Closed tamarinvs19 closed 11 months ago
Hm, test breaks now - the caldav.elements.cdav._to_utc_date_string
has a return ts.strftime("%Y%m%dT%H%M%SZ")
, so I find it strange that it returns 10101T000000Z
and not 00010101T000000Z
Hm, test breaks now - the
caldav.elements.cdav._to_utc_date_string
has areturn ts.strftime("%Y%m%dT%H%M%SZ")
, so I find it strange that it returns10101T000000Z
and not00010101T000000Z
Oh, it is a feature of linux CPython implementation. I rewrote regression test with .strftime
.
Oh, it is a feature of linux CPython implementation.
Or perhaps it's a bug? :-)
It's bad if different python implementations returns different things, so it should probably be reported upstream (but I don't have time doing any research into that now).
-- Tobias Brox Senior System Consultant Redpill Linpro AS - Changing the game Mobil: +47 917 000 50 Kontor: +47 215 441 68
Oh, it is a feature of linux CPython implementation. Or perhaps it's a bug? :-) It's bad if different python implementations returns different things, so it should probably be reported upstream (but I don't have time doing any research into that now).
I agree, it is really interesting! We will try to contact python developers about it
We are developing an open-source automatic test generation tool for Python. We have run the tool on a number of open source projects including your. We have generated a test that leads to an
OverflowError
in functioncaldav.elements.cdav._to_utc_date_string
.This PR fixes this problem.
The first test is error suite before fix, the second - regression suite after fix.