Restore exports

[garrison]

They were dropped (by mistake, I assume) in fb96177.

[garrison]

I suppose another issue is that the tests are structured such that they didn't notice this.

[codecov-io]

Codecov Report

Merging #11 into master will not change coverage.
The diff coverage is n/a.

@@          Coverage Diff          @@
##           master    #11   +/-   ##
=====================================
  Coverage     100%   100%           
=====================================
  Files           1      1           
  Lines          29     29           
=====================================
  Hits           29     29

Impacted Files	Coverage Δ
src/Humanize.jl	100% <ø> (ø)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 17178af...ca73f41. Read the comment docs.

[IainNZ]

Wasn't a mistake, want to discourage using with my packages and instead encourage explicit imports.

[IainNZ]

(The tests were changed accordingly)

[garrison]

Hmm, interesting. What is the rationale? (Is this a larger trend or is it currently limited to your packages?)

[IainNZ]

Rationale is that namespace pollution is bad.
Basically, after two years of Google-ing etc I've become quite into style guides and such, and @mlubin has been looking at doing this for the JuliaOpt-verse. Part of this style guide would be this, so I thought I'd give it a go here.

[IainNZ]

Closing.

[garrison]

Fwiw, I never really agreed with this decision. If the user doesn't want namespace pollution, they are always free to write import Humanize instead of using Humanize. IMO removing everything from export in julia makes as much sense as setting __all__ = [] in python for fear that somebody might use from xxx import *. It seems a bit patronizing.

But #21 provides a sensible way forward, too.