dockerfile – miniconda dev container – cannot install snowflake-snowpark-python package via conda but can via pip
Not entirely sure on the right level of information I need to provide here, fairly new to this, but…
Problem Statement:
I have a vscode miniconda dev container that I’m using, which I believe is using python 3.12, and I tried to install snowflake-snowpark-python package via
conda install snowflake-snowpark-python
But the error I get is as follows:
Steps attempts for Resolution:
Step 1:
I tried to adjust the the version of python being ran inside the dev container by adding the following to the “features” section within my devcontainer.json file to include, as follows:
when I query the dev container for the python version I thought it was telling me that I had the 3.10 installed:
But I get the same error message as before when I run the same ‘conda install snowflake-snowpark-python’ command.
Step 2:
I’ve looked at the documentation for the miniconda dev container (hubgw.docker.com/r/microsoft/devcontainers-miniconda) which told me to perform the following upon installation of a new version of python:
pip install --no-cache-dir pipx
pipx uninstall pipx
pipx reinstall-all
But still I am receiving the same error message when I try to then run ‘conda install snowflake-snowpark-python’
Step 3:
Running the following I am able to install the package as required:
pip install snowflake-snowpark-python
Desired Resolution:
- To understand if this will have any impact on the utilisation of the package – bearing in mind this is a development container.
- To understand if there is a fundamental error I’m making in either the setup of the dev container or the installation of a different version of python which is resulting in an inability to be able to install ‘snowflake-snowpark-package’ within this dev container using conda
Clearly I’m doing something fundamentally wrong, but I’m just not sure what it is. If anyone could help identify my error or point me toward a resolution it would be most appreciated.
Read more here: Source link