all messages for Emacs-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
From: Mike Mattie <codermattie@gmail.com>
To: help-gnu-emacs@gnu.org
Subject: Re: Help with keybinding to delete between {}
Date: Thu, 13 Dec 2007 04:46:12 -0800	[thread overview]
Message-ID: <20071213044612.43b371c4@reforged> (raw)
In-Reply-To: <20071213024400.4c65e207@reforged>


[-- Attachment #1.1: Type: text/plain, Size: 6110 bytes --]

On Thu, 13 Dec 2007 02:44:00 -0800
Mike Mattie <codermattie@gmail.com> wrote:

> On Thu, 6 Dec 2007 09:14:12 -0800 (PST)
> Xah Lee <xah@xahlee.org> wrote:
> 
> > for some reason my code in the previous post is completely non-
> > functional. (i swear i used it for few months. Perhaps when i put on
> > the website i got smart and edited it "for the better" without
> > testing)
> > 
> > Here's the correct version:
> > 
> > (defun delete-enclosed-text ()
> >   "Delete texts between any pair of delimiters.
> > Note: if you have nested matching pairs, the cursor
> > should be inside the inner most one. Else it gets confused.
> > This code should to be fixed in the future."
> >   (interactive)
> >   (save-excursion
> >     (let (p1 p2)
> >       (skip-chars-backward "^(<["<<") (setq p1 (point))
> >       (skip-chars-forward "^)>]">>") (setq p2 (point))
> >       (delete-region p1 p2)
> >     )
> >   )
> > )
> 
> I decided to fix it in regards to nested lists. My implementation
> works however it is purely for elucidation. It is a recursive
> implementation that is nth recursive where n is the number of nested
> delimiters in the list.
> 
> that means that if you use this on a list with deep nesting it *will*
> crash after exhausting the recursion limit. So don't use it for real.
> 
> A reliable solution would likely be based off something from
> thingatpoint.el ? there must be something like this in the
> code-motion code for elisp.
> 
> This sort of function/feature? is dangerous even when implemented
> procedurally because it will go wild if your delimiters aren't
> matched correctly.
> 
> Other flaws:
>   * error path in bounds scan marked by "error ?" not implemented
> because I have not studied elisp error handling yet.
> 
>   * does not handle multi-byte characters due to use of aref in
> bounds-scan-{forward,backward}
> 
> With that said it does illustrate the value of lexical-let, and
> solves the problem in a general way. It would be quite easy to add
> variations that kill instead of deleting, or whatever other features
> you want.
> 
> I implemented it recursively simply because It looked prettier to me,
> probably since I started with scheme. anyways here goes.
> 
> one last big fat warning: this code is for fun, it will eat your
> children eventually.

a small re-factor I noticed when I read the code a bit later. the check
for motion in the scan is hoisted into bounds-scan-{forward,backward} so
it isn't duplicated needlessly.

> ;; --- start of elisp

(defun bounds-scan ( seek-bounds open-bound-p close-bound-p restart-position position level )
  "scan for the delimitation of a region. This is a general form of a
   simple algorithm that counts opening and closing delimiters to scan
   past nested delimited spans."
  (progn
    (goto-char position) ;; move to the starting position before scanning.
    (funcall seek-bounds)

    (cond
      ((funcall open-bound-p)
        (bounds-scan seek-bounds open-bound-p close-bound-p restart-position
          (funcall restart-position) (+ level 1)))

      ((funcall close-bound-p)
        (if (> level 0)
          ;; when we have a positive level to start with
          ;; scan again with a decremented level.
          (bounds-scan seek-bounds open-bound-p close-bound-p restart-position
            (funcall restart-position) (- level 1))

          ;; return point as we are done
          (point)
          ))
      ;; error ?
      )))

(defun bounds-scan-forward ( delimiters position )
  "entry point for bounds-scan forward. given delimiters: a
   string containing a pair of delimiting characters, which must
   be in \"open close\" order, scan forward for the bounding
   delimiter returning the position before the delimiter"

  (lexical-let
    ((open-delimiter (aref delimiters 0))
     (close-delimiter (aref delimiters 1)))

    (let
      ((close-at (bounds-scan
                   (lambda ()
                     (skip-chars-forward (concat "^" delimiters)))

                   (lambda ()
                     (char-equal open-delimiter (char-after)))

                   (lambda ()
                     (char-equal close-delimiter (char-after)))

                   (lambda ()
                     (+ (point) 1))

                   position 0)))
      (if (> close-at position)
        (- close-at 1)
        position)
      )))

(defun bounds-scan-backward ( delimiters position )
  "entry point for bounds-scan backward. given delimiters: a
   string containing a pair of delimiting characters, which must
   be in \"open close\" order, scan backward for the bounding
   delimiter returning the position after the delimiter"

  (lexical-let
    ;; note the inversion of the order since we are looking backwards
    ((open-delimiter (aref delimiters 1))
     (close-delimiter (aref delimiters 0)))

    (let
      ((open-at (bounds-scan
                  (lambda ()
                    (skip-chars-backward (concat "^" delimiters)))

                  (lambda ()
                    (char-equal open-delimiter (char-before)))

                  (lambda ()
                    (char-equal close-delimiter (char-before)))

                  (lambda ()
                    (- (point) 1))

                  position 0)))
      (if (< open-at position)
        (+ open-at 1)
        position)
      )))

(defun scan-lisp-list-close ()
  "wrapper for bounds-scan that searches for the closing delimiter of a lisp list"
  (bounds-scan-forward "()" (point)))

(defun scan-lisp-list-open ()
  "wrapper for bounds-scan that searches for the opening delimiter of a lisp list"
  (bounds-scan-backward "()" (point)))

(defun lisp-list-delete-body ()
  "delete the body of a lisp list including any nested lists"
  (interactive)
  (let
    ((open-pos (scan-lisp-list-open))
     (close-pos (scan-lisp-list-close)))

    (delete-backward-char (- close-pos open-pos))))

> 
> ;;----- end of elisp
> 
> Cheers,
> Mike Mattie

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

[-- Attachment #2: Type: text/plain, Size: 152 bytes --]

_______________________________________________
help-gnu-emacs mailing list
help-gnu-emacs@gnu.org
http://lists.gnu.org/mailman/listinfo/help-gnu-emacs

  reply	other threads:[~2007-12-13 12:46 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-12-05  1:33 Help with keybinding to delete between {} lampshade
2007-12-05 10:48 ` Bernardo Bacic
2007-12-05 12:55 ` William Xu
2007-12-05 13:30 ` Andreas Röhler
2007-12-05 23:50 ` Ilya Zakharevich
2007-12-06  6:33 ` Xah Lee
2007-12-06 17:14   ` Xah Lee
2007-12-12  9:20     ` lampshade
2007-12-13  3:59     ` Mike Mattie
     [not found]     ` <mailman.4961.1197518641.18990.help-gnu-emacs@gnu.org>
2007-12-13 10:39       ` Xah Lee
2007-12-13 10:44     ` Mike Mattie
2007-12-13 12:46       ` Mike Mattie [this message]
2007-12-13 13:01       ` Peter Dyballa
2007-12-14 11:43   ` Xah Lee
2007-12-07 16:30 ` Stefan Monnier

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20071213044612.43b371c4@reforged \
    --to=codermattie@gmail.com \
    --cc=help-gnu-emacs@gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this external index

	https://git.savannah.gnu.org/cgit/emacs.git
	https://git.savannah.gnu.org/cgit/emacs/org-mode.git

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.