Re: [PATCH 09/12] docs: kdoc: straighten up dump_declaration()

[Mauro Carvalho Chehab <mchehab+huawei@kernel.org>]

Em Thu, 10 Jul 2025 07:27:07 -0600
Jonathan Corbet <corbet@lwn.net> escreveu:

> Mauro Carvalho Chehab <mchehab+huawei@kernel.org> writes:
> 
> > Em Wed,  2 Jul 2025 16:35:21 -0600
> > Jonathan Corbet <corbet@lwn.net> escreveu:
> >  
> >> Get rid of the excess "return" statements in dump_declaration(), along with
> >> a line of never-executed dead code.
> >> 
> >> Signed-off-by: Jonathan Corbet <corbet@lwn.net>
> >> ---
> >>  scripts/lib/kdoc/kdoc_parser.py | 15 +++++----------
> >>  1 file changed, 5 insertions(+), 10 deletions(-)
> >> 
> >> diff --git a/scripts/lib/kdoc/kdoc_parser.py b/scripts/lib/kdoc/kdoc_parser.py
> >> index 6e35e508608b..7191fa94e17a 100644
> >> --- a/scripts/lib/kdoc/kdoc_parser.py
> >> +++ b/scripts/lib/kdoc/kdoc_parser.py
> >> @@ -878,18 +878,13 @@ class KernelDoc:
> >>  
> >>          if self.entry.decl_type == "enum":
> >>              self.dump_enum(ln, prototype)
> >> -            return
> >> -
> >> -        if self.entry.decl_type == "typedef":
> >> +        elif self.entry.decl_type == "typedef":
> >>              self.dump_typedef(ln, prototype)
> >> -            return
> >> -
> >> -        if self.entry.decl_type in ["union", "struct"]:
> >> +        elif self.entry.decl_type in ["union", "struct"]:
> >>              self.dump_struct(ln, prototype)
> >> -            return
> >> -  
> >
> > The above LGTM.
> >  
> >> -        self.output_declaration(self.entry.decl_type, prototype,
> >> -                                entry=self.entry)
> >> +        else:
> >> +            # This would be a bug
> >> +            self.emit_message(ln, f'Unknown declaration type: {self.entry.decl_type}')  
> >
> > Hmm... Are you sure about that? If I'm not mistaken, this was used for
> > other types of arguments, like DOC: tags.  
> 
> DOC tags are handled in a different path entirely.  I did ensure that
> the code in question was never executed ... but then left the message in
> place just in case.

OK.


If the output didn't change neither for ReST nor for man, that's fine
for me. Besides being a port from Perl, I'm almost sure I hit this
code before during the conversion, but it it is now a dead code,
your approach is better ;-)

While I didn't test, I trust you. So feel free to add:

Reviewed-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>

Thanks,
Mauro