syntax question for defun

syntax question for defun

[David Elson]

Hello. I have been using emacs for ... a while.

For questions, I have often browsed and browsed (i.e. googled) and ... 
and usually find a good explanation. But sometimes I don't find an 
answer, even when it is staring right at me.

My question:

To emulate something akin to the PL1/C/C++ arrow notation, I have used:

*(defun -> (object method &rest args) ...)*

and it works.

*Is this valid elisp/lisp?*

If it works only coincidentally, then it might break in the future, when 
a pressing emacs issue is resolved in a manner that requires plugging 
this "loophole".

If it is formally valid, where does it document the level of flexibility 
that allows this syntax?

If it is not formally valid, then why does it work coincidentally?

I'd like to enhance/use this code, but I hesitate due to this question.

Thanks in advance?

-David

Re: syntax question for defun

[tomas]

[-- Attachment #1: Type: text/plain, Size: 2118 bytes --]

On Wed, Aug 28, 2024 at 01:01:33PM -0700, David Elson wrote:
> Hello. I have been using emacs for ... a while.
> 
> For questions, I have often browsed and browsed (i.e. googled) and ... and
> usually find a good explanation. But sometimes I don't find an answer, even
> when it is staring right at me.
> 
> My question:
> 
> To emulate something akin to the PL1/C/C++ arrow notation, I have used:
> 
> *(defun -> (object method &rest args) ...)*
> 
> and it works.
> 
> *Is this valid elisp/lisp?*

It is. From the elisp documentation:
 -- Macro: defun name args [doc] [declare] [interactive] body...
     ‘defun’ is the usual way to define new Lisp functions.  It defines
     the symbol NAME as a function with argument list ARGS ...

... so NAME is a symbol. Again, from the doc:

   A symbol name can contain any characters whatever.  Most symbol names
  are written with letters, digits, and the punctuation characters
  ‘-+=*/’.  Such names require no special punctuation; the characters of
  the name suffice as long as the name does not look like a number.  (If
  it does, write a ‘\’ at the beginning of the name to force
  interpretation as a symbol.)  The characters ‘_~!@$%^&:<>{}?’ are less
  often used but also require no special punctuation.  Any other
  characters may be included in a symbol’s name by escaping them with a
  backslash.

So "->" is not only a legal symbol name, but also one that doesn't need
special escaping to be seen as such. So go wild :)

Actually, Emacs Lisp comes with one function named "-" and another named
">".

That said, this doesn't mean that other people will find your code readable
or enjoyable. It's on you to find that out :-)

> If it works only coincidentally, then it might break in the future, when a
> pressing emacs issue is resolved in a manner that requires plugging this
> "loophole".
> 
> If it is formally valid, where does it document the level of flexibility
> that allows this syntax?

The Emacs Lisp documentation should be your go-to place for such questions.

Cheers
-- 
t

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

Re: syntax question for defun

[David Elson]

Many *many* thanks tomas !!!

The information you have provided not only answers my basic question (I 
was able to find the sections by googling effectively disambiguating 
strings from your fine reply), but gives me some good insight into more 
properly reading the elisp docs.

Pending a look at elisp lexical analysis code (would be interesting), 
I'll count on algorithmic "greediness" to disambiguate '->' from '-' 
(:-), and a new battery in my keyboard to disambiguate it from '>' (:-).

I try to write clear code, but sometimes ... always a good caution, though.

Y'all be well.

-DE

PS Also, per your excellent reply, I will go (cautiously) wild with this 
(pun intended).

PPS (please pardon the *skippable* after-the-fact context) As for 
readability, I try to keep the number of columns down, mainly due to the 
way I tend to split windows.

Some elisp code from 2018, not (yet) intended for anyone (else). I'm 
sure there are many ways, per knowledgeable folks such as yourself, to 
improve the approach/style, and I'm all ears. This is just a snippet of 
what I was starting with:

(require 'object)

...

(defun new-filter-state-machine ()
   (let*
       (
        (fsm (new-object 'filter-state-machine))
        )
     (-> fsm 'add-attr 'state 0)
     (-> fsm 'add-attr 'exp-len -1)
     (-> fsm 'add-method 'reset
         (lambda (this)
           (-> this 'set-state 0)
           (-> this 'set-exp-len -1)
           t
           )
         )
     (-> fsm 'add-method 'process-string
     ...

On 8/28/24 9:57 PM, tomas@tuxteam.de wrote:
> On Wed, Aug 28, 2024 at 01:01:33PM -0700, David Elson wrote:
>> Hello. I have been using emacs for ... a while.
>>
>> For questions, I have often browsed and browsed (i.e. googled) and ... and
>> usually find a good explanation. But sometimes I don't find an answer, even
>> when it is staring right at me.
>>
>> My question:
>>
>> To emulate something akin to the PL1/C/C++ arrow notation, I have used:
>>
>> *(defun -> (object method &rest args) ...)*
>>
>> and it works.
>>
>> *Is this valid elisp/lisp?*
> It is. From the elisp documentation:
>   -- Macro: defun name args [doc] [declare] [interactive] body...
>       ‘defun’ is the usual way to define new Lisp functions.  It defines
>       the symbol NAME as a function with argument list ARGS ...
>
> ... so NAME is a symbol. Again, from the doc:
>
>     A symbol name can contain any characters whatever.  Most symbol names
>    are written with letters, digits, and the punctuation characters
>    ‘-+=*/’.  Such names require no special punctuation; the characters of
>    the name suffice as long as the name does not look like a number.  (If
>    it does, write a ‘\’ at the beginning of the name to force
>    interpretation as a symbol.)  The characters ‘_~!@$%^&:<>{}?’ are less
>    often used but also require no special punctuation.  Any other
>    characters may be included in a symbol’s name by escaping them with a
>    backslash.
>
> So "->" is not only a legal symbol name, but also one that doesn't need
> special escaping to be seen as such. So go wild :)
>
> Actually, Emacs Lisp comes with one function named "-" and another named
> ">".
>
> That said, this doesn't mean that other people will find your code readable
> or enjoyable. It's on you to find that out :-)
>
>> If it works only coincidentally, then it might break in the future, when a
>> pressing emacs issue is resolved in a manner that requires plugging this
>> "loophole".
>>
>> If it is formally valid, where does it document the level of flexibility
>> that allows this syntax?
> The Emacs Lisp documentation should be your go-to place for such questions.
>
> Cheers

Re: syntax question for defun

[tomas]

[-- Attachment #1: Type: text/plain, Size: 440 bytes --]

On Thu, Aug 29, 2024 at 09:45:59AM -0700, David Elson wrote:
> Many *many* thanks tomas !!!
> 
> The information you have provided not only answers my basic question (I was
> able to find the sections by googling effectively disambiguating strings
> from your fine reply), but gives me some good insight into more properly
> reading the elisp docs.

Glad it helped. I see you are enjoying -- thus doubly glad :-)

Cheers
-- 
t

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

Re: syntax question for defun

[Stefan Monnier via Users list for the GNU Emacs text editor]

> Some elisp code from 2018, not (yet) intended for anyone (else). I'm sure
> there are many ways, per knowledgeable folks such as yourself, to improve
> the approach/style, and I'm all ears. This is just a snippet of what I was
> starting with:

You might enjoy

    Eoops: an object-oriented programming system for Emacs-Lisp
    Houser, Chris and Kalter, Scott D. (1992)
    https://dl.acm.org/doi/10.1145/147135.147248
    https://www.emacswiki.org/emacs/EmacsObjectOrientedProgrammingSystem


- Stefan