Miralib is the Java library that contains all the data handling and calculation algorithms used in Mirador. It is possible to run miralib calculations in a Python script by using PyJNIus. This allows to perform batch operations from the command line, in combination with any standard Python library (numpy, matplotlib).
The only requirement to run miralib scripts is to install the PyJNIus package, for example using pip:
pip install pyjniusInstallation of PyJNIus on Windows is slightly more involved, please see this brief wiki entry for more details.
One PyJNIus is installet, then the miralib classes can be accessed from Python as follows:
os.environ['CLASSPATH'] = "../miralib/dist/*"
from jnius import autoclass
Preferences = autoclass("miralib.utils.Preferences")
Project = autoclass("miralib.utils.Project")
DataRanges = autoclass("miralib.data.DataRanges")
DataSet = autoclass("miralib.data.DataSet")
The os.environ['CLASSPATH'] assignment is important so PyJNIus is able to locate the jar files containing the classes. The two dependencies of miralib, common-math3 and joda-time, need to be in the classpath folder as well.