Re: "WARNING: Duplicate C declaration" from recent Sphinx (was Re: [PATCH] docs: sphinx/requirements: Limit jinja2<3.1)

From: Akira Yokosawa <akiyks@gmail.com>
To: Mauro Carvalho Chehab <mchehab@kernel.org>
Cc: linux-doc@vger.kernel.org, Jonathan Corbet <corbet@lwn.net>,
	Akira Yokosawa <akiyks@gmail.com>
Subject: Re: "WARNING: Duplicate C declaration" from recent Sphinx (was Re: [PATCH] docs: sphinx/requirements: Limit jinja2<3.1)
Date: Sun, 22 May 2022 18:28:01 +0900	[thread overview]
Message-ID: <57a0cd2a-b4fb-2063-d3a7-09a6e292b2c7@gmail.com> (raw)
In-Reply-To: <20220522070707.313e6558@coco.lan>

On Sun, 22 May 2022 07:07:07 +0200,
Mauro Carvalho Chehab wrote:
> Em Sun, 22 May 2022 09:57:45 +0900
> Akira Yokosawa <akiyks@gmail.com> escreveu:
> 
>> On Sat, 21 May 2022 11:46:29 +0200,
>> Mauro Carvalho Chehab wrote:
>> [...]
>>>
>>> Sphinx 4.0 and above: it is now possible to add a namespace. This allows
>>> fixing things like read() system calls that have different meanings on
>>> different subsystems.
>>>
>>> On other words, only with Sphinx 4.0 and above, the cross-references
>>> for C domain symbols should all be OK.  
>>
>> So, I noticed there is a PR at https://github.com/sphinx-doc/sphinx/pull/8313
>> which is still open.
> 
> Are you sure? I always believed that this (or a variant of it) got 
> merged on 4.0.

Hmm, I'm afraid I don't think it has ever did.

In the discussion of the PR, you commented on 15 Oct 2020 at
https://github.com/sphinx-doc/sphinx/pull/8313#issuecomment-708977457:

> On a quick test, the remaining warnings I was getting with the Linux
> Kernel docs with Sphinx 3.x have gone after this patch set.
> 
> Thanks!
>
> I'm looking forward to see those patches applied to 3.2.2 ;-)

So you might have thought that the issue was already resolved.

> 
>>
>> This PR is supposed to resolve "WARNING: Duplicate C declaration"
>> due to struct and function with the same name, isn't it?
>>
>> Are you sure the issue is resolved in Sphinx 4.0 and later?
> 
> You need to ping Sphinx C domain maintainer to be sure. This was
> the author of the PR by the time I looked into it, but I'm not
> tracking Sphinx development, so things might have changed.

Later on in the discussion, @jakobandersen reported the status of
the PR on 2 Apr 2021 at:
https://github.com/sphinx-doc/sphinx/pull/8313#issuecomment-812408069.

If I understand the complex comment correctly, there is another PR at
https://github.com/sphinx-doc/sphinx/pull/8929 as a prerequisite
to PR 8313.

PR 8929 now belongs to the 5.x milestone (missed 5.0.0). PR 8313 has never
had a milestone assigned.  So it looks to me like things are progressing
very slowly and we need to wait another release or two at least.

Of course, all of this is just my interpretation without knowing any
of Sphinx development and might be wrong in many aspects.

        Thanks, Akira

> 
> Thanks,
> Mauro