reverse order of DatetimeCoder and MaskCoder in encoding/decoding, retrieve wanted fill_value in MaskCoder

Author: kmuehlbauer

xarray/coding/variables.py

@@ -9,6 +9,7 @@
 import numpy as np
 import pandas as pd
 
+from xarray.coding import times
 from xarray.core import dtypes, duck_array_ops, indexing
 from xarray.core.pycompat import is_duck_dask_array
 from xarray.core.variable import Variable
@@ -239,6 +240,16 @@ def encode(self, variable: Variable, name: T_Name = None):
             # Ensure _FillValue is cast to same dtype as data's
             encoding["_FillValue"] = dtype.type(fv)
             fill_value = pop_to(encoding, attrs, "_FillValue", name=name)
+            # retrieve _FillValue in case of np.datetime64
+            # see GH 7817
+            if np.issubdtype(data.dtype, np.datetime64):
+                units = encoding.get("units", None)
+                if isinstance(units, str) and "since" in units:
+                    delta, _ = times._unpack_netcdf_time_units(units)
+                    delta = times._netcdf_to_numpy_timeunit(delta)
+                    fill_value = np.datetime64(fill_value.item(), delta).astype(
+                        "datetime64[ns]"
+                    )
             if not pd.isnull(fill_value):
                 data = duck_array_ops.fillna(data, fill_value)
 
@@ -275,7 +286,17 @@ def decode(self, variable: Variable, name: T_Name = None):
                 )
 
             dtype, decoded_fill_value = dtypes.maybe_promote(data.dtype)
-
+            # retrieve _FillValue in case of np.datetime64
+            # see GH 7817
+            if np.issubdtype(data.dtype, np.datetime64) and decoded_fill_value.astype(
+                np.int64
+            ) == np.datetime64("NaT").astype(np.int64):
+                delta, _ = times._unpack_netcdf_time_units(encoding["units"])
+                delta = times._netcdf_to_numpy_timeunit(delta)
+                encoded_fill_values = {
+                    np.datetime64(encfill.item(), delta).astype("datetime64[ns]")
+                    for encfill in encoded_fill_values
+                }
             if encoded_fill_values:
                 transform = partial(
                     _apply_mask,

xarray/conventions.py

@@ -174,10 +174,10 @@ def encode_cf_variable(
     ensure_not_multiindex(var, name=name)
 
     for coder in [
-        times.CFDatetimeCoder(),
-        times.CFTimedeltaCoder(),
         variables.CFScaleOffsetCoder(),
         variables.CFMaskCoder(),
+        times.CFDatetimeCoder(),
+        times.CFTimedeltaCoder(),
         variables.UnsignedIntegerCoder(),
         variables.NonStringCoder(),
         variables.DefaultFillvalueCoder(),
@@ -263,6 +263,13 @@ def decode_cf_variable(
             var = strings.CharacterArrayCoder().decode(var, name=name)
         var = strings.EncodedStringCoder().decode(var)
 
+    # time decoding before masking
+    # GH 7817
+    if decode_timedelta:
+        var = times.CFTimedeltaCoder().decode(var, name=name)
+    if decode_times:
+        var = times.CFDatetimeCoder(use_cftime=use_cftime).decode(var, name=name)
+
     if mask_and_scale:
         for coder in [
             variables.UnsignedIntegerCoder(),
@@ -271,11 +278,6 @@ def decode_cf_variable(
         ]:
             var = coder.decode(var, name=name)
 
-    if decode_timedelta:
-        var = times.CFTimedeltaCoder().decode(var, name=name)
-    if decode_times:
-        var = times.CFDatetimeCoder(use_cftime=use_cftime).decode(var, name=name)
-
     if decode_endianness and not var.dtype.isnative:
         var = variables.EndianCoder().decode(var)
         original_dtype = var.dtype