Merge pull request #8843 from jtratner/support-step-in-str-slice

BUG: Implement step in slice StringMethod

Author: jorisvandenbossche

doc/source/whatsnew/v0.15.2.txt

@@ -69,6 +69,7 @@ Bug Fixes
 - ``io.data.Options`` now raises ``RemoteDataError`` when no expiry dates are available from Yahoo (:issue:`8761`).
 - ``Timedelta`` kwargs may now be numpy ints and floats (:issue:`8757`).
 - ``sql_schema`` now generates dialect appropriate ``CREATE TABLE`` statements (:issue:`8697`)
+- ``slice`` string method now takes step into account (:issue:`8754`)
 
 
 

pandas/core/strings.py

@@ -666,14 +666,15 @@ def str_split(arr, pat=None, n=None, return_type='series'):
     return res
 
 
-def str_slice(arr, start=None, stop=None, step=1):
+def str_slice(arr, start=None, stop=None, step=None):
     """
     Slice substrings from each element in array
 
     Parameters
     ----------
     start : int or None
     stop : int or None
+    step : int or None
 
     Returns
     -------
@@ -993,8 +994,8 @@ def center(self, width):
         return self._wrap_result(result)
 
     @copy(str_slice)
-    def slice(self, start=None, stop=None, step=1):
-        result = str_slice(self.series, start, stop)
+    def slice(self, start=None, stop=None, step=None):
+        result = str_slice(self.series, start, stop, step)
         return self._wrap_result(result)
 
     @copy(str_slice)

pandas/tests/test_strings.py

@@ -628,6 +628,7 @@ def test_empty_str_methods(self):
         tm.assert_series_equal(empty_str, empty.str.center(42))
         tm.assert_series_equal(empty_list, empty.str.split('a'))
         tm.assert_series_equal(empty_str, empty.str.slice(stop=1))
+        tm.assert_series_equal(empty_str, empty.str.slice(step=1))
         tm.assert_series_equal(empty_str, empty.str.strip())
         tm.assert_series_equal(empty_str, empty.str.lstrip())
         tm.assert_series_equal(empty_str, empty.str.rstrip())
@@ -922,6 +923,17 @@ def test_slice(self):
         exp = Series(['foo', 'bar', NA, 'baz'])
         tm.assert_series_equal(result, exp)
 
+        for start, stop, step in [(0, 3, -1), (None, None, -1),
+                                  (3, 10, 2), (3, 0, -1)]:
+            try:
+                result = values.str.slice(start, stop, step)
+                expected = Series([s[start:stop:step] if not isnull(s) else NA for s in
+                                   values])
+                tm.assert_series_equal(result, expected)
+            except:
+                print('failed on %s:%s:%s' % (start, stop, step))
+                raise
+
         # mixed
         mixed = Series(['aafootwo', NA, 'aabartwo', True, datetime.today(),
                         None, 1, 2.])
@@ -933,6 +945,10 @@ def test_slice(self):
         tm.assert_isinstance(rs, Series)
         tm.assert_almost_equal(rs, xp)
 
+        rs = Series(mixed).str.slice(2, 5, -1)
+        xp = Series(['oof', NA, 'rab', NA, NA,
+                     NA, NA, NA])
+
         # unicode
         values = Series([u('aafootwo'), u('aabartwo'), NA,
                          u('aabazqux')])
@@ -941,6 +957,10 @@ def test_slice(self):
         exp = Series([u('foo'), u('bar'), NA, u('baz')])
         tm.assert_series_equal(result, exp)
 
+        result = values.str.slice(0, -1, 2)
+        exp = Series([u('afow'), u('abrw'), NA, u('abzu')])
+        tm.assert_series_equal(result, exp)
+
     def test_slice_replace(self):
         pass
 
@@ -1151,6 +1171,10 @@ def test_string_slice_get_syntax(self):
         expected = s.str.slice(stop=3)
         assert_series_equal(result, expected)
 
+        result = s.str[2::-1]
+        expected = s.str.slice(start=2, step=-1)
+        assert_series_equal(result, expected)
+
     def test_string_slice_out_of_bounds(self):
         s = Series([(1, 2), (1,), (3,4,5)])