A Pytest plugin for automatically prioritizing/ranking tests to speed up failure detection
Project description
pytest-ranking
A Pytest plugin for reducing failure detection time of your Python test suite with automated test prioritization/ranking.
This pytest plugin allows you to find failures faster and receive sooner debugging feedback from CI. It speed up the failure detection of your test suite by prioritizing the execution of tests that are faster, recently failed and/or more textually related to code change.
Installation
To install pytest-ranking via pip from PyPI:
pip install pytest-ranking
Usage
Pytest will automatically find the plugin and use it when you run pytest. You can use the default prioritization heuristic, which runs tests that have shorter execution time first by passing the --rank option:
pytest --rank
The terminal output will tell you the current configurations and runtime overhead of this plugin. For example:
[pytest-ranking] Weights: 1-1-0
[pytest-ranking] History length: 30
[pytest-ranking] Number of files with new hashes: 0
[pytest-ranking] Change relatedness computation time (s): 0.0007872581481933594
[pytest-ranking] Test order computation time(s): 0.00020933151245117188
You can configure the weights of different prioritization heuristics by passing the optional --rank-weight flag with formatted values:
pytest --rank --rank-weight=0-1-0
Weights are separated by hyphens -. The 1st weight is for running faster tests, the 2nd weight is for running recently failed tests, and the 3rd weight is for running tests more textually similar to the changed *.py files in the codebase since the last run.
All weights must be integers or floats, and their sum will be normalized to 1.
A higher weight means that a corresponding heuristic is favored. The default value is 1-0-0, only prioritizes faster tests.
You can also configure the maximum window size for looking into previous test runs, which is used to compute the number of runs since a test had failed, by passing the optional --rank-hist-len flag (the default value is 50):
pytest --rank --rank-hist-len=30
You can always apply these options by adding them to the addopts setting in your pytest.ini.
For example, create pytest.ini in the codebase root folder as such:
[pytest]
addopts = --rank --rank-weight=0-1-0 --rank-hist-len=30
and run pytest on command line.
Alternatively, you can also create pytest.ini in the codebase root folder as such:
[pytest]
rank-weight=0-1-0
rank-hist-len=30
and run pytest --rank on command line.
Warning
Because pytest-ranking re-orders tests to speed up failure detection time, please disable other pytest plugins that enforeces other test orders, e.g., pytest-randomly, pytest-random-order, pytest-reverse.
Contributing
Contributions are very welcome. Tests can be run with tox.
License
Distributed under the terms of the MIT license, pytest-ranking is free and open source software.
Issues
If you encounter any problems, please file an issue or pull request along with a detailed description.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file pytest-ranking-0.2.4.tar.gz.
File metadata
- Download URL: pytest-ranking-0.2.4.tar.gz
- Upload date:
- Size: 10.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.8.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
9601eb1ba56426fabdac57911be06ff346044c9a54c757c8d09a768244946523
|
|
| MD5 |
b694ef9d0777a14962e874b5b29a4513
|
|
| BLAKE2b-256 |
504fe83d1ee79198da560e9a21a8bffea35be7ff6fdc4e18ef7fe3bd7bdc4c0c
|
File details
Details for the file pytest_ranking-0.2.4-py3-none-any.whl.
File metadata
- Download URL: pytest_ranking-0.2.4-py3-none-any.whl
- Upload date:
- Size: 8.8 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.8.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
3e860f5c8e2e462980b5bb80410a08a26b616a066ef013d201522cef02d8e89c
|
|
| MD5 |
96aef9aae030c7c738f3e6d77f29ba9f
|
|
| BLAKE2b-256 |
7f30050f07ae5227af62b50cfa0d3a9de53cc03d23dcb635d9c152ed9467771a
|