From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:470:142:3::10]:44590) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1j573H-000122-WA for guix-patches@gnu.org; Fri, 21 Feb 2020 07:05:05 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1j573G-0005Ea-50 for guix-patches@gnu.org; Fri, 21 Feb 2020 07:05:03 -0500 Received: from debbugs.gnu.org ([209.51.188.43]:40518) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1j573F-0005EW-WD for guix-patches@gnu.org; Fri, 21 Feb 2020 07:05:02 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1j573F-0002X7-Pw for guix-patches@gnu.org; Fri, 21 Feb 2020 07:05:01 -0500 Subject: [bug#39599] [PATCH 1/4] build-system: Add copy-build-system. Resent-Message-ID: From: Pierre Neidhardt In-Reply-To: <87lfowf9nq.fsf@nicolasgoaziou.fr> References: <9a61841f-e79b-469f-af02-4a739cb0c5f2@www.fastmail.com> <87ftf57itl.fsf@ambrevar.xyz> <87pne95krx.fsf@ambrevar.xyz> <2AB2B5D1-42B9-4DDF-A712-385BACF8BA60@lepiller.eu> <87sgj5t9zs.fsf@ambrevar.xyz> <875zg05kqw.fsf@ambrevar.xyz> <87pne8fcg9.fsf@nicolasgoaziou.fr> <87tv3k42r6.fsf@ambrevar.xyz> <87lfowf9nq.fsf@nicolasgoaziou.fr> Date: Fri, 21 Feb 2020 13:04:15 +0100 Message-ID: <87o8ts405c.fsf@ambrevar.xyz> MIME-Version: 1.0 Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha256; protocol="application/pgp-signature" List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-patches-bounces+kyle=kyleam.com@gnu.org Sender: "Guix-patches" To: Nicolas Goaziou Cc: Julien Lepiller , Alex Griffin , 39599@debbugs.gnu.org --=-=-= Content-Type: text/plain Content-Transfer-Encoding: quoted-printable Nicolas Goaziou writes: > Pierre Neidhardt writes: > >> Oh, yes, I see that. I thought it would help with readability. How are >> we supposed to visualize nested @itemize at the moment? > > I don't think there's a clear answer, but, IMO, for readability sake, we > should not (ab)use nested lists in a manual.=20 > > There are three levels of such lists here. I think this is not > necessary. For example > > --8<---------------cut here---------------start------------->8--- > @item When @var{source} matches a file or directory without trailing slas= h, install it to @var{target}. > @itemize > @item If @var{target} has a trailing slash, install @var{source} basena= me beneath @var{target}. > @item Otherwise install @var{source} as @var{target}. > @end itemize > --8<---------------cut here---------------end--------------->8--- > > could be written as, e.g., > > --8<---------------cut here---------------start------------->8--- > @item > When @var{source} matches a file or directory without a trailing slash, > install it to @var{target}. More accurately, if @var{target} ends with > a slash, install @var{source} basename beneath @var{target} directory. > Otherwise install @var{source} as @var{target}. > --8<---------------cut here---------------end--------------->8--- Fair enough. The idea behind the items was to structure it like a spec that would easily translate to code then. (I wrote these specs before writing the code.) > Similarly, instead of discussing about #:include and al. in a nested > list, this could happen in a subsequent paragraph, once "source" and > "target" are clarified, i.e., after "In all cases, the paths (BTW, > shouldn't it be "file names"?) I don't know. In my opinion, "file names" is often interpreted as "base names". Here I mean that the full subpath of the file is preserved. May I should use "subpath" then. > As a side note, are you sure about: "With @code{#:include}, install all > the files which (I would use "whose" here, but I'm not a native > speaker) https://en.wikipedia.org/wiki/Inanimate_whose Actually, should be "which the" or "the files the path of which". But all this is very pedantic :) > path suffix (isn't it "basename" or, possibly better, "base name" > instead?) No, it really is "path suffix" here because it matches against the parent directories, e.g. "foo/bar" is a valid suffix. > exactly matches one of the elements in the given list"? Do you > really mean that a file name matching two regexps is _not_ going to be > included? Indeed, that's a mistake! Thanks!. =2D-=20 Pierre Neidhardt https://ambrevar.xyz/ --=-=-= Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQEzBAEBCAAdFiEEUPM+LlsMPZAEJKvom9z0l6S7zH8FAl5Pxz8ACgkQm9z0l6S7 zH/0bgf+K8N7yGqE9u0i2HFqt0euxZegLyYywuisLGPsBcYpVwDvviH7NYwGEqFY ieKixuV1AIe3L+JgI3hoOh6OEJUwIJRAPKoezS/Nn8SNpndeZbVVfwH5v+OSyuj4 bHsEtJOp9q7lOCGKWT9x+eNpe0Q/neqSvZf6dIzhA6OoksbGwxb6BEqB2K48J9Yv YQNErfkJuP6WPb/dy2pNP3OPd0At/2G0diwzVRfvZUkFuCZe3lNh1/VtnB+wAlIq DoiLBHf+a9SB3xAM7ydMiXo54cgpr7cBBli/B7+Dd1zMMggs6PeCy9yJa1udTaHK Rd+xvbhXxoHGrSO6cED9plW9eY+j+Q== =nLYz -----END PGP SIGNATURE----- --=-=-=--