Home/Learn/Python A–Z/Files & the with Statement

Files & the with Statement

Beginner

Errors & Files

open() + with reads and writes files with guaranteed closing; iterate file objects line by line for constant-memory processing of files of any size.

Overview

File handling is where beginners leak resources and professionals use with: the context manager closes the file even if an exception fires mid-read. Modes: "r" read, "w" write (truncates!), "a" append, "rb"/"wb" binary; always pass encoding="utf-8" for text. The key habit: a file object is an iterator of lines — loop it directly instead of read()-ing gigabytes into memory.

Reading — Line by Line, Not All at Once

for line in f streams the file with constant memory. read() loads everything (fine for small files), readlines() gives a list. strip() the newline that each line keeps.

with + line iteration — the only pattern you need

# students.txt: one "name,marks" per line
with open("students.txt", encoding="utf-8") as f:
    for line in f:                      # streams — file can be 10 GB
        name, marks = line.strip().split(",")
        print(name, int(marks))
# file auto-closed here, even on exceptions

# Small files — grab it all
with open("config.txt", encoding="utf-8") as f:
    text = f.read()

# Multiple files in one with
with open("in.txt") as src, open("out.txt", "w") as dst:
    for line in src:
        dst.write(line.upper())

# Without with (what NOT to do):
# f = open("x.txt"); data = f.read(); f.close()
#  ^ an exception between open and close leaks the handle

Writing & Appending

"w" truncates the file the moment you open it — the classic data-loss mistake; "a" appends. write() does not add newlines; print(file=f) does.

"w" truncates immediately — append is "a"

results = {"asha": 91, "ravi": 78}

with open("report.txt", "w", encoding="utf-8") as f:   # w TRUNCATES
    f.write("Placement Report\n")
    for name, score in results.items():
        f.write(f"{name}: {score}\n")

with open("report.txt", "a", encoding="utf-8") as f:   # append
    print("generated by AiCanCode", file=f)            # print adds \n

# Exists check before overwriting precious data
from pathlib import Path
if Path("report.txt").exists():
    print("careful — already there")

# Binary mode for non-text (images, pickles)
# with open("logo.png", "rb") as f: header = f.read(8)

Key Points to Remember

1Always with open(...) — guaranteed close on success AND exception
2Iterate the file object for constant memory; read() only for small files
3Mode "w" truncates at open; "a" appends; add encoding="utf-8" for text
4write() adds no newline; print(..., file=f) does

Interview Questions

Why is with open() preferred over open()/close()? What guarantees does it give?

EasyWipro

How do you process a 10 GB log file in Python without running out of memory?

MediumFlipkart

Ask Aria about Files & the with Statement

Your personal AI tutor — ask anything about this concept

Revision Status

Personal Notes

Discussion

Loading discussion…

Raising, Custom Exceptions & Chaining

JSON, CSV & pathlib