bug#6448: `event-apply-modifier' EVENT arg accepts float then bails

bug#6448: `event-apply-modifier' EVENT arg accepts float then bails

[MON KEY]

Can `read-event' return a float?

If not, why does `event-apply-modifier' invoke the `numberp' predicate instead
of `integerp'?

e.g. these signals an error:

 (event-apply-modifier 32.3 'shift 25 "<S-")
 (event-apply-modifier (float (string-to-char "\C-s")) 'shift 25 "<S-")

This is kosher:

 (event-apply-modifier 19 'shift 25 "<S-")

--
/s_P\

bug#6448: `event-apply-modifier' EVENT arg accepts float then bails

[Chong Yidong]

MON KEY <monkey@sandpframing.com> writes:

> Can `read-event' return a float?
>
> If not, why does `event-apply-modifier' invoke the `numberp' predicate instead
> of `integerp'?
>
> e.g. these signals an error:
>
>  (event-apply-modifier 32.3 'shift 25 "<S-")
>  (event-apply-modifier (float (string-to-char "\C-s")) 'shift 25 "<S-")
>
> This is kosher:
>
>  (event-apply-modifier 19 'shift 25 "<S-")

I don't think this is a problem in real life.

bug#6448: `event-apply-modifier' EVENT arg accepts float then bails

[MON KEY]

On Sat, Jun 19, 2010 at 1:38 PM, Chong Yidong <cyd@stupidchicken.com> wrote:
>
> I don't think this is a problem in real life.

Indeed, but in fantasy-land it might be slightly more correct :)
Is there any good reason not to change numberp -> integerp?

--
/s_P\

bug#6448: `event-apply-modifier' EVENT arg accepts float then bails

[Andreas Schwab]

MON KEY <monkey@sandpframing.com> writes:

> Indeed, but in fantasy-land it might be slightly more correct :)
> Is there any good reason not to change numberp -> integerp?

Look at (numberp) as a more efficient alternative to (and (not (lisp))
(not (symbolp))).  There is no need to be conclusive.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

bug#6448: `event-apply-modifier' EVENT arg accepts float then bails

[MON KEY]

On Sun, Jun 20, 2010 at 4:01 AM, Andreas Schwab <schwab@linux-m68k.org> wrote:
>> Is there any good reason not to change numberp -> integerp?
>
> Look at (numberp) as a more efficient alternative to (and (not (lisp))
> (not (symbolp))).

I assume that in the above for `(not (lisp)' you intend `(not (listp)'?

> There is no need to be conclusive.

Thank you for taking the time to answer my query.
This addresses exactly the concern I had in mind but was afraid to ask :).

So, use of `integerp' is neither no more efficient than `numberp' nor any more
useful for debugging/signaling purposes where there is a known constraint on
the arg type?  e.g. for example w/ `event-apply-modifier' where it is known that
EVENT can only be an integer?

> Andreas.

--
/s_P\