Merge pull request #4601 from robertcv/enh/editdomain_time

[FIX] Edit Domain: Improve Text/Categorical to Time conversion

Author: lanzagar

Orange/widgets/data/oweditdomain.py

@@ -2527,8 +2527,16 @@ def mapper(arr, out=None, dtype=dtype, **kwargs):
 def time_parse(values: Sequence[str], name="__"):
     tvar = Orange.data.TimeVariable(name)
     parse_time = ftry(tvar.parse, ValueError, np.nan)
-    values = [parse_time(v) for v in values]
-    return tvar, values
+    _values = [parse_time(v) for v in values]
+    if np.all(np.isnan(_values)):
+        # try parsing it with pandas (like in transform)
+        dti = pd.to_datetime(values, errors="coerce")
+        _values = datetime_to_epoch(dti)
+        date_only = getattr(dti, "_is_dates_only", False)
+        if np.all(dti != pd.NaT):
+            tvar.have_date = True
+            tvar.have_time = not date_only
+    return tvar, _values
 
 
 as_string = np.frompyfunc(str, 1, 1)
@@ -2734,17 +2742,23 @@ def transform(self, c):
             raise TypeError
 
 
+def datetime_to_epoch(dti: pd.DatetimeIndex) -> np.ndarray:
+    """Convert datetime to epoch"""
+    data = dti.values.astype("M8[us]")
+    mask = np.isnat(data)
+    data = data.astype(float) / 1e6
+    data[mask] = np.nan
+    return data
+
+
 class ReparseTimeTransform(Transformation):
     """
     Re-parse the column's string repr as datetime.
     """
     def transform(self, c):
         c = column_str_repr(self.variable, c)
-        c = pd.to_datetime(c, errors="coerce").values.astype("M8[us]")
-        mask = np.isnat(c)
-        orangecol = c.astype(float) / 1e6
-        orangecol[mask] = np.nan
-        return orangecol
+        c = pd.to_datetime(c, errors="coerce")
+        return datetime_to_epoch(c)
 
 
 class LookupMappingTransform(Transformation):

Orange/widgets/data/tests/test_oweditdomain.py

@@ -8,6 +8,7 @@
 
 import numpy as np
 from numpy.testing import assert_array_equal
+import pandas as pd
 
 from AnyQt.QtCore import QItemSelectionModel, Qt, QItemSelection
 from AnyQt.QtWidgets import QAction, QComboBox, QLineEdit, \
@@ -33,7 +34,7 @@
     table_column_data, ReinterpretVariableEditor, CategoricalVector,
     VariableEditDelegate, TransformRole,
     RealVector, TimeVector, StringVector, make_dict_mapper, DictMissingConst,
-    LookupMappingTransform, as_float_or_nan, column_str_repr,
+    LookupMappingTransform, as_float_or_nan, column_str_repr, time_parse,
     GroupItemsDialog)
 from Orange.widgets.data.owcolor import OWColor, ColorRole
 from Orange.widgets.tests.base import WidgetTest, GuiTest
@@ -917,6 +918,19 @@ def test_column_str_repr(self):
         d = column_str_repr(v, np.array([0., np.nan, 1.0]))
         assert_array_equal(d, ["00:00:00", "?", "00:00:01"])
 
+    def test_time_parse(self):
+        """parsing additional datetimes by pandas"""
+        date = ["1/22/20", "1/23/20", "1/24/20"]
+        # we use privet method, check if still exists
+        assert hasattr(pd.DatetimeIndex, '_is_dates_only')
+
+        tval, values = time_parse(date)
+
+        self.assertTrue(tval.have_date)
+        self.assertFalse(tval.have_time)
+        self.assertListEqual(list(values),
+                             [1579651200.0, 1579737600.0, 1579824000.0])
+
 
 class TestLookupMappingTransform(TestCase):
     def setUp(self) -> None: