Add some tests

Local tests

These tests are for development purpose, to ensure that the features that are developped are working as expected. They are present in the directory src/tests and can be run with the following command, from the src directory.

python3 -m tests.test_<feature>

A file test_<feature>.py contains the plotly figures that are used to test the feature <feature>. Each desired plot should be a function that returns the figure and a string that is the expected output:

def fig1():
    fig = px.imshow([[1, 20, 30],
                    [20, 1, 60],
                    [30, 60, 1]])
    return fig, "Heatmaps with Plotly Express"

Then set a list functions that contains the list of all the plot functions that you want to test:

    functions = [
        ("1", fig1),
        # ("2", fig2),  # you can comment a line to not run the test
        ("3", fig3)
    ]

Then the main function will run all the selected tests, and generate a TeX file in a subdirectory outputs/test_<feature>, named main.tex. Compile this TeX file with your favorite LaTeX compiler to see the result.

Tests for the CI

These tests are run on the CI to check that the generated LaTeX code is still the same after the modifications. They are placed in teh directory tests at the root of the repository. When tox is run (see here), they are the tests that are run, not the ones on the present section.

Here is an example of a test that is run on the CI:

def plot_1():
    fig = px.imshow([[1, 20, 30],
                    [20, 1, 60],
                    [30, 60, 1]])
    return fig

def test_1():
    assert_equality(plot_1(), os.path.join(this_dir, test_name, test_name + "_1_reference.tex"))

The reference TeX code should be inserted in a file test_<test_name>_<number>_reference.tex in the directory tests/<test_name>.

Note

The reference code should not contain the header % This file was created with tikzplotly version 0.1.7., as the version number is bound to change. This header can be hidden with the option include_disclamer=False in the function tikzplotly.save.