kib, mib, etc for -o should be powers of 1024, not 1000

[eatsu]

According to Wikipedia, KiB, MiB, etc are units based on binary prefixes which are powers of 1024, but in Dust they seem to be treated as SI units which are powers of 1000.

https://github.com/bootandy/dust/blob/v1.0.0/src/display.rs#L412-L420

I'm afraid this is technically incorrect and is confusing.

For reference, commands du, df and ls treat K, M, KiB, MiB, etc as powers of 1024 and KB, MB, etc as powers of 1000. Quoting from their --help:

The SIZE argument is an integer and optional unit (example: 10K is 10*1024).
Units are K,M,G,T,P,E,Z,Y,R,Q (powers of 1024) or KB,MB,... (powers of 1000).
Binary prefixes can be used, too: KiB=K, MiB=M, and so on.

I would appreciate it if this were fixed and k, m, etc could be used as aliases for kib, mib, etc, like the above commands convention.

Thanks!

[bootandy]

Oh wow that's embarrassing. I've implemented it backwards, haven't I. Good find. How did you notice that ?

I think this is a recent bug only in this version, it's from when I merged the behaviour of the now redundant --si flag.

[bootandy]

#392

does this look right ?

[eatsu]

How did you notice that ?

I noticed that the values of the output K was different between with -o kib and without the flag when working on _dust_output_formats in #390 :)

does this look right ?

kib / mib / kb / mb work now as I expected as far as I've tested. Thank you! However, it seems that k / m are now powers of 1000, which is different from the behavior of du / df / ls.

[bootandy]

Noted, updated.

[eatsu]

Ah, it now works perfectly for me. Thank you again!

[bootandy]

will go out in next release