unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
From: "Drew Adams" <drew.adams@oracle.com>
To: "'Thierry Volpiatto'" <thierry.volpiatto@gmail.com>,
	<10386@debbugs.gnu.org>
Cc: 'Jari Aalto' <jari.aalto@cante.net>
Subject: bug#10386: CODE wishlist: íncludedelete-duplicates.el
Date: Wed, 28 Dec 2011 08:56:48 -0800	[thread overview]
Message-ID: <97F263C286E14A3A9D8656356D98ECD1@us.oracle.com> (raw)
In-Reply-To: <87aa6dvz8l.fsf@gmail.com>

> >     http://mwolson.org/static/dist/elisp/delete-duplicates.el
> 
> This is already provided in Emacs24+.

`delete-dups' has been in Emacs since Emacs 22, actually.

> What is not provided is a fast version of remove-duplicates.
> http://article.gmane.org/gmane.emacs.devel/139546/match=remove+dups

+1

And the cl-seq.el version of `remove-duplicates' is *particularly* slow.  Even a
classic list dups removal algorithm is much faster.

Here's a comparison using `equal' and a list of strings (`elp-results'):

hash-remove-dups     1           0.031         0.031
list-remove-dups     1           5.813         5.813
remove-duplicates    1           122.875       122.875

Where:

(defun list-remove-dups (list)
  (let ((tail  list)
        new)
    (while tail
      (unless (member (car tail) new) (push (car tail) new))
      (pop tail))
    (nreverse new)))

(defun* hash-remove-dups (seq &key (test 'equal))
  (let ((cont  (make-hash-table :test test)))
    (loop for elm in seq
       unless (gethash elm cont)
       do (puthash elm elm cont)
       finally return (loop for i being the hash-values
                                in cont collect i))))

With all 3 functions byte-compiled, using these calls:

(hash-remove-dups    B :test 'equal)
(list-remove-dups    B) ; uses `equal'
(remove-duplicates   B :test 'equal)

And with this list B (initialized anew each time):

(let ((seq (loop for i from 1 to 10000
             collect
             (format "%s" (random most-positive-fixnum)))))
  (append seq seq))

With B 10 times smaller (1000):

hash-remove-dups     1           0.0           0.0
list-remove-dups     1           0.047         0.047
remove-duplicates    1           1.172         1.172

With B 10 times bigger (100000), the difference between hash and classic list is
even greater (fuggedabowt cl-seq's `remove-duplicates' in this case):

hash-remove-dups     1           0.359         0.359
list-remove-dups     1           1209.578      1209.578
remove-duplicates    la-la...






  reply	other threads:[~2011-12-28 16:56 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-12-28  7:55 bug#10386: CODE wishlist: ínclude delete-duplicates.el Jari Aalto
2011-12-28  9:09 ` Thierry Volpiatto
2011-12-28 16:56   ` Drew Adams [this message]
2012-01-11  8:47 ` Glenn Morris

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

  List information: https://www.gnu.org/software/emacs/

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

  git send-email \
    --in-reply-to=97F263C286E14A3A9D8656356D98ECD1@us.oracle.com \
    --to=drew.adams@oracle.com \
    --cc=10386@debbugs.gnu.org \
    --cc=jari.aalto@cante.net \
    --cc=thierry.volpiatto@gmail.com \
    /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 public inbox

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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).