pandas-0.15.0 test failure test_string_index_alias_tz_aware (pandas.tseries.tests.test_timezones.TestTimeZoneSupportDateutil) #8639

jlec · 2014-10-26T19:39:28Z

ERROR: test_string_index_alias_tz_aware (pandas.tseries.tests.test_timezones.TestTimeZoneSupportDateutil)

Traceback (most recent call last):
File "/var/tmp/portage/dev-python/pandas-0.15.0/work/pandas-0.15.0-python3_3/lib/pandas/tseries/tests/test_timezones.py", line 606, in test_string_index_alias_tz_aware
self.assertAlmostEqual(result, ts[2])
File "/usr/lib64/python3.3/unittest/case.py", line 665, in assertAlmostEqual
if first == second:
File "/var/tmp/portage/dev-python/pandas-0.15.0/work/pandas-0.15.0-python3_3/lib/pandas/core/generic.py", line 692, in nonzero
.format(self.class.name))
ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

jreback · 2014-10-26T19:48:59Z

pls pd.show_versions()

jlec · 2014-12-06T08:41:22Z

pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 2.7.8.final.0
python-bits: 64
OS: Linux
OS-release: 3.17.4-lh-nvidia
machine: x86_64
processor: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
byteorder: little
LC_ALL: en_GB.UTF-8
LANG: en_GB.utf8

pandas: 0.15.0
nose: 1.3.4
Cython: 0.21.1
numpy: 1.9.1
scipy: 0.14.0
statsmodels: None
IPython: 2.3.0
sphinx: 1.2.3
patsy: None
dateutil: 2.2
pytz: 2014.10
bottleneck: 0.8.0
tables: 3.1.1
numexpr: 2.4
matplotlib: 1.4.2
openpyxl: 2.0.2
xlrd: 0.9.3
xlwt: 0.7.5
xlsxwriter: None
lxml: 3.4.1
bs4: None
html5lib: 0.999
httplib2: None
apiclient: None
rpy2: None
sqlalchemy: None
pymysql: None
psycopg2: None

jreback · 2014-12-06T16:49:17Z

this works with latest master so not sure what the problem is (I matched your installed versions).

Here are some windows builds of a test version of 0.15.2 (0.15.1 is officially the current release)

https://github.com/pydata/pandas/releases

jlec · 2014-12-07T08:59:05Z

I see the same error with current HEAD.

jreback · 2014-12-07T12:48:02Z

the error report you have doesn't match the show versions

first is in py3.3 and 2nd says py2.7

which is it??

jlec · 2014-12-07T13:38:59Z

it is with every python version

INSTALLED VERSIONS

commit: None
python: 3.4.2.final.0
python-bits: 64
OS: Linux
OS-release: 3.17.4-lh-nvidia
machine: x86_64
processor: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
byteorder: little
LC_ALL: en_GB.UTF-8
LANG: en_GB.utf8

pandas: 0.15.1-196-ge51eb9e
nose: 1.3.4
Cython: 0.21.1
numpy: 1.9.1
scipy: 0.14.0
statsmodels: 0.6.1
IPython: 2.3.1
sphinx: 1.2.3
patsy: 0.3.0
dateutil: 2.3
pytz: 2014.10
bottleneck: 0.8.0
tables: 3.1.1
numexpr: 2.4
matplotlib: 1.4.2
openpyxl: 1.6.2
xlrd: 0.9.3
xlwt: None
xlsxwriter: 0.6.4
lxml: 3.4.1
bs4: 4.3.2
html5lib: 0.999
httplib2: 0.8
apiclient: None
rpy2: None
sqlalchemy: 0.9.8
pymysql: 0.6.2.None
psycopg2: None
None

ERROR: test_string_index_alias_tz_aware (test_timezones.TestTimeZoneSupportDateutil)

Traceback (most recent call last):
File "/var/tmp/portage/dev-python/pandas-9999/work/pandas-9999-python3_4/lib/pandas/tseries/tests/test_timezones.py", line 606, in test_string_index_alias_tz_aware
self.assertAlmostEqual(result, ts[2])
File "/usr/lib64/python3.4/unittest/case.py", line 821, in assertAlmostEqual
if first == second:
File "/var/tmp/portage/dev-python/pandas-9999/work/pandas-9999-python3_4/lib/pandas/core/generic.py", line 698, in nonzero
.format(self.class.name))
ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

