matplotlib.testing
¶matplotlib.testing.compare
¶Provides a collection of utilities for comparing (image) results.
matplotlib.testing.compare.
compare_float
(expected, actual, relTol=None, absTol=None)[source]¶Deprecated since version 3.0: The compare_float function was deprecated in Matplotlib 3.0 and will be removed in 3.2.
Fail if the floating point values are not close enough, with the given message.
You can specify a relative tolerance, absolute tolerance, or both.
matplotlib.testing.compare.
compare_images
(expected, actual, tol, in_decorator=False)[source]¶Compare two "image" files checking differences within a tolerance.
The two given filenames may point to files which are convertible to
PNG via the converter
dictionary. The underlying RMS is calculated
with the calculate_rms
function.
Parameters: | expected : str
actual :str
tol : float
in_decorator : bool
|
---|
Examples
img1 = "./baseline/plot.png" img2 = "./output/plot.png" compare_images(img1, img2, 0.001):
matplotlib.testing.decorators
¶matplotlib.testing.decorators.
CleanupTest
(**kwargs)[source]¶Bases: object
Deprecated since version 3.0: The CleanupTest class was deprecated in Matplotlib 3.0 and will be removed in 3.2.
setup_class
()¶teardown_class
()¶matplotlib.testing.decorators.
CleanupTestCase
(methodName='runTest')[source]¶Bases: unittest.case.TestCase
A wrapper for unittest.TestCase that includes cleanup operations.
Create an instance of the class that will use the named test method when executed. Raises a ValueError if the instance does not have a method with the specified name.
matplotlib.testing.decorators.
ImageComparisonTest
(**kwargs)[source]¶Bases: matplotlib.testing.decorators.CleanupTest
, matplotlib.testing.decorators._ImageComparisonBase
Deprecated since version 3.0: The ImageComparisonTest class was deprecated in Matplotlib 3.0 and will be removed in 3.2.
Nose-based image comparison class
This class generates tests for a nose-based testing framework. Ideally,
this class would not be public, and the only publicly visible API would
be the image_comparison()
decorator. Unfortunately, there are
existing downstream users of this class (e.g., pytest-mpl) so it cannot yet
be removed.
matplotlib.testing.decorators.
check_figures_equal
(*, extensions=('png', 'pdf', 'svg'), tol=0)[source]¶Decorator for test cases that generate and compare two figures.
The decorated function must take two arguments, fig_test and fig_ref, and draw the test and reference images on them. After the function returns, the figures are saved and compared.
matplotlib.testing.decorators.
cleanup
(style=None)[source]¶A decorator to ensure that any global state is reset before running a test.
Parameters: | style : str, optional
|
---|
matplotlib.testing.decorators.
image_comparison
(baseline_images, extensions=None, tol=0.306, freetype_version=None, remove_text=False, savefig_kwarg=None, style='_classic_test')[source]¶Compare images generated by the test with those specified in baseline_images, which must correspond else an ImageComparisonFailure exception will be raised.
Deprecated since version 3.0: The skip_if_command_unavailable function was deprecated in Matplotlib 3.0 and will be removed in 3.2.
skips a test if a command is unavailable.
Parameters: | cmd : list of str
|
---|
matplotlib.testing.disable_internet
¶matplotlib.testing.disable_internet.
check_internet_off
(original_function)[source]¶Wraps original_function
, which in most cases is assumed
to be a socket.socket
method, to raise an IOError
for any operations
on non-local AF_INET sockets.
matplotlib.testing.disable_internet.
no_internet
(verbose=False)[source]¶Context manager to temporarily disable internet access (if not already
disabled). If it was already disabled before entering the context manager
(i.e. turn_off_internet
was called previously) then this is a no-op and
leaves internet access disabled until a manual call to turn_on_internet
.
matplotlib.testing.disable_internet.
turn_off_internet
(verbose=False)[source]¶Disable internet access via python by preventing connections from being created using the socket module. Presumably this could be worked around by using some other means of accessing the internet, but all default python modules (urllib, requests, etc.) use socket [citation needed].