{"id":125940,"date":"2022-06-23T05:52:34","date_gmt":"2022-06-23T05:52:34","guid":{"rendered":"https:\/\/blog.finxter.com\/?p=435162"},"modified":"2022-06-23T05:52:34","modified_gmt":"2022-06-23T05:52:34","slug":"convert-csv-to-list-of-tuples-in-python","status":"publish","type":"post","link":"https:\/\/sickgaming.net\/blog\/2022\/06\/23\/convert-csv-to-list-of-tuples-in-python\/","title":{"rendered":"Convert CSV to List of Tuples in Python"},"content":{"rendered":"
\n
\n
\n
\n
<\/div>\n<\/p><\/div>\n
\n
<\/div>\n<\/p><\/div>\n
\n
<\/div>\n<\/p><\/div>\n
\n
<\/div>\n<\/p><\/div>\n
\n
<\/div>\n<\/p><\/div>\n<\/p><\/div>\n
\n
\n
<\/div>\n<\/p><\/div>\n
\n
<\/div>\n<\/p><\/div>\n
\n
<\/div>\n<\/p><\/div>\n
\n
<\/div>\n<\/p><\/div>\n
\n
<\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/div>\n
4.5\/5 – (2 votes) <\/div>\n<\/div>\n
\n
\"Convert<\/figure>\n<\/div>\n

Problem Formulation<\/h2>\n

Given a CSV file (e.g., stored in the file with name 'my_file.csv'<\/code>). <\/p>\n

INPUT<\/strong>: file 'my_file.csv'<\/code>\n9,8,7\n6,5,4\n3,2,1<\/code><\/pre>\n
Challenge<\/strong>: How to convert the CSV file to a list of tuples, i.e., putting the row values into the inner tuples<\/a>?<\/p>\n
OUTPUT<\/strong>: Python list of tuples<\/code>\n[(9, 8, 7), (6, 5, 4), (3, 2, 1)]<\/code><\/pre>\n
Method 1: csv.reader()<\/h2>\n
<\/span>\"\"<\/p>\n
\n
Method 1: csv.reader()<\/strong><\/p>\n<\/div>\n<\/div>\n
To convert a CSV file 'my_file.csv'<\/code> into a list of tuples in Python, use csv.reader(file_obj)<\/code> to create a CSV file reader that holds an iterable of lists, one per row. Now, use the list(tuple(line) for line in reader)<\/code><\/a> expression with a generator expression<\/a> to convert each inner list to a tuple.<\/p>\n
Here’s a simple example that converts our CSV file to a nested list using this approach:<\/p>\n
import csv csv_filename = 'my_file.csv' with open(csv_filename) as f: reader = csv.reader(f) lst = list(tuple(line) for line in reader)<\/pre>\n
Output:<\/p>\n
print(lst)\n# [('9', '8', '7'), ('6', '5', '4'), ('3', '2', '1')]<\/pre>\n
Method 2: One-Liner<\/h2>\n
<\/span>\"\"<\/p>\n
\n
Method 2: One-Liner<\/strong><\/p>\n<\/div>\n<\/div>\n
You can also convert a CSV to a list of tuples using the following Python one-liner<\/a> idea:<\/p>\n
Open the file using open()<\/a><\/code>, pass the file object into csv.reader()<\/a><\/code>, and convert the CSV reader object to a list using the list()<\/a><\/code> built-in function in Python with a generator expression<\/a> to convert each inner list to a tuple.<\/p>\n
Here’s how that looks:<\/p>\n
import csv; lst=list(tuple(line) for line in csv.reader(open('my_file.csv'))); print(lst)<\/pre>\n
Concise, isn’t it? \"\ud83d\ude42\"<\/p>\n
The output is the same:<\/p>\n
[('9', '8', '7'), ('6', '5', '4'), ('3', '2', '1')]<\/code><\/pre>\n
By the way: Do you need all inner elements converted to a certain data type such as integer?<\/em><\/p>\n
If so, use the following approach on top of the previously shown list lst<\/code>:<\/p>\n
new_lst = [tuple(int(x) for x in inner) for inner in lst]<\/pre>\n
The converted new_lst<\/code> now contains a list of tuples of ints:<\/p>\n
print(new_lst)\n# [('9', '8', '7'), ('6', '5', '4'), ('3', '2', '1')]<\/pre>\n
Method 3: Pandas<\/h2>\n
<\/span>\"\"<\/p>\n
\n
Method 3: Pandas<\/strong><\/p>\n<\/div>\n<\/div>\n
You can convert a CSV to a list of tuples with Pandas by first readin<\/a>g<\/a> the CSV<\/a> without header line using pd.read_csv('my_file.csv', header=None)<\/code> function and second converting the resulting DataFrame to a nested list using df.values.tolist()<\/a><\/code>. Third, convert<\/a> the nested list to a list of tuples and you’re done.<\/p>\n
Here’s an example that converts the CSV to a Pandas DataFrame and then to a nested raw Python list and then to a list of tuples:<\/p>\n
import pandas as pd # CSV to DataFrame\ndf = pd.read_csv('my_file.csv', header=None) # DataFrame to List of Lists\nlst = df.values.tolist() # List of Lists to List of Tuples:\nnew_lst = [tuple(x) for x in lst] print(new_lst)\n# [(9, 8, 7), (6, 5, 4), (3, 2, 1)]<\/pre>\n
This was easy, wasn’t it? \"\ud83d\ude42\"<\/p>\n
Of course, you can also one-linerize it by chaining commands like so:<\/p>\n
# One-Liner to convert CSV to list of tuples:\nlst = [tuple(x) for x in pd.read_csv('my_file.csv', header=None).values.tolist()]<\/pre>\n
Method 4: Raw Python No Dependency<\/h2>\n
<\/span>\"\"<\/p>\n
\n
Method 4: Raw Python No Dependency<\/strong><\/p>\n<\/div>\n<\/div>\n
If you’re like me, you try to avoid using dependencies if they are not needed. Raw Python is often more efficient and simple enough anyways. Also, you don’t open yourself up to unnecessary risks and complexities.<\/p>\n
Question<\/strong>: So, is there a simple way to read a CSV to a list of tuples in raw Python without external dependencies? <\/p>\n
Sure!<\/p>\n
To read a CSV to a list of tuples in pure Python, open the file using open('my_file.csv')<\/a><\/code>, read all lines into a variable using f.readlines()<\/a><\/code>. Iterate over all lines, strip them from whitespace using strip()<\/a><\/code>, split them on the delimiter ','<\/code> using split(',')<\/a><\/code>, and pass everything in the tuple()<\/a><\/code> function.<\/p>\n
You can accomplish this in a simple list comprehension<\/a> statement like so:<\/p>\n
csv_filename = 'my_file.csv' with open(csv_filename) as f: lines = f.readlines() lst = [tuple(line.strip().split(',')) for line in lines] print(lst)<\/pre>\n
Feel free to check out my detailed video in case you need a refresher on the powerful Python concept list comprehension:<\/p>\n
\n
\n