How do I read and write CSV files with Python?

Question

I have a file example.csv with the contents

1,\"A towel,\",1.0
42,\" it says, \",2.0
1337,is about the most ,-1
0,massively useful thing ,123
-2         


        

Answer 1

Here are some minimal complete examples how to read CSV files and how to write CSV files with Python.

Python 3: Reading a CSV file

Pure Python

import csv

# Define data
data = [
    (1, "A towel,", 1.0),
    (42, " it says, ", 2.0),
    (1337, "is about the most ", -1),
    (0, "massively useful thing ", 123),
    (-2, "an interstellar hitchhiker can have.", 3),
]

# Write CSV file
with open("test.csv", "wt") as fp:
    writer = csv.writer(fp, delimiter=",")
    # writer.writerow(["your", "header", "foo"])  # write header
    writer.writerows(data)

# Read CSV file
with open("test.csv") as fp:
    reader = csv.reader(fp, delimiter=",", quotechar='"')
    # next(reader, None)  # skip the headers
    data_read = [row for row in reader]

print(data_read)

After that, the contents of data_read are

[['1', 'A towel,', '1.0'],
 ['42', ' it says, ', '2.0'],
 ['1337', 'is about the most ', '-1'],
 ['0', 'massively useful thing ', '123'],
 ['-2', 'an interstellar hitchhiker can have.', '3']]

Please note that CSV reads only strings. You need to convert to the column types manually.

A Python 2+3 version was here before (link), but Python 2 support is dropped. Removing the Python 2 stuff massively simplified this answer.

Related

• How do I write data into csv format as string (not file)?
• How can I use io.StringIO() with the csv module?: This is interesting if you want to serve a CSV on-the-fly with Flask, without actually storing the CSV on the server.

mpu

Have a look at my utility package mpu for a super simple and easy to remember one:

import mpu.io
data = mpu.io.read('example.csv', delimiter=',', quotechar='"', skiprows=None)
mpu.io.write('example.csv', data)

Pandas

import pandas as pd

# Read the CSV into a pandas data frame (df)
#   With a df you can do many things
#   most important: visualize data with Seaborn
df = pd.read_csv('myfile.csv', sep=',')
print(df)

# Or export it in many ways, e.g. a list of tuples
tuples = [tuple(x) for x in df.values]

# or export it as a list of dicts
dicts = df.to_dict().values()

See read_csv docs for more information. Please note that pandas automatically infers if there is a header line, but you can set it manually, too.

If you haven't heard of Seaborn, I recommend having a look at it.

Other

Reading CSV files is supported by a bunch of other libraries, for example:

• dask.dataframe.read_csv
• spark.read.csv

Created CSV file

1,"A towel,",1.0
42," it says, ",2.0
1337,is about the most ,-1
0,massively useful thing ,123
-2,an interstellar hitchhiker can have.,3

Common file endings

.csv

Working with the data

After reading the CSV file to a list of tuples / dicts or a Pandas dataframe, it is simply working with this kind of data. Nothing CSV specific.

Alternatives

• JSON: Nice for writing human-readable data; VERY commonly used (read & write)
• CSV: Super simple format (read & write)
• YAML: Nice to read, similar to JSON (read & write)
• pickle: A Python serialization format (read & write)
• MessagePack (Python package): More compact representation (read & write)
• HDF5 (Python package): Nice for matrices (read & write)
• XML: exists too *sigh* (read & write)

For your application, the following might be important:

• Support by other programming languages
• Reading / writing performance
• Compactness (file size)

See also: Comparison of data serialization formats

In case you are rather looking for a way to make configuration files, you might want to read my short article Configuration files in Python