Installing flepiMoP For Development
Instructions on how to install `flepiMoP` for development purposes, which uses a specific utility script that installs extras and force reinstall.
When developing flepiMoP
it is helpful to install it in a way that gives developers more control over the development environment. To assist developers with this there is the bin/flepimop-install-dev
helper script which wraps the bin/flepimop-install
script with some defaults.
First obtain a clone of
flepiMoP
where you plan on doing development work:
$ git clone git@github.com:HopkinsIDD/flepiMoP.git
Cloning into 'flepiMoP'...
remote: Enumerating objects: 28538, done.
remote: Counting objects: 100% (3449/3449), done.
remote: Compressing objects: 100% (857/857), done.
Receiving objects: 100% (28538/28538), 146.00 MiB | 37.49 MiB/s, done.
remote: Total 28538 (delta 2915), reused 2808 (delta 2589), pack-reused 25089 (from 2)
Resolving deltas: 100% (14847/14847), done.
Or replace git@github.com:HopkinsIDD/flepiMoP.git
with the appropriate URI for a fork if applicable.
Then change directory into this clone, checkout a branch to do development on, and then run the
bin/flepimop-install-dev
script. For more details on branch naming and GitHub usage please refer to the Git and GitHub Usage documentation.
$ cd flepiMoP
$ git checkout -b feature/XYZ/my-cool-new-thing
$ ./bin/flepimop-install-dev
The bin/flepimop-install-dev
script is a thin wrapper around the standard bin/flepimop-install
script that:
Installs
flepiMoP
to a conda environment located inside of this clone in a directory namedvenv/
,Force reinstalls
flepiMoP
which will wipe out an existing conda environment with a new one, andInstall all extra/optional dependencies for
gempyor
(likepytest
,black
,pylint
, etc.) that are particularly useful for python development.
Then you can activate this conda environment with:
conda activate venv/
You can verify that the installation was made to this local conda environment with:
$ which flepimop
/path/to/your/dev/flepiMoP/venv/bin/flepimop
If you need to refresh your development install, for example if you edit the R packages or add a new dependency, you can simply rerun the
bin/flepimop-install-dev
script which will wipe the previous conda environment and create a fresh new one.
Last updated