Closed
Description
dateutil/dateutil#217 was fixed in 2.5.2
similar to fixes in #12613
so need to adjust our tests based on version >= 2.5.2 to handle the new behavior.
the OSX build uses pip to pick up the new version so currently is failing
the 3.5 build uses latest conda, so its still using 2.5.1
======================================================================
ERROR: test_parsers_dayfirst_yearfirst (pandas.tseries.tests.test_tslib.TestDatetimeParsingWrappers)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/Users/jreback/pandas/pandas/tseries/tests/test_tslib.py", line 608, in test_parsers_dayfirst_yearfirst
yearfirst=yearfirst)
File "/Users/jreback/pandas/pandas/tseries/tools.py", line 523, in parse_time_string
yearfirst=yearfirst)
File "pandas/tslib.pyx", line 1765, in pandas.tslib.parse_datetime_string_with_reso (pandas/tslib.c:32262)
raise DateParseError(e)
DateParseError: month must be in 1..12