Update to dada2 1.8.0; drop osx pins

[epruesse]

BTW: 1.2.0a0 is < than 1.2.0, so pinning should use >=1.1.23,<1.2a0 or similar. Just in case someone pushes alphas.

If in doubt, use

from conda.exports import VersionOrder
[str(x) for x in sorted(VersionOrder(v) for v in 
('1.2.0', '1.2.0a0', '1.2a0', '1.2', '1.1.99', '1.2.1', 
'1.2p1', '1.2post0', '1.2.12', '1.2.0.post0', '1.2.p0'))]

['1.1.99', '1.2a0', '1.2p1', '1.2.0a0', '1.2.p0', '1.2.0', 
'1.2', '1.2.0.post0', '1.2.1', '1.2.12', '1.2post0']

to verify what will actually happen. It's mostly PEP440 but not completely so, and not really intuitive (but then the intuition on that topic varies).

I really like 1.2.0 < 1.2 < 1.2.1 ...

[epruesse]

ping @thermokarst

See https://github.com/qiime2/q2-alignment/pull/62

[thermokarst]

@benjjneb - any reason to hold off on updating the plugin to DADA2 1.8? I recall we discussed in at least one other issue thread that there were some planned script updates to go with this, but I am not sure if those are necessary, or just nice-to-have.

[epruesse]

I can change this back to 1.6 - just want to know whether busywork will come up with a plan that has the right glib so that SINA will comply on OSX.

[thermokarst]

Let's stick with 1.6 for now.

[benjjneb]

@benjjneb - any reason to hold off on updating the plugin to DADA2 1.8?

Adding the script updates in is a reason, but perhaps more importantly is that I know there is one edge-case bug that will arise in the current q2-dada2 scripts once it updates to 1.8, which needs a workaround added to the plugin R scripts.

BTW, am I right to think that the next release of Q2 will be around the end of April?

[thermokarst]

Adding the script updates in is a reason, but perhaps more importantly is that I know there is one edge-case bug that will arise in the current q2-dada2 scripts once it updates to 1.8, which needs a workaround added to the plugin R scripts.

👍 @epruesse --- let's keep it at 1.6 for now, let's just drop the bioconductor pins.

BTW, am I right to think that the next release of Q2 will be around the end of April?

Yep! Project page has specific dates for the next two releases: https://github.com/orgs/qiime2/projects

[epruesse]

@thermokarst done

[epruesse]

Uh-oh ... lots of failures: https://busywork.qiime2.org/teams/main/pipelines/master-distribution/jobs/unit-test-q2-dada2/builds/33

[epruesse]

This looks like it doesn't like the biostrings version installed.

BTW: We can re-build older versions of dada2 on Bioconda, too. Just need to add bioconductor-dada2/1.6.0/....

[thermokarst]

@epruesse --- see this PR for context: https://github.com/qiime2/q2-dada2/pull/103

We won't be able to drop these pins until we are able to update DADA2 to 1.8. I will revert these changes. Until then, you can run an env without the DADA2 stack in it, check out this gist from @ebolyen: https://gist.github.com/ebolyen/7cff6868d1fac82f6b5dae7a70f15858

@benjjneb --- anything we can do on this end to help? I am happy to take a stab at a PR, just let me know.