You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 2a89e10
Browse filesBrowse the repository at this point in the historyBrowse files
composes a transform and returns a new view without reading data, and
17
17
`result()` materializes it into owned system memory. `LazyArray(source)` uses
18
-
the conservative basic reader; `from_numpy`explicitly selects NumPy's
19
-
optimized reader. Device arrays require an explicit custom reader responsible
20
-
for transferring values into the supplied system-memory output buffer.
18
+
the basic reader; `from_numpy` selects `numpy_reader`, which currently uses
19
+
the same slab-and-gather implementation. Device sources that refuse NumPy
20
+
conversion need a custom reader to transfer values into the output buffer.
21
21
-`Reader` — the explicit backend execution boundary: transforms say which
22
22
values belong in the result, while readers say how a backend obtains them
23
23
-`IndexDomain` — a rectangular region of integer coordinates
@@ -32,9 +32,11 @@ Key types:
32
32
dimension can depend on the input
33
33
-`compose` — chain two transforms into one
34
34
35
-
The package depends only on NumPy and the standard library; it does not import
36
-
`zarr`. It is developed in the [zarr-python](https://github.com/zarr-developers/zarr-python)
37
-
repository and consumed by `zarr` to resolve array indexing operations.
35
+
The base package depends on NumPy and the standard library; its optional testing
36
+
module also requires Hypothesis. The package does not import `zarr`. It is developed
37
+
in the [zarr-python](https://github.com/zarr-developers/zarr-python) repository,
38
+
and its examples include reading Zarr arrays through Dask. Installing it does not
Reject invalid wire index-array values and unrepresentable normalized bounds/ranks, and raise an explicit error for unsupported intersections sharing an affine and lookup input axis instead of returning incorrect coordinates. Group negative chunk-coordinate tuples without merging distinct chunks.
Correct indexing, reader, serialization, cache, and integration descriptions to match supported behavior; qualify NumPy/TensorStore compatibility and performance claims.
2
+
3
+
Describe current contracts in source and test docstrings instead of narrating prior implementations. Clarify that immutable index coordinates do not snapshot source values.
Expand planner property tests across signed origins and chunk IDs, custom grids, mixed affine and lookup dependencies, duplicate coordinates, and empty domains. Verify exact request coverage and storage mapping with an independent pointwise oracle.
0 commit comments