Let us create a new environment called fenics and install under this.
conda create -n fenics -c conda-forge fenics mshr
You may also want to install some other packages
conda install -n fenics -c conda-forge ipython notebook matplotlib scipy
Now activate the environment and do your Fenics work
source activate fenics
After you are done working with Fenics
source deactivate
Show list of installed packages in this environment
conda list -n fenics
Update the packages in this environment
conda update -n fenics -c conda-forge --all
NOTE: Fenics installed via Anaconda does not provide full functionality of Fenics, and Docker images are the best option for this.
First install Docker; this is easy on a Mac and needs some configuration on Linux, see here. After this, start Docker and install the fenics script
curl -s https://get.fenicsproject.org | bash
Then install a Fenics image
fenicsproject pull
Note that this will install the latest image of Fenics. You are now ready to work with Fenics. Go into some directory where you want to do your work and do
fenicsproject run
To update the Fenics version at any time, just do fenicsproject pull. Many things like plot windows will not work in Docker, so you may want to use the Conda option to install Fenics.
To install an older version using Docker
docker pull quay.io/fenicsproject/stable:2017.2.0
or
fenicsproject pull stable:2017.2.0
Now, run this version like this
docker run -ti -p 127.0.0.1:8000:8000 -v $(pwd):/home/fenics/shared \
-w /home/fenics/shared quay.io/fenicsproject/stable:2017.2.0
or
fenicsproject run stable:2017.2.0
Install latest version
docker pull dolfinx/real
Run in docker
docker run -ti -p 127.0.0.1:8000:8000 -v $(pwd):/home/fenics/shared \
-w /home/fenics/shared dolfinx/real