How to import using a path that is a variable in Python

Question

I am trying to make a program that will go through and visit an array of directories and run a program and create a file inside.

I have everything working except that I need to figure out a way to import from a new path each time to get to a new directory.

For example:

L =["directory1", "directory2", "directory3"]
for i in range(len(L)):
   #I know this is wrong, but just to give an idea
   myPath = "parent."+L[i]
   from myPath import file
   #make file... etc.

Obviously when I use myPath as a variable for the path to import, I get an error. I have tried several different ways by searching online through Stack Overflow and reading OS and Sys documentation, but have come to no working result.

Answer 1

You can use 'imp' module to load source code of python scrips

import imp
root_dir = '/root/'
dirs =["directory1", "directory2", "directory3"]

for _dir in dirs:
    module_path = os.path.join(root_dir,_dir,'module.py')

    mod = imp.load_source("module_name", module_path)

    # now you can call function in regular way, like mod.some_func()

Answer 2

The imp module (suggested in the accepted answer) has been deprecated since version 3.4 in favor of importlib.

Instead of using imp.load_source, you can now use importlib.machinery.SourceFileLoader.load_module:

import os
from importlib.machinery import SourceFileLoader


module_path = os.path.join('root_dir', 'parent_dir', 'my_module.py')

# Load your module using the module's path
my_module = SourceFileLoader('my_module', module_path).load_module()

# Now you can access the module's functions by using the dot operator
my_module.my_function()