pytest-needle is a pytest implementation of needle.
It's fairly similar to needle and shares much of the same functionality, except it uses pytest-selenium for handling the webdriver and implements needle as a fixture instead of having test cases inherit from needle's base test class.
Install through pip:
pip install pytest-needle
Install from source:
cd /path/to/source/pytest-needle python setup.py install
Example needle pytest implementation:
"""test_example.py """ from selenium.webdriver.common.by import By import pytest @pytest.mark.element def test_example_element(needle): """Example for comparing individual elements :param NeedleDriver needle: NeedleDriver instance :return: """ # Navigate to web page needle.driver.get('https://www.google.com') # Take an element screen diff needle.assert_screenshot('search_field', (By.ID, 'tsf'))
To create a baseline for all subsequent test run:
pytest --driver Chrome --needle-save-baseline test_example.py
After we have a baseline, to run test use:
pytest --driver Chrome test_example.py
Selecting a WebDriver
To control which browser to use, use
--driver <BROWSER> from pytest-selenium. For example to change to browser to Firefox:
pytest --driver Firefox test_example.py
Setting the viewport's size
You may set the size of the browser's viewport using the
set_viewport_size() on the needle fixture
def test_example_viewport(needle): # Navigate to web page needle.set_viewport_size(width=1024, height=768) # Rest of the test ...
You may also set the default viewport size for all your tests by using the command line argument
pytest --driver Chrome --needle-viewport-size "1024 x 768" test_example.py
Sometimes areas on a web page may contain dynamic content and cause false negatives, or worse convince testers to raise the threshold at which changes are acceptable. You can instead choose to mask these areas to avoid the issue of consistently failing tests:
"""test_example.py """ from selenium.webdriver.common.by import By import pytest @pytest.mark.mask def test_example_page_with_mask(needle): """Example for comparing page with a mask :param NeedleDriver needle: NeedleDriver instance :return: """ # Navigate to web page needle.driver.get('https://www.google.com') # Take a entire page screen diff, ignore the doodle banner needle.assert_screenshot('search_page', threshold=60, exclude=[(By.ID, 'hplogo'), (By.ID, 'prm')])
In the case with Google's home page the doodle banner frequently changes, so to visually regress day-to-day requires generating new baselines every time the banner is updated. Masking allows only the banner to be ignored while the rest of the page can be evaluated.
By default Needle uses the PIL engine (
needle.engines.pil_engine.Engine) to take screenshots. Instead of PIL, you may also use PerceptualDiff or ImageMagick.
Example with PerceptualDiff:
pytest --driver Chrome --needle-engine perceptualdiff test_example.py
Example with ImageMagick:
pytest --driver Chrome --needle-engine imagemagick test_example.py
Besides being much faster than PIL, PerceptualDiff and ImageMagick also generate a diff PNG file when a test fails, highlighting the differences between the baseline image and the new screenshot.
Note that to use the PerceptualDiff engine you will first need to download the perceptualdiff binary and place it in your PATH.
To use the ImageMagick engine you will need to install a package on your machine (e.g. sudo apt-get install imagemagick on Ubuntu or brew install imagemagick on OSX).
Each time you run tests, Needle will create new screenshot images on disk, for comparison with the baseline screenshots. It’s then up to you whether you want to delete them or archive them. To remove screenshots from successful test use:
pytest --driver Chrome --needle-cleanup-on-success test_example.py
Any unsuccessful tests will remain on the file system.
To specify a path for baseline image path use:
pytest --driver Chrome --needle-baseline-dir /path/to/baseline/images
Default path is ./screenshots/baseline
To specify a path for output image path use:
pytest --driver Chrome --needle-output-dir /path/to/output/images
Default path is ./screenshots
Generating HTML reports
To generate html reports use:
pytest --driver Chrome --html=report.html --self-contained-html
Special thanks to BrowserStack for providing automated browser testing, at no charge, for this project and other open source projects like this. With over 1000+ device, browser and os versions combinations to choose from and integrations with Travis CI this project could not be successful without the hard work of the BrowserStack team and their continued support of the open source community.