rename v2_dtype_kind_to_default_filters_and_compressor to v2_default_compressors

Author: brokkoli71

src/zarr/core/config.py

@@ -64,10 +64,10 @@ def reset(self) -> None:
             },
             "buffer": "zarr.core.buffer.cpu.Buffer",
             "ndbuffer": "zarr.core.buffer.cpu.NDBuffer",
-            "v2_dtype_kind_to_default_filters_and_compressor": {
-                "biufcmM": ["zstd"],
-                "U": ["vlen-utf8"],
-                "OSV": ["vlen-bytes"],
+            "v2_default_compressors": {
+                "numeric": ["zstd"],
+                "unicode": ["vlen-utf8"],
+                "bytes": ["vlen-bytes"],
             },
         }
     ],

src/zarr/core/metadata/v2.py

@@ -341,13 +341,16 @@ def _default_filters_and_compressor(
 ) -> tuple[list[dict[str, str]], dict[str, str] | None]:
     """Get the default filters and compressor for a dtype.
 
-    The config contains a mapping from numpy dtype kind to the default compressor.
     https://numpy.org/doc/2.1/reference/generated/numpy.dtype.kind.html
     """
-    dtype_kind_to_default_compressor = config.get("v2_dtype_kind_to_default_filters_and_compressor")
-    for dtype_kinds, filters_and_compressor in dtype_kind_to_default_compressor.items():
-        if dtype.kind in dtype_kinds:
-            filters = [{"id": f} for f in filters_and_compressor]
-            compressor = None
-            return filters, compressor
-    return [], None
+    default_compressors = config.get("v2_default_compressors")
+    if dtype.kind in "biufcmM":
+        dtype_key = "numeric"
+    elif dtype.kind in "U":
+        dtype_key = "unicode"
+    elif dtype.kind in "OSV":
+        dtype_key = "bytes"
+    else:
+        raise ValueError(f"Unsupported dtype kind {dtype.kind}")
+
+    return [{"id": f} for f in default_compressors[dtype_key]], None

tests/test_config.py

@@ -63,10 +63,10 @@ def test_config_defaults_set() -> None:
                 "vlen-utf8": "zarr.codecs.vlen_utf8.VLenUTF8Codec",
                 "vlen-bytes": "zarr.codecs.vlen_utf8.VLenBytesCodec",
             },
-            "v2_dtype_kind_to_default_filters_and_compressor": {
-                "biufcmM": ["zstd"],
-                "U": ["vlen-utf8"],
-                "OSV": ["vlen-bytes"],
+            "v2_default_compressors": {
+                "numeric": ["zstd"],
+                "unicode": ["vlen-utf8"],
+                "bytes": ["vlen-bytes"],
             },
         }
     ]

tests/test_v2.py

@@ -84,8 +84,8 @@ def test_codec_pipeline() -> None:
 async def test_v2_encode_decode(dtype):
     with config.set(
         {
-            "v2_dtype_kind_to_default_filters_and_compressor": {
-                "SV": ["vlen-bytes"],
+            "v2_default_compressors": {
+                "bytes": ["vlen-bytes"],
             },
         }
     ):
@@ -126,9 +126,9 @@ def test_v2_encode_decode_with_data(dtype_value):
     dtype, value = dtype_value
     with config.set(
         {
-            "v2_dtype_kind_to_default_filters_and_compressor": {
-                "U": ["vlen-utf8"],
-                "OSV": ["vlen-bytes"],
+            "v2_default_compressors": {
+                "unicode": ["vlen-utf8"],
+                "bytes": ["vlen-bytes"],
             },
         }
     ):
@@ -171,9 +171,9 @@ def test_default_filters_and_compressor(dtype_expected: Any) -> None:
     with config.set(
         {
             "v2_dtype_kind_to_default_filters_and_compressor": {
-                "biufcmM": ["zstd"],
-                "U": ["vlen-utf8"],
-                "OSV": ["vlen-bytes"],
+                "numeric": ["zstd"],
+                "unicode": ["vlen-utf8"],
+                "bytes": ["vlen-bytes"],
             },
         }
     ):