Python Arrays

Python Arrays: Lists, array.array, and NumPy Compared

Python does not have a single built-in keyword called array like Java or C. Instead, Python gives you three array-like tools depending on your job: list for general data, array.array for compact typed sequences, and NumPy ndarray for fast math.

For 90% of everyday scripting, a standard Python list is the right choice because it integrates everywhere with zero imports.

Comparing Python's Three Array Options

Using the Standard Library array Module

When you store millions of uniform integers or floats without installing third-party libraries, array.array packs raw C values tightly in memory. It supports indexing, slicing, .append(), and .extend() just like a list, while enforcing a strict single-byte type code across every element.

from array import array

# 'i' = signed integer, 'd' = double-precision float
sensor_ids = array("i", [101, 102, 103])
sensor_ids.append(104)

temperatures = array("d", [21.5, 22.1, 20.8])
print(sensor_ids[1], sum(temperatures) / len(temperatures))

Common Mistakes to Avoid

• Inserting the wrong type into array.array: Appending "105" or 3.5 to an integer array("i") raises TypeError.
• Looping over huge numeric lists in pure Python: For heavy math on thousands of numbers, switch to numpy vectorized operations instead of Python for loops.
• Mutating slices inconsistently: list slicing returns a new list, whereas slicing a NumPy array returns a view sharing the same memory.

Frequently Asked Questions

Are Python lists technically dynamic arrays?

Yes. Under the hood, CPython stores a list as a contiguous C array of pointers to Python objects, resizing automatically as you append items.

Do I need to install anything for array.array?

No. The array module is built directly into the Python standard library.

How do I inspect type codes for array.array?

Common codes include 'i' (signed int), 'I' (unsigned int), 'f' (32-bit float), and 'd' (64-bit float).