array Module Complexity

The array module provides an efficient array type for storing homogeneous data with lower memory overhead than lists.

Complexity Reference

Operation	Time	Space	Notes
array.array()	O(n)	O(n)	Create array
append()	O(1) amortized	O(1)	O(n) worst case when resizing
extend()	O(k)	O(k)	Add multiple
Indexing	O(1)	O(1)	Access by index
Search	O(n)	O(1)	Linear search
Insert	O(n)	O(1)	Insert at position
Remove	O(n)	O(1)	Remove element
typecodes	O(1)	O(1)	String of supported type codes

Basic Usage

import array

# Create array - O(n)
arr = array.array('i', [1, 2, 3, 4, 5])  # O(5) - integer type

# Type codes: 'i' = int, 'f' = float, 'd' = double, 'b' = byte
arr_float = array.array('f', [1.0, 2.5, 3.14])  # O(3)

# Append - O(1) amortized
arr.append(6)  # O(1)

# Access - O(1)
value = arr[0]  # O(1)

# Length - O(1)
length = len(arr)  # O(1)

Array Operations

Creation Methods

import array

# From iterable - O(n)
arr = array.array('i', range(10))  # O(10)

# From bytes - O(n)
arr = array.array('i')
arr.frombytes(b'\x01\x00\x00\x00\x02\x00\x00\x00')  # O(n)

# From list - O(n)
arr.fromlist([1, 2, 3])  # O(3)

Modification

import array

arr = array.array('i', [1, 2, 3, 4, 5])

# Insert - O(n)
arr.insert(2, 99)  # O(5) - shift elements

# Remove - O(n)
arr.remove(99)  # O(5) - shift elements

# Pop - O(1) at end, O(n) elsewhere
arr.pop()  # O(1) - remove last
arr.pop(0)  # O(5) - remove first, shift rest

Conversion

import array

arr = array.array('i', [1, 2, 3])

# To list - O(n)
lst = arr.tolist()  # O(3)

# To bytes - O(n)
bytes_data = arr.tobytes()  # O(3)

Performance Comparison

Every byte count below is from a 64-bit CPython build. Object headers, pointer width, int size and the C type behind each type code all vary with implementation, build and architecture, so treat them as one worked example rather than as constants.

import array
import sys

# An array's header is larger - 80 bytes against a list's 56 here - so for a
# handful of elements the array is the bigger object. The saving arrives at
# a few dozen, and grows from there.
lst = [1, 2, 3, 4, 5]
arr = array.array('i', [1, 2, 3, 4, 5])
print(sys.getsizeof(lst), sys.getsizeof(arr))  # too close to call at n=5

# List of references (8 bytes each) vs packed elements (4 for 'i', 8 for 'd')
lst = list(range(10_000))
arr = array.array('i', range(10_000))
print(sys.getsizeof(lst), sys.getsizeof(arr))  # 80056 vs 40420 bytes

# But that 2x understates it: getsizeof() counts the list's pointers, not the
# int objects they point at, at 28 bytes each. Summing those gives the
# reachable deep size - not memory the list owns, since CPython shares small
# ints (-5..256), which is 257 of these 10,000 and about 2% of the total.
deep = sys.getsizeof(lst) + sum(sys.getsizeof(x) for x in lst)
print(deep // sys.getsizeof(arr))  # 8x reachable, against 2x by getsizeof

Type Codes

import array

# Every constructor below is O(n) in the number of items; the type code
# fixes the bytes per item, which is what array buys over list
# Available type codes
# 'b' = signed byte (1 byte)
arr_b = array.array('b', [-128, 0, 127])

# 'B' = unsigned byte (1 byte)
arr_B = array.array('B', [0, 128, 255])

# 'i' = signed integer (2-4 bytes)
arr_i = array.array('i', [-1000, 0, 1000])

# 'I' = unsigned integer (2-4 bytes)
arr_I = array.array('I', [0, 1000, 2000])

# 'f' = float (4 bytes)
arr_f = array.array('f', [1.0, 2.5, 3.14])

# 'd' = double (8 bytes)
arr_d = array.array('d', [1.0, 2.5, 3.14])

When to Use Array

Good For:

• Large collections of numeric data
• Memory-constrained environments
• Binary file I/O with numeric data
• Raw byte operations

Not Good For:

• Mixed types (use list or tuple)
• Complex objects (use list)
• Frequent insertions (use list)
• Type flexibility needed

Version Notes

• Python 2.x: array available
• Python 3.x: Same functionality
• All versions: O(n) memory savings vs lists

Related Modules

• list - Flexible container
• struct - Binary data packing

Best Practices

✅ Do:

• Use for numeric data collections
• Use for memory efficiency
• Convert to list for type mixing

❌ Avoid:

• Mixed types in array
• Frequent insertions/deletions
• When flexibility needed