INSTALLED VERSIONS

commit: None
python: 2.7.8.final.0
python-bits: 64
OS: Linux
OS-release: 3.17.4-lh-nvidia
machine: x86_64
processor: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
byteorder: little
LC_ALL: en_GB.UTF-8
LANG: en_GB.utf8

pandas: 0.15.1-196-ge51eb9e
nose: 1.3.4
Cython: 0.21.1
numpy: 1.9.1
scipy: 0.14.0
statsmodels: 0.6.1
IPython: 2.3.1
sphinx: 1.2.3
patsy: 0.3.0
dateutil: 2.3
pytz: 2014.10
bottleneck: 0.8.0
tables: 3.1.1
numexpr: 2.4
matplotlib: 1.4.2
openpyxl: 1.6.2
xlrd: 0.9.3
xlwt: 0.7.5
xlsxwriter: 0.6.4
lxml: 3.4.1
bs4: 4.3.2
html5lib: 0.999
httplib2: 0.8
apiclient: None
rpy2: None
sqlalchemy: 0.9.8
pymysql: 0.6.2.None
psycopg2: None
None

ERROR: test_string_index_alias_tz_aware (test_timezones.TestTimeZoneSupportDateutil)

Traceback (most recent call last):
File "/var/tmp/portage/dev-python/pandas-9999/work/pandas-9999-python2_7/lib/pandas/tseries/tests/test_timezones.py", line 606, in test_string_index_alias_tz_aware
self.assertAlmostEqual(result, ts[2])
File "/usr/lib64/python2.7/unittest/case.py", line 537, in assertAlmostEqual
if first == second:
File "/var/tmp/portage/dev-python/pandas-9999/work/pandas-9999-python2_7/lib/pandas/core/generic.py", line 698, in nonzero
.format(self.class.name))
ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

INSTALLED VERSIONS

commit: None
python: 3.3.5.final.0
python-bits: 64
OS: Linux
OS-release: 3.17.4-lh-nvidia
machine: x86_64
processor: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
byteorder: little
LC_ALL: en_GB.UTF-8
LANG: en_GB.utf8

pandas: 0.15.1-196-ge51eb9e
nose: 1.3.4
Cython: 0.21.1
numpy: 1.9.1
scipy: 0.14.0
statsmodels: 0.6.1
IPython: 2.3.1
sphinx: 1.2.3
patsy: 0.3.0
dateutil: 2.3
pytz: 2014.10
bottleneck: 0.8.0
tables: 3.1.1
numexpr: 2.4
matplotlib: 1.4.2
openpyxl: 1.6.2
xlrd: 0.9.3
xlwt: None
xlsxwriter: 0.6.4
lxml: 3.4.1
bs4: 4.3.2
html5lib: 0.999
httplib2: 0.8
apiclient: None
rpy2: None
sqlalchemy: 0.9.8
pymysql: 0.6.2.None
psycopg2: None

None

ERROR: test_string_index_alias_tz_aware (test_timezones.TestTimeZoneSupportDateutil)

Traceback (most recent call last):
File "/var/tmp/portage/dev-python/pandas-9999/work/pandas-9999-python3_3/lib/pandas/tseries/tests/test_timezones.py", line 606, in test_string_index_alias_tz_aware
self.assertAlmostEqual(result, ts[2])
File "/usr/lib64/python3.3/unittest/case.py", line 665, in assertAlmostEqual
if first == second:
File "/var/tmp/portage/dev-python/pandas-9999/work/pandas-9999-python3_3/lib/pandas/core/generic.py", line 698, in nonzero
.format(self.class.name))
ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

jreback · 2014-12-07T14:43:11Z

for dateutil 2.3 might be related to #9021

@jlec you are showing several different versions of dateutil here, so very confusing. I think that this may be broken on your linux version. You can provide a debug output (e.g. run nosetests with --pdb --pdb-failure and show the variables that are being compared (you have to go up a couple of levels).

jreback · 2014-12-07T20:52:46Z

ok, this is ONLY on >= 2.3. The attached PR #9036 now shows the issues (was being suppressed somehow on windows).

