SClib, a hack for straightforward embedded C functions in Python

TL;DR

SClib is a simple method that enables seamless integration of C functions into Python, significantly improving performance and flexibility for scientific computing and engineering applications.

Contribution

It introduces SClib, a straightforward approach for embedding C functions in Python, demonstrated through high-speed scientific and engineering case studies.

Findings

150x speed-up in Schrödinger equation solver

Effective integration of C code with IPython for visualization

Enhanced simulation speed and accuracy in control systems

Abstract

We present SClib, a simple hack that allows easy and straightforward evaluation of C functions within Python code, boosting flexibility for better trade-off between computation power and feature availability, such as visualization and existing computation routines in SciPy. We also present two cases were SClib has been used. In the first set of applications we use SClib to write a port to Python of a Schr\"odinger equation solver that has been extensively used the literature, the resulting script presents a speed-up of about 150x with respect to the original one. A review of the situations where the speeded-up script has been used is presented. We also describe the solution to the related problem of solving a set of coupled Schr\"odinger-like equations where SClib is used to implement the speed-critical parts of the code. We argue that when using SClib within IPython we can use NumPy and Matplotlib for the manipulation and visualization of the solutions in an interactive environment with no performance compromise. The second case is an engineering application. We use SClib to evaluate the control and system derivatives in a feedback control loop for electrical motors. With this and the integration routines available in SciPy, we can run simulations of the control loop a la Simulink. The use of C code not only boosts the speed of the simulations, but also enables to test the exact same code that we use in the test rig to get experimental results. Again, integration with IPython gives us the flexibility to analyze and visualize the data.