GNU bug report logs - #59328
29.0.50; `seq-keep' implementation only valid for lists

Previous Next

Full log

Message #8 received at 59328 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Michael Heerdegen <michael_heerdegen <at> web.de>
Cc: larsi <at> gnus.org, jonas <at> bernoul.li, 59328 <at> debbugs.gnu.org
Subject: Re: bug#59328: 29.0.50; `seq-keep' implementation only valid for lists
Date: Sat, 19 Nov 2022 15:12:10 +0200

> Cc: Lars Ingebrigtsen <larsi <at> gnus.org>, Jonas Bernoulli <jonas <at> bernoul.li>
> From: Michael Heerdegen <michael_heerdegen <at> web.de>
> Date: Thu, 17 Nov 2022 03:17:46 +0100
> 
> The current implementation of the (non-generic!) function `seq-keep':
> 
> #+begin_src emacs-lisp
> (defun seq-keep (function sequence)
>   (delq nil (seq-map function sequence)))
> ;; ^^^^
> #+end_src
> 
> obviously only works when `seq-map' returns a list.  This is the case
> for the default implementation of the generic function `seq-map' but not
> necessarily for other implementations of `seq-map'.
> 
> We need to filter out the `nil' elements with a way appropriate for any
> sequence type supported by "seq.el" (i.e. with a generic function
> defined in this lib), e.g.
> 
> #+begin_src emacs-lisp
> (defun seq-keep (function sequence)
>   (seq-filter #'identity (seq-map function sequence)))
> #+end_src

This makes sense to me, so please go ahead and install, preferably
with a test for non-list cases.

> BTW, is the name a good one?  Why "keep"?  It returns a sequence of
> potentially all completely different elements.  And is the function that
> useful and a good abstraction at all (I don't have thought about it
> too long...)?

FWIW, "keep" doesn't sound problematic to me.

Thanks.

Previous Next

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