CODEWITHSUNDEEP
pythonunit-testing
nicksoft
nicksoft

Reputation: 23

Python unit testing for long running functions

I have a long running python method, this function will go inside folders and read all the documents and extract text. Number of output will not going to be a constant as it can change. This process might take from several hours to days. How to write a unit test for such a function, how to check whether its pass or fail as the number of output might change and it takes many hours

Upvotes: 0

Views: 1422

Answers (1)

Laurent LAPORTE
Laurent LAPORTE

Reputation: 22942

This kind of problem can be solve by using test resources. This is a classic problem you can have when your need to test a function on file data.

In your tests directory (you have one, isn’t it?), you can create a resources package.

All you need is to put your test data here. Of course, choose a minimal tree structure: the only necessary files and directories required for your test.

To access your resources directory I recommend using a RESOURCES_DIR variable define as follow:

In your __init__.py file, you can write:

import os

RESOURCES_DIR = os.path.dirname(__file__)

You can easily import this variable in your unit test and use it to access your test files.

Here is a more detailed example:

tests
+-- resources
|   +-- __init__.py
|   \-- test_func1
|       +-- scenario1
|       |   +-- <data for scenario1>
|       \-- scenario2
|           \-- <data for scenario2>
|
+-- test_func1.py

Your test_func1.py can look like this:

import unittest
import os

from tests.resources import RESOURCES_DIR


class TestFunc1(unittest.TestCase):
    def test_func1_scenario1(self):
        data_dir = os.path.join(RESOURCES_DIR, 'test_func1/scenario1')
        ...

    def test_func1_scenario2(self):
        data_dir = os.path.join(RESOURCES_DIR, 'test_func1/scenario2')
        ...

Upvotes: 2

Related Questions