QuICCPython

QuICCPython is a postprocessing software developed (almost natively) in Python. It provides routines for the manipulation of stored HDF5 files generated from QuICC runs (https://github.com/QuICC/QuICC.git). Several routines are also available to perform computation of kinetic and magnetic energy, uniform vorticity of the flow. Generation of slices for the different geometries supported by QuICC is also supported.

Instalation:

Install dependencies

QuICCPython requires a C++ compiled module to compute associated Legendre functions, Worland polynomials (Marti and Jackson, 2016), and perform rotation of the spectrum (Gimbutas and Greengard, 2009). To provide this requirements, please follow this steps

install SHTns (Schäffer, 2013):

You are gonna need mercurial for this task. Download SHTns in your workspace or a download folder.

hg clone https://bitbucket.org/nschaeff/shtns
cd shtns
./configure 
make 
sudo make install

install pybind11

Pybind11 is required to link C++ executable into a Python module. Mac user using python through anaconda require to set the following environment variable:

export MACOSX_DEPLOYMENT_TARGET=10.9

Pybind11 is available through pip

python3 -m pip install pybind11

download QuICCPython

Next you will download QuICCPython and obtain the submodules

git clone https://github.com/jeromeNoir/QuICCPython.git
cd QuICCPython
From Jerome: the following doesn't work for me
	-git submodule init
	-git submodule update
	instead I do:
	-sudo git submodule init
	-I upload manually the folder "eigen-git-mirror @ 1456fe2" on our git 
	-copy it into the QuICCPyhton directory and rename it "eigen-git-mirror"

install quicc-bind

You're gonna need to compile the C++ sources and link them to a python module using pybind11. To do this, run

python3 -m pip install ./

Usage

Reading files

HDF5 files can be opened with:

from QuICCPython import read

myDataSpectral = read.SpectralState(<fileName>)
myDataPhysical = read.PhysicalState(<fileName>)

Once opened

myData.parameters. ...

contains the parameters used to run the simulation, and

myData.fields. ...

contains the states, either in spectral coefficient of physical values, of the simulation.

Obtaining slices and line cuts

Once the data are open it is possible to obtain slices of the various field components for plotting. The specific implementation is dependent firstly on whether the data is in its spectral or physical representation. Secondly, it depends on the geometry. The exact module can be found in QuICCPython..<specrtal/physical>. For example one can obtain meridional slices of the simulation state with:

from QuICCPython.shell.spectral import getMeridionalSlice
meridionalSlice = getMeridionalSlice(myData, phi0 = np.pi/2)

The returned value is a python dictionary with keys

meridionalSlice['x',
	'y',
	'uR',
	'uTheta',
	'uPhi']

thanks to which it is possible to plot with ease, with

from matplotlib import pyplot as plt
plt.contourf(meridionalSlice['x'], meridionalSlice['y'], meridionalSlice['uPhi'])

Furthermore it is possible to store the results to memory with

from scipy.io import savemat
savemat('myFile.mat', mdict = meridionalSlice)

For further examples, please consult the notebooks shellExample.ipynb, sphereExample.ipynb and cartesianExample.ipynb in the main folder.