Source: astropy Version: 4.0.1+post1-3 Severity: serious Tags: ftbfs Justification: fails to build from source (but built successfully in the past) Control: found -1 4.1~rc1-2
Hi,
astropy recently started to FTBFS in sid (and experimental, but not yet
in bullseye). I could reproduce the failure in bullseye with python3-scipy/sid:
=================================== FAILURES ===================================
______________ [doctest] astropy.stats.funcs.binom_conf_interval _______________
631 --------
632 Integer inputs return an array with shape (2,):
633
634 >>> binom_conf_interval(4, 5, interval='wilson')
635 array([0.57921724, 0.92078259])
636
637 Arrays of arbitrary dimension are supported. The Wilson and Jeffreys
638 intervals give similar results, even for small k, N:
639
640 >>> binom_conf_interval([0, 1, 2, 5], 5, interval='wilson')
Expected:
array([[0. , 0.07921741, 0.21597328, 0.83333304],
[0.16666696, 0.42078276, 0.61736012, 1. ]])
Got:
array([[1.38777878e-17, 7.92174125e-02, 2.15973276e-01, 8.33333042e-01],
[1.66666958e-01, 4.20782762e-01, 6.17360116e-01, 1.00000000e+00]])
/build/astropy-4.0.1+post1/.pybuild/cpython3_3.8/build/astropy/stats/funcs.py:640:
DocTestFailure
= 1 failed, 13907 passed, 412 skipped, 49 xfailed, 12 xpassed in 169.79 seconds
=
Andreas
astropy_bullseye+scipy_sid.build.gz
Description: application/gzip