I think the API changed in 2.3

jlec · 2014-12-08T07:51:33Z

Do you need anything from me, or is the issue clear?

jlec · 2014-12-08T14:46:09Z

This issue is also present with dateutil-2.2

https://gist.github.com/jlec/3a196d98bdf540fdda44

jlec · 2014-12-08T14:56:45Z

The requested values from the test:

result -> 2000-01-03 00:00:00-05:00 -1.566048 Freq: D, dtype: float64
ts[2] -> -1.56604761311

jreback · 2014-12-09T01:23:18Z

@jlec ok, give #9047 a try.

This fixes everything I have tried on windows (e.g. 2.2/2.3).

There were certain operations using dateutil timezones that were giving incorrect results (in certain cases) as the tests were not exercising all the cases. and their are a lot!.

lmk

jreback · 2014-12-09T23:51:58Z

@jlec did you get a chance to try #9047?

jlec · 2014-12-10T10:02:51Z

Latest HEAD plus #9047 with python-dateutil-2.3: https://gist.github.com/jlec/ca1f2fb9d383faacaef6
Latest HEAD plus #9047 with python-dateutil-2.2: https://gist.github.com/jlec/a4a3930d174327458c1e

jlec · 2014-12-10T10:05:11Z

With dateutil-2.3 there is

test_append_with_timezones_dateutil (pandas.io.tests.test_pytables.TestHDFStore) ... /usr/lib64/python2.7/site-packages/dateutil/zoneinfo/init.py:36: UserWarning: I/O error(2): No such file or directory
warnings.warn("I/O error({0}): {1}".format(e.errno, e.strerror))
FAIL

which is a simple FAIL with 2.2

Beside that the testfailures are the same for both dateutil versions

jlec · 2014-12-10T10:41:36Z

PR #9047 doesn't change anything in the test result. I tested all combinations of HEAD+/-#9047 with dateutil-2.2/2.3. The failures are always the same.

jreback · 2014-12-10T11:09:37Z

import dateutil
print dateutil.__version__

from dateutil.tz import gettz
from dateutil.zoneinfo import gettz as gettz2

print gettz('US/Eastern')
print gettz2('US/Eastern')

pls run this and report on 2.2 and 2.3 under python 2.7

jreback · 2014-12-11T22:42:11Z

@jlec see this report as well: #9059

I suspect that your kernel moved around tz files and dateutil is not finding them (as compared to an older version of the kernel).

zoneinfo.gettz() seems to have problems (1 & 2) on system which do not install the zoninfo tarball (e.g. Debian, Gentoo and Fedora) but rely on the system zoneinfo files. This results in test failures (3 & 4) tz.gettz() doesn't suffer from this problem. 1 dateutil/dateutil#8 2 dateutil/dateutil#11 3 pandas-dev#9059 4 pandas-dev#8639 Signed-off-by: Justin Lecher <jlec@gentoo.org>

python-dateutil provides two implementations for gettz(), tz.gettz() and zoneinfo.gettz(). The former tries first to use system provided timezone data, where as the later always uses a bundled tarball. Upstreams recommandation for library consumers is only using tz.gettz() (1 & 2). Further more, on system which do not install the zoninfo tarball (e.g. Debian, Gentoo and Fedora) but rely on the system zoneinfo files the direct usage of zoneinfo.gettz() creates problems which result in test failures (3 - 6). For compatibility in pandas code pandas.tslib._dateutil_gettz() should be used. 1 dateutil/dateutil#8 2 dateutil/dateutil#11 3 pandas-dev#9059 4 pandas-dev#8639 5 pandas-dev#10121 6 pandas-dev#9663 Signed-off-by: Justin Lecher <jlec@gentoo.org>

