Re: Please backport commit 00d95fcc4dee ("docs: kdoc: handle the obsolescensce of docutils.ErrorString()") to v6.17.y

[Jonathan Corbet <corbet@lwn.net>]

Salvatore Bonaccorso <carnil@debian.org> writes:

> Hi,
>
> On Mon, Oct 27, 2025 at 10:06:33AM -0600, Jonathan Corbet wrote:
>> Salvatore Bonaccorso <carnil@debian.org> writes:
>> 
>> > Hi,
>> >
>> > On Sun, Oct 26, 2025 at 08:36:00AM +0100, Andreas Radke wrote:
>> >> For kernel 6.12 there's just one more place required to add the fix:
>> >> 
>> >> --- a/Documentation/sphinx/kernel_abi.py        2025-10-23 16:20:48.000000000 +0200
>> >> +++ b/Documentation/sphinx/kernel_abi.py.new    2025-10-26 08:08:33.168985951 +0100
>> >> @@ -42,9 +42,11 @@
>> >>  from docutils import nodes, statemachine
>> >>  from docutils.statemachine import ViewList
>> >>  from docutils.parsers.rst import directives, Directive
>> >> -from docutils.utils.error_reporting import ErrorString
>> >>  from sphinx.util.docutils import switch_source_input
>> >> 
>> >> +def ErrorString(exc):  # Shamelessly stolen from docutils
>> >> +    return f'{exc.__class__.__name}: {exc}'
>> >> +
>> >>  __version__  = '1.0'
>> >> 
>> >>  def setup(app):
>> >
>> > Yes this is why I asked Jonathan, how to handle backports to older
>> > series, if it is wanted to pick specifically as well faccc0ec64e1
>> > ("docs: sphinx/kernel_abi: adjust coding style") or a partial backport
>> > of it, or do a 6.12.y backport of 00d95fcc4dee with additional
>> > changes (like you pointed out).
>> >
>> > I'm just not sure what is preferred here. 
>> 
>> I'm not sure it matters that much...the additional change suggested by
>> Andreas seems fine.  It's just a backport, and it shouldn't break
>> anything, so doesn't seem worth a lot of worry.
>
> Okay here is a respective backported change for the 6.12.y series as
> well.
>
> Does that look good for you?

I haven't actually tried it, but it looks OK to me.

jon