Follow-up Comment #2, bug #66410 (group groff): Hi,
[comment #1 comment #1:]
> [comment #0 original submission:]
>
> In my opinion, the `Lb` _mdoc_ feature is not properly orthogonalized.
Thanks for the explanation. I suspect you are right!
>> Could this warning perhaps be restricted to use of the .Lb macro in sections
>> where it would be expected to be expanded with system library details?
>
> Hmm. How do you suggest that the package is going to know that a section--I
> assume you mean a `Sh` section--is going to mention only system libraries?
> Does the man page author need to tell it somehow, perhaps via a macro
> argument?
That is a good question. I inferred from groff_mdoc(7)
"In a section titled “Library”, ‘Lb’ causes a break before and after
its arguments."
and the external mandoc_mdoc(7) page that it were normal for special treatment
of at least this macro between sections.
However, that does seem fragile. I was looking for a compromise with the
intent of the warning but it just seems like a misguided warning to have at
all to me because mandoc_mdoc(7) implies a "custom library" without definition
is a legitimate use case:
"The name parameter may be a system library, such as z or pam, in which case
a small library description is printed next to the linker invocation; or a
custom library, in which case the library name is printed in quotes."
But that does bring us on to the next question beyond the warning, namely,
would the result be reasonable if there _were_ a definition but the man page
author didn't expect it to be present. If it was expanded out in a section
other than 'synopsis' it probably would be undesirable as it would break the
flow of text.
What do you think the best next step would be? One option is not to use this
macro!
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?66410>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/
signature.asc
Description: PGP signature
