Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Feat/write empty chunks #2429

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from 4 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
27 changes: 15 additions & 12 deletions src/zarr/codecs/pipeline.py
Original file line number Diff line number Diff line change
Expand Up @@ -331,6 +331,7 @@ async def write_batch(
value: NDBuffer,
drop_axes: tuple[int, ...] = (),
) -> None:
write_empty_chunks = config.get("array.write_empty_chunks") == True # noqa: E712
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have some concerns about unpacking this config value so deep in the stack. I'd rather make this a property of the Array so that we can guarantee consistent write behavior after an Array has been initialized.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'd rather make this a property of the Array so that we can guarantee consistent write behavior after an Array has been initialized.

If we do that, then we also require that users create a brand new array if they want to write just some parts of the data with different empty chunks handling (or we introduce write_empty_chunks as a mutable attribute, which I would rather avoid)

if self.supports_partial_encode:
await self.encode_partial_batch(
[
Expand Down Expand Up @@ -360,7 +361,7 @@ async def _read_key(
_read_key,
config.get("async.concurrency"),
)
chunk_array_batch = await self.decode_batch(
chunk_array_decoded = await self.decode_batch(
[
(chunk_bytes, chunk_spec)
for chunk_bytes, (_, chunk_spec, _, _) in zip(
Expand All @@ -369,23 +370,25 @@ async def _read_key(
],
)

chunk_array_batch = [
chunk_array_merged = [
self._merge_chunk_array(
chunk_array, value, out_selection, chunk_spec, chunk_selection, drop_axes
)
for chunk_array, (_, chunk_spec, chunk_selection, out_selection) in zip(
chunk_array_batch, batch_info, strict=False
)
]

chunk_array_batch = [
None
if chunk_array is None or chunk_array.all_equal(chunk_spec.fill_value)
else chunk_array
for chunk_array, (_, chunk_spec, _, _) in zip(
chunk_array_batch, batch_info, strict=False
chunk_array_decoded, batch_info, strict=False
)
]
chunk_array_batch: list[NDBuffer | None] = []
for chunk_array, (_, chunk_spec, _, _) in zip(
chunk_array_merged, batch_info, strict=False
):
if chunk_array is None:
chunk_array_batch.append(None) # type: ignore[unreachable]
else:
if not write_empty_chunks and chunk_array.all_equal(chunk_spec.fill_value):
chunk_array_batch.append(None)
else:
chunk_array_batch.append(chunk_array)

chunk_bytes_batch = await self.encode_batch(
[
Expand Down
2 changes: 1 addition & 1 deletion src/zarr/core/config.py
Original file line number Diff line number Diff line change
Expand Up @@ -42,7 +42,7 @@ def reset(self) -> None:
defaults=[
{
"default_zarr_version": 3,
"array": {"order": "C"},
"array": {"order": "C", "write_empty_chunks": False},
"async": {"concurrency": 10, "timeout": None},
"threading": {"max_workers": None},
"json_indent": 2,
Expand Down
37 changes: 37 additions & 0 deletions tests/test_array.py
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@
from zarr.core.array import chunks_initialized
from zarr.core.buffer.cpu import NDBuffer
from zarr.core.common import JSON, MemoryOrder, ZarrFormat
from zarr.core.config import config
from zarr.core.group import AsyncGroup
from zarr.core.indexing import ceildiv
from zarr.core.sync import sync
Expand Down Expand Up @@ -436,3 +437,39 @@ def test_array_create_order(
assert vals.flags.f_contiguous
else:
raise AssertionError


@pytest.mark.parametrize("store", ["memory"], indirect=True)
@pytest.mark.parametrize("write_empty_chunks", [True, False])
@pytest.mark.parametrize("fill_value", [0, 5])
def test_write_empty_chunks(
zarr_format: ZarrFormat, store: MemoryStore, write_empty_chunks: bool, fill_value: int
) -> None:
"""
Check that the write_empty_chunks value of the config is applied correctly. We expect that
when write_empty_chunks is True, writing chunks equal to the fill value will result in
those chunks appearing in the store.

When write_empty_chunks is False, writing chunks that are equal to the fill value will result in
those chunks not being present in the store. In particular, they should be deleted if they were
already present.
"""
arr = Array.create(
store=store,
shape=(2,),
zarr_format=zarr_format,
dtype="i4",
fill_value=fill_value,
chunk_shape=(1,),
)

# initialize the store with some non-fill value chunks
arr[:] = fill_value + 1
assert arr.nchunks_initialized == arr.nchunks

with config.set({"array.write_empty_chunks": write_empty_chunks}):
arr[:] = fill_value
if not write_empty_chunks:
assert arr.nchunks_initialized == 0
else:
assert arr.nchunks_initialized == arr.nchunks
2 changes: 1 addition & 1 deletion tests/test_config.py
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,7 @@ def test_config_defaults_set() -> None:
assert config.defaults == [
{
"default_zarr_version": 3,
"array": {"order": "C"},
"array": {"order": "C", "write_empty_chunks": False},
"async": {"concurrency": 10, "timeout": None},
"threading": {"max_workers": None},
"json_indent": 2,
Expand Down