This is an automated email from the ASF dual-hosted git repository.
kou pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow.git
The following commit(s) were added to refs/heads/main by this push:
new 221484804f GH-47263: [MATLAB] Add `NumNulls` property to
`arrow.array.ChunkedArray` class (#47264)
221484804f is described below
commit 221484804ffae19bf55d7198e5773d9e956b7b42
Author: Sarah Gilmore <[email protected]>
AuthorDate: Wed Aug 6 20:51:36 2025 -0400
GH-47263: [MATLAB] Add `NumNulls` property to `arrow.array.ChunkedArray`
class (#47264)
### Rationale for this change
This is a followup to #38422. Now that `NumNulls` is a property on
`arrow.array.Array`, we should add `NumNulls` as a property on
`arrow.array.ChunkedArray`.
### What changes are included in this PR?
Added `NumNulls` as a property to `arrow.array.ChunkedArray`.
**Example**:
```matlab
>> a1 = arrow.array(1:10);
>> a2 = arrow.array([11 12 NaN 14 NaN]);
>> a3 = arrow.array([16 17 NaN 18 19]);
>> a4 = arrow.array(20:30);
>> C1 = arrow.array.ChunkedArray.fromArrays(a1, a2, a3)
C1 =
ChunkedArray with properties:
Type: [1×1 arrow.type.Float64Type]
NumChunks: 4
NumElements: 31
NumNulls: 3
>> C1.NumNulls
ans =
int64
3
```
### Are these changes tested?
1. Added a `NumNullsNoSetter` test case to `tChunkedArray.m`.
2. Updated `tChunkedArray/verifyChunkedArray` helper method to verify the
`NumNulls` property value is set to the expected value.
### Are there any user-facing changes?
Yes. `arrow.array.ChunkedArray` has a new public property called `NumNulls`.
* GitHub Issue: #47263
Authored-by: Sarah Gilmore <[email protected]>
Signed-off-by: Sutou Kouhei <[email protected]>
---
.../cpp/arrow/matlab/array/proxy/chunked_array.cc | 8 ++++++++
.../cpp/arrow/matlab/array/proxy/chunked_array.h | 2 ++
matlab/src/matlab/+arrow/+array/ChunkedArray.m | 9 +++++++--
matlab/test/arrow/array/tChunkedArray.m | 21 +++++++++++++++++++--
4 files changed, 36 insertions(+), 4 deletions(-)
diff --git a/matlab/src/cpp/arrow/matlab/array/proxy/chunked_array.cc
b/matlab/src/cpp/arrow/matlab/array/proxy/chunked_array.cc
index 00410f83e3..c5f188b8df 100644
--- a/matlab/src/cpp/arrow/matlab/array/proxy/chunked_array.cc
+++ b/matlab/src/cpp/arrow/matlab/array/proxy/chunked_array.cc
@@ -52,6 +52,7 @@
ChunkedArray::ChunkedArray(std::shared_ptr<arrow::ChunkedArray> chunked_array)
: chunked_array{std::move(chunked_array)} {
// Register Proxy methods.
REGISTER_METHOD(ChunkedArray, getNumElements);
+ REGISTER_METHOD(ChunkedArray, getNumNulls);
REGISTER_METHOD(ChunkedArray, getNumChunks);
REGISTER_METHOD(ChunkedArray, getChunk);
REGISTER_METHOD(ChunkedArray, getType);
@@ -94,6 +95,13 @@ void
ChunkedArray::getNumElements(libmexclass::proxy::method::Context& context)
context.outputs[0] = num_elements_mda;
}
+void ChunkedArray::getNumNulls(libmexclass::proxy::method::Context& context) {
+ namespace mda = ::matlab::data;
+ mda::ArrayFactory factory;
+ auto num_nulls_mda = factory.createScalar(chunked_array->null_count());
+ context.outputs[0] = num_nulls_mda;
+}
+
void ChunkedArray::getNumChunks(libmexclass::proxy::method::Context& context) {
namespace mda = ::matlab::data;
mda::ArrayFactory factory;
diff --git a/matlab/src/cpp/arrow/matlab/array/proxy/chunked_array.h
b/matlab/src/cpp/arrow/matlab/array/proxy/chunked_array.h
index 1ba0feb572..34da4324ae 100644
--- a/matlab/src/cpp/arrow/matlab/array/proxy/chunked_array.h
+++ b/matlab/src/cpp/arrow/matlab/array/proxy/chunked_array.h
@@ -37,6 +37,8 @@ class ChunkedArray : public libmexclass::proxy::Proxy {
protected:
void getNumElements(libmexclass::proxy::method::Context& context);
+ void getNumNulls(libmexclass::proxy::method::Context& context);
+
void getNumChunks(libmexclass::proxy::method::Context& context);
void getChunk(libmexclass::proxy::method::Context& context);
diff --git a/matlab/src/matlab/+arrow/+array/ChunkedArray.m
b/matlab/src/matlab/+arrow/+array/ChunkedArray.m
index f0754bf391..b1d07d76bb 100644
--- a/matlab/src/matlab/+arrow/+array/ChunkedArray.m
+++ b/matlab/src/matlab/+arrow/+array/ChunkedArray.m
@@ -26,6 +26,7 @@ classdef ChunkedArray < matlab.mixin.CustomDisplay & ...
