Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Stefan Monnier]

>     cl-fill: Rename arguments to cl-seq and cl-item

Why?

The "cl-" prefix is for global vars.  It used to be used "everywhere"
for local vars as well because dynamic binding made the local vars
visible in other scopes as well, but nowadays that CL uses
lexical-binding this should not be needed any more and the tendency is
rather to eliminate use of the "cl-" prefix for args and local vars.

        Stefan


>     * lisp/emacs-lisp/cl-seq.el (cl-fill):
>     Rename arguments to 'cl-seq' and 'cl-item' as elsewhere.
> ---
>  lisp/emacs-lisp/cl-seq.el |   18 +++++++++---------
>  1 file changed, 9 insertions(+), 9 deletions(-)

> diff --git a/lisp/emacs-lisp/cl-seq.el b/lisp/emacs-lisp/cl-seq.el
> index 98e5f4e..ed27b7c 100644
> --- a/lisp/emacs-lisp/cl-seq.el
> +++ b/lisp/emacs-lisp/cl-seq.el
> @@ -144,24 +144,24 @@ called.
>        cl-accum)))
 
>  ;;;###autoload
> -(defun cl-fill (seq item &rest cl-keys)
> +(defun cl-fill (cl-seq cl-item &rest cl-keys)
>    "Fill the elements of SEQ with ITEM.
>  \nKeywords supported:  :start :end
>  \n(fn SEQ ITEM [KEYWORD VALUE]...)"
>    (cl--parsing-keywords ((:start 0) :end) ()
> -    (if (listp seq)
> -	(let ((p (nthcdr cl-start seq))
> +    (if (listp cl-seq)
> +	(let ((p (nthcdr cl-start cl-seq))
>  	      (n (if cl-end (- cl-end cl-start) 8000000)))
>  	  (while (and p (>= (setq n (1- n)) 0))
> -	    (setcar p item)
> +	    (setcar p cl-item)
>  	    (setq p (cdr p))))
> -      (or cl-end (setq cl-end (length seq)))
> -      (if (and (= cl-start 0) (= cl-end (length seq)))
> -	  (fillarray seq item)
> +      (or cl-end (setq cl-end (length cl-seq)))
> +      (if (and (= cl-start 0) (= cl-end (length cl-seq)))
> +	  (fillarray cl-seq cl-item)
>  	(while (< cl-start cl-end)
> -	  (aset seq cl-start item)
> +	  (aset cl-seq cl-start cl-item)
>  	  (setq cl-start (1+ cl-start)))))
> -    seq))
> +    cl-seq))
 
>  ;;;###autoload
>  (defun cl-replace (cl-seq1 cl-seq2 &rest cl-keys)

> _______________________________________________
> Emacs-diffs mailing list
> Emacs-diffs@gnu.org
> https://lists.gnu.org/mailman/listinfo/emacs-diffs

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Leo Liu]

On 2016-08-13 02:06 -0400, Stefan Monnier wrote:
>>     cl-fill: Rename arguments to cl-seq and cl-item
>
> Why?
>
> The "cl-" prefix is for global vars.  It used to be used "everywhere"
> for local vars as well because dynamic binding made the local vars
> visible in other scopes as well, but nowadays that CL uses
> lexical-binding this should not be needed any more and the tendency is
> rather to eliminate use of the "cl-" prefix for args and local vars.
>
>         Stefan

Apparently the functions surrounding cl-fill use cl- prefix. so the
change appear to be more cosmetic than technical.

Leo

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Stefan Monnier]

> Apparently the functions surrounding cl-fill use cl- prefix. so the
> change appear to be more cosmetic than technical.

Indeed.  But since the "cl-" prefix has to be used for global (aka
dynamically scoped) variables, it's good practice to *not* use it for
lexically scoped variables.

IOW, I'm all for such cosmetic changes that harmonize the code, but we
should do that by removing (rather than adding) the "cl-" prefix where
it's not needed.


        Stefan

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[John Wiegley]

>>>>> "SM" == Stefan Monnier <monnier@IRO.UMontreal.CA> writes:

SM> IOW, I'm all for such cosmetic changes that harmonize the code, but we
SM> should do that by removing (rather than adding) the "cl-" prefix where
SM> it's not needed.

Yay!

