Propagate attrs with unary, binary functions (#4195)

* Propagate attrs with unary, binary functions

Closes #3490
Closes #4065
Closes #3433
Closes #3595

* Un xfail test

* bugfix

* Some progress. Still need keep_attrs in DataArray._unary_op

* Fix dataset attrs

* whats-new

* small fix

* Fix imag, real

* fix variable tests

* fix multiple return variables.

* review comments

* Update doc/whats-new.rst

* Propagate attrs with DataArray unary ops

* More tests

* Small cleanup

* Review comments.

* Fix duplication

Co-authored-by: Maximilian Roos <[email protected]>

doc/whats-new.rst

@@ -30,6 +30,8 @@ New Features
 - :py:func:`open_dataset` and :py:func:`open_mfdataset`
   now works with ``engine="zarr"`` (:issue:`3668`, :pull:`4003`, :pull:`4187`).
   By `Miguel Jimenez <https://github.com/Mikejmnez>`_ and `Wei Ji Leong <https://github.com/weiji14>`_.
+- Unary & binary operations follow the ``keep_attrs`` flag (:issue:`3490`, :issue:`4065`, :issue:`3433`, :issue:`3595`, :pull:`4195`).
+  By `Deepak Cherian <https://github.com/dcherian>`_.
 
 Bug fixes
 ~~~~~~~~~

xarray/core/arithmetic.py

@@ -3,7 +3,7 @@
 
 import numpy as np
 
-from .options import OPTIONS
+from .options import OPTIONS, _get_keep_attrs
 from .pycompat import dask_array_type
 from .utils import not_implemented
 
@@ -77,6 +77,7 @@ def __array_ufunc__(self, ufunc, method, *inputs, **kwargs):
             dataset_fill_value=np.nan,
             kwargs=kwargs,
             dask="allowed",
+            keep_attrs=_get_keep_attrs(default=True),
         )
 
     # this has no runtime function - these are listed so IDEs know these

xarray/core/computation.py

@@ -42,6 +42,14 @@
 _JOINS_WITHOUT_FILL_VALUES = frozenset({"inner", "exact"})
 
 
+def _first_of_type(args, kind):
+    """ Return either first object of type 'kind' or raise if not found. """
+    for arg in args:
+        if isinstance(arg, kind):
+            return arg
+    raise ValueError("This should be unreachable.")
+
+
 class _UFuncSignature:
     """Core dimensions signature for a given function.
 
@@ -252,8 +260,9 @@ def apply_dataarray_vfunc(
             args, join=join, copy=False, exclude=exclude_dims, raise_on_invalid=False
         )
 
-    if keep_attrs and hasattr(args[0], "name"):
-        name = args[0].name
+    if keep_attrs:
+        first_obj = _first_of_type(args, DataArray)
+        name = first_obj.name
     else:
         name = result_name(args)
     result_coords = build_output_coords(args, signature, exclude_dims)
@@ -270,6 +279,14 @@ def apply_dataarray_vfunc(
         (coords,) = result_coords
         out = DataArray(result_var, coords, name=name, fastpath=True)
 
+    if keep_attrs:
+        if isinstance(out, tuple):
+            for da in out:
+                # This is adding attrs in place
+                da._copy_attrs_from(first_obj)
+        else:
+            out._copy_attrs_from(first_obj)
+
     return out
 
 
@@ -390,15 +407,16 @@ def apply_dataset_vfunc(
     """
     from .dataset import Dataset
 
-    first_obj = args[0]  # we'll copy attrs from this in case keep_attrs=True
-
     if dataset_join not in _JOINS_WITHOUT_FILL_VALUES and fill_value is _NO_FILL_VALUE:
         raise TypeError(
             "to apply an operation to datasets with different "
             "data variables with apply_ufunc, you must supply the "
             "dataset_fill_value argument."
         )
 
+    if keep_attrs:
+        first_obj = _first_of_type(args, Dataset)
+
     if len(args) > 1:
         args = deep_align(
             args, join=join, copy=False, exclude=exclude_dims, raise_on_invalid=False
@@ -417,9 +435,11 @@ def apply_dataset_vfunc(
         (coord_vars,) = list_of_coords
         out = _fast_dataset(result_vars, coord_vars)
 
-    if keep_attrs and isinstance(first_obj, Dataset):
+    if keep_attrs:
         if isinstance(out, tuple):
-            out = tuple(ds._copy_attrs_from(first_obj) for ds in out)
+            for ds in out:
+                # This is adding attrs in place
