Python context managers are one of the language’s most elegant and underused features. The with statement guarantees cleanup code runs even when exceptions occur. In 2026, context managers are used everywhere from file handling to database connections, HTTP sessions, and async operations. This complete guide covers all patterns.
📋 Table of Contents
How Context Managers Work
# The with statement calls __enter__ and __exit__
with open("file.txt") as f:
data = f.read()
# f.close() called automatically, even on exception
# Equivalent to:
f = open("file.txt")
try:
data = f.read()
finally:
f.close()
# Multiple context managers
with open("input.txt") as infile, open("output.txt", "w") as outfile:
for line in infile:
outfile.write(line.upper())
# Nested (older style, avoid)
with open("input.txt") as infile:
with open("output.txt", "w") as outfile:
pass
Creating Context Managers
class DatabaseConnection:
def __init__(self, url: str):
self.url = url
self.conn = None
def __enter__(self):
import psycopg2
self.conn = psycopg2.connect(self.url)
self.cursor = self.conn.cursor()
return self.cursor # what 'as' variable receives
def __exit__(self, exc_type, exc_val, exc_tb):
if exc_type is None:
self.conn.commit() # commit if no exception
else:
self.conn.rollback() # rollback on exception
self.cursor.close()
self.conn.close()
return False # False = don't suppress the exception
# Usage
with DatabaseConnection("postgresql://localhost/mydb") as cursor:
cursor.execute("INSERT INTO users (name) VALUES (%s)", ("Alice",))
# commit() called automatically on exit
# rollback() called if exception occurs
contextlib.contextmanager — The Decorator Way
from contextlib import contextmanager
import time, logging
logger = logging.getLogger(__name__)
@contextmanager
def timer(name: str = ""):
start = time.perf_counter()
try:
yield # code inside 'with' block runs here
finally:
elapsed = time.perf_counter() - start
logger.info(f"{name}: {elapsed*1000:.2f}ms")
# Usage
with timer("database query"):
results = db.execute_slow_query()
@contextmanager
def transaction(db):
try:
yield db
db.commit()
except Exception:
db.rollback()
raise
@contextmanager
def temporary_file(suffix=".tmp"):
import tempfile, os
path = None
try:
fd, path = tempfile.mkstemp(suffix=suffix)
os.close(fd)
yield path
finally:
if path and os.path.exists(path):
os.unlink(path)
with temporary_file(".json") as tmp:
with open(tmp, "w") as f:
json.dump(data, f)
process_file(tmp)
# file is deleted here, even on exception
Async Context Managers
import asyncio
from contextlib import asynccontextmanager
import httpx
@asynccontextmanager
async def http_session(base_url: str, timeout: float = 30.0):
async with httpx.AsyncClient(base_url=base_url, timeout=timeout) as client:
yield client
async def fetch_data():
async with http_session("https://api.example.com") as client:
r = await client.get("/users")
return r.json()
# Async database context manager
@asynccontextmanager
async def async_transaction(pool):
async with pool.acquire() as conn:
async with conn.transaction():
yield conn
async def create_user(pool, email: str):
async with async_transaction(pool) as conn:
user_id = await conn.fetchval(
"INSERT INTO users (email) VALUES ($1) RETURNING id", email
)
await conn.execute(
"INSERT INTO user_settings (user_id) VALUES ($1)", user_id
)
return user_id
# Transaction auto-commits or rolls back
FastAPI Lifespan — Application Context Manager
from fastapi import FastAPI
from contextlib import asynccontextmanager
import asyncpg
db_pool = None
@asynccontextmanager
async def lifespan(app: FastAPI):
# Startup
global db_pool
db_pool = await asyncpg.create_pool(
dsn="postgresql://localhost/mydb",
min_size=5, max_size=20
)
print("Database pool created")
yield # App runs here
# Shutdown
await db_pool.close()
print("Database pool closed")
app = FastAPI(lifespan=lifespan)
@app.get("/users")
async def list_users():
async with db_pool.acquire() as conn:
return await conn.fetch("SELECT id, name FROM users")
contextlib Utilities
from contextlib import suppress, redirect_stdout, nullcontext, ExitStack
import io, os
# suppress — ignore specific exceptions
with suppress(FileNotFoundError):
os.remove("temp_file.txt") # OK if doesn't exist
# redirect_stdout — capture print output
buffer = io.StringIO()
with redirect_stdout(buffer):
print("This goes to buffer, not terminal")
output = buffer.getvalue()
# nullcontext — conditional context manager
def process(data, lock=None):
ctx = lock if lock is not None else nullcontext()
with ctx:
process_data(data)
# ExitStack — dynamic number of context managers
with ExitStack() as stack:
files = [stack.enter_context(open(f)) for f in filenames]
# All files open, all closed on exit regardless of errors
data = [f.read() for f in files]
# Re-entrant context manager
from contextlib import AbstractContextManager
import threading
class ThreadSafeDB(AbstractContextManager):
def __init__(self):
self._lock = threading.RLock() # Reentrant lock
def __enter__(self):
self._lock.acquire()
return self
def __exit__(self, *args):
self._lock.release()
return False
Real-World Patterns
# Feature flag context manager
@contextmanager
def feature_flag(flag_name: str, user_id: int):
enabled = check_feature_flag(flag_name, user_id)
try:
yield enabled
finally:
if enabled:
track_feature_usage(flag_name, user_id)
with feature_flag("new_checkout", user.id) as new_checkout:
if new_checkout:
result = new_checkout_flow()
else:
result = legacy_checkout_flow()
# Retry context manager
@contextmanager
def retry(max_attempts: int = 3, exceptions=(Exception,)):
last_error = None
for attempt in range(max_attempts):
try:
yield attempt
break # success — exit loop
except exceptions as e:
last_error = e
if attempt < max_attempts - 1:
time.sleep(2 ** attempt)
else:
raise last_error
# Note: retry CM is tricky because yield can't be resumed after exception
# Better with a function decorator (see decorators guide)
Python context managers are the right tool for any resource that needs cleanup — files, connections, locks, timers, transactions. Use @contextmanager for simple cases, implement __enter__/__exit__ for reusable classes, and ExitStack for dynamic resources. Async context managers work identically with async with and @asynccontextmanager.
📚 You might also like
🔗 Share this article




✍️ Leave a Comment