wip debugging

Author: brokkoli71

src/zarr/codecs/_v2.py

@@ -46,7 +46,10 @@ async def _decode_single(
         # special case object dtype, because incorrect handling can lead to
         # segfaults and other bad things happening
         if chunk_spec.dtype != object:
+            print(chunk_spec.dtype, chunk.dtype)
             chunk = chunk.view(chunk_spec.dtype)
+            print("worked")
+
         elif chunk.dtype != object:
             # If we end up here, someone must have hacked around with the filters.
             # We cannot deal with object arrays unless there is an object

src/zarr/core/config.py

@@ -66,7 +66,8 @@ def reset(self) -> None:
             "ndbuffer": "zarr.core.buffer.cpu.NDBuffer",
             "v2_dtype_kind_to_default_filters_and_compressor": {
                 "biufcmM": ["zstd"],
-                "OSUV": ["vlen-utf8"],
+                "SV": ["vlen-bytes"],
+                "OU": ["vlen-utf8"],
             },
         }
     ],

tests/test_properties.py

@@ -6,7 +6,7 @@
 
 import hypothesis.extra.numpy as npst  # noqa: E402
 import hypothesis.strategies as st  # noqa: E402
-from hypothesis import assume, given  # noqa: E402
+from hypothesis import assume, given, reproduce_failure  # noqa: E402
 
 from zarr.testing.strategies import arrays, basic_indices, numpy_arrays, zarr_formats  # noqa: E402
 

tests/test_v2.py

@@ -141,18 +141,19 @@ def test_v2_filters_codecs(filters: Any) -> None:
 
 @pytest.mark.parametrize(
     "dtype_expected",
-    [["b", "zstd"], ["i", "zstd"], ["f", "zstd"], ["|S1", "vlen-utf8"], ["|U1", "vlen-utf8"]],
+   # [["b", "zstd"], ["i", "zstd"], ["f", "zstd"], ["|S1", "vlen-utf8"], ["|U1", "vlen-utf8"]],
+    [["|S1", "vlen-bytes"]],
 )
 def test_default_filters_and_compressor(dtype_expected: Any) -> None:
     with config.set(
         {
             "v2_dtype_kind_to_default_filters_and_compressor": {
                 "biufcmM": ["zstd"],
-                "OSUV": ["vlen-utf8"],
+                "OSUV": ["vlen-bytes"],
             },
         }
     ):
         dtype, expected = dtype_expected
-        arr = zarr.create(shape=(10,), path="foo", store={}, zarr_format=2, dtype=dtype)
+        arr = zarr.create(shape=(3,), path="foo", store={}, zarr_format=2, dtype=dtype)
         assert arr.metadata.filters[0].codec_id == expected
-        print(arr.metadata)
+        arr[:] = np.array(["a", "bb", "ccc"], dtype=dtype)