read_excel_benchmark.py

from datetime import date
from timeit import timeit

import pandas as pd
import polars as pl

df = pl.DataFrame(
    {
        "v": range(1_000_000),
        "w": 999.999,
        "x": -42.0,
        "y": date.today(),
        "z": "acbdefghijlmnop",
    }
)

wb = df.write_excel(xl := "test.xlsx")


def pandas_read_excel():
    pd.read_excel(xl)


def polars_xlsx2csv():
    pl.read_excel(xl, engine="xlsx2csv")


def polars_openpyxl():
    pl.read_excel(xl, engine="openpyxl")


def polars_calamine():
    pl.read_excel(xl, engine="calamine")


def read_excel_performance(csv_filename):
    pandas_version = pd.__version__

    results = [
        {
            "Engine": f"Pandas {pandas_version}",
            "Time": timeit(pandas_read_excel, number=1),
        },
        {"Engine": "xlsx2csv", "Time": timeit(polars_xlsx2csv, number=1)},
        {"Engine": "openpyxl", "Time": timeit(polars_openpyxl, number=1)},
        {"Engine": "calamine (new)", "Time": timeit(polars_calamine, number=1)},
    ]

    # Write results to CSV
    pl.DataFrame(results).write_csv(f"output/{csv_filename}-asdfkjsahfkjash.csv")


read_excel_performance("read_excel_results")