From: Mauro Carvalho Chehab <mchehab@kernel.org>
To: Jonathan Corbet <corbet@lwn.net>
Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
Akira Yokosawa <akiyks@gmail.com>
Subject: Re: [PATCH] docs: Raise the minimum Sphinx requirement to 2.4.4
Date: Tue, 28 Nov 2023 16:56:45 +0100 [thread overview]
Message-ID: <20231128165645.2dbe416c@coco.lan> (raw)
In-Reply-To: <877cm2uegr.fsf@meer.lwn.net>
Em Tue, 28 Nov 2023 07:42:12 -0700
Jonathan Corbet <corbet@lwn.net> escreveu:
> Mauro Carvalho Chehab <mchehab@kernel.org> writes:
>
> > Em Mon, 27 Nov 2023 16:31:39 -0700
> > Jonathan Corbet <corbet@lwn.net> escreveu:
>
> >> diff --git a/scripts/sphinx-pre-install b/scripts/sphinx-pre-install
> >> index 1fb88fdceec3..db75b1b86086 100755
> >> --- a/scripts/sphinx-pre-install
> >> +++ b/scripts/sphinx-pre-install
> >> @@ -32,8 +32,7 @@ my $python_cmd = "";
> >> my $activate_cmd;
> >> my $min_version;
> >> my $cur_version;
> >> -my $rec_version = "1.7.9"; # PDF won't build here
> >> -my $min_pdf_version = "2.4.4"; # Min version where pdf builds
> >> +my $rec_version = "3.0";
> >
> > Please don't. 3.0 version has a broken C domain, not properly supported.
> > The fixes arrived only starting at 3.1 (I guess some went to 3.2, but
> > 3.1 is usable, as far as I remember).
>
> So you're asking for 3.1 or 3.2 instead of 3.0?
Yes.
>
> Honestly, I just picked 3.0 out of the air in the hopes of eventually
> deprecating 2.x. Not lots of thought has gone into that
> number...perhaps we should recommend higher yet?
Well, we could recommend a higher version, but I can't see much
differences between 3.2 and the latest version: for what we use,
both will work on a similar way. Ok, layout may be different, there
were some improvements on PDF output, etc. but they will all produce
a decent documentation.
Yet, while most C domain bugs introduced on 3.0 were solved in 3.1
and 3.2, there's one still pending issue[1].
Once C domain finally gets rid from this long term bug that having:
.. c:struct:: foo
.. c:function:: void foo(void)
Produce warnings that "foo" id duplicated, then we'll have, IMO,
our next recommended version :-)
While Sphinx developers don't fix such bug, it doesn't really matter
what version user will pick, so I would just pick the fastest one
as a recommendation, starting from 3.1 or 3.2 as our currently
recommended version.
[1] https://github.com/sphinx-doc/sphinx/pull/8313
While I didn't make any benchmarks, I remember people reported
poor performance with newer versions, so, without thinking to
much, 3.1 or 3.2 seems a good candidate for the recommended
version.
Regards,
Mauro
next prev parent reply other threads:[~2023-11-28 15:56 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-27 23:31 [PATCH] docs: Raise the minimum Sphinx requirement to 2.4.4 Jonathan Corbet
2023-11-28 1:30 ` Mauro Carvalho Chehab
2023-11-28 14:42 ` Jonathan Corbet
2023-11-28 15:56 ` Mauro Carvalho Chehab [this message]
2023-11-30 7:33 ` Akira Yokosawa
2023-11-30 8:27 ` Mauro Carvalho Chehab
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20231128165645.2dbe416c@coco.lan \
--to=mchehab@kernel.org \
--cc=akiyks@gmail.com \
--cc=corbet@lwn.net \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).