reassign 985737 pymeeus found 985737 0.3.7-1 fixed 985737 0.4.3+dfsg1-1 thanks
On 2021-03-22 21:29:43, Sebastian Ramacher wrote: > Source: convertdate > Version: 2.3.2-1 > Severity: serious > X-Debbugs-Cc: [email protected] > > The autopkgtests of convertdate fail everywhere: > | test_jwday (tests.test_utils.TestConvertdate) ... ok > | > | ====================================================================== > | ERROR: test_equinox_jd (tests.test_persian.testPersian) (y=1000, jd=2086381) > | ---------------------------------------------------------------------- > | Traceback (most recent call last): > | File > "/tmp/autopkgtest-lxc.ctbufrn7/downtmp/autopkgtest_tmp/tests/test_persian.py", > line 43, in test_equinox_jd > | self.assertEqual(persian.equinox_jd(gyear), jd) > | File "/usr/lib/python3/dist-packages/convertdate/persian.py", line 51, in > equinox_jd > | deltat_jd = mean_jd - Epoch.tt2ut(gyear, 3) / (24 * 60 * 60.) > | File "/usr/lib/python3/dist-packages/pymeeus/Epoch.py", line 1403, in > tt2ut > | dt = 1574.2 + u * ( > | UnboundLocalError: local variable 'u' referenced before assignment > > See > https://ci.debian.net/data/autopkgtest/testing/amd64/c/convertdate/11090578/log.gz Hum. The failure is actually in pymeeus, and, as far as I can tell, the code in the *unstable* version of pymeeus doesn't have that bug. Here's `tt2ut` where `dt` is assigned `1574.2 + u`, with context: elif year >= 500 and year < 1600: u = (year - 1000) / 100.0 dt = 1574.2 + u * ( So I think this bug is actually fixed already, and CI will pass when pymeeus migrates. A. -- Si l'image donne l'illusion de savoir C'est que l'adage pretend que pour croire, L'important ne serait que de voir - Lofofora
