Read CSV with Pandas from subdirectory on Windows and UNIX

Question

I am writing a Python script that I need to run on Windows and UNIX systems.

I have two folders, let's call them 'Main' and 'Data'. 'Main' contains 'Data' and my script, and 'Data' contains my data in the form of CSV files.

'Main' --> 'Data' ---> data1.csv, data2.csv....

Currently my code below works if I keep the CSV files in 'Main'. How can I get it to work on both Windows and UNIX, keeping the CSV files in 'Data', if the path naming conventions are different?

Here is the code I am running

import pandas as pd
myfile = "data1.csv" 
df = pd.read_csv(myfile,sep=',')

Possible duplicate of [How to open my files in \`data\_folder\` with pandas using relative path?](https://stackoverflow.com/questions/35384358/how-to-open-my-files-in-data-folder-with-pandas-using-relative-path) — arbitrarystringofletters, Feb 21 '18 at 14:49

score 4 · Accepted Answer · answered Feb 21 '18 at 14:51

4

Python 3 has a library to handle paths (https://docs.python.org/3.6/library/pathlib.html).

So in both cases you would do

from pathlib import Path
import pandas as pd
root = Path('main', 'data')
df = pd.read_csv(root / 'data1.csv')

answered Feb 21 '18 at 14:51

KPLauritzen

1,519
11
20

Could you explain the "root / " syntax? – max Feb 21 '18 at 15:05
1

See https://docs.python.org/3/library/pathlib.html#basic-use. You could also just do `os.path.join('main', 'data', 'data1.csv')` – KPLauritzen Feb 21 '18 at 19:56

Read CSV with Pandas from subdirectory on Windows and UNIX

1 Answers1