Python interface¶

Cantera's Python interface is available from two channels:

1. The official cantera channel. This channel should be used if you installed Python from the default channel in conda. This channel also has pre-release versions of Cantera for testing. Cantera packages are available in this channel for the following platforms:

   • Windows (32- and 64-bit Intel)

   • Linux (64-bit Intel)

   • macOS (64-bit Intel)

2. The conda-forge channel. This channel should be used if you installed Python from the conda-forge channel or if your OS/processor combination is not supported by the cantera channel. Cantera packages are available in this channel for the following platforms:

   • Linux (64-bit Intel, 64-bit ARM, and 64-bit PPCLE)

   • macOS (64-bit Intel and 64-bit ARM (M1))

Option 1: Create a new environment for Cantera

The following instructions will create a conda environment where you can use Cantera from Python. For this example, the environment is named ct-env. From the command line (or the Anaconda Prompt on Windows), run:

conda create --name ct-env --channel cantera cantera ipython matplotlib jupyter

This will create an environment with Cantera, IPython, Matplotlib, and all their dependencies installed. In this case, we want to install Cantera from the cantera channel, so we add --channel cantera and to tell Conda to look at the cantera channel in addition to the default channels.

If you want to use the conda-forge channel, replace --channel cantera with --channel conda-forge.

To use the scripts and modules installed in the ct-env environment, including Jupyter, you must activate it it by running:

conda activate ct-env

Option 2: Create a new environment using an environment file

This option is similar to Option 1 but includes a few other packages that you may find helpful as you're working with Cantera. Copy and paste the contents of the file shown below into a file called environment.yaml. Then, save the the file somewhere and remember that location.

name: ct-env
channels:
- cantera  # or use cantera/label/dev for alpha/beta packages
- defaults
dependencies:
- python  # Cantera supports Python 3.7 and up
- cantera
- ipython  # optional (needed for nicer interactive command line)
- jupyter  # optional (needed for Jupyter Notebook)
- matplotlib  # optional (needed for plots)
- python-graphviz  # optional (needed for reaction path diagrams)
- h5py  # optional (needed for HDF/H5 output)
- pandas  # optional (needed for pandas interface)

From the command line (or the Anaconda Prompt on Windows), change directory into the folder where you saved environment.yaml:

cd folder/where/you/saved

and then run:

conda env create -f environment.yaml

This will create an environment called ct-env. Once you've done that, you need to activate the environment before using any scripts or modules that you just installed:

conda activate ct-env

Option 3: Install the development version of Cantera

To install a recent development snapshot (that is, an alpha or beta version) of Cantera, use the cantera/label/dev channel. Assuming you have an environment named ct-dev, you can type:

conda activate ct-dev
conda install --channel cantera/label/dev cantera

If you later want to revert back to the stable version in that environment, first remove and then reinstall Cantera:

conda activate ct-dev
conda remove cantera
conda install --channel cantera cantera

Alternatively, you can remove the ct-dev environment and follow Options 1 or 2 above to create a new environment.

conda activate ct-dev
conda update --channel cantera cantera

MATLAB interface¶

Cantera's MATLAB interface can be installed from the cantera channel. In this example, the command will create a new conda environment named ct-dev. From the command line (or the Anaconda Prompt on Windows), run:

conda create --name ct-dev --channel cantera cantera cantera-matlab

This will create an environment with Cantera's Python and MATLAB interfaces. Even if you plan to use Cantera via MATLAB, the Python interface is required to convert input files. In this case, Cantera must be installed from the cantera channel, so we add --channel cantera and to tell Conda to look at the cantera channel in addition to the default channels.

conda activate ct-dev
conda update --channel cantera cantera cantera-matlab

Development (C++ & Fortran 90) Interface¶

Warning: The conda package for Cantera's development interface is experimental. For the most recent version of the package, use the cantera/label/dev channel instead of the stable cantera channel. The development interface for stable releases is also available from the conda-forge channel.

In the following example, Cantera's development interface is installed from the cantera/label/dev channel. From the command line (or the Anaconda Prompt on Windows), create a new conda environment named ct-dev using:

conda create --name ct-dev --channel cantera/label/dev libcantera-devel

This will create an environment with Cantera's development interface. In this case, the addition of --channel cantera/label/dev ensures that the package is pulled from the most recent available Cantera version. Note that label/dev refers to the experimental development channel of Cantera, and not the development interface.

C++ header and libraries are installed within the ct-dev environment folder, which itself depends on the type of conda installation, and is abbreviated as path/to/conda/envs below. Within the ct-dev folder, locations follow conda recommendations for a given operating system.

library files               path/to/conda/envs/ct-dev/lib
pkg-config                  path/to/conda/envs/ct-dev/lib/pkgconfig
C++ headers                 path/to/conda/envs/ct-dev/include
samples                     path/to/conda/envs/ct-dev/share/cantera/samples
data files                  path/to/conda/envs/ct-dev/share/cantera/data

$ conda activate ct-dev
$ conda install make cmake scons pkg-config

$ cd /path/to/conda/envs/ct-dev/share/cantera/samples/cxx/demo
$ make  # uses Makefile; or
$ scons  # uses SConstruct; or
$ cmake . && cmake --build .  # uses CMakeLists.txt

$ export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/path/to/conda/envs/ct-dev/lib
$ export PKG_CONFIG_PATH=$PKG_CONFIG_PATH:/path/to/conda/envs/ct-dev/lib/pkgconfig
$ g++ demo.cpp -o demo $(pkg-config --cflags --libs cantera)

$ ./demo

library files               path/to/conda/envs/ct-dev/Library/lib
C++ headers                 path/to/conda/envs/ct-dev/Library/include
samples                     path/to/conda/envs/ct-dev/share/cantera/samples
data files                  path/to/conda/envs/ct-dev/share/cantera/data

conda activate ct-dev
conda update --channel cantera/label/dev libcantera-devel