Why there is no char type?

Why there is no char type?

[Leo]

Hello all,

As the subject line suggested.

Leo

-- 
Oracle is the new evil

Why there is no char type?

[Stephen J. Turnbull]

Leo writes:

 > As the subject line suggested.

You're using the wrong 'macs, that's why.

Anyway, there will be one in Emacs 24, if I read the tea leaves
correctly.

Re: Why there is no char type?

[Eli Zaretskii]

> From: Leo <sdl.web@gmail.com>
> Date: Tue, 14 Dec 2010 13:04:39 +0000
> 
> As the subject line suggested.

Ouch!  Do us all a favor: please read past discussions on this.  If
you still think there's something to discuss after that, ask more
specific questions, to keep the bike-shedding (and the flames) at bay.

Okay?

Re: Why there is no char type?

[Eli Zaretskii]

> From: "Stephen J. Turnbull" <stephen@xemacs.org>
> Date: Tue, 14 Dec 2010 22:57:40 +0900
> Cc: emacs-devel@gnu.org
> 
> Leo writes:
> 
>  > As the subject line suggested.
> 
> You're using the wrong 'macs, that's why.

Here we go again.  Brace for it...

> Anyway, there will be one in Emacs 24, if I read the tea leaves
> correctly.

I don't see how could you conclude something like this, but maybe I'm
missing something.

Re: Why there is no char type?

[David Kastrup]

"Stephen J. Turnbull" <stephen@xemacs.org> writes:

> Leo writes:
>
>  > As the subject line suggested.
>
> You're using the wrong 'macs, that's why.

The right one, you mean.

> Anyway, there will be one in Emacs 24, if I read the tea leaves
> correctly.

I harbored tea like that once as well, but you won't find any on me
nowadays.

-- 
David Kastrup

Re: Why there is no char type?

[Leo]

On 2010-12-14 14:11 +0000, Eli Zaretskii wrote:
> Ouch!  Do us all a favor: please read past discussions on this.

Could you point to them? I have tried google, gmane and lists.gnu.org
and can't find any.

> If you still think there's something to discuss after that, ask more
> specific questions, to keep the bike-shedding (and the flames) at bay.
>
> Okay?

OK.

Re: Why there is no char type?

[alin soare]

> > Ouch!  Do us all a favor: please read past discussions on this.
> 
> Could you point to them? I have tried google, gmane and lists.gnu.org
> and can't find any.



you can also read an introductory book of lisp. there are lots.

Re: Why there is no char type?

[David Kastrup]

alin soare <as1789@gmail.com> writes:

>> > Ouch!  Do us all a favor: please read past discussions on this.
>> 
>> Could you point to them? I have tried google, gmane and lists.gnu.org
>> and can't find any.
>
>
>
> you can also read an introductory book of lisp. there are lots.

Huh?  Whether or not to make char a separate type is more an Emacs
decision than a Lisp one.

-- 
David Kastrup

Re: Why there is no char type?

[Stephen J. Turnbull]

Eli Zaretskii writes:
 > > From: "Stephen J. Turnbull" <stephen@xemacs.org>
 > > Date: Tue, 14 Dec 2010 22:57:40 +0900
 > > Cc: emacs-devel@gnu.org
 > > 
 > > Leo writes:
 > > 
 > >  > As the subject line suggested.
 >
 > > [T]here will be one in Emacs 24, if I read the tea leaves correctly.
 > 
 > I don't see how could you conclude something like this, but maybe I'm
 > missing something.

Handa-san said so a few months ago, IIRC.  Of course I may have
misunderstood what he was saying.

Re: Why there is no char type?

[Stephen J. Turnbull]

David Kastrup writes:
 > "Stephen J. Turnbull" <stephen@xemacs.org> writes:
 > 
 > > Leo writes:
 > >
 > >  > As the subject line suggested.
 > >
 > > You're using the wrong 'macs, that's why.
 > 
 > The right one, you mean.

Not if he wants a char type in Lisp.  Both XEmacs and SXEmacs have
such a type, and it far more frequently leads to detection of bugs
than it causes incompatibilities with code written for Emacs.  Big win
for us IMO (vs. XEmacs not having a char type; I'm not comparing the
different 'macs *except* from the point of view of having a char type,
and that's not arguable.  S?XEmacs has it, GNU don't.)

Re: Why there is no char type?

[Eli Zaretskii]

> From: "Stephen J. Turnbull" <stephen@xemacs.org>
> Cc: sdl.web@gmail.com,
>     emacs-devel@gnu.org
> Date: Wed, 15 Dec 2010 13:55:36 +0900
> 
>  > > [T]here will be one in Emacs 24, if I read the tea leaves correctly.
>  > 
>  > I don't see how could you conclude something like this, but maybe I'm
>  > missing something.
> 
> Handa-san said so a few months ago, IIRC.  Of course I may have
> misunderstood what he was saying.

