GNU bug report logs - #25995
26.0.50; Mismatch between documented and actual behaviour of icomplete

Previous Next

Full log

Message #29 received at 25995 <at> debbugs.gnu.org (full text, mbox):

From: Dmitry Gutov <dgutov <at> yandex.ru>
To: npostavs <at> users.sourceforge.net
Cc: 25995 <at> debbugs.gnu.org, Alexis <flexibeast <at> gmail.com>
Subject: Re: bug#25995: 26.0.50; Mismatch between documented and actual
 behaviour of icomplete
Date: Thu, 22 Jun 2017 01:17:46 +0300

On 6/21/17 5:50 AM, npostavs <at> users.sourceforge.net wrote:

> Ah, you're measuring interpreted code which will penalize the open-coded
> loop more than the `cl-delete-if' call which has the loop tucked away
> (and compiled).

Indeed, thanks. Since eager macro-expansion landed, I don't recall 
seeing more than a 2x difference between byte-compiled and interpreted 
code, but here it is. :)

>> In any case, nreverse barely affects the runtime, so please go ahead
>> and push the patch. Thanks!
> 
> Pushed to master [1: 1ed2086a03], also fixes the same issue for
> `completion-pcm--all-completions' (Bug#24676).

Indeed!

Previous Next

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