Skip to content

pynvl

MIT License PyPI

General purpose data utilities, inspired by PL/SQL.

This package contains a suite of functions which flatten out data handling operations.

🔹 Features

  • nvl(expr, default) => Substitute a value when None is encountered
  • decode(expr, search1, result1, ..., default) => if-elif-else functionality
  • sign(n) => Returns a value indicating the sign of a number
  • noneif(expr1, expr2) => If expr1 and expr2 are equal then return None
  • nvl2(expr, value_if_not_null, value_if_null) => Extending NVL to substitute when non-None value is encountered
  • decode(None, None, ...) treats None == None (follows PL/SQL)
  • coalesce(expr1, expr2, ..., expr_n) => Returns first non-None value in list
  • Fully tested with pytest
  • MIT Licensed

🔹 Pandas Integration

pynvl also provides optional pandas-native helpers for use with pd.Series and operates element-wise.

import pandas as pd
from pynvl import pd_sign, pd_nvl, pd_nvl2, pd_noneif, pd_decode

s = pd.Series([-5, 0, 3, None])

print(pd_sign(s).tolist())
# [-1, 0, 1, None]

print(pd_nvl(s, 99).tolist())
# [-5, 0, 3, 99]

print(pd_nvl2(s, "not-null", "is-null").tolist())
# ['not-null', 'not-null', 'not-null', 'is-null']

print(pd_noneif(s, 0).tolist())
# [-5, None, 3, None]

print(pd_decode(s, -5, "neg", 0, "zero", 3, "pos", default="other").tolist())
# ['neg', 'zero', 'pos', 'other']

🔹 Installation From PyPI

pip install pynvl-lib

🔹 Demo page