Skip to content

NaN-package: v.3.7.2#743

Open
schloegl wants to merge 2 commits intognu-octave:mainfrom
schloegl:main
Open

NaN-package: v.3.7.2#743
schloegl wants to merge 2 commits intognu-octave:mainfrom
schloegl:main

Conversation

@schloegl
Copy link
Copy Markdown
Contributor

2022-08-16: Release of NaN-toolbox 3.7.1

  • kurtosis, skewness:
    add support for FLAG input argument, supporting
    biased and unbiased estimation.
    Compatibility tests with Octave and Matlab are added.
    It will break backwards compatibility, the previous
    version is supported by FLAG=-1.

  • fishers_exact_test:
    extend to non-square contingency tables according to
    https://mathworld.wolfram.com/FishersExactTest.html

@mmuetzel
Copy link
Copy Markdown
Member

Two of the tests fail (unexpectedly?):

>>>>> processing /usr/share/octave/packages/nan-3.7.1/histo2.m
***** assert(getfield(histo2([]),'N'), 0)
!!!!! test failed
ASSERT errors for:  assert (getfield (histo2 ({}), 'N'),0)

  Location  |  Observed  |  Expected  |  Reason
     .          O(1x0)       E(1x1)      Dimensions don't match
>>>>> processing /usr/share/octave/packages/nan-3.7.1/histo3.m
***** assert(getfield(histo3([]),'N'), 0)
!!!!! test failed
ASSERT errors for:  assert (getfield (histo3 ({}), 'N'),0)

  Location  |  Observed  |  Expected  |  Reason
     .          O(1x0)       E(1x1)      Dimensions don't match

Should we ignore these errors? Or would you like to look into these before the release?

@schloegl
Copy link
Copy Markdown
Contributor Author

I'll look into this.

@mmuetzel
Copy link
Copy Markdown
Member

Thank you for looking into the errors.

Fwiw, these tests already failed with version 3.7.0 of the package.
See, e.g, https://github.com/gnu-octave/octave-buildbot/actions/runs/24915756227/job/72967363943#step:15:11902

  >>>>> processing D:\a\octave-buildbot\octave-buildbot\octave-2026-04-24-03-01-w64\mingw64\share\octave\packages\nan-3.7.0\histo2.m
  ***** assert(getfield(histo2([]),'N'), 0)
  !!!!! test failed
  ASSERT errors for:  assert (getfield (histo2 ({}), 'N'),0)
  
    Location  |  Observed  |  Expected  |  Reason
       .          O(1x0)       E(1x1)      Dimensions don't match
  >>>>> processing D:\a\octave-buildbot\octave-buildbot\octave-2026-04-24-03-01-w64\mingw64\share\octave\packages\nan-3.7.0\histo3.m
  ***** assert(getfield(histo3([]),'N'), 0)
  !!!!! test failed
  ASSERT errors for:  assert (getfield (histo3 ({}), 'N'),0)
  
    Location  |  Observed  |  Expected  |  Reason
       .          O(1x0)       E(1x1)      Dimensions don't match

@schloegl
Copy link
Copy Markdown
Contributor Author

It seems the failed tests appear only with Octave 11 - due to change in behaviour of sum([],1). Admittingly, it improves compatibility with Matlab.

I've now fixed these tests and some other issues, and release v3.7.2

@mmuetzel mmuetzel changed the title NaN-package: v.3.7.1 NaN-package: v.3.7.2 Apr 29, 2026
@mmuetzel
Copy link
Copy Markdown
Member

Thank you for the update.

There are two new errors that I didn't see in the test run for 3.7.1:

  >>>>> processing /usr/share/octave/packages/nan-3.7.2/kurtosis.m
  ***** assert (kurtosis (1, [], 3), NaN)
  !!!!! test failed
  sz(3): out of bound 2 (dimensions are 1x2)

And:

  >>>>> processing /usr/share/octave/packages/nan-3.7.2/skewness.m
  ***** assert (skewness (1, [], 3), NaN)
  !!!!! test failed
  sz(3): out of bound 2 (dimensions are 1x2)

@mmuetzel
Copy link
Copy Markdown
Member

Forgot to ask yesterday, sorry:
Should we ignore these new errors and proceed with the release anyway? Or would you prefer to look into them?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants