Please make consumable as a library

[DidgetyTech]

This is precisely the algorithm I'm looking for to use in my production system! But to use it, there's a few things to change to make it useful as a library:

1. Return the sample result in sample_n(). It implicitly returns None now.
2. Use logging and single line printouts instead of print.
   • With logging, the consuming code turn on or off logging or set the levels to log. With single lines, logs are easily greppable. I think the printout is fine for a script, but it's not neighborly as a library.

If this works for you, I can submit a pull request tomorrow.

[asmith26]

Hi @hoganbc,

Thanks for your great suggestions. I particularly think getting sample_n() to return something instead of just None is a great idea. Pull requests very welcome!

Thanks again

[DidgetyTech]

PR #4

[asmith26]

Thanks very much @hoganbc - a great and very clear pull request!

Updated PyPi accordingly.

[DidgetyTech]

Thanks!

[DidgetyTech]

Do you mind adding a tag to track the versions? This makes it really nice to know which commits were released with which version.

[asmith26]

Happily will - I haven't really used git tags before, what command would you recommend?

[DidgetyTech]

Looks like it's tagged now! The tag subcommand should be pretty straightforward, similar to using the branch subcommand.