+                ds._copy_attrs_from(first_obj)
         else:
             out._copy_attrs_from(first_obj)
     return out
@@ -595,6 +615,8 @@ def apply_variable_ufunc(
     """Apply a ndarray level function over Variable and/or ndarray objects."""
     from .variable import Variable, as_compatible_data
 
+    first_obj = _first_of_type(args, Variable)
+
     dim_sizes = unified_dim_sizes(
         (a for a in args if hasattr(a, "dims")), exclude_dims=exclude_dims
     )
@@ -734,8 +756,8 @@ def func(*arrays):
                     )
                 )
 
-        if keep_attrs and isinstance(args[0], Variable):
-            var.attrs.update(args[0].attrs)
+        if keep_attrs:
+            var.attrs.update(first_obj.attrs)
         output.append(var)
 
     if signature.num_outputs == 1:

xarray/core/dataarray.py

@@ -55,7 +55,7 @@
 from .indexes import Indexes, default_indexes, propagate_indexes
 from .indexing import is_fancy_indexer
 from .merge import PANDAS_TYPES, MergeError, _extract_indexes_from_coords
-from .options import OPTIONS
+from .options import OPTIONS, _get_keep_attrs
 from .utils import Default, ReprObject, _check_inplace, _default, either_dict_or_kwargs
 from .variable import (
     IndexVariable,
@@ -2734,13 +2734,19 @@ def __rmatmul__(self, other):
     def _unary_op(f: Callable[..., Any]) -> Callable[..., "DataArray"]:
         @functools.wraps(f)
         def func(self, *args, **kwargs):
+            keep_attrs = kwargs.pop("keep_attrs", None)
+            if keep_attrs is None:
+                keep_attrs = _get_keep_attrs(default=True)
             with warnings.catch_warnings():
                 warnings.filterwarnings("ignore", r"All-NaN (slice|axis) encountered")
                 warnings.filterwarnings(
                     "ignore", r"Mean of empty slice", category=RuntimeWarning
                 )
                 with np.errstate(all="ignore"):
-                    return self.__array_wrap__(f(self.variable.data, *args, **kwargs))
+                    da = self.__array_wrap__(f(self.variable.data, *args, **kwargs))
+                if keep_attrs:
+                    da.attrs = self.attrs
+                return da
 
         return func
 

xarray/core/dataset.py

@@ -4403,12 +4403,15 @@ def map(
             foo      (dim_0, dim_1) float64 1.764 0.4002 0.9787 2.241 1.868 0.9773
             bar      (x) float64 1.0 2.0
         """
+        if keep_attrs is None:
+            keep_attrs = _get_keep_attrs(default=False)
         variables = {
             k: maybe_wrap_array(v, func(v, *args, **kwargs))
             for k, v in self.data_vars.items()
         }
-        if keep_attrs is None:
-            keep_attrs = _get_keep_attrs(default=False)
+        if keep_attrs:
+            for k, v in variables.items():
+                v._copy_attrs_from(self.data_vars[k])
         attrs = self.attrs if keep_attrs else None
         return type(self)(variables, attrs=attrs)
 
@@ -4939,15 +4942,20 @@ def from_dict(cls, d):
         return obj
 
     @staticmethod
-    def _unary_op(f, keep_attrs=False):
+    def _unary_op(f):
         @functools.wraps(f)
         def func(self, *args, **kwargs):
             variables = {}
+            keep_attrs = kwargs.pop("keep_attrs", None)
+            if keep_attrs is None:
+                keep_attrs = _get_keep_attrs(default=True)
             for k, v in self._variables.items():
                 if k in self._coord_names:
                     variables[k] = v
                 else:
                     variables[k] = f(v, *args, **kwargs)
+                    if keep_attrs:
+                        variables[k].attrs = v._attrs
             attrs = self._attrs if keep_attrs else None
             return self._replace_with_new_dims(variables, attrs=attrs)
 
@@ -5684,11 +5692,11 @@ def _integrate_one(self, coord, datetime_unit=None):
 
     @property
     def real(self):
-        return self._unary_op(lambda x: x.real, keep_attrs=True)(self)
+        return self.map(lambda x: x.real, keep_attrs=True)
 
