Add drop_duplicates for dims (#5239)

* Add drop_duplicates for dims

* Update PR # and fix lint

* Remove dataset

* Remove references to ds

* Update dataarray.py

* Update xarray/core/dataarray.py

Co-authored-by: keewis <keewis@users.noreply.github.com>

* Update dataarray.py

* Single dim

* Update xarray/core/dataarray.py

Co-authored-by: Maximilian Roos <5635139+max-sixty@users.noreply.github.com>

* Update xarray/core/dataarray.py

Co-authored-by: Maximilian Roos <5635139+max-sixty@users.noreply.github.com>

* Update xarray/core/dataarray.py

Co-authored-by: Maximilian Roos <5635139+max-sixty@users.noreply.github.com>

* [skip-ci]

Co-authored-by: ahuang11 <ahuang11@illinois.edu>
Co-authored-by: keewis <keewis@users.noreply.github.com>
Co-authored-by: Maximilian Roos <5635139+max-sixty@users.noreply.github.com>
Co-authored-by: dcherian <deepak@cherian.net>

Author: 5 people

doc/api.rst

@@ -292,6 +292,7 @@ DataArray contents
    DataArray.swap_dims
    DataArray.expand_dims
    DataArray.drop_vars
+   DataArray.drop_duplicates
    DataArray.reset_coords
    DataArray.copy
 

doc/whats-new.rst

@@ -21,6 +21,10 @@ v0.18.1 (unreleased)
 
 New Features
 ~~~~~~~~~~~~
+
+- Implement :py:meth:`DataArray.drop_duplicates`
+  to remove duplicate dimension values (:pull:`5239`).
+  By `Andrew Huang <https://github.com/ahuang11>`_.
 - allow passing ``combine_attrs`` strategy names to the ``keep_attrs`` parameter of
   :py:func:`apply_ufunc` (:pull:`5041`)
   By `Justus Magin <https://github.com/keewis>`_.

xarray/core/dataarray.py

@@ -4572,6 +4572,33 @@ def curvefit(
             kwargs=kwargs,
         )
 
+    def drop_duplicates(
+        self,
+        dim: Hashable,
+        keep: Union[
+            str,
+            bool,
+        ] = "first",
+    ):
+        """Returns a new DataArray with duplicate dimension values removed.
+        Parameters
+        ----------
+        dim : dimension label, optional
+        keep : {"first", "last", False}, default: "first"
+            Determines which duplicates (if any) to keep.
+            - ``"first"`` : Drop duplicates except for the first occurrence.
+            - ``"last"`` : Drop duplicates except for the last occurrence.
+            - False : Drop all duplicates.
+
+        Returns
+        -------
+        DataArray
+        """
+        if dim not in self.dims:
+            raise ValueError(f"'{dim}' not found in dimensions")
+        indexes = {dim: ~self.get_index(dim).duplicated(keep=keep)}
+        return self.isel(indexes)
+
     # this needs to be at the end, or mypy will confuse with `str`
     # https://mypy.readthedocs.io/en/latest/common_issues.html#dealing-with-conflicting-names
     str = utils.UncachedAccessor(StringAccessor)

xarray/tests/test_dataarray.py

@@ -7434,3 +7434,24 @@ def test_clip(da):
     # Unclear whether we want this work, OK to adjust the test when we have decided.
     with pytest.raises(ValueError, match="arguments without labels along dimension"):
         result = da.clip(min=da.mean("x"), max=da.mean("a").isel(x=[0, 1]))
+
+
+@pytest.mark.parametrize("keep", ["first", "last", False])
+def test_drop_duplicates(keep):
+    ds = xr.DataArray(
+        [0, 5, 6, 7], dims="time", coords={"time": [0, 0, 1, 2]}, name="test"
+    )
+
+    if keep == "first":
+        data = [0, 6, 7]
+        time = [0, 1, 2]
+    elif keep == "last":
+        data = [5, 6, 7]
+        time = [0, 1, 2]
+    else:
+        data = [6, 7]
+        time = [1, 2]
+
+    expected = xr.DataArray(data, dims="time", coords={"time": time}, name="test")
+    result = ds.drop_duplicates("time", keep=keep)
+    assert_equal(expected, result)