Add #_ reader form

[agzam]

Clojure's #_ (Ignore next form reader macro) is very useful. Would be very cool to have a function and a dedicated key binding to toggle it.

My initial attempt looks like this:

(defun cljr-toggle-ignore-next-form ()
  "Ignore next form (#_)"
  (interactive)
  (if (search-backward "#_" 2 t 1)
  (delete-char 2)
(progn
  (let ((fc (following-char)))
    (cond ((-contains? '( ?\) ?\] ?\} ) fc) (paredit-backward-up))
      ((-contains? '( ?\( ?\[ ?\: ?\{ ) fc) nil)
      (t (beginning-of-thing 'sexp)))
    (insert "#_")))))

What do you guys, think?

[expez]

I personally use this a lot, but I guess I can see this being useful.

@benedekfazekas has done some fantastic work in moving our elisp-only refactorings over to clojure-mode. We might considering implementing this feature over there straight away.

[benedekfazekas]

I like the idea although I use @Malabarba's hack too quoted by Lars. Agree
with him: factor out paredit and see what @Malabarba and @bbatsov think
about it.
On 21 May 2016 1:38 pm, "Lars Andersen" notifications@github.com wrote:

I personally use this
http://endlessparentheses.com/a-comment-or-uncomment-sexp-command.html
a lot, but I guess I can see this being useful.

@benedekfazekas https://github.com/benedekfazekas has done some
fantastic work in moving our elisp-only refactorings over to clojure-mode.
We might considering implementing this feature over there straight away.

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
#334 (comment)

[Malabarba]

I too use the comment-sexp function linked above, but I know a lot of people prefer the #_ reader macro, so I'd be fine with adding something like this to Clojure mode.
It needs a bit of work though.
Currently if you try to comment a form but there's already a commented form higher up in the file it will go there and uncomment it instead.

[bbatsov]

I'm fine with adding something like this to clojure-mode, but it definitely needs some polish. I'm guessing we can add a variant of it for the comment macro as well (although using paredit wrapping & unwrapping with some form is not complex at all).

[agzam]

Guys, just a reminder: comment-sexp and #_ are different things. This guy would force Clojure's Reader to ignore any form that follows this macro.

[Malabarba]

Commenting a sexp with semi colons is the same thing as commenting it with #_, isn't it?

[bbatsov]

Yep, it is. Commenting with comment results in a nil return value.

[agzam]

according to the docs http://clojure.org/reference/reader

The form following #_ is completely skipped by the reader. (This is a more complete removal than the comment macro which yields nil)

[Malabarba]

@agzam I think you misunderstood us. :-)
The comment-sexp command we all talked about uses semi-colons to comment out (or uncomment) a sexp. It has exactly the same effect as using #_. It has nothing to do with Clojure's (comment ...) macro. You can follow the link expez posted to see it in action.

Nonetheless, like I said, your proposal would be a fine addition to Clojure-mode, just as long as we fix the issue when removing the #_.

[agzam]

@Malabarba no, I get it, it works for sexps but for something like this:

(foo [a b #_c] 
...)

semicolons make it difficult

(foo [a b
;c
]
...)

Honestly I'm starting to think that default comment operator in clj/cljs/cider-repl modes should always comment with #_ and only use semicolons when block of lines region is selected

[Malabarba]

Honestly I'm starting to think that default comment operator in clj/cljs/cider-repl modes should always comment with #_ and only use semicolons when block of lines region is selected

If you're referring to the behaviour of commands like M-; (which is an Emacs built-in), then I think it would be nice to have a command that uses #_ if the region exactly encompasses a sexp, while using ;; if the region encompasses something more general.
Unfortunately, I think that would be a little too complicated to implement (mostly because it would have to seamlessly support UNcommenting both of those syntaxes).

OTOH, having a separate command to comment/uncomment a sexp with #_ is much simpler and seems like a fine compromise to me.

[agzam]

@Malabarba oh, I'm excited, I guess it's possible to make it smart enough. Is there a way to check if the thing-at-point commented without having to search backwards for either ; or #_?