Create a .csv file with values from a Python list

python

csv-writing

pandas

data-manipulation

byAlex Kataev·Feb 3, 2025

A fast solution to creating a .csv file from a Python list is by using the csv module. You can employ a writer() to set up the CSV writer object. After that, use writerows() with your list to output your rows directly:

import csv

# Your data list
data = [('Name', 'Age'), ('Alice', 25), ('Bob', 22)]

# Life's not complex. Write to your CSV!
with open('output.csv', 'w', newline='') as f:
    csv.writer(f).writerows(data)

This script produces an 'output.csv' file with the list's content.

Python 3.x CSV: unlock potential tricks

Python 3.x has some tucked-away treasures for enhancing your CSV writing:

Use 'w' mode for writing with the added distinction of newline='' to skip extra line spaces.
Set the quoting parameter to customize CSV values, enveloping all fields with quotes using csv.QUOTE_ALL is also an option.

# Example with quoting
with open('output.csv', 'w', newline='') as f:
    writer = csv.writer(f, quoting=csv.QUOTE_ALL)
    writer.writerows(data)

The mighty pandas: beyond simple CSV writing

When data complexity escalates, your ally should be Pandas and its tools for smooth CSV writing:

Apply DataFrame.to_csv() to export your data with ease. Use index=False to drop row indexes.
Use pandas to handle data types and encoding, assuring your CSV's wellbeing.

import pandas as pd

# Data in a DataFrame
df = pd.DataFrame(data[1:], columns=data[0])

# Unload the Pandas to CSV!
df.to_csv('output_with_pandas.csv', index=False)

Remember, Pandas is your one-stop-shop for all data manipulation and provides immense support for CSV nuances.

Simplifying with numpy for sturdy data

Scientific computing demands uncomplicated user experiences, hence Numpy:

numpy.savetxt() swiftly outputs data, separated by a user-defined delimiter and molded by fmt.
Summon numpy.column_stack() to conjure a multi-column CSV from multiple lists.

import numpy as np

# Numeric data lists
column1 = [1, 2, 3]
column2 = [4, 5, 6]

# Conjure the lists into CSV
np.savetxt('output_with_numpy.csv', np.column_stack((column1, column2)), fmt='%d', delimiter=',', header='col1,col2', comments='')

Full control by advanced customization

Some of us like to be in control; Python's stdlib, Pandas, and Numpy respect that when dealing with CSV manipulation:

Gear towards your preference for headers, formatting, or delimiters.
Diagnose your quoting strategies for text fields.
Grasp tight onto varied data types and encodings.

A spoonful of CSV cases

For the `writerows()` lovers: nested lists to multi-row CSV

Each inner list would translate to a row:

# Nested like a Russian doll
nested_data = [['John', 28], ['Sally', 24]]

# A row for each doll, just the way you like it
with open('nested_data.csv', 'w', newline='') as f:
    writer = csv.writer(f)
    writer.writerows(nested_data)

Pandas with structure: DataFrame to CSV

Reflecting more structured data to a CSV is Pandas delight:

# Columnar for the column lovers
data = {
    'Name': ['John', 'Sally'],
    'Age': [28, 24]
}
df = pd.DataFrame(data)

# Keep it neat and tabulate CSV!
df.to_csv('structured_data.csv', index=False)

Do you need to handle special characters and encodings?

Pandas is ready for the challenge:

# DataFrame with a special character
df = pd.DataFrame({
    'Name': ['Jöhn', 'Sälly'],
    'Age': [28, 24]
})

# Confronting special characters in style
df.to_csv('special_chars.csv', index=False, encoding='utf-8')