I know of no such plans.

Re: Why there is no char type?

[David Kastrup]

"Stephen J. Turnbull" <stephen@xemacs.org> writes:

> David Kastrup writes:
>  > "Stephen J. Turnbull" <stephen@xemacs.org> writes:
>  > 
>  > > Leo writes:
>  > >
>  > >  > As the subject line suggested.
>  > >
>  > > You're using the wrong 'macs, that's why.
>  > 
>  > The right one, you mean.
>
> Not if he wants a char type in Lisp.

He asked why there wasn't one.  If he unconditionally wanted one without
wanting to hear reasons, he supposedly would have said so (likely
including asking where to get a different Emacs).

> Both XEmacs and SXEmacs have such a type, and it far more frequently
> leads to detection of bugs than it causes incompatibilities with code
> written for Emacs.

I have a hard time imagining that since automatic conversions in both
directions are in place for compatibility IIRC.  Could you point to a
few instances of these frequently detected bugs from your bug tracker?

-- 
David Kastrup

Re: Why there is no char type?

[Leo]

On 2010-12-15 05:03 +0000, Stephen J. Turnbull wrote:
> Both XEmacs and SXEmacs have such a type, and it far more frequently
> leads to detection of bugs than it causes incompatibilities with code
> written for Emacs. Big win for us IMO (vs. XEmacs not having a char
> type; I'm not comparing the different 'macs *except* from the point of
> view of having a char type, and that's not arguable.

It is very convenient to have a native char type. I have tried to view a
few variables, they show up nicely in XEmacs. This is a really good
feature of XEmacs. But unfortunately, it would take years for XEmacs to
sort out the license, then the lack of active development. By then GNU
Emacs will be riding on the guile ship.

Leo

-- 
Oracle is the new evil

Re: Why there is no char type?

[Richard Stallman]

Then main reason I did not add a char type is that it would require a
bunch of new primitives, adding permanently to the complexity of Emacs
Lisp and its documentation.  As well as complexity of Lisp code that
uses characters.

--
Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org, www.gnu.org

Re: Why there is no char type?

[Andy Wingo]

On Wed 15 Dec 2010 19:40, Richard Stallman <rms@gnu.org> writes:

> Then main reason I did not add a char type is that it would require a
> bunch of new primitives, adding permanently to the complexity of Emacs
> Lisp and its documentation.  As well as complexity of Lisp code that
> uses characters.

FWIW, I have not felt that a character type in Scheme was a burden in
any way; and it's nice to have a human-readable and -writable external
representation.

From a selfish, wanting-emacs-on-guile POV, it would be a good thing for
emacs to have a character type. Obviously that's predicated on there
being someone willing to do the work, and suitable back-compatibility
support.

Guile's characters have space for buckybits, I think (space we don't
currently use, I mean).

Andy
-- 
http://wingolog.org/

Re: Why there is no char type?

[Stephen J. Turnbull]

David Kastrup writes:

 > I have a hard time imagining that since automatic conversions in both
 > directions are in place for compatibility IIRC.  Could you point to a
 > few instances of these frequently detected bugs from your bug tracker?

They were eliminated from XEmacs itself over a decade ago, long before
there was a tracker.

Re: Why there is no char type?

[David Kastrup]

"Stephen J. Turnbull" <stephen@xemacs.org> writes:

> David Kastrup writes:
>
>  > I have a hard time imagining that since automatic conversions in
>  > both directions are in place for compatibility IIRC.  Could you
>  > point to a few instances of these frequently detected bugs from
>  > your bug tracker?
>
> They were eliminated from XEmacs itself over a decade ago, long before
> there was a tracker.

It would appear that our notions of "frequently" are somewhat
incompatible.

-- 
David Kastrup

Re: Why there is no char type?

[Stephen J. Turnbull]

David Kastrup writes:
 > "Stephen J. Turnbull" <stephen@xemacs.org> writes:
 > 
 > > They were eliminated from XEmacs itself over a decade ago, long before
 > > there was a tracker.
 > 
 > It would appear that our notions of "frequently" are somewhat
 > incompatible.

Not at all.  I still commit them occasionally in one-off scripts, and
I see them a few times a year in other users' help requests.  They're
just not going to be found in the ITS, that's all.

Re: Why there is no char type?

[Stefan Monnier]

Because,


        Stefan


PS: That could be changed, but there isn't much pressure for it now that
we learned to live without them.