Hi folks! <@U0802KLPRD0> needs help on Baysor: <ht...
# bioconda
p
Hi folks! @Quentin Blampey needs help on Baysor: https://github.com/bioconda/bioconda-recipes/pull/64488#issuecomment-4294172925 I tried and failed. Works with a magic binary, but Björn G is asking for a build if possible. Anyone able to help out, or fight back against Björn? 🙏🏻
n
It's crazy that it takes so long to compile 😮
p
n
Then it shouldn't take that long though... very weird
m
Have you seen that there is a current branch called
cpp
with a compiler upgrade and some dependency fixes? Maybe this would work?
q
Hi @Matthias Zepper, thanks, I didn't know he was working on this. I'm not sure when it will be ready to be merged and released though - it may take a while, without being sure it will fix the recipe...
p
Yeah I think that's yet another reason to go with the magic binaries now, rather than pissing around trying to fix it.
n
I'd rather see it implemented in a way we don't piss off the other bioconda maintainers though. I think this might be a better case where we clone the docker container to our repo for this release of the pipeline and add the conda recipe once it can be properly created.
p
Fair. You think it's piss-off worthy? I was more thinking that we needed to demonstrate reasonable effort 😅 Quite a few people have asked for this conda package already, so I think it'd be useful for the community.
The binaries are right there..
n
If it were up to me we would have already merged it with the binary, just don't want to cause trouble over there
👍🏻 1
q
Hi, so do think we should merge it with the binaries? Do you have an idea to convince Björn G?
p
Looks like @Sameesh Kher has an alternative PR here? https://github.com/bioconda/bioconda-recipes/pull/65052
Ah, it's also a magic binary
@Quentin Blampey can you revert back to the static binary again in your PR?
✅ 1
42aeca2
q
I'm also fine with the alternative PR, it looks to be really similar
Hi, this PR has been merged, but it had some permission issues, so a second PR with a new built was done there. I created a custom container using
wave
on the env.yaml below, but I still get the same error. Do you think it’s because it used the first build and not the second? If so, how to use the second baysor build?
Copy code
channels:
  - conda-forge
  - bioconda
dependencies:
  - python=3.12
  - pip=24.3.1
  - baysor=0.7.1=*_1 # I tried to use the second build
  - pip:
      - sopa[baysor]==2.2.8
p
Yes this is being discussed on the Nextflow slack (
#nextflow-containers
channel)
👍 1