This lesson's interactive features are locked, please to keep using them

CSV Files

CSV is the default "just export it to a file" data format.

It's simple, everywhere, and kind of dumb. That's not an insult; dumb formats are useful formats because everyone understands them.

If you do a lot of work with non-technical stakeholders, you'll work with CSVs via Microsoft Excel or Google Sheets constantly.

This is what CSV data looks like:

title,author,year
Dune,Frank Herbert,1965
The Hobbit,J.R.R. Tolkien,1937
Neuromancer,William Gibson,1984

Each line is a row, and rows are separated by newlines. Fields are separated by commas.

The first row usually contains column names.

Why CSV Is So Common

CSV sticks around because it's:

Easy to export
Easy to share
Easy to open in any spreadsheet software
Easy to parse in almost any programming language

One downside is that CSV has almost no built-in type information. Unlike JSON, everything is text.

Reading a CSV File

Python's built-in csv module can parse CSV files using csv.DictReader:

import csv

with open("books.csv", "r", newline="") as file:
    reader = csv.DictReader(file)
    for row in reader:
        print(row)

Each row comes back as a dictionary:

{"title": "Dune", "author": "Frank Herbert", "year": "1965"}

Working With Open Files

csv.DictReader expects a file-like object: something it can read text from one line at a time.

When you call open(), Python gives you a file object. DictReader reads from that and parses each CSV row as a dictionary.

Always use a with statement when opening files. It closes the file automatically, even if something goes wrong. I also recommend setting newline="" when opening CSV files. This lets the csv module handle newlines itself, which avoids weird blank-line behavior on some systems.

Loading Rows Into a List

If you want the whole file in memory, you can build a list of dictionaries, one for each row:

def load_book_data(filename: str) -> list[dict]:
    books = []
    with open(filename, "r", newline="") as file:
        reader = csv.DictReader(file)
        for row in reader:
            books.append(row)
    return books

Assignment

SnackStack's device telemetry team exports sensor readings as CSV files. Let's make sure we can parse them.

Complete the load_device_readings function. It accepts a CSV file name and returns a list of dictionaries, one for each row.
1. Open the file with a with statement and newline="".
2. Use csv.DictReader to read each row as a dictionary.
3. Return a list of those row dictionaries.
Complete the get_device_count function. It accepts a CSV file name and returns the number of devices in the file.
1. Use your load_device_readings function to load the CSV.
2. Return the length of the list.