A pytest plugin for image snapshot management and comparison.
- Image Comparison: Automatically compares a test-generated image with a pre-stored snapshot, identifying any visual discrepancies.
- Snapshot Creation: If a reference snapshot doesn't exist, the plugin will create it during the test run, making initial setup effortless.
- Verbose Mode Display: Capable of displaying the difference image for quick visual feedback in case of mismatches when running tests with
-v
. - Snapshot Update Option: Includes a
--image-snapshot-update
flag to update existing snapshots or create new ones, accommodating visual changes in your project. - Threshold-Based Comparison: Utilizes the
threshold
argument for enhanced image comparison with thepixelmatch
library, enabling anti-aliasing pixel detection.
- Pillow
- pixelmatch (optional)
You can install "pytest-image-snapshot" via pip from PyPI:
$ pip install pytest-image-snapshot
pytest-image-snapshot
offers enhanced functionality with the optional pixelmatch package, suitable for advanced image comparison scenarios. To install pytest-image-snapshot
along with this optional feature, use the following command:
$ pip install pytest-image-snapshot[pixelmatch]
The image_snapshot
fixture is designed for visual regression testing in pytest. It compares a generated image in your tests with a stored reference image (snapshot). If the snapshot doesn't exist, it will be automatically created. This makes the fixture ideal for both creating initial snapshots and for ongoing comparison in visual tests.
Here's a example of how to utilize the image_snapshot
fixture:
from PIL import Image
def test_image(image_snapshot):
# Create a new white image of 100x100 pixels
image = Image.new('RGB', (100, 100), 'white')
# Compare it to the snapshot stored in test_snapshots/test.png
# If test_snapshots/test.png does not exist, it will be created
image_snapshot(image, "test_snapshots/test.png")
The image_snapshot
function includes an optional threshold
argument. When set, and if the image does not match the snapshot, the pixelmatch
library is used for a detailed comparison with anti-aliasing pixel detection.
- Default Threshold: If
threshold
is set toTrue
, a default threshold value is utilized. - Custom Threshold: If
threshold
is a numeric value, it is passed to thepixelmatch
library to specify the tolerance level for image comparison.
image_snapshot(image, "test_snapshots/test.png", True)
image_snapshot(image, "test_snapshots/test.png", 0.2)
⚠️ Warning:The
image_snapshot
fixture does not automatically create directories for storing image snapshots. Ensure that the necessary directories (e.g.,test_snapshots/
) are created in your project structure before running tests.
The verbose mode enhances the output detail for image_snapshot
tests:
-v
: Displays the 'diff' image when there's a mismatch.-vv
: Shows all three images - 'diff', 'original', and 'current' for a comprehensive comparison.
This feature assists in quickly identifying and analyzing visual differences during test failures.
Use the --image-snapshot-save-diff
flag to save the actual image and the diff image when there's a mismatch. This is particularly useful for debugging in a CI environment.
pytest --image-snapshot-save-diff
Use the --image-snapshot-update
flag to update or create new reference snapshots. This is useful for incorporating intentional visual changes into your tests, ensuring that your snapshots always reflect the current expected state.
pytest --image-snapshot-update
Use the --image-snapshot-fail-if-missing
flag to fail the test when the snapshot is missing. This is particularly useful in CI check to ensure that all snapshots are present and up-to-date.
pytest --image-snapshot-fail-if-missing
Visual regression test for Django application home page with playwright:
from PIL import Image
from io import BytesIO
def test_homepage(live_server, page: Page, image_snapshot):
page.goto(f"{live_server}")
# convert screenshot to image
screenshot = Image.open(BytesIO(page.screenshot()))
image_snapshot(screenshot, "test_snapshots/homepage.png", threshold=True)
Contributions are very welcome. Tests can be run with tox, please ensure the coverage at least stays the same before you submit a pull request.
Distributed under the terms of the MIT license, "pytest-image-snapshot" is free and open source software
If you encounter any problems, please file an issue along with a detailed description.
This pytest plugin was generated with Cookiecutter along with @hackebrot's cookiecutter-pytest-plugin template.