Difficulties to install miniconda properly

[gandroz]

I've got this buildspec

version: 0.2

phases:
  install:
    runtime-versions:
      python: 3.8
    commands:
      - export PATH=/opt/conda/bin:$PATH  # should not be mandatory, but needed to call `conda` command
      - wget --quiet https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda.sh
      - sh ~/miniconda.sh -b -p /opt/conda
      - rm ~/miniconda.sh
      - ln -s /opt/conda/etc/profile.d/conda.sh /etc/profile.d/conda.sh
      - echo ". /opt/conda/etc/profile.d/conda.sh" >> ~/.bashrc
      - echo "conda activate" >> ~/.bashrc



  pre_build:
    commands:
      # Initialize Conda for the current shell
      - conda init bash
      # Restart shell for changes to take effect. Does not seem to work :(
      - exec $SHELL
      - conda create -n myenv python=3.6
      # command `conda activate myenv` does not work, call `activate` the standard way
      - . activate myenv && which python  # returns /opt/conda/bin/python instead of /opt/conda/envs/myenv/bin/python

Conda is well installed but I have trouble to make it work properly. As stated in the doc, conda need to be initialized. However, the shell needs to be restarted afterwards, and the exec $SHELL does not seems to do it as I am not able to activate the virtual environment with conda activate myenv. Although, once activated (or I think it was), the python binary called is the one on the root, not the virtual environment.

[JelsB]

I've had a very similar problem. Activating a conda environment in a CI server or Docker image isn't straightforward. I've encountered this issues several times. The documentation of conda does provide an example with using Travis CI, but it didn't work properly when I tried it on AWS CodeBuild

# part of conda docs for Travis CI
- wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh
- bash miniconda.sh -b -p $HOME/miniconda
- source "$HOME/miniconda/etc/profile.d/conda.sh"
- hash -r
- conda config --set always_yes yes --set changeps1 no
- conda update -q conda
# Useful for debugging any issues with conda
- conda info -a

# Replace dep1 dep2 ... with your dependencies
- conda create -q -n test-environment python=$TRAVIS_PYTHON_VERSION dep1 dep2 ...
- conda activate test-environment
- python setup.py install

For AWS CodeCuild I had to change source "$HOME/miniconda/etc/profile.d/conda.sh" into . "$HOME/miniconda/etc/profile.d/conda.sh" since the source command is not a POSIX standard.
Next, it failed when trying to conda activate test-environment, an error that most people encounter:

CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'. To initialize your shell, run

  $ conda init <SHELL_NAME>

Currently supported shells are:
- bash
- fish
- tcsh
- xonsh
- zsh
- powershell

See 'conda init --help' for more information and options.

IMPORTANT: You may need to close and restart your shell after running 'conda init'.

And like you tried, the conda init command doesn't really work...
But via this blog, I found a working solution!
You can just use an environment without activating it by running conda run -n my_env python file.py

Translating this to AWS CodeCommit my buildspecs (conda part) look like this:

{
  "version": "0.2",
  "phases": {
    "pre_build": {
      "commands": [
        "wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh -q",
        "bash miniconda.sh -b -p $HOME/miniconda",
        ". \"$HOME/miniconda/etc/profile.d/conda.sh\"",
        "hash -r",
        "conda config --set always_yes yes --set changeps1 no",
        "conda update -q conda",
        "conda info -a",
        "conda env create -q -p ./py_env -f environment.yml"
      ]
    },
    "build": {
      "commands": [
        "conda run -p ./py_env python app.py"
      ]
    }
  }
}

[subinataws]

Thank @JelsB. @gandroz - Marking this resolved, assuming steps listed by @JelsB works for you.