Question about `cond*'

Question about `cond*'

[Michelangelo Rodriguez]

Is there a reason to not provide, in the file "dict-star.el", a feature?
Best regards,
Michelangelo Rodriguez

Re: Question about `cond*'

[Eli Zaretskii]

> From: Michelangelo Rodriguez <michelangelo.rodriguez@gmail.com>
> Date: Tue, 01 Oct 2024 10:35:14 +0200
> 
> Is there a reason to not provide, in the file "dict-star.el", a feature?

What is dict-star.el? where did you see it?

Re: Question about `cond*'

[Michelangelo Rodriguez]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Michelangelo Rodriguez <michelangelo.rodriguez@gmail.com>
>> Date: Tue, 01 Oct 2024 10:35:14 +0200
>> 
>> Is there a reason to not provide, in the file "dict-star.el", a feature?
>
> What is dict-star.el? where did you see it?
Sorry, the file is "cond-star.el".
Anyway the new feature `cond*' is not available as feature.

Thanks.

Re: Question about `cond*'

[Philip Kaludercic]

Michelangelo Rodriguez <michelangelo.rodriguez@gmail.com> writes:

> Eli Zaretskii <eliz@gnu.org> writes:
>
>>> From: Michelangelo Rodriguez <michelangelo.rodriguez@gmail.com>
>>> Date: Tue, 01 Oct 2024 10:35:14 +0200
>>> 
>>> Is there a reason to not provide, in the file "dict-star.el", a feature?
>>
>> What is dict-star.el? where did you see it?
> Sorry, the file is "cond-star.el".
> Anyway the new feature `cond*' is not available as feature.

I can only assume it was a mistake.  The question is should the feature
be called `cond-star' or `cond*'.

> Thanks.

-- 
	Philip Kaludercic on siskin

Re: Question about `cond*'

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > I can only assume it was a mistake.  The question is should the feature
  > be called `cond-star' or `cond*'.

I think `cond*' is a more natural name for the feature.
That is what users will think of -- to do (require 'cont-star)
would require a little extra bit of trivial knowledge.

I called the file `cond-star.el' to avoid the inconvenience that
an asterisk in the file name could cause.  But it woukd be better
if users did not have to know that file name.

`cond*' really ought to be preloaded.

-- 
Dr Richard Stallman (https://stallman.org)
Chief GNUisance of the GNU Project (https://gnu.org)
Founder, Free Software Foundation (https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)

Re: Question about `cond*'

[Sean Whitton]

Hello,

On Wed 02 Oct 2024 at 11:35pm -04, Richard Stallman wrote:

> [[[ To any NSA and FBI agents reading my email: please consider    ]]]
> [[[ whether defending the US Constitution against all enemies,     ]]]
> [[[ foreign or domestic, requires you to follow Snowden's example. ]]]
>
>   > I can only assume it was a mistake.  The question is should the feature
>   > be called `cond-star' or `cond*'.
>
> I think `cond*' is a more natural name for the feature.
> That is what users will think of -- to do (require 'cont-star)
> would require a little extra bit of trivial knowledge.
>
> I called the file `cond-star.el' to avoid the inconvenience that
> an asterisk in the file name could cause.  But it woukd be better
> if users did not have to know that file name.
>
> `cond*' really ought to be preloaded.

Given that it isn't used in a single place yet, and we have repeatedly
refused to preload cl-lib, that does not seem appropriate.

-- 
Sean Whitton

Re: Question about `cond*'

[Alfred M. Szmidt]

   >   > I can only assume it was a mistake.  The question is should the feature
   >   > be called `cond-star' or `cond*'.
   >
   > I think `cond*' is a more natural name for the feature.
   > That is what users will think of -- to do (require 'cont-star)
   > would require a little extra bit of trivial knowledge.
   >
   > I called the file `cond-star.el' to avoid the inconvenience that
   > an asterisk in the file name could cause.  But it woukd be better
   > if users did not have to know that file name.
   >
   > `cond*' really ought to be preloaded.

   Given that it isn't used in a single place yet, and we have repeatedly
   refused to preload cl-lib, that does not seem appropriate.

cl-lib isn't preloaded for other reasons (Emacs Lisp isn't Common
Lisp), cond* is part of the Emacs Lisp language, and that would be a
good reason to preloaded it so that it can be more widley used.

Re: Question about `cond*'

[Sean Whitton]

Hello,

On Thu 03 Oct 2024 at 12:51am -04, Alfred M. Szmidt wrote:

> cl-lib isn't preloaded for other reasons (Emacs Lisp isn't Common
> Lisp), cond* is part of the Emacs Lisp language, and that would be a
> good reason to preloaded it so that it can be more widley used.

