Modeling

Modeling in Python  

The best practice for modeling in Python is to create a model class by calling the opensees.openseespy.Model(...) constructor (note the capital “M”). All standard OpenSees functions, as documented here  can be called as methods on the object that is retured. For example:

import opensees.openseespy as ops
model = ops.Model(ndm=2, ndf=3)
model.node(1, 2.0, 3.0)

Evaluate Tcl models from Python  

The first scripting interface to OpenSees used a programming language called Tcl, which continues to be supported. To execute Tcl commands from a Python script, just create an instance of the opensees.openseespy.Model class and call its eval() method:

import opensees.openseespy as ops
model = ops.Model()
model.eval("model Basic -ndm 2")
model.eval("print -json")

Full Tcl files can be conveniently executed in this way. For example, if a Tcl file called model.tcl exists in the current working directory:

import opensees.openseespy as ops
model = ops.Model()
with open("model.tcl", "r") as f:
    model.eval(f.read())