fix: update multiqc version to 1.12

[vinisalazar]

Hi,

I've been having constant problems with this wrapper. My Snakemake execution almost always hangs in the "Solving environment step". I was supportive of merging #422 but considering MultiQC is on version 1.12, I believe it would be better to drop this pinned version. I tried creating the environment from the file as it is in this PR and it correctly installs v1.12 of MultiQC.

Ping @fgvieira @thomasbtf, would you agree?

Thank you for any assistance you can provide.

Cheers,
Vini

Summary of changes:

• Remove '= 1.11' from multiqc environment yaml

MultiQC has been updated to 1.12 and pinning v1.11 is hampering environment solving.

@thomasbtf @fgvieira would you be okay with this?

[fgvieira]

I agree about updating multiqc version, but not sure about completely unpinning its version. Why not just update to 1.12?

[vinisalazar]

Hi,

@fgvieira thank you for the reply. I assumed that most wrappers did not specify a version, instead letting conda solve for the latest version available in the channel. Upon a second look I now see that most if not all wrappers set a specific version. I will update the PR accordingly.

However, this did get me thinking that all wrappers must be manually updated every time a new version is out? This seems very laborious. Is there a reason to avoid not pinning a version (i.e. attempting to solve for the latest one)? If there is, I believe this repository would benefit from some sort of automation, such as a GitHub Action or Bot, to update the version based on the Bioconda repository? Or is manual updating a deliberate practice?

Apologies if I'm confused about any aspects of best practices regarding the usage of wrappers, and thank you for the support.

Cheers,
V

[fgvieira]

As far as I know, all wrappers set a specific version so the results of a workflow are reproducible; this way every wrapper version is associated with a specific program version.

As far as implementing a GitHub Action or Bot, it could be a good idea (at least for most wrappers), but maybe we should have some feedback from @johanneskoester.

For this PR, can you fix the two failed tests?

[vinisalazar]

Hi @fgvieira thank you for your response.

I fixed the title of the PR. The Code quality / formatting check however fails on code that is unrelated to this PR. I even tried running black on the files indicated in the test but they were left unchanged.

This is what I get when running Black:

$ black --check */*/*/wrapper.py */*/wrapper.py
All done! ✨ 🍰 ✨
288 files would be left unchanged.

Would you have any advice or would it be okay to merge as is?

Thank you,
V

[thomasbtf]

It's also totally fine with me to update the multiQC version 👍

Regarding the code quality issue, it seems that some other files would be changed:

would reformat bio/biobambam2/bamsormadup/wrapper.py
would reformat bio/gatk/applybqsrspark/wrapper.py

Maybe it helps to update this PR with the current master of the snakemake-wrappers repo?

[vinisalazar]

Hi @thomasbtf, thank you for your response.

I just double-checked and the branch is up-to-date with master (at d66c264).

Cheers,
V

[thomasbtf]

Okay, no luck there.

Another idea: maybe your local black version doesn't match the one from the GitHub checks? On the GitHub Runners, black is used in version 22.1.0.

[vinisalazar]

@thomasbtf, thanks, that nailed it. Just pushed c9e489e that fixes the tests.