Update temporal accessor to use specific cftime objects

- Update format of `decode_non_cf_time()` conditional statement
- Update `time_cf` with encoding dict attributes (xarray behavior)
- Update `test_dataset.py` and `test_temporal.py` with specific `cftime` date_type objects
- Add `xesmf` to `dev.yml`
- Remove `Jupyter` setting from VS Code workspace

.vscode/xcdat.code-workspace

@@ -43,7 +43,6 @@
         // Extension settings
         // ===================
         "jupyter.notebookFileRoot": "${workspaceFolder}",
-        "jupyter.runStartupCommands": ["%load_ext autoreload", "%autoreload 2"],
-        "autoDocstring.docstringFormat": "numpy"
+        "autoDocstring.docstringFormat": "numpy",
     }
 }

conda-env/dev.yml

@@ -17,6 +17,7 @@ dependencies:
     - numpy=1.22.4
     - pandas=1.4.3
     - xarray=2022.3.0
+    - xesmf=0.6.3
     - python-dateutil=2.8.2
     - types-python-dateutil=2.8.19
     # Quality Assurance

tests/fixtures.py

@@ -41,17 +41,25 @@
         "standard_name": "time",
     },
 )
+time_cf.encoding["calendar"] = "standard"
+time_cf.encoding["units"] = "days since 2000-01-01"
+
+# NOTE: When opening a dataset with `decode_times=False``, the encoding attrs
+# such as 'calendar' and 'units' are saved in the `.attrs` dict rather than the
+# `.encoding` dict. After the decoding process is done, these attributes are
+# stored in the encoding dictionary (same behavior as xarray).
 time_non_cf = xr.DataArray(
     data=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14],
     dims=["time"],
     attrs={
-        "units": "months since 2000-01-01",
-        "calendar": "standard",
         "axis": "T",
         "long_name": "time",
         "standard_name": "time",
+        "calendar": "standard",
+        "units": "months since 2000-01-01",
     },
 )
+
 time_non_cf_unsupported = xr.DataArray(
     data=np.arange(1850 + 1 / 24.0, 1851 + 3 / 12.0, 1 / 12.0),
     dims=["time"],