linux-doc.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
To: Akira Yokosawa <akiyks@gmail.com>
Cc: Linux Doc Mailing List <linux-doc@vger.kernel.org>,
	Jonathan Corbet <corbet@lwn.net>,
	Breno Leitao <leitao@debian.org>,
	"David S. Miller" <davem@davemloft.net>,
	Donald Hunter <donald.hunter@gmail.com>,
	Eric Dumazet <edumazet@google.com>,
	Ignacio Encinas Rubio <ignacio@iencinas.com>,
	Jan Stancek <jstancek@redhat.com>, Marco Elver <elver@google.com>,
	Paolo Abeni <pabeni@redhat.com>,
	Ruben Wauters <rubenru09@aol.com>,
	Shuah Khan <skhan@linuxfoundation.org>,
	joel@joelfernandes.org, linux-kernel-mentees@lists.linux.dev,
	linux-kernel@vger.kernel.org, lkmm@lists.linux.dev,
	netdev@vger.kernel.org, peterz@infradead.org,
	stern@rowland.harvard.edu
Subject: Re: [PATCH v5 01/15] docs: conf.py: properly handle include and exclude patterns
Date: Wed, 18 Jun 2025 13:44:58 +0200	[thread overview]
Message-ID: <20250618134458.10ee8412@foz.lan> (raw)
In-Reply-To: <1adba2c6-e4c3-4da2-874e-a304a1fdfd25@gmail.com>

Hi Akira,

Em Wed, 18 Jun 2025 11:42:14 +0900
Akira Yokosawa <akiyks@gmail.com> escreveu:

> Hi Mauro,
> 
> A comment on compatibility with earlier Sphinx.
> 
> On Tue, 17 Jun 2025 10:01:58 +0200, Mauro Carvalho Chehab wrote:
> > When one does:
> > 	make SPHINXDIRS="foo" htmldocs
> > 
> > All patterns would be relative to Documentation/foo, which
> > causes the include/exclude patterns like:
> > 
> > 	include_patterns = [
> > 		...
> > 		f'foo/*.{ext}',
> > 	]
> > 
> > to break. This is not what it is expected. Address it by
> > adding a logic to dynamically adjust the pattern when
> > SPHINXDIRS is used.
> > 
> > That allows adding parsers for other file types.
> > 
> > Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
> > ---
> >  Documentation/conf.py | 52 +++++++++++++++++++++++++++++++++++++++----
> >  1 file changed, 48 insertions(+), 4 deletions(-)
> > 
> > diff --git a/Documentation/conf.py b/Documentation/conf.py
> > index 12de52a2b17e..e887c1b786a4 100644
> > --- a/Documentation/conf.py
> > +++ b/Documentation/conf.py
> > @@ -17,6 +17,54 @@ import os
> >  import sphinx
> >  import shutil
> >  
> > +# Location of Documentation/ directory
> > +doctree = os.path.abspath('.')
> > +
> > +# List of patterns that don't contain directory names, in glob format.
> > +include_patterns = ['**.rst']
> > +exclude_patterns = []
> > +  
> 
> Where "exclude_patterns" has been with us ever since Sphinx 1.0,
> "include_patterns" was added fairly recently in Sphinx 5.1 [1].
> 
> [1]: https://www.sphinx-doc.org/en/master/usage/configuration.html#confval-include_patterns
> 
> So, this breaks earlier Sphinx versions.

Heh, testing against old versions is harder with python 3.13 (Fedora
42 default), as one library used by older Sphinx versions were dropped.

I found a way to make it backward compatible up to 3.4.3, with a
backward-compatible logic at conf.py. I'll send the new version in a few.

> Also, after applying all of v5 on top of docs-next, I see these new
> warnings with Sphinx 7.2.6 (of Ubuntu 24.04):
> 
> /<srcdir>/Documentation/output/ca.h.rst: WARNING: document isn't included in any toctree
> /<srcdir>/Documentation/output/cec.h.rst: WARNING: document isn't included in any toctree
> /<srcdir>/Documentation/output/dmx.h.rst: WARNING: document isn't included in any toctree
> /<srcdir>/Documentation/output/frontend.h.rst: WARNING: document isn't included in any toctree
> /<srcdir>/Documentation/output/lirc.h.rst: WARNING: document isn't included in any toctree
> /<srcdir>/Documentation/output/media.h.rst: WARNING: document isn't included in any toctree
> /<srcdir>/Documentation/output/net.h.rst: WARNING: document isn't included in any toctree
> /<srcdir>/Documentation/output/videodev2.h.rst: WARNING: document isn't included in any toctree

We should likely use a Sphinx extension for those as well. Building those
are also made via some Makefile tricks that predates the time we start
adding our own extensions at the tree.

> Sphinx 7.3.7 and later are free of them.  I have no idea which change in
> Sphinx 7.3 got rid of them.
> 
> Now that the parallel build performance regression has be resolved in
> Sphinx 7.4, I don't think there is much demand for keeping Sphinx versions
> compatible.
> These build errors and extra warnings would encourage people to upgrade
> there Sphinx.  So I'm not going to nack this.
> 
> Of course, getting rid of above warnings with < Sphinx 7.3 would be ideal.

