Repository navigation
Travis Failing in many PRs #17037
Description
Bug report
Travis is showing following errors
======================FAILURES===================================
______________________ test_bbox_inches_tight_raster[pdf] ______________________
[gw1] darwin -- Python 3.7.7 /usr/local/opt/python/bin/python3.7
expected = '/Users/travis/build/matplotlib/matplotlib/result_images/test_bbox_tight/bbox_inches_tight_raster-expected.pdf'
actual = PosixPath('/Users/travis/build/matplotlib/matplotlib/result_images/test_bbox_tight/bbox_inches_tight_raster.pdf')
tol = 0
def _raise_on_image_difference(expected, actual, tol):
tracebackhide = True
err = compare_images(expected, actual, tol, in_decorator=True)
if err:
for key in ["actual", "expected"]:
err[key] = os.path.relpath(err[key])
raise ImageComparisonFailure(
'images not close (RMS %(rms).3f):\n\t%(actual)s\n\t%(expected)s '
% err)
E matplotlib.testing.exceptions.ImageComparisonFailure: images not close (RMS 6.457):
E result_images/test_bbox_tight/bbox_inches_tight_raster_pdf.png
E result_images/test_bbox_tight/bbox_inches_tight_raster-expected_pdf.png
Bug summary
The error is separate from work done in the prs. Maybe we need to change /bbox_inches_tight_raster-expected_pdf.png ?
Code for reproduction
Check travis build for above prs
Activity
timhoffm commented on Apr 5, 2020
This is only failing on OSX. Before changing any reference images, we need to make sure that‘s the correct thing to do. First thing to do is check the generated images to see what‘s different. Somehow we should be able to get those images from Travis, but I don‘t know how.
tacaswell commented on Apr 5, 2020
It is also worth looking through the build history (particularly on the master branch) to see if there is a particular commit where it broke. In general the tests are presumed to be correct and the code is presumed to be wrong. Changing the tests generally indicates an API changed an requires a justification.
QuLogic commented on Apr 7, 2020
From gitter:
@tacaswell:
@QuLogic:
@tacaswell:
@QuLogic
@jklymak:
@QuLogic:
@jklymak:
QuLogic commented on Apr 8, 2020
I can reproduce in a Fedora 32 container with ghostscript 9.52 , but it only changes the result from 5.534 to 6.457. There must be something else different as well. Scratch that, it was just the test image cache.