Skip to content

Different results with tzlib and dateutil_tz with "ambiguous" argument to tz_localize  #16234

Closed
@toobaz

Description

@toobaz
Member

Code Sample, a copy-pastable example if possible

In [2]: import pytz

In [3]: pytz_tz = pytz.timezone('US/Eastern')
   ...: dateutil_tz = pd._libs.tslib.maybe_get_tz('dateutil/US/Eastern')
   ...: 

In [4]: idx = pd.DatetimeIndex(['2011-11-06 01:00:00'])

In [5]: l1 = idx.tz_localize(pytz_tz, ambiguous=[1])

In [6]: l2 = idx.tz_localize(dateutil_tz, ambiguous=[1])

In [7]: l1 == l2
Out[7]: array([ True], dtype=bool)

In [8]: l1[0] == l2[0]
Out[8]: True

In [9]: str(l1[0]) == str(l2[0])
Out[9]: False

In [10]: l1, l2
Out[10]: 
(DatetimeIndex(['2011-11-06 01:00:00-04:00'], dtype='datetime64[ns, US/Eastern]', freq=None),
 DatetimeIndex(['2011-11-06 01:00:00-05:00'], dtype='datetime64[ns, tzfile('/usr/share/zoneinfo/US/Eastern')]', freq=None))

Problem description

I guess (and the existing tests suggest) the two should produce the same results.

Expected Output

l1 and l2 should be perfectly equivalent I guess.

Output of pd.show_versions()

INSTALLED VERSIONS ------------------ commit: 1002cc3 python: 3.5.3.final.0 python-bits: 64 OS: Linux OS-release: 4.7.0-1-amd64 machine: x86_64 processor: byteorder: little LC_ALL: None LANG: it_IT.utf8 LOCALE: it_IT.UTF-8

pandas: 0.20.0rc1+53.g1002cc339
pytest: 3.0.6
pip: 9.0.1
setuptools: 33.1.1
Cython: 0.25.2
numpy: 1.12.1
scipy: 0.18.1
xarray: 0.9.1
IPython: 5.1.0.dev
sphinx: 1.4.9
patsy: 0.3.0-dev
dateutil: 2.5.3
pytz: 2016.7
blosc: None
bottleneck: 1.2.0
tables: 3.3.0
numexpr: 2.6.1
feather: 0.3.1
matplotlib: 2.0.0
openpyxl: 2.3.0
xlrd: 1.0.0
xlwt: 1.1.2
xlsxwriter: 0.9.6
lxml: 3.7.1
bs4: 4.5.3
html5lib: 0.999999999
sqlalchemy: 1.0.15
pymysql: None
psycopg2: None
jinja2: 2.8
s3fs: None
pandas_gbq: None
pandas_datareader: 0.2.1

Activity

jreback

jreback commented on May 6, 2017

@jreback
Contributor

no, these are not equivalent in reality. yes in theory they should be but I doubt that there is actual equivalence between these different timezone databases. nor do we actually care. you can either use one or the other. anything else would be way too complicated.

if you want to offer a doc note somewhere I guess that would be ok.

added this to the won't fix milestone on May 6, 2017
modified the milestone: won't fix on Jul 6, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    TimezonesTimezone data dtype

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

      Development

      No branches or pull requests

        Participants

        @jreback@toobaz@TomAugspurger

        Issue actions

          Different results with tzlib and dateutil_tz with "ambiguous" argument to tz_localize · Issue #16234 · pandas-dev/pandas