The kernel-doc script uses the $type_constant2 variable to match
expressions used to find embedded type information. The current
implementation of $type_constant2 does not include the "*" unicode
character, which is used to highlight inline literals in the
documentation. This causes a Sphinx error when the inline literal
end-string is used in the documentation.

This commit follows the pattern of the commit
8aaf297a0dd6 ("docs: scripts: kernel-doc: accept bitwise negation like ~@var")
and takes inspiration from the following commit
69fc23efc7e5 ("kernel-doc: Add unary operator * to $type_param_ref").

Thanks Akira, for your suggestions, I have made the required changes.
I am fairly new to the kernel community, so if I am making 
any mistakes while making patches and replying to mails,
please let me know, it will be very helpful.

Signed-off-by: Utkarsh Tripathi <utripathi2002@gmail.com>
Reviewed-by: Akira Yokosawa <akiyks@gmail.com>
Suggested-by: Akira Yokosawa <akiyks@gmail.com>
---
 scripts/kernel-doc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/kernel-doc b/scripts/kernel-doc
index cb1be22afc65..58129b1cf3f4 100755
--- a/scripts/kernel-doc
+++ b/scripts/kernel-doc
@@ -62,7 +62,7 @@ my $anon_struct_union = 0;
 
 # match expressions used to find embedded type information
 my $type_constant = '\b``([^\`]+)``\b';
-my $type_constant2 = '\%([-_\w]+)';
+my $type_constant2 = '\%([-_*\w]+)';
 my $type_func = '(\w+)\(\)';
 my $type_param = '\@(\w*((\.\w+)|(->\w+))*(\.\.\.)?)';
 my $type_param_ref = '([\!~\*]?)\@(\w*((\.\w+)|(->\w+))*(\.\.\.)?)';

base-commit: 4d2008430ce87061c9cefd4f83daf2d5bb323a96
-- 
2.34.1

Hi Utkarsh,

First of all, thank you for taking the time!

Besides Jon's comments on the summary and changelog, please find my
suggestion below.

On Wed,  1 May 2024 23:27:30 +0530, Utkarsh Tripathi wrote:
> The kernel-doc script uses the $type_constant2 variable to match
> expressions used to find embedded type information. The current
> implementation of $type_constant2 does not include the "*" unicode
> character, which is used to highlight inline literals in the
> documentation. This causes a Sphinx error when the inline literal
> end-string is used in the documentation.

I'm afraid your description of what is wrong is not clear enough ...
Let me talk using some examples.

Current kernel-doc (script) conversion to reST:

    %WQ_* -->  ``WQ_``*

Against which Sphinx complains:

    WARNING: Inline literal start-string without end-string.

, because ``* is not recognized as end-string (of inline literal).

With your change applied, conversion to reST becomes:

    %WQ_* --> ``WQ_*``

, and it is a proper inline literal.

Please update the changelog accordingly.

This is not urgent at all.  Please take your time and read through
Documentation/process/submitting-patches.rst (among others) before
submitting v3.

Feel free to add (in v3):

Reviewed-by: Akira Yokosawa <akiyks@gmail.com>

    Thanks, Akira

Utkarsh Tripathi <utripathi2002@gmail.com> writes:

> The kernel-doc script uses the $type_constant2 variable to match
> expressions used to find embedded type information. The current
> implementation of $type_constant2 does not include the "*" unicode
> character, which is used to highlight inline literals in the
> documentation. This causes a Sphinx error when the inline literal
> end-string is used in the documentation.

So I need to look a bit further at the actual change, but I do have a
couple of comments on the patch itself.  First, the text above is a
reasonable description of the problem, as a changelog should have.  That
said, the subject line could be a bit shorter and to the point.

This text below:

> This commit follows the pattern of the commit
> 8aaf297a0dd6 ("docs: scripts: kernel-doc: accept bitwise negation like ~@var")
> and takes inspiration from the following commit
> 69fc23efc7e5 ("kernel-doc: Add unary operator * to $type_param_ref").
>
> Thanks Akira, for your suggestions, I have made the required changes.
> I am fairly new to the kernel community, so if I am making 
> any mistakes while making patches and replying to mails,
> please let me know, it will be very helpful.

...doesn't belong in the changelog.  If you put comments like this below
the "---" line, then the maintainer won't have to edit them out when
applying the patch.

> Signed-off-by: Utkarsh Tripathi <utripathi2002@gmail.com>
> Reviewed-by: Akira Yokosawa <akiyks@gmail.com>
> Suggested-by: Akira Yokosawa <akiyks@gmail.com>

Did Akira offer you that Reviewed-by tag?  I haven't seen it (which
doesn't mean it didn't happen).  If it was not explicitly given to you,
though, you cannot put it here.

Thanks,

jon