-- 
John Wiegley                  GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com                          60E1 46C4 BD1A 7AC1 4BA2

RE: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Drew Adams]

>> IOW, I'm all for such cosmetic changes that harmonize the code, but we
>> should do that by removing (rather than adding) the "cl-" prefix where
>> it's not needed.
> 
> Yay!

To clarify things more, we should prefix all `cl-'-prefixed names
with another `cl-': `cl-cl-cdr', etc.  If we later find that names
are still confusing we can try `cl-cl-cl-cdr', etc. ;-)

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Stefan Monnier]

> To clarify things more, we should prefix all `cl-'-prefixed names
> with another `cl-': `cl-cl-cdr', etc.  If we later find that names

That would be safer, yes.

> are still confusing we can try `cl-cl-cl-cdr', etc. ;-)

I think 3 "cl-" should be enough for everyone,


        Stefan

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[John Wiegley]

>>>>> Stefan Monnier <monnier@IRO.UMontreal.CA> writes:

>> are still confusing we can try `cl-cl-cl-cdr', etc. ;-)

> I think 3 "cl-" should be enough for everyone,

Along with the 512K that I run my computer on? :)  Never underestimate how
many cl-'s the future will need!!

-- 
John Wiegley                  GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com                          60E1 46C4 BD1A 7AC1 4BA2

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Robert Weiner]

[-- Attachment #1: Type: text/plain, Size: 232 bytes --]

On Tue, Aug 16, 2016 at 1:09 PM, Stefan Monnier <monnier@iro.umontreal.ca>
wrote:

> I think 3 "cl-" should be enough for everyone,


​Would not an infinite recursion of cl-'s help show peopl​e the
possibilities?

Bob

[-- Attachment #2: Type: text/html, Size: 833 bytes --]

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Lars Ingebrigtsen]

Robert Weiner <rsw@gnu.org> writes:

> ​Would not an infinite recursion of cl-'s help show peopl​e the possibilities?

Then we need a syntax for circular symbol names.  What about:

#1=cl-#1cadr

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

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Tino Calancha]

On 08/17/2016 03:15 AM, Lars Ingebrigtsen wrote:
> Robert Weiner <rsw@gnu.org> writes:
>
>> ​Would not an infinite recursion of cl-'s help show peopl​e the possibilities?
> Then we need a syntax for circular symbol names.  What about:
>
> #1=cl-#1cadr
>

At least my commit served to increase the moral of the troops
right before the inmminent 25.1 release.  Not to bad.

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Stefan Monnier]

>> ​Would not an infinite recursion of cl-'s help show peopl​e the possibilities?
> Then we need a syntax for circular symbol names.  What about:
> #1=cl-#1cadr

OMG, and then we could finally have a (recursive) symbol
for #1=GNU#1's Not Unix!


        Stefan

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Clément Pit--Claudel]

[-- Attachment #1.1: Type: text/plain, Size: 562 bytes --]

On 2016-08-16 11:04, Stefan Monnier wrote:
>> Apparently the functions surrounding cl-fill use cl- prefix. so the
>> change appear to be more cosmetic than technical.
> 
> Indeed.  But since the "cl-" prefix has to be used for global (aka
> dynamically scoped) variables, it's good practice to *not* use it for
> lexically scoped variables.
> 
> IOW, I'm all for such cosmetic changes that harmonize the code, but we
> should do that by removing (rather than adding) the "cl-" prefix where
> it's not needed.

Should the change be reverted, then?


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: [Emacs-diffs] master 50548fd: cl-fill: Rename arguments to cl-seq and cl-item

[Tino Calancha]

On 08/16/2016 11:00 PM, Leo Liu wrote:
> On 2016-08-13 02:06 -0400, Stefan Monnier wrote:
>>>      cl-fill: Rename arguments to cl-seq and cl-item
>> Why?
>>
>> The "cl-" prefix is for global vars.  It used to be used "everywhere"
>> for local vars as well because dynamic binding made the local vars
>> visible in other scopes as well, but nowadays that CL uses
>> lexical-binding this should not be needed any more and the tendency is
>> rather to eliminate use of the "cl-" prefix for args and local vars.
>>
>>          Stefan
> Apparently the functions surrounding cl-fill use cl- prefix. so the
> change appear to be more cosmetic than technical.
>
> Leo

I confirm Leo explanation is 100% is right about my intentions.
I am sorry for the troubles.
Tino