/pandas/tseries/tests/test_tslib.py

import nose

import numpy as np

from pandas import tslib
import datetime

from pandas.core.api import Timestamp, Series
from pandas.tslib import period_asfreq, period_ordinal
from pandas.tseries.index import date_range
from pandas.tseries.frequencies import get_freq
import pandas.tseries.offsets as offsets
from pandas import _np_version_under1p7
import pandas.util.testing as tm
from pandas.util.testing import assert_series_equal

class TestTimestamp(tm.TestCase):
    def test_repr(self):
        date = '2014-03-07'
        tz = 'US/Eastern'
        freq = 'M'

        date_only = Timestamp(date)
        self.assertIn(date, repr(date_only))
        self.assertNotIn(tz, repr(date_only))
        self.assertNotIn(freq, repr(date_only))
        self.assertEqual(date_only, eval(repr(date_only)))

        date_tz = Timestamp(date, tz=tz)
        self.assertIn(date, repr(date_tz))
        self.assertIn(tz, repr(date_tz))
        self.assertNotIn(freq, repr(date_tz))
        self.assertEqual(date_tz, eval(repr(date_tz)))

        date_freq = Timestamp(date, offset=freq)
        self.assertIn(date, repr(date_freq))
        self.assertNotIn(tz, repr(date_freq))
        self.assertIn(freq, repr(date_freq))
        self.assertEqual(date_freq, eval(repr(date_freq)))

        date_tz_freq = Timestamp(date, tz=tz, offset=freq)
        self.assertIn(date, repr(date_tz_freq))
        self.assertIn(tz, repr(date_tz_freq))
        self.assertIn(freq, repr(date_tz_freq))
        self.assertEqual(date_tz_freq, eval(repr(date_tz_freq)))

        # this can cause the tz field to be populated, but it's redundant to information in the datestring
        date_with_utc_offset = Timestamp('2014-03-13 00:00:00-0400', tz=None)
        self.assertIn('2014-03-13 00:00:00-0400', repr(date_with_utc_offset))
        self.assertNotIn('tzoffset', repr(date_with_utc_offset))
        self.assertEqual(date_with_utc_offset, eval(repr(date_with_utc_offset)))

    def test_bounds_with_different_units(self):
        out_of_bounds_dates = (
            '1677-09-21',
            '2262-04-12',
        )

        time_units = ('D', 'h', 'm', 's', 'ms', 'us')

        for date_string in out_of_bounds_dates:
            for unit in time_units:
                self.assertRaises(
                    ValueError,
                    Timestamp,
                    np.datetime64(date_string, dtype='M8[%s]' % unit)
                )

        in_bounds_dates = (
            '1677-09-23',
            '2262-04-11',
        )

        for date_string in in_bounds_dates:
            for unit in time_units:
                Timestamp(
                    np.datetime64(date_string, dtype='M8[%s]' % unit)
                )

    def test_tz(self):
        t = '2014-02-01 09:00'
        ts = Timestamp(t)
        local = ts.tz_localize('Asia/Tokyo')
        self.assertEqual(local.hour, 9)
        self.assertEqual(local, Timestamp(t, tz='Asia/Tokyo'))
        conv = local.tz_convert('US/Eastern')
        self.assertEqual(conv,
                         Timestamp('2014-01-31 19:00', tz='US/Eastern'))
        self.assertEqual(conv.hour, 19)

        # preserves nanosecond
        ts = Timestamp(t) + offsets.Nano(5)
        local = ts.tz_localize('Asia/Tokyo')
        self.assertEqual(local.hour, 9)
        self.assertEqual(local.nanosecond, 5)
        conv = local.tz_convert('US/Eastern')
        self.assertEqual(conv.nanosecond, 5)
        self.assertEqual(conv.hour, 19)

    def test_barely_oob_dts(self):
        one_us = np.timedelta64(1).astype('timedelta64[us]')

        # By definition we can't go out of bounds in [ns], so we
        # convert the datetime64s to [us] so we can go out of bounds
        min_ts_us = np.datetime64(Timestamp.min).astype('M8[us]')
        max_ts_us = np.datetime64(Timestamp.max).astype('M8[us]')