There are criteria for preloading, and they do not, to my knowledge,
include ideas like these, which, in any case, aren't shared by everyone.

Anyway, I'll leave this to the head maintainers.

-- 
Sean Whitton

Re: Question about `cond*'

[Eli Zaretskii]

> From: Richard Stallman <rms@gnu.org>
> Cc: michelangelo.rodriguez@gmail.com, emacs-devel@gnu.org
> Date: Wed, 02 Oct 2024 23:35:20 -0400
> 
>   > I can only assume it was a mistake.  The question is should the feature
>   > be called `cond-star' or `cond*'.
> 
> I think `cond*' is a more natural name for the feature.
> That is what users will think of -- to do (require 'cont-star)
> would require a little extra bit of trivial knowledge.
> 
> I called the file `cond-star.el' to avoid the inconvenience that
> an asterisk in the file name could cause.  But it woukd be better
> if users did not have to know that file name.

If we provide 'cond*' as the feature, (require 'cond*) will not work
unless Lisp programs also specify the file name.  So this gets us the
need to know the file name back through a back-door anyway.

> `cond*' really ought to be preloaded.

I don't see a need to preload it at this time, for the same reason we
don't preload pcase and other macros: they are not necessarily needed
right at the beginning of a session, and not used in any preloaded
code.  I think we should first consider autoloading cond*, before we
talk about preloading it.

Re: Question about `cond*'

[Andrea Corallo]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Richard Stallman <rms@gnu.org>
>> Cc: michelangelo.rodriguez@gmail.com, emacs-devel@gnu.org
>> Date: Wed, 02 Oct 2024 23:35:20 -0400

[...]

>> `cond*' really ought to be preloaded.
>
> I don't see a need to preload it at this time, for the same reason we
> don't preload pcase and other macros: they are not necessarily needed
> right at the beginning of a session, and not used in any preloaded
> code.  I think we should first consider autoloading cond*, before we
> talk about preloading it.

+1

  Andrea

Re: Question about `cond*'

[Richard Stallman]

[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  >   I think we should first consider autoloading cond*, before we
  > talk about preloading it.

You're right, autoloading it would be good enough.  How about it?

-- 
Dr Richard Stallman (https://stallman.org)
Chief GNUisance of the GNU Project (https://gnu.org)
Founder, Free Software Foundation (https://fsf.org)
Internet Hall-of-Famer (https://internethalloffame.org)

Re: Question about `cond*'

[Sean Whitton]

Hello,

On Sat 05 Oct 2024 at 11:18pm -04, Richard Stallman wrote:

> [[[ To any NSA and FBI agents reading my email: please consider    ]]]
> [[[ whether defending the US Constitution against all enemies,     ]]]
> [[[ foreign or domestic, requires you to follow Snowden's example. ]]]
>
>   >   I think we should first consider autoloading cond*, before we
>   > talk about preloading it.
>
> You're right, autoloading it would be good enough.  How about it?

Sounds good to me too.

-- 
Sean Whitton

Re: Question about `cond*'

[Eli Zaretskii]

> From: Richard Stallman <rms@gnu.org>
> Cc: philipk@posteo.net, michelangelo.rodriguez@gmail.com,
> 	emacs-devel@gnu.org
> Date: Sat, 05 Oct 2024 23:18:55 -0400
> 
>   >   I think we should first consider autoloading cond*, before we
>   > talk about preloading it.
> 
> You're right, autoloading it would be good enough.  How about it?

I have no objections to that.

Re: Question about `cond*'

[Stefan Kangas]

Michelangelo Rodriguez <michelangelo.rodriguez@gmail.com> writes:

> Anyway the new feature `cond*' is not available as feature.

This was fixed in 4197a8450669e08c2d26f4a7d77dcb2771bd9492.

Re: Question about `cond*'

[Eli Zaretskii]

> From: Michelangelo Rodriguez <michelangelo.rodriguez@gmail.com>
> Date: Tue, 01 Oct 2024 19:05:30 +0200
> 
> Anyway the new feature `cond*' is not available as feature.

The feature's name is 'cond-star', which is consistent with what we do
elsewhere in Emacs: the name of the feature is the same as the name of
the file which provides the feature.

Re: Question about `cond*'

[Alfred M. Szmidt]

   Eli Zaretskii <eliz@gnu.org> writes:

   >> From: Michelangelo Rodriguez <michelangelo.rodriguez@gmail.com>
   >> Date: Tue, 01 Oct 2024 10:35:14 +0200
   >> 
   >> Is there a reason to not provide, in the file "dict-star.el", a feature?
   >
   > What is dict-star.el? where did you see it?
   Sorry, the file is "cond-star.el".
   Anyway the new feature `cond*' is not available as feature.

A (macrop #'cond*) would satisfy any feature requirements.