Reading and writing files

Writing and reading a file

A filefileData saved under a name, like notes.txt, that stays after the program ends. keeps text after the program that wrote it has ended. This program writes two lines to notes.txt, then reads the file back:

with open("notes.txt", "w", encoding="utf-8") as f:
    f.write("ቡና 25\n")
    f.write("ሻይ 10\n")

with open("notes.txt", encoding="utf-8") as f:
    text = f.read()
print(text)

Run it and open Files: notes.txt is marked new, with both lines in it. The empty line at the end of the output is the file's last \n, followed by the one print adds.

open() takes the file's name, a modemodeHow open() opens a file: "r" to read (the default), "w" to write from empty, "a" to add at the end., and an encodingencodingThe rule for turning characters into bytes and back (lesson 6.3). For files, always write encoding="utf-8".. "w" means write. With no mode, as in the second open(), the mode is "r", read. The encoding is the rule from lesson 6.3 for turning characters into bytes; "utf-8" saves Amharic correctly.

The withwithA statement that runs the block under it and cleans up when the block ends, even if an error happens. with open(...) as f: closes the file at the end of the block. statement does the rest: open() opens the file, as f names the open file f for the indented block, and when the block ends, with closes it. f.write() writes exactly the text you give it and nothing more, so each line needs its own \n. f.read() returns the whole file as one string.

Write, or add?

This time notes.txt already has two lines (shown under the code). Predict what this prints:

with open("notes.txt", "w", encoding="utf-8") as f:
    f.write("three")

with open("notes.txt", encoding="utf-8") as f:
    print(f.read())

one
two

three

If you picked the first answer, you read "write" as "add". If you picked the last, you pictured the new text replacing only the first line. In fact, "w" empties the whole file as soon as it opens it, before anything is written. That's what you want for saving a fresh copy, and a disaster for a log you meant to keep.

To add to the end, open with "a", append:

with open("notes.txt", "a", encoding="utf-8") as f:
    f.write("three\n")

with open("notes.txt", encoding="utf-8") as f:
    print(f.read())

one
two

"w" and "a" both create the file if it doesn't exist. Change "a" to "w" and run it again to see the difference in Files.

Reading line by line

A file isn't a list you can read again and again. Predict:

with open("notes.txt", encoding="utf-8") as f:
    first = f.read()
    second = f.read()
print(repr(first))
print(repr(second))

one
two

'one\ntwo\n'
''

If you expected the text twice, you pictured the file like a list. An open file keeps a position: reading moves it forward, and after the first read() it's at the end. To read again, open the file again.

readlines() reads every line into a list, and the first line of output below shows that each line keeps its \n. To work with the lines one at a time, loop over the file instead; the lines keep their \n there too, so strip() each one (lesson 6.1):

with open("visitors.txt", encoding="utf-8") as f:
    print(f.readlines())

with open("visitors.txt", encoding="utf-8") as f:
    for line in f:
        print(line.strip())

Hana
Abel
Selam

Closing, and files that aren't there

with closes the file even when an error happens inside the block, like finally from lesson 7.2. Here int() fails, and the file is closed anyway (f.closed is True once a file is closed):

try:
    with open("count.txt", encoding="utf-8") as f:
        count = int(f.read())
except ValueError:
    print("Not a number")
print(f.closed)

twelve

A file you never close may not have all its text saved, and other programs may not be able to use it. With with, you can't forget.

Reading a file that doesn't exist is the first error every program with files meets:

Why encoding="utf-8" every time? Without it, Python uses the computer's default encoding. On macOS and Linux that's UTF-8, but on Windows, up to Python 3.14, it's usually an older one that has no Ge'ez letters: writing Amharic stops with a UnicodeEncodeError, and reading a UTF-8 file usually stops with a UnicodeDecodeError (or shows garbage). Python 3.15, due in October 2026, makes UTF-8 the default everywhere, but writing it yourself works on every version and every computer.