: 29.2; Valid key "<TAB>" not accepted by `keymap-global-set'

[tpeplt]

Recipe for reproducing this problem:

1. Start Emacs at a shell prompt using "emacs -Q".

2. Evaluate the following expressions in the *scratch* buffer.

The following keys evaluate as valid (as expected):

(key-valid-p "TAB")
(key-valid-p "<TAB>")
(key-valid-p "<tab>")

The following key evaluates as invalid (as expected):

(key-valid-p "tab")

3. So, the following expressions should evaluate to
‘indent-for-tab-command’:

(keymap-global-set "<tab>" 'indent-for-tab-command)
(keymap-global-set "TAB" 'indent-for-tab-command)
(keymap-global-set "<TAB>" 'indent-for-tab-command)

The first two expressions evaluate as expected, but when the third
expression is evaluated, then the following message is reported:

keymap-global-set: To bind the key TAB, use [?\t], not [TAB]

Because (key-valid-p "<TAB>") evaluates to t, that error message appears
to be invalid.

4. Also, the following expressions all evaluate to nil:

(key-valid-p "[?\t]")
(key-valid-p "[?\\t]")

(key-valid-p "?\t")
(key-valid-p "?\\t")

(key-valid-p "<?\t>")
(key-valid-p "<?\\t>")

So the error message appears to direct a user to change the key string
to an invalid key.

--

bug#69893: 29.2; Valid key "<TAB>" not accepted by `keymap-global-set'

[Eli Zaretskii]

> From: tpeplt <tpeplt@gmail.com>
> Date: Mon, 18 Mar 2024 17:14:16 -0400
> 
> Recipe for reproducing this problem:
> 
> 1. Start Emacs at a shell prompt using "emacs -Q".
> 
> 2. Evaluate the following expressions in the *scratch* buffer.
> 
> The following keys evaluate as valid (as expected):
> 
> (key-valid-p "TAB")
> (key-valid-p "<TAB>")
> (key-valid-p "<tab>")
> 
> The following key evaluates as invalid (as expected):
> 
> (key-valid-p "tab")
> 
> 3. So, the following expressions should evaluate to
> ‘indent-for-tab-command’:
> 
> (keymap-global-set "<tab>" 'indent-for-tab-command)
> (keymap-global-set "TAB" 'indent-for-tab-command)
> (keymap-global-set "<TAB>" 'indent-for-tab-command)
> 
> The first two expressions evaluate as expected, but when the third
> expression is evaluated, then the following message is reported:
> 
> keymap-global-set: To bind the key TAB, use [?\t], not [TAB]
> 
> Because (key-valid-p "<TAB>") evaluates to t, that error message appears
> to be invalid.
> 
> 4. Also, the following expressions all evaluate to nil:
> 
> (key-valid-p "[?\t]")
> (key-valid-p "[?\\t]")
> 
> (key-valid-p "?\t")
> (key-valid-p "?\\t")
> 
> (key-valid-p "<?\t>")
> (key-valid-p "<?\\t>")
> 
> So the error message appears to direct a user to change the key string
> to an invalid key.

Stefan, is the below the right fix for this?

If not, why does key-parse omit TAB from the list of specially-handled
keys?

diff --git a/lisp/keymap.el b/lisp/keymap.el
index d2544e3..b2b475c 100644
--- a/lisp/keymap.el
+++ b/lisp/keymap.el
@@ -260,7 +260,7 @@ key-parse
                         (setq word (concat (match-string 1 word)
                                            (match-string 3 word)))
                         (not (string-match
-                              "\\<\\(NUL\\|RET\\|LFD\\|ESC\\|SPC\\|DEL\\)$"
+                              "\\<\\(NUL\\|RET\\|LFD\\|TAB\\|ESC\\|SPC\\|DEL\\)$"
                               word))))
                  (setq key (list (intern word))))
                 ((or (equal word "REM") (string-match "^;;" word))

bug#69893: 29.2; Valid key "<TAB>" not accepted by `keymap-global-set'

[Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors]

> Stefan, is the below the right fix for this?

Looks correct to me.
That code dates back to

    commit 629d4dcd2a184da6a0b246d31f152a84327db51a
    Author: Richard M. Stallman <rms@gnu.org>
    Date:   Tue Sep 21 03:44:04 1993 +0000
    
        Total rewrite by Gillespie.

and I suspect it was just an oversight.


        Stefan

bug#69893: 29.2; Valid key "<TAB>" not accepted by `keymap-global-set'

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Cc: tpeplt <tpeplt@gmail.com>,  69893@debbugs.gnu.org
> Date: Thu, 21 Mar 2024 15:46:08 -0400
> 
> > Stefan, is the below the right fix for this?
> 
> Looks correct to me.
> That code dates back to
> 
>     commit 629d4dcd2a184da6a0b246d31f152a84327db51a
>     Author: Richard M. Stallman <rms@gnu.org>
>     Date:   Tue Sep 21 03:44:04 1993 +0000
>     
>         Total rewrite by Gillespie.
> 
> and I suspect it was just an oversight.

Thanks, installed on the emacs-29 branch, and closing the bug.