[finalize] release_notes_file = release_notes.md release_notes_contents = readme_file = README.md readme_data = Currently, conda is required to manage installation of the environment. A fresh installation of Anaconda3 or Miniconda3 is not required for each {meta:name} release, however. An existing conda installation may be used. The method described here allows for multiple, entirely segregated, pipeline installations. A `bash`-compatible shell is required for all steps below. Python dependencies are taken directly from PyPI or development repositories as needed, so conda's ability to install hybrid conda/pip environments from a YAML specification is used. This requires first downloading the specification file and then creating the environment with it. ## For existing conda installations: ### Install the pipeline environment ``` $ curl -O https://ssb.stsci.edu/releases/{meta:name}/{cycle}/latest-linux.yml $ conda env create --file latest-linux.yml ``` ### Activate the pipeline environment ``` $ conda activate {release_name} ``` Each new delivery iteration of the calibration environment will need to be installed using the above commands. ------ ## If conda is not yet installed If the conda tool is not available on the target system, it will need to be installed before the calibration environment can be installed. ### Installing conda For detailed instructions of this step, please visit: https://conda.io/projects/conda/en/latest/user-guide/install/index.html **Miniconda:** ``` # TODO: UPDATE THIS $ wget https://repo.anaconda.com/miniconda/Miniconda3-py37_4.8.2-Linux-x86_64.sh $ bash Miniconda3-py37_4.8.2-Linux-x86_64.sh $ export PATH=$HOME/miniconda3/bin:$PATH ``` **Anaconda (if preferred):** ``` # TODO: UPDATE THIS $ wget https://repo.continuum.io/archive/Anaconda3-2019.10-Linux-x86_64.sh $ bash Anaconda3-2019.10-Linux-x86_64.sh $ export PATH=$HOME/anaconda3/bin:$PATH ```