Type
NumChunks
NumElements
+ NumNulls
end
methods
@@ -45,6 +46,10 @@ classdef ChunkedArray < matlab.mixin.CustomDisplay & ...
numElements = obj.Proxy.getNumElements();
end
+ function numNulls = get.NumNulls(obj)
+ numNulls = obj.Proxy.getNumNulls();
+ end
+
function type = get.Type(obj)
typeStruct = obj.Proxy.getType();
traits =
arrow.type.traits.traits(arrow.type.ID(typeStruct.TypeID));
@@ -123,5 +128,5 @@ classdef ChunkedArray < matlab.mixin.CustomDisplay & ...
chunkedArray = arrow.array.ChunkedArray(proxy);
end
- end
-end
\ No newline at end of file
+ end
+end
diff --git a/matlab/test/arrow/array/tChunkedArray.m
b/matlab/test/arrow/array/tChunkedArray.m
index c37206241b..412e575a9c 100644
--- a/matlab/test/arrow/array/tChunkedArray.m
+++ b/matlab/test/arrow/array/tChunkedArray.m
@@ -113,8 +113,9 @@ classdef tChunkedArray < matlab.unittest.TestCase
%
% 1. Their Type properties are equal
% 2. Their NumElements properties are equal
- % 3. The same elements are considered null
- % 4. All corresponding valid elements have the same values
+ % 3. Their NumNulls properties are equal
+ % 4. The same elements are considered null
+ % 5. All corresponding valid elements have the same values
%
% NOTE: Having the same "chunking" is not a requirement for two
% ChunkedArrays to be equal. ChunkedArrays are considered equal
@@ -204,6 +205,18 @@ classdef tChunkedArray < matlab.unittest.TestCase
testCase.verifyError(fcn, "MATLAB:class:SetProhibited");
end
+ function NumNullsNoSetter(testCase)
+ % Verify an error is thrown when trying to set the value
+ % of the NumNulls property.
+ import arrow.array.ChunkedArray
+
+ arrays = {testCase.Float64Array1, testCase.Float64Array2,
testCase.Float64Array3};
+ chunkedArray = ChunkedArray.fromArrays(arrays{:});
+
+ fcn = @() setfield(chunkedArray, "NumNulls", int64(100));
+ testCase.verifyError(fcn, "MATLAB:class:SetProhibited");
+ end
+
function ChunkNonNumericIndexError(testCase)
% Verify that an error is thrown when a non-numeric index value
% is provided to the chunk() method.
@@ -476,10 +489,14 @@ classdef tChunkedArray < matlab.unittest.TestCase
end
testCase.assertTrue(numel(opts.Arrays) == opts.NumChunks);
allNumElements = cellfun(@(a) a.NumElements, opts.Arrays,
UniformOutput=true);
+ allNumNulls = cellfun(@(a) a.NumNulls, opts.Arrays,
UniformOutput=true);
+
expectedNumElements = int64(sum(allNumElements));
+ expectedNumNulls = int64(sum(allNumNulls));
testCase.verifyEqual(chunkedArray.NumChunks, opts.NumChunks);
testCase.verifyEqual(chunkedArray.NumElements,
expectedNumElements);
+ testCase.verifyEqual(chunkedArray.NumNulls, expectedNumNulls);
testCase.verifyEqual(chunkedArray.Type, opts.Type);
for ii = 1:opts.NumChunks
