bug#9666: 24.0.50; C-d has unexpected behaviour in CUA rectangle mode

["Stefan Reichör" <stefan@xsteve.at>]

Stephen Berman <stephen.berman@gmx.net> writes:

> On Tue, 04 Oct 2011 09:55:00 +0200 Stefan Reichör <stefan@xsteve.at> wrote:
>
>> Stephen Berman <stephen.berman@gmx.net> writes:
>>
>>> On Tue, 04 Oct 2011 08:30:29 +0200 Stefan Reichör <stefan@xsteve.at> wrote:
>>>
>>>> Just activate cua mode, select an rectangle and hit DEL or C-d:
>>>> M-x cua-mode
>>>> C-RET
>>>> select rectangle
>>>> DEL
>>>>
>>>> I expect this command to delete the whole rectangle
>>>>
>>>> But it deletes only one character and deactivates the marked
>>>> rectangle
>>> [...]
>>>> In GNU Emacs 24.0.50.1 (i386-mingw-nt5.1.2600)
>>>>  of 2011-09-12 on 3249CTO
>>>> Windowing system distributor `Microsoft Corp.', version 5.1.2600
>>>> configured using `configure --with-gcc (4.5) --no-opt'
>>>
>>> I see somewhat different behavior with the above recipe: DEL (backspace
>>> key), deletes the one-column rectangle immediately following the
>>> selected rectangle, and does not deactivate (unhighlight) the latter;
>>> C-d, on the other hand, does delete the selected rectangle.  This is on
>>> GNU Emacs 24.0.90.3 (i686-suse-linux-gnu, GTK+ Version 2.22.1) of
>>> 2011-10-04 on escher.
>>
>> Digging further into this problem.
>> There are some modes that bind delete (not BS) to delete-forward-char (e.g. the
>> *scratch* buffer).
>
> Ah, I guess you also meant <delete> above, not DEL.  With <delete> I
> also see what you reported.  This is indeed different from the behavior
> of <delete> with an active (highlighted) region.
>
>> When this is the case the rectangle is not deleted.
>
>> When delete is bound to delete-char then the rectangle is deleted as expected.
>
> But the standard binding of delete-char is C-d, which you (meant to)
> report above behaves like <delete> (assuming you meant that key instead
> of DEL); see also your Subject header.  Whereas, as I said, I do get
> deletion with C-d.  So the unexpected behavior in cua-mode is (just) with
> <delete> (bound to delete-forward-char), is that what you're saying?  If
> so, I agree (assuming, that is, that <delete> should behave the same
> with rectangles in cua-mode and with active regions outside of cua-mode;
> I don't know, since I haven't made much use of cua-mode).

I assumed that C-d and <delete> are bound to the same function.
I see that this is no longer the case - and this also the reason
for the different behaviour in cua rectangle mode.

I confirm that C-d does the expected thing. <delete> does not work as expected.

cua rectangle mode should treat delete-forward-char the same way as delete-char.