Marcel Bargull

> On Linux, after upgrading mamba from 1.5.3 to 1.5.4, "mamba update --all" worked again, but "mamba clean --yes --all" still had an error similar to what reported in #534....

> I really meant the content of `CONDA_PKGS_DIRS`, there is a `repodata_record.json` and an `index.json` file contianing the dependencies, and I think that mamba is using one of them when...

(An aside regarding the question "Should some of the core Numba folks request access to CF feedstocks for LLVM/llvmlite/Numba so as to be able to help with/fix issues more easily?"...

Option 3 has been working well for us at Bioconda since March/April or so. (I.e., using `quay.io/condaforge/linux-anvil-cos7-x86_64` as the base but not requiring `sysroot_linux-64=2.17` by default.) ---- One minor thing...

Hi @Takadonet, I took a closer look at the patch and I'd asses (without looking at the entirety of the code) it really does only fix shortcomings without other behavioral...

Ah, ok, I was only looking at the call at https://github.com/katholt/srst2/blob/v0.2.0/scripts/srst2.py#L1422 which can't work (I think). But for the other calls `create_allele_pileup` is used beforehand which adds the `.`, right.

pre-commit.ci autofix

gh-4850 just disabled this code path by default without any attempted fixes. Seeing the code in gh-5228 again reminded me of this PR and since I don't believe anything has...

@BiocondaBot please fetch artifacts

@BiocondaBot please fetch artifacts