From mboxrd@z Thu Jan 1 00:00:00 1970 From: Kyle Meyer Subject: Re: About org-sort -> org-sort-list with custom sort function Date: Sun, 07 May 2017 10:20:16 -0400 Message-ID: <87shkg695r.fsf@kyleam.com> References: <87inlhhgvf.fsf@gmail.com> <87vapd5qb0.fsf@kyleam.com> <87bmr57zqp.fsf@nicolasgoaziou.fr> Mime-Version: 1.0 Content-Type: text/plain Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:51918) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1d7N2s-00028t-Hr for emacs-orgmode@gnu.org; Sun, 07 May 2017 10:20:23 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1d7N2p-00064r-Cv for emacs-orgmode@gnu.org; Sun, 07 May 2017 10:20:22 -0400 Received: from pb-smtp2.pobox.com ([64.147.108.71]:57726 helo=sasl.smtp.pobox.com) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1d7N2p-00064c-7k for emacs-orgmode@gnu.org; Sun, 07 May 2017 10:20:19 -0400 In-Reply-To: <87bmr57zqp.fsf@nicolasgoaziou.fr> List-Id: "General discussions about Org-mode." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-orgmode-bounces+geo-emacs-orgmode=m.gmane.org@gnu.org Sender: "Emacs-orgmode" To: Nicolas Goaziou Cc: Zhitao Gong , emacs-orgmode@gnu.org Nicolas Goaziou writes: > Kyle Meyer writes: > >> I'm a bit confused about why org-call-with-arg is necessary because I >> think call-interactively already propagates the current prefix argument, >> but perhaps I'm missing some subtlety here. Either way ... > > I thought the same. I think we can replace `org-call-with-arg' with > `call-interactively' in master. OK, I'll have a closer look at which org-call-with-arg calls are unnecessary. > Actually, there's some confusion in `org-sort-list' about how > getkey-func and compare-func are handled. For example, the question > asked to bind GETKEY-FUNC is "Sort using function: ", which should > really be asked for COMPARE-FUNC. > > IOW, we need to move this question to COMPARE-FUNC and ask a new one for > GETKEY-FUNC. True, at the very least, the prompt is confusingly phrased, but it's probably an indication that the intended interface wasn't fully implemented. >> And I think it's OK to not expose compare-func to the interactive >> caller. In cases where sort-subr's default behavior won't do and a user >> wants to supply a value for compare-func, they can create their own >> command that wraps a non-interactive org-sort-list call. > > I disagree. getkey-func and compare-func work hand in hand. You cannot > let users provide one but not the other without crippling functionality. The interactive caller is certainly losing functionality, but my guess was that sort-subr's default behavior is sufficient for handling the keys returned by most custom key functions. Probably a bad guess. -- Kyle