use-operator-module

When to Use

• Building interpreters or expression evaluators
• Dict mapping symbols to operations
• Reduce with arithmetic operations
• Cleaner than lambda a, b: a + b
• Functional programming with operators

When NOT to Use

• Simple inline expression is clearer
• Not doing functional programming
• Custom operator behavior needed

The Pattern

Use operator module for named function versions of operators.

import operator as op

# Instead of lambdas
add = op.add        # lambda a, b: a + b
mul = op.mul        # lambda a, b: a * b
neg = op.neg        # lambda x: -x
getitem = op.getitem  # lambda obj, key: obj[key]

# Reduce with operators
from functools import reduce
product = reduce(op.mul, [1, 2, 3, 4])  # 24
total = reduce(op.add, numbers)

# Comparison operators
op.lt(3, 5)  # True (3 < 5)
op.eq(3, 3)  # True (3 == 3)

# Attribute access
op.attrgetter('name')  # lambda obj: obj.name
op.itemgetter(0)       # lambda x: x[0]

Example (from pytudes)

# Lisp interpreter (lis.py)
import operator as op

env = {
    '+': op.add,
    '-': op.sub,
    '*': op.mul,
    '/': op.truediv,
    '>': op.gt,
    '<': op.lt,
    '>=': op.ge,
    '<=': op.le,
    '=': op.eq,
    'equal?': op.eq,
    'eq?': op.is_,
    'not': op.not_,
}

# Cleaner than:
# '+': lambda a, b: a + b,

# Extended Lisp interpreter (lispy.py)
import math

def add_globals(env):
    env.update(vars(math))  # sin, cos, sqrt, pi, ...
    env.update({
        '+': op.add, '-': op.sub, '*': op.mul, '/': op.truediv,
        '>': op.gt, '<': op.lt, '>=': op.ge, '<=': op.le, '=': op.eq,
        'not': op.not_,
        'cons': lambda x, y: [x] + y,
        'car': lambda x: x[0],
        'cdr': lambda x: x[1:],
    })
    return env

# Product function (AdventUtils.ipynb)
def prod(numbers):
    """Product of numbers."""
    return reduce(op.mul, numbers, 1)

Key Principles

1. Named is clearer: op.add beats lambda a,b: a+b
2. Works with reduce: Standard pattern for aggregation
3. Attrgetter/itemgetter: Create key functions for sorting
4. methodcaller: op.methodcaller('strip') for method calls
5. Consistent interface: All binary operators have same signature