From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marius Bakke Subject: bug#37850: Glib documentation is missing Date: Sun, 10 Nov 2019 00:20:14 +0100 Message-ID: <8736ewmyc1.fsf@devup.no> References: <87pniq9y2e.fsf@ambrevar.xyz> <8736fjwcd2.fsf@devup.no> <87wocvwbmy.fsf@ambrevar.xyz> <87pnikldtt.fsf@gnu.org> <87v9sbj2ru.fsf@ambrevar.xyz> <87h83ret3a.fsf@ambrevar.xyz> <87k18gncbn.fsf@ambrevar.xyz> <8736f4ok77.fsf@devup.no> <87ftj4m0s0.fsf@ambrevar.xyz> Mime-Version: 1.0 Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha512; protocol="application/pgp-signature" Return-path: Received: from eggs.gnu.org ([2001:470:142:3::10]:38534) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1iTa2S-0003IO-0e for bug-guix@gnu.org; Sat, 09 Nov 2019 18:21:05 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1iTa2Q-00051D-HS for bug-guix@gnu.org; Sat, 09 Nov 2019 18:21:03 -0500 Received: from debbugs.gnu.org ([209.51.188.43]:41655) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1iTa2Q-000515-CE for bug-guix@gnu.org; Sat, 09 Nov 2019 18:21:02 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1iTa2Q-0001NZ-6d for bug-guix@gnu.org; Sat, 09 Nov 2019 18:21:02 -0500 Sender: "Debbugs-submit" Resent-Message-ID: In-Reply-To: <87ftj4m0s0.fsf@ambrevar.xyz> List-Id: Bug reports for GNU Guix List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-guix-bounces+gcggb-bug-guix=m.gmane.org@gnu.org Sender: "bug-Guix" To: Pierre Neidhardt , Ludovic =?UTF-8?Q?Court=C3=A8s?= Cc: 37850@debbugs.gnu.org --=-=-= Content-Type: text/plain Pierre Neidhardt writes: > Marius Bakke writes: > >> Pierre Neidhardt writes: >> >>> From 893613a3b99c20688cc331d2926dbee28cc143d7 Mon Sep 17 00:00:00 2001 >>> From: Pierre Neidhardt >>> Date: Sun, 3 Nov 2019 17:36:17 +0100 >>> Subject: [PATCH] gnu: Add glib-minimal and build glib doc. >>> >>> glib documentation must be built with gtk-doc which in turn depends on glib, >>> so we need to define glib-minimal which does not depend on gtk-doc. >>> >>> * gnu/packages/glib.scm (glib-minimal): New variable. >>> (glib)[source]: Don't use `name'. >>> * gnu/packages/avahi.scm: Use glib-minimal when necessary. >>> * gnu/packages/cups.scm: Use glib-minimal when necessary. >>> * gnu/packages/gnome.scm: Use glib-minimal when necessary. >>> * gnu/packages/graphviz.scm: Use glib-minimal when necessary. >>> * gnu/packages/gtk.scm: Use glib-minimal when necessary. >>> * gnu/packages/inkscape.scm: Use glib-minimal when necessary. >>> * gnu/packages/pdf.scm: Use glib-minimal when necessary. >> >> Please mention all changed variables and inputs here, as we always do. > > Sure, I didn't because I wasn't sure it was a good idea either :p > >> That said, I'm not certain this is a good solution. Why do some >> packages use glib-minimal and others not? What does "necessary" mean in >> this context? >> >> What about 'hiding' the normal glib package, and expose a >> 'glib-with-documentation' variant to end users, similar to how the >> 'cmake' package works? > > cmake-minimal builds the doc already, I don't see a > cmake-with-documentation. Did you mean something else? > I see a couple packages with the "-documentation" prefix, so we could > use "glib-documentation". I meant this part: (define-public cmake (package (inherit cmake-minimal) (name "cmake") (arguments (substitute-keyword-arguments (package-arguments cmake-minimal) ((#:configure-flags configure-flags ''()) `(append ,configure-flags ;; Extra configure flags used to generate the documentation. '("--sphinx-info" "--sphinx-man" "--sphinx-html"))) Do you think it would make sense to do something similar with GLib, and "hiding" the normal variant? > In a previous email, I explained that I wanted to put the documentation > in the "doc" output of glib for 2 reasons: > > - For consistency with the other packages from the GTK family. > - To restore the missing "gtk:doc" output. If I understand correctly, > we don't have a provision to deprecate outputs, only packages. I > think this is a bug. --=-=-= Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQEzBAEBCgAdFiEEu7At3yzq9qgNHeZDoqBt8qM6VPoFAl3HSa4ACgkQoqBt8qM6 VPql9wf7BvCPou3hu5kWaNT+h4w0u5N6FPAKsp0Yq2eildSbBkTVlDepXMy6SVD4 IYp8RaINYlR3o+BaefZjrfFWSpHzSyksVR+6Qa0s8EmMr9VDPIp3qcFMp0RrynIl erXPoo9pibhdyPcKXZA5WGGpewPTuI9PKA2N330enzPkphvbZoBFwVwsAw7Eifvz oDJPGRjuzajID+HFknWWoiQvDTWoPCts3lyHNESEPP9+DS5GBItNfiNyksxvX0p3 /FEMELNhbtO6r1vNxLAaHCdM6E0AvU7l7+nSj+S0pbBMrkqCmKTO2w3JwbnAJlHb 0dC4owNmj+eUFtV5iUVIC53+fd0mrg== =AE57 -----END PGP SIGNATURE----- --=-=-=--