A reconstructed mini-skeleton of LiquidLib written in Python
Liquidlib is a post-processing package for molecular simulations written in C++. It takes a trajectory (containing atomic coordinates, etc.) output from simulation, and compute some physical quantities as instructed by an input file, which defines all the necessary parameters for the calculation.
This repo is one of my attempts to apply design patterns to some problems we have worked on before. I am pretty sure there are still a few places that can be improved. The code may not be that Pythonic as we may need to rewrite the C++ version if it works. So the code is more like a demo.
NOTE:
This is NOT a python version of LiquidLib, but a demo to reconstruct LiquidLib! The code does not compute anything yet!
Design patterns applied:
- Simple factory: to instantiate trajectory of specific type
- Strategy: to select atoms of interest
- Decorator: to validate inputs based on variable dependency
Useful references: