Overview

The analysis (or crunching) of multi-dimensional, semantically grouped data is typically performed by database systems (relational, column-store, schema-free). Dynamically-typed languages like Python introduced feature-rich frameworks (e.g. pandas) with an easy-to-use syntax to perform complex data analysis operations with a small amount of code. While dynamically-typed languages have some advantages about statically-compiled languages in terms of flexibility, they often lack premature (type) correctness checks such that operations fail at runtime.

The goal of this project to a C++ data analysis framework that combines the two worlds of syntatically easy-to-use API with operation correctness checks at compile-time.