        # No error for the min/max datetimes
        Timestamp(min_ts_us)
        Timestamp(max_ts_us)

        # One us less than the minimum is an error
        self.assertRaises(ValueError, Timestamp, min_ts_us - one_us)

        # One us more than the maximum is an error
        self.assertRaises(ValueError, Timestamp, max_ts_us + one_us)


class TestDatetimeParsingWrappers(tm.TestCase):
    def test_does_not_convert_mixed_integer(self):
        bad_date_strings = (
            '-50000',
            '999',
            '123.1234',
            'm',
            'T'
        )

        for bad_date_string in bad_date_strings:
            self.assertFalse(
                tslib._does_string_look_like_datetime(bad_date_string)
            )

        good_date_strings = (
            '2012-01-01',
            '01/01/2012',
            'Mon Sep 16, 2013',
            '01012012',
            '0101',
            '1-1',
        )

        for good_date_string in good_date_strings:
            self.assertTrue(
                tslib._does_string_look_like_datetime(good_date_string)
            )


class TestArrayToDatetime(tm.TestCase):
    def test_parsing_valid_dates(self):
        arr = np.array(['01-01-2013', '01-02-2013'], dtype=object)
        self.assert_numpy_array_equal(
            tslib.array_to_datetime(arr),
            np.array(
                    [
                        '2013-01-01T00:00:00.000000000-0000',
                        '2013-01-02T00:00:00.000000000-0000'
                    ],
                    dtype='M8[ns]'
            )
        )

        arr = np.array(['Mon Sep 16 2013', 'Tue Sep 17 2013'], dtype=object)
        self.assert_numpy_array_equal(
            tslib.array_to_datetime(arr),
            np.array(
                    [
                        '2013-09-16T00:00:00.000000000-0000',
                        '2013-09-17T00:00:00.000000000-0000'
                    ],
                    dtype='M8[ns]'
            )
        )

    def test_number_looking_strings_not_into_datetime(self):
        # #4601
        # These strings don't look like datetimes so they shouldn't be
        # attempted to be converted
        arr = np.array(['-352.737091', '183.575577'], dtype=object)
        self.assert_numpy_array_equal(tslib.array_to_datetime(arr), arr)

        arr = np.array(['1', '2', '3', '4', '5'], dtype=object)
        self.assert_numpy_array_equal(tslib.array_to_datetime(arr), arr)

    def test_coercing_dates_outside_of_datetime64_ns_bounds(self):
        invalid_dates = [
            datetime.date(1000, 1, 1),
            datetime.datetime(1000, 1, 1),
            '1000-01-01',
            'Jan 1, 1000',
            np.datetime64('1000-01-01'),
        ]

        for invalid_date in invalid_dates:
            self.assertRaises(
                ValueError,
                tslib.array_to_datetime,
                np.array([invalid_date], dtype='object'),
                coerce=False,
                raise_=True,
            )
            self.assertTrue(
                np.array_equal(
                    tslib.array_to_datetime(
                        np.array([invalid_date], dtype='object'), coerce=True
                    ),
                    np.array([tslib.iNaT], dtype='M8[ns]')
                )
            )

        arr = np.array(['1/1/1000', '1/1/2000'], dtype=object)
        self.assert_numpy_array_equal(
            tslib.array_to_datetime(arr, coerce=True),
            np.array(
                    [
                        tslib.iNaT,
                        '2000-01-01T00:00:00.000000000-0000'
                    ],
                    dtype='M8[ns]'
            )
        )

    def test_coerce_of_invalid_datetimes(self):
        arr = np.array(['01-01-2013', 'not_a_date', '1'], dtype=object)

        # Without coercing, the presence of any invalid dates prevents
        # any values from being converted
        self.assert_numpy_array_equal(tslib.array_to_datetime(arr), arr)

        # With coercing, the invalid dates becomes iNaT
        self.assert_numpy_array_equal(
            tslib.array_to_datetime(arr, coerce=True),
            np.array(
                    [
                        '2013-01-01T00:00:00.000000000-0000',
                        tslib.iNaT,
                        tslib.iNaT
                    ],
                    dtype='M8[ns]'
            )
        )

    def test_parsing_timezone_offsets(self):
        # All of these datetime strings with offsets are equivalent
        # to the same datetime after the timezone offset is added
        dt_strings = [
            '01-01-2013 08:00:00+08:00',
            '2013-01-01T08:00:00.000000000+0800',
            '2012-12-31T16:00:00.000000000-0800',
            '12-31-2012 23:00:00-01:00',
        ]

