du(1): Add --si option and test

[kevans]

Summary

Add an --si option to du(1) to display in terms of powers of 1000 rather
than powers of 1024

[cem]

It looks ok to me. Ideally this matches some kind of existing convention. But if not, it's an ok convention to start, too.

usr.bin/du/du.c

499

misindent?

[kevans]

In D11748#243528, @cem wrote:

It looks ok to me. Ideally this matches some kind of existing convention. But if not, it's an ok convention to start, too.

This and its df(1) brother-review follow a kind of convention from their GNU counterparts: df exposes -H/--si, du exposes --si

usr.bin/du/du.c

499

Whoops, will fix in morning or before commit

[jilles]

As a side effect, this change allows options and operands to be mixed in GNU style. Either this should be disabled by prepending "+" to the optstring or this should be mentioned in the commit message.

[kevans]

Address difference in behavior between getopt(3) and getopt_long(3)

[kevans]

Fix indentation

[kevans]

In D11748#244126, @jilles wrote:

As a side effect, this change allows options and operands to be mixed in GNU style. Either this should be disabled by prepending "+" to the optstring or this should be mentioned in the commit message.

Thanks for that, I was unfortunately unaware of that detail. I haven't personally used getopt{,_long} all that much, and getopt_long(3) doesn't seem to discuss leading optstring possibilities other than to mention that a leading '+'/'-' are ignored with POSIXLY_CORRECT. =(

I have no intention of changing existing behavior here -- that's a discussion for a different day. =)

[emaste]

usr.bin/du/du.1

31

Don't forget to update .Dd before commit.