Open cghAshes opened 2 months ago
I triggered conda-forge
google-compute-engine
auto migration, it failed onlinux-ppc64le
but ok onlinux-aarch64
.
TravisCI is very unreliable :-/ You have two options:
linux_ppc64le: linux_64
For both you will need help from someone who has permissions to push to that repository.
The lowest version pandas is 0.24.2 in conda-forge, which not satisfy svtools recipe(0.19.2). Should I write a recipe for pandas 0.19.2 and submit a PR to conda-forge?
There are many versions - https://anaconda.org/conda-forge/pandas/files
You could try to remove 0.19.2
at https://github.com/bioconda/bioconda-recipes/blob/master/recipes/svtools/meta.yaml#L25 and see whether using a newer version will still work.
I think the first version with support for linux-aarch64 is 0.24.2 - https://anaconda.org/conda-forge/pandas/files?page=34
but the svtools recipe require
python < 3
.
This is a problem! You could try to migrate the package with 2to3.py
I executed
conda build
svtools in linux-aarch64 env. I need to solve two dependencies problem.google-compute-engine
I triggered conda-forge
google-compute-engine
auto migration, it failed onlinux-ppc64le
but ok onlinux-aarch64
. LINK I found that conda-forge auto migration just providegoogle-compute-engine
above py38, but the svtools recipe requirepython < 3
.Can it work properly when the conda-forge provide
google-compute-engine
version above python 3? (Also, should I go to solve the problem why it fail on linux_ppc64le which hard for me?)pandas-0.19.2
The lowest version pandas is 0.24.2 in conda-forge, which not satisfy svtools recipe(0.19.2). Should I write a recipe for pandas 0.19.2 and submit a PR to conda-forge?
Any suggestion would be appreciated! Many thanks!