        expected_output = tslib.array_to_datetime(
            np.array(['01-01-2013 00:00:00'], dtype=object)
        )

        for dt_string in dt_strings:
            self.assert_numpy_array_equal(
                tslib.array_to_datetime(
                    np.array([dt_string], dtype=object)
                ),
                expected_output
            )

class TestTimestampNsOperations(tm.TestCase):
    def setUp(self):
        tm._skip_if_not_numpy17_friendly()
        self.timestamp = Timestamp(datetime.datetime.utcnow())

    def assert_ns_timedelta(self, modified_timestamp, expected_value):
        value = self.timestamp.value
        modified_value = modified_timestamp.value

        self.assertEqual(modified_value - value, expected_value)

    def test_timedelta_ns_arithmetic(self):
        self.assert_ns_timedelta(self.timestamp + np.timedelta64(-123, 'ns'), -123)

    def test_timedelta_ns_based_arithmetic(self):
        self.assert_ns_timedelta(self.timestamp + np.timedelta64(1234567898, 'ns'), 1234567898)

    def test_timedelta_us_arithmetic(self):
        self.assert_ns_timedelta(self.timestamp + np.timedelta64(-123, 'us'), -123000)

    def test_timedelta_ms_arithmetic(self):
        time = self.timestamp + np.timedelta64(-123, 'ms')
        self.assert_ns_timedelta(time, -123000000)

    def test_nanosecond_string_parsing(self):
        self.timestamp = Timestamp('2013-05-01 07:15:45.123456789')
        self.assertEqual(self.timestamp.value, 1367392545123456000)

    def test_nanosecond_timestamp(self):
        # GH 7610
        expected = 1293840000000000005
        t = Timestamp('2011-01-01') + offsets.Nano(5)
        self.assertEqual(repr(t), "Timestamp('2011-01-01 00:00:00.000000005')")
        self.assertEqual(t.value, expected)
        self.assertEqual(t.nanosecond, 5)

        t = Timestamp(t)
        self.assertEqual(repr(t), "Timestamp('2011-01-01 00:00:00.000000005')")
        self.assertEqual(t.value, expected)
        self.assertEqual(t.nanosecond, 5)

        t = Timestamp(np.datetime64('2011-01-01 00:00:00.000000005Z'))
        self.assertEqual(repr(t), "Timestamp('2011-01-01 00:00:00.000000005')")
        self.assertEqual(t.value, expected)
        self.assertEqual(t.nanosecond, 5)

        expected = 1293840000000000010
        t = t + offsets.Nano(5)
        self.assertEqual(repr(t), "Timestamp('2011-01-01 00:00:00.000000010')")
        self.assertEqual(t.value, expected)
        self.assertEqual(t.nanosecond, 10)

        t = Timestamp(t)
        self.assertEqual(repr(t), "Timestamp('2011-01-01 00:00:00.000000010')")
        self.assertEqual(t.value, expected)
        self.assertEqual(t.nanosecond, 10)

        t = Timestamp(np.datetime64('2011-01-01 00:00:00.000000010Z'))
        self.assertEqual(repr(t), "Timestamp('2011-01-01 00:00:00.000000010')")
        self.assertEqual(t.value, expected)
        self.assertEqual(t.nanosecond, 10)

    def test_nat_arithmetic(self):
        # GH 6873
        nat = tslib.NaT
        t = Timestamp('2014-01-01')
        dt = datetime.datetime(2014, 1, 1)
        delta = datetime.timedelta(3600)

        # Timestamp / datetime
        for (left, right) in [(nat, nat), (nat, t), (dt, nat)]:
            # NaT + Timestamp-like should raise TypeError
            with tm.assertRaises(TypeError):
                left + right
            with tm.assertRaises(TypeError):
                right + left

            # NaT - Timestamp-like (or inverse) returns NaT
            self.assertTrue((left - right) is tslib.NaT)
            self.assertTrue((right - left) is tslib.NaT)

