Pytest follow-up #1751

djarecka · 2016-12-16T04:58:05Z

This is a follow-up to my previous Pytes PR (#1722), this time I've read the Satra review :)

removed some files from tools directory that are not used anymore
removed fixes
changed .coveragerc (workflows is not omitted)
removed coverage from travis, so it is now calculated from CircleCI only
added one auto tests test_auto_ACompCor.py and changing test_auto_TCompCor.py, so it agrees with checkspec.py output
recovered test_BEDPOSTX.py and test_Xfibres.py
removed nosetes from __init__.py
changed one test from test_math.py that was failing
some smaller changes suggested in the review

Still TODO:

tests test_BEDPOSTX.py and test_Xfibres.py don't affect anything IMO. Should I create test_BEDPOSTX5.py and test_Xfibres5.py, and delete test_auto_BEDPOSTX5.py and test_auto_Xfibres5.py?
__init__.py is not finished, nipype.test() doesn't work

…g the same assert as the previous one

…cle CI only)

…for now

…me lines from travis

codecov-io · 2016-12-16T12:45:22Z

Current coverage is 72.41% (diff: 96.77%)

Merging #1751 into master will increase coverage by 1.30%

@@             master      #1751   diff @@
==========================================
  Files          1028       1055     +27   
  Lines         50616      52708   +2092   
  Methods           0          0           
  Messages          0          0           
  Branches       7333       7665    +332   
==========================================
+ Hits          35994      38170   +2176   
+ Misses        13495      13323    -172   
- Partials       1127       1215     +88

Powered by Codecov. Last update ff36588...3786640

…RT and COVERAGE_DATA_FILE from travis again

satra · 2016-12-16T14:13:04Z

nipype/interfaces/fsl/tests/test_maths.py

-    #NOTE_dj: not sure if this should pass, it uses cmdline from interface.base.CommandLine
-    #assert stder.cmdline == "fslmaths a.nii -Tstd %s"%os.path.join(testdir, "a_std.nii")
+    stder = fsl.StdImage(in_file="a.nii", output_type='NIFTI')
+    assert stder.cmdline == "fslmaths a.nii -Tstd %s"%os.path.join(testdir, "a_std.nii")


use .format()

…re used only locally in test functions)

satra · 2016-12-18T15:01:25Z

nipype/algorithms/tests/test_icc_anova.py

@@ -19,4 +19,4 @@ def test_ICC_rep_anova():
    assert round(icc, 2) == 0.71
    assert dfc == 3
    assert dfe == 15
-    assert r_var / (r_var + e_var) == icc
+    assert np.isclose(r_var / (r_var + e_var), icc)


prefer allclose for compatibility with earlier numpy versions.

satra · 2016-12-18T15:07:10Z

@djarecka - test_bedpostx and test_xfibres simply prevent the autocreation of the auto tests for those which would duplicate the test_auto_*5 at a point in the past we had two versions of these tools. so your current changes are fine and it doesn't need test_*5.py

satra · 2016-12-18T15:08:11Z

@djarecka - if you can finish the nipype.test() this should be ready. i'll also send a PR to your branch with some updated other tests.

djarecka · 2016-12-18T16:21:55Z

@satra even without test_bedpostx and test_xfibres auto tests are NOT created.

djarecka · 2016-12-18T16:27:24Z

@satra I'm a bit stuck with nipype.test(). If you want to merge this ASAP, it might be better if I move nipype.test() to another PR, it somehow independent issue.

satra · 2016-12-18T16:33:23Z

@djarecka - you are correct - the autotests are not created because they are the same exact class as the *5.py classes in this case. so yes, in this case those two tests can be deleted.

satra · 2016-12-18T16:33:50Z

test_BEDPOSTX.py and test_Xfibres.py

…ated anyway)

djarecka · 2016-12-18T16:46:57Z

@satra ok, I've removed the extra tests. Just wasn't sure, if you want to disable auto tests for BEDPOSTX5 and XFibres5.

I also changed earlier %s to format() in test_math in the places where I could do it without changing the code itself (you're using %s style in nipype.interfaces.fsl.maths)

djarecka · 2016-12-18T16:53:41Z

@satra can you also check this assert? https://github.com/nipy/nipype/blob/master/nipype/interfaces/tests/test_base.py#L137
I've removed it since I'm only getting a UserWarning

satra · 2016-12-18T17:35:18Z

@djarecka - i can't seem to figure out why that should raise a TypeError - ok to remove for now.

satra · 2016-12-18T17:36:07Z

regarding .format to slowly start moving towards current python standards - so a new change should always reflect current coding standards.

djarecka added 16 commits December 14, 2016 22:55

restoring 2 tests that had only import statement, changing comments

bb24c59

fixing one assert in test_math

9fb7730

removing one test from interfaces/tests/test_base.py; it was repeatin…

b261208

…g the same assert as the previous one

restring one skipif in interfaces/tests/test_io.py

baa349b

cleaning:removing pdb

74bd3b8

changing a simple assert to np.isclose in test_icc_anova

688774b

adding the auto test for ACompCor

54f496c

changing test_auto_TCompCor.py to the output of checkspecs.py

e988b97

removing scripts from tools directory that are not used anymore

547a386

removing fixes

ce1b3e2

removing coverage from travis (so the coverage is calculated from Cir…

c8ee538

…cle CI only)

fixing the travis file

623f20f

Removing nosetest from nipype/__init__.py; pytest version is missing …

f353fbb

…for now

restore get_info in nipype/__init__.py

d7c6a54

removing workflows from the directories that are omitted

7d77343

trying to fix coverage (codecov doesnt report anything), restoring so…

e981a53

…me lines from travis

it looks like codecov works at the end, removing COVERAGE_PROCESS_STA…

a3cf60d

…RT and COVERAGE_DATA_FILE from travis again

satra reviewed Dec 16, 2016

View reviewed changes

changing %s to format in test_maths (in all places that the strings a…

7d529e0

…re used only locally in test functions)

satra reviewed Dec 18, 2016

View reviewed changes

removing test_BEDPOSTX.py and test_XFibres.py (auto tests wont be cre…

3786640

…ated anyway)

satra mentioned this pull request Dec 18, 2016

Fix/misc test fixes #1757

Merged

satra merged commit 3786640 into nipy:master Dec 24, 2016

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Pytest follow-up #1751

Pytest follow-up #1751

Uh oh!

djarecka commented Dec 16, 2016

Uh oh!

codecov-io commented Dec 16, 2016 •

edited

Loading

Uh oh!

satra Dec 16, 2016

Uh oh!

satra Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

djarecka commented Dec 18, 2016

Uh oh!

djarecka commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

djarecka commented Dec 18, 2016

Uh oh!

djarecka commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

Uh oh!

Pytest follow-up #1751

Pytest follow-up #1751

Uh oh!

Conversation

djarecka commented Dec 16, 2016

Uh oh!

codecov-io commented Dec 16, 2016 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Current coverage is 72.41% (diff: 96.77%)

Uh oh!

satra Dec 16, 2016

Choose a reason for hiding this comment

Uh oh!

satra Dec 18, 2016

Choose a reason for hiding this comment

Uh oh!

satra commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

djarecka commented Dec 18, 2016

Uh oh!

djarecka commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

djarecka commented Dec 18, 2016

Uh oh!

djarecka commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

satra commented Dec 18, 2016

Uh oh!

Uh oh!

codecov-io commented Dec 16, 2016 •

edited

Loading