Slight refactor of TrialParameters to localize everything needed to b…

…uild a ts spec

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "czpeedy"
-version = "0.2.1"
+version = "0.2.2"
 description = "A command-line tool used to determine the tensorstore settings which yield the fastest write speed on a given machine."
 authors = [
     { name = "Seth Hinz", email = "[email protected]" }

src/czpeedy/czpeedy.py

@@ -230,6 +230,7 @@ def main() -> None:
         parameter_space = ParameterSpace(
             data.shape,
             args.chunk_size,
+            args.dest,
             data.dtype,
             args.zarr_version,
             args.clevel,
@@ -243,7 +244,6 @@ def main() -> None:
         runner = Runner(
             parameter_space.all_combinations(),
             data,
-            args.dest,
             args.repetitions,
             parameter_space.num_combinations,
         )

src/czpeedy/parameter_space.py

@@ -1,5 +1,6 @@
 from typing import Iterable, Iterator
 from itertools import product
+from pathlib import Path
 
 import numpy as np
 from numpy.typing import ArrayLike
@@ -15,6 +16,7 @@ class ParameterSpace:
 
     shape: tuple[int, ...]
     dtype: np.dtype
+    dest: Path
     zarr_versions: list[int]
     clevels: list[int]
     compressors: list[str]
@@ -27,6 +29,7 @@ def __init__(
         self,
         shape: ArrayLike,
         chunk_sizes: Iterable[ArrayLike],
+        dest: Path,
         dtype: np.dtype,
         # Default parameters would be nice, but the user needs to be able to explicitly pass None and
         # have the parameters still be set when needed (i.e. in ParameterSpace(..., clevels=args.clevels, ...),
@@ -90,6 +93,7 @@ def __init__(
 
         self.shape = tuple(shape)
         self.chunk_sizes = [tuple(chunk_size) for chunk_size in chunk_sizes]
+        self.dest = dest
         self.dtype = dtype
         self.zarr_versions = zarr_versions
         self.clevels = list(clevels)
@@ -150,6 +154,7 @@ def to_trial_parameters(
             return TrialParameters(
                 self.shape,
                 chunk_size,
+                self.dest,
                 self.dtype,
                 zarr_version=zarr_version,
                 clevel=clevel,

src/czpeedy/runner.py

@@ -15,7 +15,6 @@
 class Runner:
     trial_params: Iterable[TrialParameters]
     data: np.ndarray
-    dest: Path
     repetitions: int
     batch_count: int
     results: dict[TrialParameters, list[float]]
@@ -24,13 +23,11 @@ def __init__(
         self,
         trial_params: Iterable[TrialParameters],
         data: np.ndarray,
-        dest: Path,
         repetitions: int,
         batch_count: int | None = None,
     ):
         self.trial_params = trial_params
         self.data = data
-        self.dest = dest
         self.repetitions = repetitions
         self.batch_count = batch_count
         self.results = {}
@@ -53,7 +50,7 @@ def run_all(self):
 
         for batch_id, trial_param in enumerate(self.trial_params):
             result = []
-            spec = trial_param.to_spec(self.dest)
+            spec = trial_param.to_spec()
             codecs = ts.CodecSpec(trial_param.codecs())
 
             dataset = ts.open(spec, codec=codecs).result()

src/czpeedy/trial_parameters.py

@@ -15,13 +15,15 @@ class TrialParameters:
     compressor: str
     shuffle: int
     endianness: int
+    output_path: Path
 
     # All parameters are either obvious or can be referenced in tensorstore's spec documentation, with the exception of `endianness`.
     # `endianness` is -1 for little endian, 0 for indeterminate endianness (only applies for 1 byte values), and +1 for big endian.
     def __init__(
         self,
         shape: ArrayLike[int],
         chunk_size: ArrayLike,
+        output_path: Path,
         dtype: np.dtype,
         zarr_version: int,
         clevel: int,
@@ -31,6 +33,7 @@ def __init__(
     ):
         self.shape = shape
         self.chunk_size = list(chunk_size)
+        self.output_path = output_path
         self.dtype = dtype
         self.zarr_version = zarr_version
         self.clevel = clevel
@@ -55,13 +58,13 @@ def dtype_json_v3(self) -> str:
 
     # Produces a jsonable dict that communicates all the trial parameters to tensorstore.
     # Usage: `ts.open(trial_parameters.to_spec()).result()`
-    def to_spec(self, output_path: Path) -> dict:
+    def to_spec(self) -> dict:
         if self.zarr_version == 2:
             return {
                 "driver": "zarr",
                 "kvstore": {
                     "driver": "file",
-                    "path": str(output_path.absolute()),
+                    "path": str(self.output_path.absolute()),
                 },
                 "metadata": {
                     "compressor": {
@@ -119,7 +122,7 @@ def to_spec(self, output_path: Path) -> dict:
                 "driver": "zarr3",
                 "kvstore": {
                     "driver": "file",
-                    "path": str(output_path.absolute()),
+                    "path": str(self.output_path.absolute()),
                 },
                 "metadata": {
                     "shape": self.shape,