Re: ucs input method bug.

[Kenichi Handa]

In article <address@hidden>, address@hidden (Michaël Cadilhac) writes:

> Test case:

> emacs -Q
> M-x set-input-method RET ucs RET
> Type « ut » it will produce « tut ».

> IMO, ucs strives to be Emacs by doing the command of the faulty
> (unexpected) key (here « t ») while it is not needed.

> I propose just to remove the part that does that.

I agree.  I've just installed your change, thank you.

---
Kenichi Handa
address@hidden

> [1.1.2 uni-input.patch <text/x-patch; iso-8859-1 (quoted-printable)>]
> Index: leim/quail/uni-input.el
> ===================================================================
> RCS file: /sources/emacs/emacs/leim/quail/uni-input.el,v
> retrieving revision 1.11
> diff -c -r1.11 uni-input.el
> *** leim/quail/uni-input.el   20 Apr 2006 08:38:14 -0000      1.11
> --- leim/quail/uni-input.el   7 Sep 2006 11:25:07 -0000
> ***************
> *** 100,110 ****
>                     (progn
>                       (push key events)
>                       (ucs-input-insert-char key))
> -                 (let ((last-command-char key)
> -                       (current-prefix-arg))
> -                   (condition-case err
> -                       (call-interactively (key-binding seq))
> -                     (quail-error (message "%s" (cdr err)) (beep))))
>                   (quail-delete-region)
>                   (throw 'non-digit (append (reverse events)
>                                             (listify-key-sequence seq))))))
> --- 100,105 ----
> Index: leim/ChangeLog
> ===================================================================
> RCS file: /sources/emacs/emacs/leim/ChangeLog,v
> retrieving revision 1.193
> diff -c -0 -r1.193 ChangeLog
> *** leim/ChangeLog    12 Jul 2006 00:33:19 -0000      1.193
> --- leim/ChangeLog    7 Sep 2006 11:25:15 -0000
> ***************
> *** 0 ****
> --- 1,6 ----
> + 2006-09-06  Michaël Cadilhac  <address@hidden>
> + 
> +     * quail/uni-input.el (ucs-input-method): Don't make the action of
> +     a key not in [0-9a-zA-Z] when it was expected to be.  Let the Emacs
> +     mechanism do it.
> + 
> [1.1.3  <text/plain; iso-8859-1 (quoted-printable)>]

> (Note: ucs  input method triggered  the bug I  proposed to fix  in the
> thread « The order input events are  processed. » : just type « uabt »
> in a flyspell-ized buffer, and it will introduce the bug.)

> -- 
>  |      Michaël `Micha' Cadilhac   |  Un certain Blaise Pascal              |
>  |         Epita/LRDE Promo 2007   |    etc... etc...                       |
>  | http://www.lrde.org/~cadilh_m   |  -- Prévert (Les paris stupides)       |
>  `--  -   JID: address@hidden --'                                   -  --'
> [1.2  <application/pgp-signature (7bit)>]

> [2  <text/plain; us-ascii (7bit)>]
> _______________________________________________
> Emacs-devel mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/emacs-devel