Move default chunk sizes to constant

Author: benjeffery

bio2zarr/icf.py

@@ -1044,10 +1044,10 @@ def generate_schema(
 
         dimensions = {
             "variants": vcz.VcfZarrDimension(
-                size=m, chunk_size=variants_chunk_size or 1000
+                size=m, chunk_size=variants_chunk_size or vcz.DEFAULT_VARIANT_CHUNK_SIZE
             ),
             "samples": vcz.VcfZarrDimension(
-                size=n, chunk_size=samples_chunk_size or 10000
+                size=n, chunk_size=samples_chunk_size or vcz.DEFAULT_SAMPLE_CHUNK_SIZE
             ),
             # ploidy added conditionally below
             "alleles": vcz.VcfZarrDimension(

bio2zarr/plink.py

@@ -72,10 +72,10 @@ def generate_schema(
         # Define dimensions with sizes and chunk sizes
         dimensions = {
             "variants": vcz.VcfZarrDimension(
-                size=m, chunk_size=variants_chunk_size or 1000
+                size=m, chunk_size=variants_chunk_size or vcz.DEFAULT_VARIANT_CHUNK_SIZE
             ),
             "samples": vcz.VcfZarrDimension(
-                size=n, chunk_size=samples_chunk_size or 10000
+                size=n, chunk_size=samples_chunk_size or vcz.DEFAULT_SAMPLE_CHUNK_SIZE
             ),
             "ploidy": vcz.VcfZarrDimension(size=2),
             "alleles": vcz.VcfZarrDimension(size=2),

bio2zarr/vcz.py

@@ -15,6 +15,8 @@
 logger = logging.getLogger(__name__)
 
 ZARR_SCHEMA_FORMAT_VERSION = "0.5"
+DEFAULT_VARIANT_CHUNK_SIZE = 1000
+DEFAULT_SAMPLE_CHUNK_SIZE = 10_000
 DEFAULT_ZARR_COMPRESSOR = numcodecs.Blosc(cname="zstd", clevel=7)
 DEFAULT_ZARR_COMPRESSOR_GENOTYPES = numcodecs.Blosc(
     cname="zstd", clevel=7, shuffle=numcodecs.Blosc.BITSHUFFLE
@@ -273,8 +275,12 @@ def __init__(
         self.defaults = defaults
         if dimensions is None:
             dimensions = {
-                "variants": VcfZarrDimension(size=0, chunk_size=1000),
-                "samples": VcfZarrDimension(size=0, chunk_size=10_000),
+                "variants": VcfZarrDimension(
+                    size=0, chunk_size=DEFAULT_VARIANT_CHUNK_SIZE
+                ),
+                "samples": VcfZarrDimension(
+                    size=0, chunk_size=DEFAULT_SAMPLE_CHUNK_SIZE
+                ),
             }
         self.dimensions = dimensions