xlnumba 0.1.0

A package for converting excel files to Numba optimized python code

xlNumba

xlNumba is utility to convert Excel spreadsheets into optimized Python code using Numba. This idea is to enable taking complex models built in Excel and converting them into high-performance Python code.

This library does not attempt to provide full Excel compatability. Instead it focusses on providing a subset of capabilities which can be compiled by Numba and optimized to provide high-performance execution.

In general numerical (and in the future date) related functions are supported while text functions have very limited support. Excel has some fairly complex exception handling and type coercion rules which are not supported.

Basic Example

from xlnumba import Compiler

ctx = Compiler("tests/fixtures/basic")
ctx.add_input('Raw', 'A1')
ctx.add_output('Adjusted', 'B1')
fn = ctx.compile()
adjusted = fn(Raw=3.0)

User Defined Functions example.

User defined functions can be created using the @xlnumba_function decorator. This decorator takes the name of the excel function. Note that the function must be numba compilable.

@xlnumba_function("EXCEL_FNC_NAME")
def custom_func(x):
    return 3 * x + 5

Acknowledgements

The original idea for this was inspired from the excellent pycel library.