jreback closed this as completed Dec 6, 2014

jreback added the Timezones Timezone data dtype label Dec 6, 2014

This was referenced Dec 7, 2014

TST: dateutil 2.3 #9021

Closed

COMPAT: dateutil fixups for 2.3 (GH9021, GH8639) #9036

Merged

jreback added this to the 0.15.2 milestone Dec 7, 2014

jreback reopened this Dec 7, 2014

jreback mentioned this issue Dec 9, 2014

TST: dateutil fixes (GH8639) #9047

Merged

jarondl mentioned this issue Dec 10, 2014

pandas claim we broke the API dateutil/dateutil#5

Closed

jreback modified the milestones: 0.16.0, 0.15.2 Dec 10, 2014

jreback mentioned this issue Dec 12, 2014

COMPAT: dateutil 2.2/2.3 on linux 3.17 dateutil/dateutil#8

Closed

jlec changed the title ~~veriosn 0.15.0 test failure test_string_index_alias_tz_aware (pandas.tseries.tests.test_timezones.TestTimeZoneSupportDateutil)~~ pandas-0.15.0 test failure test_string_index_alias_tz_aware (pandas.tseries.tests.test_timezones.TestTimeZoneSupportDateutil) Dec 19, 2014

jlec mentioned this issue Dec 21, 2014

Use tz.gettz() instead of zoneinfo.gettz() #9123

Merged

jreback modified the milestones: 0.16.0, 0.16.1 Mar 6, 2015

jreback mentioned this issue Mar 17, 2015

BUG: Time zone information lost for some dateutil time zones #9663

Closed

jreback added Difficulty Intermediate labels Apr 4, 2015

jreback modified the milestones: 0.17.0, 0.16.1 Apr 21, 2015

jreback closed this as completed in #9123 May 15, 2015

jorisvandenbossche modified the milestones: 0.17.0, 0.16.2 Jun 2, 2015

jlec mentioned this issue Nov 24, 2015

Use tz.gettz() instead of zoneinfo.gettz() #11688

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

pandas-0.15.0 test failure test_string_index_alias_tz_aware (pandas.tseries.tests.test_timezones.TestTimeZoneSupportDateutil) #8639

pandas-0.15.0 test failure test_string_index_alias_tz_aware (pandas.tseries.tests.test_timezones.TestTimeZoneSupportDateutil) #8639

jlec commented Oct 26, 2014

jreback commented Oct 26, 2014

jlec commented Dec 6, 2014

jreback commented Dec 6, 2014

jlec commented Dec 7, 2014

jreback commented Dec 7, 2014

jlec commented Dec 7, 2014

jreback commented Dec 7, 2014

jreback commented Dec 7, 2014

jlec commented Dec 8, 2014

jlec commented Dec 8, 2014

jlec commented Dec 8, 2014

jreback commented Dec 9, 2014

jreback commented Dec 9, 2014

jlec commented Dec 10, 2014

jlec commented Dec 10, 2014

jlec commented Dec 10, 2014

jreback commented Dec 10, 2014

jreback commented Dec 11, 2014

pandas-0.15.0 test failure test_string_index_alias_tz_aware (pandas.tseries.tests.test_timezones.TestTimeZoneSupportDateutil) #8639

pandas-0.15.0 test failure test_string_index_alias_tz_aware (pandas.tseries.tests.test_timezones.TestTimeZoneSupportDateutil) #8639

Comments

jlec commented Oct 26, 2014

ERROR: test_string_index_alias_tz_aware (pandas.tseries.tests.test_timezones.TestTimeZoneSupportDateutil)

jreback commented Oct 26, 2014

jlec commented Dec 6, 2014

INSTALLED VERSIONS

jreback commented Dec 6, 2014

jlec commented Dec 7, 2014

jreback commented Dec 7, 2014

jlec commented Dec 7, 2014

INSTALLED VERSIONS

ERROR: test_string_index_alias_tz_aware (test_timezones.TestTimeZoneSupportDateutil)

INSTALLED VERSIONS

ERROR: test_string_index_alias_tz_aware (test_timezones.TestTimeZoneSupportDateutil)

INSTALLED VERSIONS

None

ERROR: test_string_index_alias_tz_aware (test_timezones.TestTimeZoneSupportDateutil)

jreback commented Dec 7, 2014

jreback commented Dec 7, 2014

jlec commented Dec 8, 2014

jlec commented Dec 8, 2014

jlec commented Dec 8, 2014

jreback commented Dec 9, 2014

jreback commented Dec 9, 2014

jlec commented Dec 10, 2014

jlec commented Dec 10, 2014

jlec commented Dec 10, 2014

jreback commented Dec 10, 2014

jreback commented Dec 11, 2014