sysctl: Add -F option to display sysctl format, refactor for a function to display info options for and added test cases

[ota_j.email.ne.jp]

Summary

I'd like to add an option to display fmt.
Before doing so, make a function for displaing info for ease of refactoring.
Added kyua test cases.

Test Plan

% make install, etc...
% kyua test -k /usr/tests/Kyuafile sbin/sysctl

[delphij]

sbin/sysctl/sysctl.c

986

i is assigned but not used. I think it's quite likely that this sysctl call would always succeed as buf is sufficiently large, so if I was you I'd check the return value or maybe convert it to an assertion here.

[ota_j.email.ne.jp]

sbin/sysctl/sysctl.c

1128

Looking at how another sysctl failure is handled, it silently skips printing. I suppose this is for -a, all option.

I will follow the same error handling.

[ota_j.email.ne.jp]

• Return error code from show_info()

[ota_j.email.ne.jp]

• Update error description of show_info()

[ota_j.email.ne.jp]

• systat: Added -F option to show "fmt"

[ota_j.email.ne.jp]

• Updated man page.

[debdrup]

Have you run mandoc -Tlint and igor on the manual page?

At the very least, the latter would hint that you need to bump .Dd :)

[pauamma_gundo.com]

sbin/sysctl/sysctl.8

102–103 ↗

(On Diff #102286)

Also, I'm not sure "structured" is the best word. Maybe "compound"?

[ota_j.email.ne.jp]

• Bump the date, use "struct" as referenced in other parts of man sysctl, and give few examples of struct types.

[ota_j.email.ne.jp]

In D34012#772496, @debdrup wrote:

Have you run mandoc -Tlint and igor on the manual page?

At the very least, the latter would hint that you need to bump .Dd :)

No, I hadn't known.
Thanks for the info and I had ran.

[ota_j.email.ne.jp]

• Bump the date, use "struct" as referenced in other parts of man sysctl, and give few examples of struct types.

[ota_j.email.ne.jp]

sbin/sysctl/sysctl.8

102–103 ↗

(On Diff #102286)

I didn't realize your suggested changes in the first round of changes.

When I looked at lower pages of sysctl.8, I found 'struct' and used it.
You seemed to question about "structured" and didn't pick up this time. If "structured" is a better word, I don't have an object against it.

[pauamma_gundo.com]

Manual page language looks good to me. Can't speak about source code or consistency between them.

sbin/sysctl/sysctl.8

102–103 ↗

(On Diff #102286)

No, "struct" is fine for consistency, I think.

[ota_j.email.ne.jp]

• sysctl -l to display the lenth of returned data.

[bcr]

Confirm man page looks good.

[mckusick]

This all looks good to me. Addition of tests is helpful and appreciated.

[bcr]

Who's going to commit this?

[markj]

sbin/sysctl/sysctl.8

113 ↗

(On Diff #102648)

113 ↗

(On Diff #102648)

I think "size" (in bytes?) is more accurate than "length".

114 ↗

(On Diff #102648)

sbin/sysctl/sysctl.c

1001

error is always 0, can we get rid of either error or i?

1063

[mckusick]

In D34012#841017, @bcr wrote:

Who's going to commit this?

I am happy to do the commit once it is finalized.

[mckusick]

@ota_j.email.ne.jp:
Is this ready to be committed?

[ota_j.email.ne.jp]

In D34012#851339, @mckusick wrote:

@ota_j.email.ne.jp:
Is this ready to be committed?

Yes, of course. Thank you for pick up the change-set and proceeding.