        # timedelta-like
        # offsets are tested in test_offsets.py
        for (left, right) in [(nat, delta)]:
            # NaT + timedelta-like returns NaT
            self.assertTrue((left + right) is tslib.NaT)
            # timedelta-like + NaT should raise TypeError
            with tm.assertRaises(TypeError):
                right + left

            self.assertTrue((left - right) is tslib.NaT)
            with tm.assertRaises(TypeError):
                right - left

        if _np_version_under1p7:
            self.assertEqual(nat + np.timedelta64(1, 'h'), tslib.NaT)
            with tm.assertRaises(TypeError):
                np.timedelta64(1, 'h') + nat

            self.assertEqual(nat - np.timedelta64(1, 'h'), tslib.NaT)
            with tm.assertRaises(TypeError):
                np.timedelta64(1, 'h') - nat


class TestTslib(tm.TestCase):

    def test_intraday_conversion_factors(self):
        self.assertEqual(period_asfreq(1, get_freq('D'), get_freq('H'), False), 24)
        self.assertEqual(period_asfreq(1, get_freq('D'), get_freq('T'), False), 1440)
        self.assertEqual(period_asfreq(1, get_freq('D'), get_freq('S'), False), 86400)
        self.assertEqual(period_asfreq(1, get_freq('D'), get_freq('L'), False), 86400000)
        self.assertEqual(period_asfreq(1, get_freq('D'), get_freq('U'), False), 86400000000)
        self.assertEqual(period_asfreq(1, get_freq('D'), get_freq('N'), False), 86400000000000)

        self.assertEqual(period_asfreq(1, get_freq('H'), get_freq('T'), False), 60)
        self.assertEqual(period_asfreq(1, get_freq('H'), get_freq('S'), False), 3600)
        self.assertEqual(period_asfreq(1, get_freq('H'), get_freq('L'), False), 3600000)
        self.assertEqual(period_asfreq(1, get_freq('H'), get_freq('U'), False), 3600000000)
        self.assertEqual(period_asfreq(1, get_freq('H'), get_freq('N'), False), 3600000000000)

        self.assertEqual(period_asfreq(1, get_freq('T'), get_freq('S'), False), 60)
        self.assertEqual(period_asfreq(1, get_freq('T'), get_freq('L'), False), 60000)
        self.assertEqual(period_asfreq(1, get_freq('T'), get_freq('U'), False), 60000000)
        self.assertEqual(period_asfreq(1, get_freq('T'), get_freq('N'), False), 60000000000)

        self.assertEqual(period_asfreq(1, get_freq('S'), get_freq('L'), False), 1000)
        self.assertEqual(period_asfreq(1, get_freq('S'), get_freq('U'), False), 1000000)
        self.assertEqual(period_asfreq(1, get_freq('S'), get_freq('N'), False), 1000000000)

        self.assertEqual(period_asfreq(1, get_freq('L'), get_freq('U'), False), 1000)
        self.assertEqual(period_asfreq(1, get_freq('L'), get_freq('N'), False), 1000000)

        self.assertEqual(period_asfreq(1, get_freq('U'), get_freq('N'), False), 1000)

    def test_period_ordinal_start_values(self):
        # information for 1.1.1970
        self.assertEqual(0, period_ordinal(1970, 1, 1, 0, 0, 0, 0, 0, get_freq('Y')))
        self.assertEqual(0, period_ordinal(1970, 1, 1, 0, 0, 0, 0, 0, get_freq('M')))
        self.assertEqual(1, period_ordinal(1970, 1, 1, 0, 0, 0, 0, 0, get_freq('W')))
        self.assertEqual(0, period_ordinal(1970, 1, 1, 0, 0, 0, 0, 0, get_freq('D')))
        self.assertEqual(0, period_ordinal(1970, 1, 1, 0, 0, 0, 0, 0, get_freq('B')))

    def test_period_ordinal_week(self):
        self.assertEqual(1, period_ordinal(1970, 1, 4, 0, 0, 0, 0, 0, get_freq('W')))
        self.assertEqual(2, period_ordinal(1970, 1, 5, 0, 0, 0, 0, 0, get_freq('W')))

        self.assertEqual(2284, period_ordinal(2013, 10, 6, 0, 0, 0, 0, 0, get_freq('W')))
        self.assertEqual(2285, period_ordinal(2013, 10, 7, 0, 0, 0, 0, 0, get_freq('W')))

