Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Stefan Monnier]

> +After a command is executed, if point moved into a region that has
> +special properties (e.g. composition, display), Emacs adjusts point to
> +the boundary of the region.  But when a command binds this variable to
> +non-nil, this point adjustment is suppressed.
 
Why did you replace "sets" with "binds"?
AFAIK let-binding this variable has no effect because the variable is
consulted only after the command finishes, i.e. after the binding has
been undone.


        Stefan

Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Eli Zaretskii]

> From: Stefan Monnier <monnier@IRO.UMontreal.CA>
> Cc: Eli Zaretskii <eliz@gnu.org>
> Date: Mon, 22 Feb 2016 13:50:37 -0500
> 
> > +After a command is executed, if point moved into a region that has
> > +special properties (e.g. composition, display), Emacs adjusts point to
> > +the boundary of the region.  But when a command binds this variable to
> > +non-nil, this point adjustment is suppressed.
>  
> Why did you replace "sets" with "binds"?
> AFAIK let-binding this variable has no effect because the variable is
> consulted only after the command finishes, i.e. after the binding has
> been undone.

Who said anything about let-binding?

Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Stefan Monnier]

>> > +After a command is executed, if point moved into a region that has
>> > +special properties (e.g. composition, display), Emacs adjusts point to
>> > +the boundary of the region.  But when a command binds this variable to
>> > +non-nil, this point adjustment is suppressed.
>> 
>> Why did you replace "sets" with "binds"?
>> AFAIK let-binding this variable has no effect because the variable is
>> consulted only after the command finishes, i.e. after the binding has
>> been undone.

> Who said anything about let-binding?

To me the difference between "sets" and "binds" is that the former
clearly means "setq" while the other can either mean "change it from
unbound to bound" or "let-binds".

So, I find "binds" to be less precise than "sets", hence my question:
Why did you replace "sets" with "binds"?


        Stefan

Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Eli Zaretskii]

> From: Stefan Monnier <monnier@IRO.UMontreal.CA>
> Cc: emacs-devel@gnu.org
> Date: Mon, 22 Feb 2016 15:45:35 -0500
> 
> To me the difference between "sets" and "binds" is that the former
> clearly means "setq" while the other can either mean "change it from
> unbound to bound" or "let-binds".
> 
> So, I find "binds" to be less precise than "sets", hence my question:
> Why did you replace "sets" with "binds"?

Because that's how I tend to think about the act of giving a value to
a variable in Emacs Lisp.

Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Stefan Monnier]

>> To me the difference between "sets" and "binds" is that the former
>> clearly means "setq" while the other can either mean "change it from
>> unbound to bound" or "let-binds".
>> So, I find "binds" to be less precise than "sets", hence my question:
>> Why did you replace "sets" with "binds"?
> Because that's how I tend to think about the act of giving a value to
> a variable in Emacs Lisp.

OK.  To me, the fact that disable-point-adjustment needs to be setq'd by
the command rather than let-bound is an important point (I'd expect this
to surprise many coders since the usual way for a command to affect the
behavior of Emacs "for this command only" is to let-bind a variable), so
I find the new wording to encourage confusion rather than trying to
avoid it.

But clearly you have a different opinion, and I honestly can't claim to
know better, especially when it comes to writing docs, so I'll shut up
now,


        Stefan

Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Eli Zaretskii]

> From: Stefan Monnier <monnier@IRO.UMontreal.CA>
> Cc: emacs-devel@gnu.org
> Date: Mon, 22 Feb 2016 16:13:36 -0500
> 
> >> Why did you replace "sets" with "binds"?
> > Because that's how I tend to think about the act of giving a value to
> > a variable in Emacs Lisp.
> 
> OK.  To me, the fact that disable-point-adjustment needs to be setq'd by
> the command rather than let-bound is an important point (I'd expect this
> to surprise many coders since the usual way for a command to affect the
> behavior of Emacs "for this command only" is to let-bind a variable), so
> I find the new wording to encourage confusion rather than trying to
> avoid it.

If this is confusing, I will changing it back.  What do others think?

Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Lars Ingebrigtsen]

Eli Zaretskii <eliz@gnu.org> writes:

> If this is confusing, I will changing it back.  What do others think?

I think "set" is clearer in this instance.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Kaushal Modi]

[-- Attachment #1: Type: text/plain, Size: 123 bytes --]

On Feb 22, 2016 10:41 PM, "Lars Ingebrigtsen" <larsi@gnus.org> wrote:
>
> I think "set" is clearer in this instance.

Yes.

[-- Attachment #2: Type: text/html, Size: 233 bytes --]

RE: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Drew Adams]

> > >> Why did you replace "sets" with "binds"?
> > > Because that's how I tend to think about the act of giving a value to
> > > a variable in Emacs Lisp.
> >
> > OK.  To me, the fact that disable-point-adjustment needs to be setq'd by
> > the command rather than let-bound is an important point (I'd expect this
> > to surprise many coders since the usual way for a command to affect the
> > behavior of Emacs "for this command only" is to let-bind a variable), so
> > I find the new wording to encourage confusion rather than trying to
> > avoid it.
> 
> If this is confusing, I will changing it back.  What do others think?

I agree with Stefan here.  Generally use "set" for assignment 
(regardless of whether the symbol has already been bound in any
way).  Use "bind" for a `let' binding.

It's true that we also speak of a symbol being unbound or bound
depending on whether its value cell is void.  But most of the
uses of "bound" and "bind" for symbols in Emacs have to do with
`let' bindings.

However, if it is important in this particular context to make
clear that "disable-point-adjustment needs to be setq'd by the
command rather than let-bound" then I'd say that, to remove any
ambiguity.

Re: [Emacs-diffs] emacs-25 6bd9d69: Fix documentation of 'global-disable-point-adjustment'

[Eli Zaretskii]

OK, I clarified the doc strings wrt this aspect.

Thanks for the feedback.