Ensure encoding["source"] is available for a pathlib.Path object (#6974)

* Ensure 'source' is encoded even when its a Path

Resolves Issue #5888.

With this change, the "source" encoding will be stored whether the filename is
a string or a Path object.

* Add a test for encoding source with Path object

* Update whats-new.rst

* Ensure only string and Path filenames encode source

Co-authored-by: Mathias Hauser <[email protected]>

* Update doc/whats-new.rst

* fix link

Co-authored-by: Mathias Hauser <[email protected]>
Co-authored-by: Deepak Cherian <[email protected]>

Author: 3 people

doc/whats-new.rst

@@ -65,6 +65,8 @@ Bug fixes
   By `András Gunyhó <https://github.com/mgunyho>`_.
 - Avoid use of random numbers in `test_weighted.test_weighted_operations_nonequal_coords` (:issue:`6504`, :pull:`6961`).
   By `Luke Conibear <https://github.com/lukeconibear>`_.
+- ``Dataset.encoding['source']`` now exists when reading from a Path object (:issue:`5888`, :pull:`6974`)
+  By `Thomas Coleman <https://github.com/ColemanTom>`_.
 
 Documentation
 ~~~~~~~~~~~~~

xarray/backends/api.py

@@ -358,9 +358,9 @@ def _dataset_from_backend_dataset(
 
     ds.set_close(backend_ds._close)
 
-    # Ensure source filename always stored in dataset object (GH issue #2550)
-    if "source" not in ds.encoding and isinstance(filename_or_obj, str):
-        ds.encoding["source"] = filename_or_obj
+    # Ensure source filename always stored in dataset object
+    if "source" not in ds.encoding and isinstance(filename_or_obj, (str, os.PathLike)):
+        ds.encoding["source"] = _normalize_path(filename_or_obj)
 
     return ds
 

xarray/tests/test_backends.py

@@ -5068,6 +5068,17 @@ def test_source_encoding_always_present() -> None:
             assert ds.encoding["source"] == tmp
 
 
+@requires_scipy_or_netCDF4
+def test_source_encoding_always_present_with_pathlib() -> None:
+    # Test for GH issue #5888.
+    rnddata = np.random.randn(10)
+    original = Dataset({"foo": ("x", rnddata)})
+    with create_tmp_file() as tmp:
+        original.to_netcdf(tmp)
+        with open_dataset(Path(tmp)) as ds:
+            assert ds.encoding["source"] == tmp
+
+
 def _assert_no_dates_out_of_range_warning(record):
     undesired_message = "dates out of range"
     for warning in record: