GNU bug report logs - #57079
29.0.50; Performance of seq-uniq is not very good

Previous Next

Full log

View this message in rfc822 format

From: Stefan Kangas <stefan <at> marxist.se>
To: 57079 <at> debbugs.gnu.org
Subject: bug#57079: 29.0.50; Performance of seq-uniq is not very good
Date: Tue, 9 Aug 2022 09:11:07 -0700

Severity: minor

`seq-uniq' is not very performant compared to `-uniq' (from dash.el) and
even slower compared to the recently removed `gnus-delete-duplicates':

    (benchmark-run 10000 (seq-uniq '(a c b c c a d)))
    => (0.355001481 1 0.2518970439999748)

    (benchmark-run 10000 (-uniq '(a c b c c a d)))
    => (0.006599549 0 0.0)

    (benchmark-run 10000 (gnus-delete-duplicates '(a c b c c a d)))
    => (0.0034537929999999997 0 0.0)

Could we improve the performance of `seq-uniq' for lists?


PS. `gnus-delete-duplicates' was recently removed but looks like this:

    (defun gnus-delete-duplicates (list)
      "Remove duplicate entries from LIST."
      (let ((result nil))
        (while list
          (unless (member (car list) result)
    	(push (car list) result))
          (pop list))
        (nreverse result)))

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.