pytest approval
July 3, 2026 · View on GitHub
A simple approval test library utilizing external diff programs such as Meld, PyCharm and Visual Studio Code to compare approved and received output.
About
Approval tests capture the output (a snapshot) of a piece of code and compare it with a previously approved version of the output (the expected result).
It's most useful in environments where frequent changes are common or where the output is of a complex nature but can be easily verified by humans, aided for example by a diff-tool or a visual representation of the output (think of an image).
Once the output has been approved then as long as the output stays the same the test will pass. A test fails if the received output is not identical to the approved version. In that case, the difference between the received and the approved output is reported to the tester.
For outputs that can be represented by text, a report can be as simple as printing the difference to the terminal. Using diff programs with a graphical user interface such as Meld, PyCharm or Visual Studio Code as reporter not only helps to visualize the difference, but they can also be used as approver by applying the changes of the received output to the approved output.
Not all data can or should be represented by text. In many cases an image is the best and most easily verifiable representation. PyCharm and Visual Studio Code can work with images as well.
A picture’s worth a 1000 tests (approvaltests.com).
Requirements
OS
- Linux/Unix
- macOS
One of following programs installed:
- PyCharm
- Visual Studio Code
- Meld
- GNU Diffutils (
diff)
Installation
uv add pytest-approval
# Including image support
uv add --optional image pytest-approval
# Including plotly support
uv add --optional plotly pytest-approval
Usage
Verify Text
from pytest_approval import verify, verify_json
def test_verify_as_string():
assert verify("Hello World!")
def test_verify_as_json():
# automatic conversion to JSON
assert verify_json({"msg": "Hello World!"})
# works with string as well
assert verify_json('{"msg": "Hello World!"}')
Verify Images
To report images visually make sure PyCharm or Visual Studio Code is installed.
from PIL import Image
from pytest_approval import verify_image, verify_image_pillow
def test_verify_image(image):
image = Image.open("my_image.jpg")
assert verify_image(image, extension=".jpg", content_only=True)
def test_verify_image_pillow(image):
image = Image.open("my_image.jpg")
assert verify_image_pillow(image, extension=".jpg")
Verify Plotly Figures
For comparison the JSON representation of a Plotly figure is used and for reporting the image representation.
from pytest_approval import verify_plotly
import plotly.graph_objects as go
FIGURE = go.Figure(
data=go.Contour(
z=[
[10, 10.625, 12.5, 15.625, 20],
[5.625, 6.25, 8.125, 11.25, 15.625],
[2.5, 3.125, 5.0, 8.125, 12.5],
[0.625, 1.25, 3.125, 6.25, 10.625],
[0, 0.625, 2.5, 5.625, 10],
]
)
)
def test_verify_plotly():
assert verify_plotly(FIGURE)
Force Reporting
During development its sometimes helpful to report even though both are equal:
from pytest_approval import verify
def test_verify_string():
assert verify("Hello World!", report_always=True)
Auto Approval
It is possible to automatically approve every tests:
uv run pytest --auto-approve
This is useful for elimination of approval files which are not in use anymore.
- Make sure tests are green.
- Then remove all approval files.
- Run pytest in auto approval mode.
Configuration
Approved and received files are stored next to the test file per default.
If you want to save those files in a specific directory instead, please set the approvals-dir key in your pyproject.toml:
[tool.pytest-approval]
"approvals-dir"="tests/approvals"
The path is relative to pytest root (usually pyproject.toml).
Development
uv sync --all-extras
uv run prek install # pre-commit hooks
uv run pytest
uv run pytest --markdown-docs -m markdown-docs README.md
Release
This project uses SemVer.
To make a new release run ./scripts/release.sh <version>.
Alternatives
- Syrupy is a zero-dependency pytest snapshot plugin. It enables developers to write tests which assert immutability of computed results.
- ApprovalTests.Python is an open source assertion/verification library to aid testing.
Comparison with ApprovalTests.Python
ApprovalTests.Python and ApprovalTests in general are the main inspiration for this library. The goal of this library is to provide a much smaller, simpler and maintainable code base while at the same time providing a simpler interface.
In contrast to ApprovalTests.Python this library features:
- Default naming of files are based on PyTest node ID and works with multiple calls to
verifyand with parametrized tests out of the box: ApprovalTests.Python Documentation - Reporters are blocking: ApprovalTests.Python Issue
- A list of programs for different operating systems are provided as reporters and the first working reporter is used without any configuration: -> ApprovalTests.Python Issue
- On approve the test goes green immediately instead of failing first and succeeding only after another run.
- No HTTP request to fetch empty binary files during testing: ApprovalTests.Python Call to EmptyFiles & EmptyFiles HTTP Request
- Modern Python project: Usage of
uv,ruffandpytest(No shell scripts needed for setup or running tests.) - Configuration happens through
pyproject.toml: ApprovalTests.Python Documentation - Auto approve mode via parameter to PyTest:
pytest --auto-approve