Page MenuHomeFreeBSD

lockf.1: Minor polish
Needs ReviewPublic

Authored by ziaee on Wed, Jan 15, 8:15 PM.
Tags
None
Referenced Files
F109362017: D48470.id149342.diff
Tue, Feb 4, 1:56 AM
Unknown Object (File)
Sat, Feb 1, 8:36 PM
Unknown Object (File)
Sat, Feb 1, 8:35 AM
Unknown Object (File)
Fri, Jan 31, 5:57 AM
Unknown Object (File)
Mon, Jan 27, 4:05 PM
Unknown Object (File)
Sun, Jan 26, 8:15 AM
Unknown Object (File)
Sat, Jan 25, 7:24 PM
Unknown Object (File)
Sat, Jan 25, 6:36 PM
Subscribers

Details

Reviewers
carlavilla
mhorne
Group Reviewers
manpages
Summary

+ tag SPDX + align lists + fix one sentance per line warning
+ alphabetize options + add a line break to examples
+ organize exits, and provide error numbers

Reported by: Antranig Vartanian <antranigv@freebsd.am>
MFC after: 3 days

Diff Detail

Repository
rG FreeBSD src repository
Lint
Lint Skipped
Unit
Tests Skipped
Build Status
Buildable 61919
Build 58803: arc lint + arc unit

Event Timeline

ziaee requested review of this revision.Wed, Jan 15, 8:15 PM

Revert s/Dv/Sy/g, I did that to align with sysexits(3), however this was
a mistake as I think apropos Dv=EX_USAGE is an outstanding workflow. I
now believe that error messages should always be Dv.

Use Er to markup error constants according to mdoc(7), and use
Dv to markup the exit numbers to enable apropos Dv=64.

usr.bin/lockf/lockf.1
171

Okay, now I see where you were coming from with D48619...

It seems clear to me that apropos Er=EX_USAGE should lead to this page. But I am not sure that apropos Dv=64 needs to.

If the status code I am investigating comes from lockf(1), then I surely know which man page to check, no?

185–186

.Po/.Pc save some headaches here.

Honestly, IMHO what would be best is Er=$actual_returned_number, but mandoc isn't written that way. The workflow for Er is actually purely imagined for me at this point since it doesn't exist yet for anything I've checked, which is normally completely unacceptable, but my rationale is that confirming to the standard will create allowance for the workflow.

One idea for Dv contributed by kevans@ is debugging complex shell scripts with $?, which I think would be reinforcing design consistency.