Exposing unifrac.meta

[wasade]

This is a work in progress pull request that exposes unifrac.meta. Would it be possible to get a very quick comment on the approach to handling the passthrough? I wanted to make sure this approach was reasonable before getting the rest of the bits finalized.

The TODO items that I'm aware of are:

• [x] adjust plugin_setup text to better reflect the method (currently effectively a copy/paste of beta_phylogenetic)
• [x] add support for meta specific arguments
• [x] add support for checking empty tables

@ebolyen and @thermokarst, per the discussion on handling paired inputs on Slack, I'd be happy to revise this to reflect the : syntax but also wanted to get something tentatively in while that discussion solidifies. I'd also be happy to submit a PR to q2cli to support the : syntax if that is considered a blocker for this PR, although I may need some guidance there as (at the moment) am not very familiar with that codebase.

[thermokarst]

BTW, I sorted out the CI issues, you might need a refresh from master.

[wasade]

Thanks @thermokarst!!

[thermokarst]

Hey @wasade - I got the CI back in order, looks like there is a legit test failure. Let me know if you want me to take a look at it.

[thermokarst]

Actually, looks like maybe that is unrelated to the changeset here? @wasade, can you comment on this - do you know if the phylo/table mis-match error msg changed in the latest release of unifrac?

cc @ChrisKeefe

[wasade]

biocore/unifrac#106 I think? These extra validations were added

[thermokarst]

Thanks for the reminder @wasade, looks like @ChrisKeefe left some breadcrumbs here - I'll push a commit up to this feature branch sometime soon to address.

[thermokarst]

https://github.com/qiime2/q2-diversity-lib/blob/5c277ae800c002a31b01de19c48a22cfe621c217/q2_diversity_lib/tests/test_beta.py#L69-L72

[wasade]

Excellent!!