    def test_period_ordinal_business_day(self):
        # Thursday
        self.assertEqual(11415, period_ordinal(2013, 10, 3, 0, 0, 0, 0, 0, get_freq('B')))
        # Friday
        self.assertEqual(11416, period_ordinal(2013, 10, 4, 0, 0, 0, 0, 0, get_freq('B')))
        # Saturday
        self.assertEqual(11417, period_ordinal(2013, 10, 5, 0, 0, 0, 0, 0, get_freq('B')))
        # Sunday
        self.assertEqual(11417, period_ordinal(2013, 10, 6, 0, 0, 0, 0, 0, get_freq('B')))
        # Monday
        self.assertEqual(11417, period_ordinal(2013, 10, 7, 0, 0, 0, 0, 0, get_freq('B')))
        # Tuesday
        self.assertEqual(11418, period_ordinal(2013, 10, 8, 0, 0, 0, 0, 0, get_freq('B')))

class TestTimestampOps(tm.TestCase):
    def test_timestamp_and_datetime(self):
        self.assertEqual((Timestamp(datetime.datetime(2013, 10, 13)) - datetime.datetime(2013, 10, 12)).days, 1)
        self.assertEqual((datetime.datetime(2013, 10, 12) - Timestamp(datetime.datetime(2013, 10, 13))).days, -1)

    def test_timestamp_and_series(self):
        timestamp_series = Series(date_range('2014-03-17', periods=2, freq='D', tz='US/Eastern'))
        first_timestamp = timestamp_series[0]

        if not _np_version_under1p7:
            delta_series = Series([np.timedelta64(0, 'D'), np.timedelta64(1, 'D')])
            assert_series_equal(timestamp_series - first_timestamp, delta_series)
            assert_series_equal(first_timestamp - timestamp_series, -delta_series)

    def test_addition_subtraction_types(self):
        # Assert on the types resulting from Timestamp +/- various date/time objects
        datetime_instance = datetime.datetime(2014, 3, 4)
        timedelta_instance = datetime.timedelta(seconds=1)
        # build a timestamp with a frequency, since then it supports addition/subtraction of integers
        timestamp_instance = date_range(datetime_instance, periods=1, freq='D')[0]

        self.assertEqual(type(timestamp_instance + 1), Timestamp)
        self.assertEqual(type(timestamp_instance - 1), Timestamp)

        # Timestamp + datetime not supported, though subtraction is supported and yields timedelta
        self.assertEqual(type(timestamp_instance - datetime_instance), datetime.timedelta)

        self.assertEqual(type(timestamp_instance + timedelta_instance), Timestamp)
        self.assertEqual(type(timestamp_instance - timedelta_instance), Timestamp)

        if not _np_version_under1p7:
            # Timestamp +/- datetime64 not supported, so not tested (could possibly assert error raised?)
            timedelta64_instance = np.timedelta64(1, 'D')
            self.assertEqual(type(timestamp_instance + timedelta64_instance), Timestamp)
            self.assertEqual(type(timestamp_instance - timedelta64_instance), Timestamp)

    def test_addition_subtraction_preserve_frequency(self):
        timestamp_instance = date_range('2014-03-05', periods=1, freq='D')[0]
        timedelta_instance = datetime.timedelta(days=1)
        original_freq = timestamp_instance.freq
        self.assertEqual((timestamp_instance + 1).freq, original_freq)
        self.assertEqual((timestamp_instance - 1).freq, original_freq)
        self.assertEqual((timestamp_instance + timedelta_instance).freq, original_freq)
        self.assertEqual((timestamp_instance - timedelta_instance).freq, original_freq)

        if not _np_version_under1p7:
            timedelta64_instance = np.timedelta64(1, 'D')
            self.assertEqual((timestamp_instance + timedelta64_instance).freq, original_freq)
            self.assertEqual((timestamp_instance - timedelta64_instance).freq, original_freq)

    def test_resolution(self):

        for freq, expected in zip(['A', 'Q', 'M', 'D', 'H', 'T', 'S', 'L', 'U'],
                                  [tslib.D_RESO, tslib.D_RESO, tslib.D_RESO, tslib.D_RESO,
                                   tslib.H_RESO, tslib.T_RESO,tslib.S_RESO, tslib.MS_RESO, tslib.US_RESO]):
            for tz in [None, 'Asia/Tokyo', 'US/Eastern']:
                idx = date_range(start='2013-04-01', periods=30, freq=freq, tz=tz)
                result = tslib.resolution(idx.asi8, idx.tz)
                self.assertEqual(result, expected)


if __name__ == '__main__':
    nose.runmodule(argv=[__file__, '-vvs', '-x', '--pdb', '--pdb-failure'],
                   exit=False)