     @property
     def imag(self):
-        return self._unary_op(lambda x: x.imag, keep_attrs=True)(self)
+        return self.map(lambda x: x.imag, keep_attrs=True)
 
     plot = utils.UncachedAccessor(_Dataset_PlotMethods)
 

xarray/core/variable.py

@@ -2102,8 +2102,14 @@ def __array_wrap__(self, obj, context=None):
     def _unary_op(f):
         @functools.wraps(f)
         def func(self, *args, **kwargs):
+            keep_attrs = kwargs.pop("keep_attrs", None)
+            if keep_attrs is None:
+                keep_attrs = _get_keep_attrs(default=True)
             with np.errstate(all="ignore"):
-                return self.__array_wrap__(f(self.data, *args, **kwargs))
+                result = self.__array_wrap__(f(self.data, *args, **kwargs))
+                if keep_attrs:
+                    result.attrs = self.attrs
+                return result
 
         return func
 

xarray/tests/test_dataarray.py

@@ -9,7 +9,15 @@
 import pytest
 
 import xarray as xr
-from xarray import DataArray, Dataset, IndexVariable, Variable, align, broadcast
+from xarray import (
+    DataArray,
+    Dataset,
+    IndexVariable,
+    Variable,
+    align,
+    broadcast,
+    set_options,
+)
 from xarray.coding.times import CFDatetimeCoder
 from xarray.convert import from_cdms2
 from xarray.core import dtypes
@@ -2486,6 +2494,21 @@ def test_assign_attrs(self):
         assert_identical(new_actual, expected)
         assert actual.attrs == {"a": 1, "b": 2}
 
+    @pytest.mark.parametrize(
+        "func", [lambda x: x.clip(0, 1), lambda x: np.float64(1.0) * x, np.abs, abs]
+    )
+    def test_propagate_attrs(self, func):
+        da = DataArray(self.va)
+
+        # test defaults
+        assert func(da).attrs == da.attrs
+
+        with set_options(keep_attrs=False):
+            assert func(da).attrs == {}
+
+        with set_options(keep_attrs=True):
+            assert func(da).attrs == da.attrs
+
     def test_fillna(self):
         a = DataArray([np.nan, 1, np.nan, 3], coords={"x": range(4)}, dims="x")
         actual = a.fillna(-1)

xarray/tests/test_dataset.py

@@ -4473,6 +4473,28 @@ def test_fillna(self):
         assert actual.a.name == "a"
         assert actual.a.attrs == ds.a.attrs
 
+    @pytest.mark.parametrize(
+        "func", [lambda x: x.clip(0, 1), lambda x: np.float64(1.0) * x, np.abs, abs]
+    )
+    def test_propagate_attrs(self, func):
+
+        da = DataArray(range(5), name="a", attrs={"attr": "da"})
+        ds = Dataset({"a": da}, attrs={"attr": "ds"})
+
+        # test defaults
+        assert func(ds).attrs == ds.attrs
+        with set_options(keep_attrs=False):
+            assert func(ds).attrs != ds.attrs
+            assert func(ds).a.attrs != ds.a.attrs
+
+        with set_options(keep_attrs=False):
+            assert func(ds).attrs != ds.attrs
+            assert func(ds).a.attrs != ds.a.attrs
+
+        with set_options(keep_attrs=True):
+            assert func(ds).attrs == ds.attrs
+            assert func(ds).a.attrs == ds.a.attrs
+
     def test_where(self):
         ds = Dataset({"a": ("x", range(5))})
         expected = Dataset({"a": ("x", [np.nan, np.nan, 2, 3, 4])})

xarray/tests/test_variable.py

@@ -342,7 +342,8 @@ def test_1d_math(self):
         assert_array_equal(y - v, 1 - v)
         # verify attributes are dropped
         v2 = self.cls(["x"], x, {"units": "meters"})
-        assert_identical(base_v, +v2)
+        with set_options(keep_attrs=False):
+            assert_identical(base_v, +v2)
         # binary ops with all variables
         assert_array_equal(v + v, 2 * v)
         w = self.cls(["x"], y, {"foo": "bar"})

xarray/tests/test_weighted.py

@@ -320,7 +320,6 @@ def test_weighted_operations_keep_attr(operation, as_dataset, keep_attrs):
     assert not result.attrs
 
 
-@pytest.mark.xfail(reason="xr.Dataset.map does not copy attrs of DataArrays GH: 3595")
 @pytest.mark.parametrize("operation", ("sum", "mean"))
 def test_weighted_operations_keep_attr_da_in_ds(operation):
     # GH #3595