I'm all for using newer versions, but we need to check what LTS distros
are using those days.

On my machine, with -jauto, 3.4.3 is taking 11 minutes to build, which
is twice the time of 8.2.3. IMO, this is a very good reason for people
stop using legacy versions when possible :-)

Regards,
Mauro

  reply	other threads:[~2025-06-18 11:45 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-17  8:01 [PATCH v5 00/15] Don't generate netlink .rst files inside $(srctree) Mauro Carvalho Chehab
2025-06-17  8:01 ` [PATCH v5 01/15] docs: conf.py: properly handle include and exclude patterns Mauro Carvalho Chehab
2025-06-17 10:38   ` Donald Hunter
2025-06-18  6:59     ` Mauro Carvalho Chehab
2025-06-18  2:42   ` Akira Yokosawa
2025-06-18 11:44     ` Mauro Carvalho Chehab [this message]
2025-06-17  8:01 ` [PATCH v5 02/15] docs: Makefile: disable check rules on make cleandocs Mauro Carvalho Chehab
2025-06-17  9:53   ` Breno Leitao
2025-06-17 10:38   ` Donald Hunter
2025-06-17  8:02 ` [PATCH v5 03/15] tools: ynl_gen_rst.py: create a top-level reference Mauro Carvalho Chehab
2025-06-17 10:40   ` Donald Hunter
2025-06-17  8:02 ` [PATCH v5 04/15] docs: netlink: netlink-raw.rst: use :ref: instead of :doc: Mauro Carvalho Chehab
2025-06-17  8:02 ` [PATCH v5 05/15] tools: ynl_gen_rst.py: make the index parser more generic Mauro Carvalho Chehab
2025-06-17 10:55   ` Donald Hunter
2025-06-17 11:59   ` Simon Horman
2025-06-18  6:57     ` Mauro Carvalho Chehab
2025-06-18 12:40       ` Simon Horman
2025-06-17  8:02 ` [PATCH v5 06/15] tools: ynl_gen_rst.py: Split library from command line tool Mauro Carvalho Chehab
2025-06-17 11:08   ` Donald Hunter
2025-06-17  8:02 ` [PATCH v5 07/15] scripts: lib: netlink_yml_parser.py: use classes Mauro Carvalho Chehab
2025-06-17  8:02 ` [PATCH v5 08/15] docs: netlink: index.rst: add a netlink index file Mauro Carvalho Chehab
2025-06-17 10:43   ` Donald Hunter
2025-06-17  8:02 ` [PATCH v5 09/15] tools: ynl_gen_rst.py: clanup coding style Mauro Carvalho Chehab
2025-06-17  9:49   ` Breno Leitao
2025-06-17  9:50   ` Breno Leitao
2025-06-17 11:12   ` Donald Hunter
2025-06-17  8:02 ` [PATCH v5 10/15] docs: sphinx: add a parser for yaml files for Netlink specs Mauro Carvalho Chehab
2025-06-17 12:35   ` Donald Hunter
2025-06-17 13:40     ` Mauro Carvalho Chehab
2025-06-17 16:00       ` Mauro Carvalho Chehab
2025-06-17 17:23         ` Donald Hunter
2025-06-18  8:21           ` Mauro Carvalho Chehab
2025-06-17  8:02 ` [PATCH v5 11/15] docs: use parser_yaml extension to handle " Mauro Carvalho Chehab
2025-06-17 12:45   ` Donald Hunter
2025-06-17  8:02 ` [PATCH v5 12/15] docs: uapi: netlink: update netlink specs link Mauro Carvalho Chehab
2025-06-17 12:46   ` Donald Hunter
2025-06-17  8:02 ` [PATCH v5 13/15] tools: ynl_gen_rst.py: drop support for generating index files Mauro Carvalho Chehab
2025-06-17  8:02 ` [PATCH v5 14/15] docs: netlink: remove obsolete .gitignore from unused directory Mauro Carvalho Chehab
2025-06-17 12:38   ` Donald Hunter
2025-06-17  8:02 ` [PATCH v5 15/15] MAINTAINERS: add netlink_yml_parser.py to linux-doc Mauro Carvalho Chehab
2025-06-17  9:50   ` Breno Leitao
2025-06-17 12:58 ` [PATCH v5 00/15] Don't generate netlink .rst files inside $(srctree) Donald Hunter

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=20250618134458.10ee8412@foz.lan \
    --to=mchehab+huawei@kernel.org \
    --cc=akiyks@gmail.com \
    --cc=corbet@lwn.net \
    --cc=davem@davemloft.net \
    --cc=donald.hunter@gmail.com \
    --cc=edumazet@google.com \
    --cc=elver@google.com \
    --cc=ignacio@iencinas.com \
    --cc=joel@joelfernandes.org \
    --cc=jstancek@redhat.com \
    --cc=leitao@debian.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel-mentees@lists.linux.dev \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lkmm@lists.linux.dev \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rubenru09@aol.com \
    --cc=skhan@linuxfoundation.org \
    --cc=stern@rowland.harvard.edu \
    /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).