Test that we forbid inlined refs when parsing kerchunk parquet (#865)

* add empty release notes

* interpret a kerchunk reference containing a nan as meaning an uninitialized zarr chunk

* raise a clearer error if kerchunk doesn't specify codecs properly

* test

* test: add test for inlined refs in kerchunk parquet files

Adds test_notimplemented_read_inline_refs_parquet to verify that
kerchunk parquet files with inlined chunk data (populated 'raw' column)
raise NotImplementedError, matching the existing JSON behavior.

Related: #489

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix indentation bug

* add unit test for nan behaviour

* add integration test for sparse kerchunk parquet arrays

Tests that kerchunk parquet files with sparse chunks (some missing)
can be read correctly by VirtualiZarr.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* add release notes for kerchunk parquet bug fixes

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: 3 people

virtualizarr/tests/test_parsers/test_kerchunk.py

@@ -356,6 +356,31 @@ def test_notimplemented_read_inline_refs(tmp_path, netcdf4_inlined_ref, local_re
             pass
 
 
+@requires_kerchunk
+@pytest.mark.skipif(not has_fastparquet, reason="fastparquet not installed")
+def test_notimplemented_read_inline_refs_parquet(
+    tmp_path, netcdf4_inlined_ref, local_registry
+):
+    # Test that parquet references with inlined data raise NotImplementedError
+    # https://github.com/zarr-developers/VirtualiZarr/issues/489
+    from kerchunk.df import refs_to_dataframe
+
+    ref_filepath = tmp_path / "ref.parquet"
+    refs_to_dataframe(fo=netcdf4_inlined_ref, url=ref_filepath.as_posix())
+
+    parser = KerchunkParquetParser()
+    with pytest.raises(
+        NotImplementedError,
+        match="Reading inlined reference data is currently not supported",
+    ):
+        with open_virtual_dataset(
+            url=ref_filepath.as_posix(),
+            registry=local_registry,
+            parser=parser,
+        ) as _:
+            pass
+
+
 @pytest.mark.parametrize("skip_variables", ["a", ["a"]])
 def test_skip_variables(refs_file_factory, skip_variables, local_registry):
     refs_file = refs_file_factory()