List Comprehensions — Pythonic Collection Building

A list comprehension is Python’s concise syntax for creating a new list by applying an expression to every item in an iterable, with an optional filter condition. Instead of writing a for loop that appends to an empty list, a comprehension does it in a single, readable line. They are one of the most distinctly Pythonic constructs — you will see them everywhere in Python code, including FastAPI response transformations, SQLAlchemy result processing, and data pipeline logic. The key is knowing when a comprehension improves readability and when a regular loop is clearer — comprehensions should be simple enough to read in a single glance.

List Comprehension Syntax #

# Syntax: [expression for item in iterable if condition]
# The 'if condition' part is optional

# ── Basic: transform every item ────────────────────────────────────────────────
numbers = [1, 2, 3, 4, 5]

# Traditional for loop
squares = []
for n in numbers:
    squares.append(n ** 2)
# squares = [1, 4, 9, 16, 25]

# List comprehension — one line
squares = [n ** 2 for n in numbers]
# [1, 4, 9, 16, 25]

# ── With filter: only include some items ───────────────────────────────────────
# Traditional
evens = []
for n in numbers:
    if n % 2 == 0:
        evens.append(n)

# Comprehension
evens = [n for n in numbers if n % 2 == 0]
# [2, 4]

# ── Transform AND filter ───────────────────────────────────────────────────────
even_squares = [n ** 2 for n in numbers if n % 2 == 0]
# [4, 16]

Practical FastAPI Examples #

# ── Extracting fields from database results ────────────────────────────────────
db_posts = [
    {"id": 1, "title": "First",  "body": "...", "author_id": 1, "published": True},
    {"id": 2, "title": "Second", "body": "...", "author_id": 2, "published": False},
    {"id": 3, "title": "Third",  "body": "...", "author_id": 1, "published": True},
]

# Get titles of published posts
published_titles = [p["title"] for p in db_posts if p["published"]]
# ["First", "Third"]

# Build a list of IDs
post_ids = [p["id"] for p in db_posts]
# [1, 2, 3]

# ── Transforming response data ─────────────────────────────────────────────────
# Add a "slug" field based on the title
def to_slug(title: str) -> str:
    return title.lower().replace(" ", "-")

posts_with_slugs = [
    {**p, "slug": to_slug(p["title"])}   # unpack dict and add slug field
    for p in db_posts
]
# [{"id":1, "title":"First", ..., "slug":"first"}, ...]

# ── String processing ──────────────────────────────────────────────────────────
tags_input = "  python , fastapi , postgresql  "
tags = [tag.strip() for tag in tags_input.split(",")]
# ["python", "fastapi", "postgresql"]

# Normalise tags: lowercase, strip, remove empty
raw_tags = ["Python", " FastAPI ", "", "PostgreSQL", "python"]
clean_tags = list({tag.strip().lower() for tag in raw_tags if tag.strip()})
# ["python", "fastapi", "postgresql"] (unique, sorted)

Comprehension with Conditionals — if/else Inside #

# The filter 'if' goes at the END of the comprehension
# An if/else INSIDE the expression goes BEFORE the for

numbers = [1, 2, 3, 4, 5]

# Filter (if at end — keeps only some items)
odds = [n for n in numbers if n % 2 != 0]
# [1, 3, 5]

# Transform with ternary (if/else before for — transforms ALL items)
labels = ["even" if n % 2 == 0 else "odd" for n in numbers]
# ["odd", "even", "odd", "even", "odd"]

# Combine: transform then filter
result = [n * 10 if n > 2 else n for n in numbers if n != 4]
# n=1: 1 (not >2), n=2: 2 (not >2), n=3: 30 (>2), n=5: 50 (>2)
# [1, 2, 30, 50]  (4 was filtered out)

Nested List Comprehensions #

# Nested comprehension — for 2D structures
matrix = [[1, 2, 3], [4, 5, 6], [7, 8, 9]]

# Flatten a 2D list to 1D
flat = [num for row in matrix for num in row]
# [1, 2, 3, 4, 5, 6, 7, 8, 9]
# Read: "for each row in matrix, for each num in row, take num"

# Transpose a matrix
transposed = [[row[i] for row in matrix] for i in range(3)]
# [[1, 4, 7], [2, 5, 8], [3, 6, 9]]

# Simple nested — all combinations
colours = ["red", "blue"]
sizes   = ["S", "M", "L"]
combos  = [f"{c}-{s}" for c in colours for s in sizes]
# ["red-S", "red-M", "red-L", "blue-S", "blue-M", "blue-L"]

Common Mistakes #

Mistake 1 — Confusing filter position with ternary position #

❌ Wrong — putting ternary after the for (syntax error) or filter before for:

[n for n in numbers if n % 2 == 0 else n * 10]  # SyntaxError

✅ Correct — ternary goes BEFORE the for, filter goes AFTER:

[n if n % 2 == 0 else n * 10 for n in numbers]  # transform all ✓
[n for n in numbers if n % 2 == 0]              # filter only ✓

Mistake 2 — Over-complex comprehensions #

❌ Wrong — unreadable one-liner:

result = [f(x) for x in data if g(x) and h(x) for y in x.items() if y[0] != "skip"]

✅ Correct — use a regular for loop when logic is complex:

result = []
for x in data:
    if g(x) and h(x):
        for key, val in x.items():
            if key != "skip":
                result.append(f(x))
# Clear intent, easy to debug ✓

Mistake 3 — Using a comprehension for side effects #

❌ Wrong — comprehension just to call a function (wastes memory building a list):

[print(item) for item in items]   # builds a [None, None, ...] list for nothing

✅ Correct — use a for loop for side effects:

for item in items:
    print(item)   # ✓ no wasted list creation

Quick Reference #