How to mock functions in streamlit unit tests?

Question

I have a streamlit app, for which I'd like to create unit tests. I am unable to mock objects though. I have a foo.py with the following content:

import streamlit as st

def dialog_func():
    result = dummy_func()
    st.write(result)

def dummy_func():
    return 'foo'

dialog_func()

and the following corresponding test_foo.py:

from unittest.mock import patch

from streamlit.testing.v1 import AppTest

def test_foo():
    with patch("foo.dummy_func") as mock_dummy_func:
        mock_dummy_func.return_value ='bar'
        at = AppTest.from_file("foo.py").run()
        assert at.markdown[0].value == 'bar'

Running this shows, that patching the method does not work and 'foo' is returned instead of 'bar':

$ pytest test_foo.py 
...
======================================= short test summary info =======================================
FAILED test_foo.py::test_foo - AssertionError: assert 'foo' == 'bar'
========================================== 1 failed in 0.26s ==========================================

94621 · Accepted Answer · 2025-04-09 20:45:03Z

1

You can use some form of dependency injection but it might be difficult for real apps other than this trivial example. Notice the use of from_function() instead, so you have the option to inject the mocked functions.

test_foo.py:

from unittest.mock import patch

import foo
import streamlit as st
from streamlit.testing.v1 import AppTest

def test_foo():
    with patch("foo.dummy_func") as mock_dummy_func:
        mock_dummy_func.return_value ='bar'
        at = AppTest.from_function(foo.main, args=(foo.dialog_func, foo.dummy_func)).run()
        assert at.markdown[0].value == 'bar'

and foo.py:

import streamlit as st

def dialog_func():
    result = dummy_func()
    st.write(result)


def dummy_func():
    return 'foo'


def main(dialog_func, dummy_func):
    import streamlit as st
    dialog_func()

main(dialog_func, dummy_func)

edited Apr 9 at 20:45

94621

11312 bronze badges

answered Apr 8 at 18:49

Diego Torres Milano

69.9k9 gold badges116 silver badges145 bronze badges

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

2 Comments

94621 Apr 9 at 8:32

Thank you very much! I have not tried it on my side yet, but is the modification that you have done to foo.py necessary to make the test case work? Or would the AppTest.from_function(...) approach work with the original foo.py too?

Diego Torres Milano Apr 9 at 15:50

You won't be able to inject the function in the original foo.py

Collectives™ on Stack Overflow

How to mock functions in streamlit unit tests?

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related