Files not being saved in the right directory in Python

Question

I have 2 files:

First file test1.py:

from dir2.test2 import func2

def func1():
   with open("test1.txt", "w") as f:
       f.write("some text")

func1()
func2()

Second file test2.py:

def func2():
    with open("test2.txt", "w") as f:
        f.write("some text")

And my directory structure looks like this before running test1.py:

dir1
 |
 - test1.py
 |
 - dir2
    |
    - test2.py

After running the test1.py the directory structure look like this:

dir1
 |
 - test1.txt
 | 
 - test2.txt
 |
 - test1.py
 |
 - dir2
    |
    - test2.py

But after running the script I was expecting the directory structure to look like this:

dir1
 |
 - test1.txt
 |
 - test1.py
 |
 - dir2
    | 
    - test2.txt
    |
    - test2.py

I have tried searching ways to fix this but haven't found any.

So is there anyway to get something like I was expecting after running test1.py.

You need to handle the path yourself. Look at stackoverflow.com/questions/32470543/… — kumarchandresh
– kumarchandresh, Commented Nov 23, 2020 at 5:37
@chandr3sh Since my code is being deployed on a different server on Heroku, I cannot get the full path — wind
– wind, Commented Nov 23, 2020 at 5:38
The path is all relative (as per your code) and depends on where you are running the script from. In "test2.py", you may want to specify the path as ./dir2/test2.txt considering that you are running the script from within dir1 — Melvin Abraham
– Melvin Abraham, Commented Nov 23, 2020 at 5:39
When you first run the script from main, it frames you in a specific location - the current working directory. You may be calling test1.py like so: python /home/dev/dir1/test1.py or C:\Users\You\dev\dir1\test1.py. Python takes in func2 and reads the relative path in the open() call relative to the current working directory (test1.py). This is why it saves the test2.txt file in dir1, not dir2. — PeptideWitch
– PeptideWitch, Commented Nov 23, 2020 at 5:40

Chase · Accepted Answer · 2020-11-23 05:51:11Z

2

Relative file paths like "test1.txt" are relative to the directory you run the script from. Since, you run test1.py from the same directory as itself - the relative path is resolved to the same directory as well. So test1.txt becomes /path/to/dir1/test1.txt

If you ran test1.py from within dir2, you'd see both text files in /path/to/dir2/.

This is how relative paths work in pretty much every language. If you want reliable path building, use os.path functions

cwd = os.path.dirname(os.path.abspath(__file__))

This will give you an absolute path to the file you use this on. So if you use this on test1.py - you get /path/to/dir1. Now you can use os.path.join to build your desired paths

Just for a complete example, to make test2.txt inside dir2 (which is inside dir1), assuming that cwd line is resolved in test1.py (i.e __file__ points to test1.py) you simply do

test2_path = os.path.join(cwd, 'dir2', 'test2.txt')
with open(test2_path, 'w') as f:
    ....

And, if cwd is resolved in test2.py (__file__ points to test2.py), you do

test2_path = os.path.join(cwd, 'test2.txt')

edited Nov 23, 2020 at 5:51

answered Nov 23, 2020 at 5:41

Chase

5,6552 gold badges21 silver badges46 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

wind Over a year ago

Thanks for the detailed explanation, but what shall my code look like with cwd = os.path.dirname(os.path.abspath(__file__)) implemented?

PeptideWitch Over a year ago

cwd stands for current working directory, by the way - this is a useful way to think about where code is being called and where relative paths point from

Chase Over a year ago

@AbhigyanJaiswal As I said, cwd is path to the file you call it in. You can now use os.path.join(cwd, 'dir2') to get /path/to/dir1/dir2, which you can use in open

PeptideWitch Over a year ago

os.getcwd() could also work assuming the code always starts off from test1.py as the main call.

Chase Over a year ago

@AbhigyanJaiswal I explicitly mentioned you need to join with dir2 if __file__ is resolved in test1.py, but you put it in test2.py. Please read on how os.path.dirname works - if you do os.path.dirname(os.path.abspath(__file__)) on test2.py, it returns /path/to/dir1/dir2. So you should be doing os.path.join(cwd, 'test2.txt')

Collectives™ on Stack Overflow

Files not being saved in the right directory in Python

1 Answer 1

5 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related