ARROW-15277: [C++][Python] Use ChunkedArray::Make for chunked_array (#…

…13950) Supersedes and will close #12096 [ARROW-15277](https://issues.apache.org/jira/browse/ARROW-15277) Lead-authored-by: Miles Granger <[email protected]> Co-authored-by: Eduardo Ponce <[email protected]> Co-authored-by: Antoine Pitrou <[email protected]> Signed-off-by: Antoine Pitrou <[email protected]>
apache · Aug 25, 2022 · dd0988b · dd0988b
1 parent 897c186
commit dd0988b
Show file tree

Hide file tree

Showing 6 changed files with 24 additions and 25 deletions.
diff --git a/cpp/src/arrow/chunked_array.cc b/cpp/src/arrow/chunked_array.cc
@@ -72,7 +72,7 @@ Result<std::shared_ptr<ChunkedArray>> ChunkedArray::Make(ArrayVector chunks,
   }
   for (const auto& chunk : chunks) {
     if (!chunk->type()->Equals(*type)) {
-      return Status::Invalid("Array chunks must all be same type");
+      return Status::TypeError("Array chunks must all be same type");
     }
   }
   return std::make_shared<ChunkedArray>(std::move(chunks), std::move(type));

diff --git a/cpp/src/arrow/chunked_array_test.cc b/cpp/src/arrow/chunked_array_test.cc
@@ -65,8 +65,8 @@ TEST_F(TestChunkedArray, Make) {
   ASSERT_OK_AND_ASSIGN(auto result2, ChunkedArray::Make({chunk0, chunk0}, int8()));
   AssertChunkedEqual(*result, *result2);
 
-  ASSERT_RAISES(Invalid, ChunkedArray::Make({chunk0, chunk1}));
-  ASSERT_RAISES(Invalid, ChunkedArray::Make({chunk0}, int16()));
+  ASSERT_RAISES(TypeError, ChunkedArray::Make({chunk0, chunk1}));
+  ASSERT_RAISES(TypeError, ChunkedArray::Make({chunk0}, int16()));
 }
 
 TEST_F(TestChunkedArray, MakeEmpty) {

diff --git a/python/pyarrow/includes/libarrow.pxd b/python/pyarrow/includes/libarrow.pxd
@@ -760,6 +760,10 @@ cdef extern from "arrow/api.h" namespace "arrow" nogil:
         CChunkedArray(const vector[shared_ptr[CArray]]& arrays)
         CChunkedArray(const vector[shared_ptr[CArray]]& arrays,
                       const shared_ptr[CDataType]& type)
+
+        @staticmethod
+        CResult[shared_ptr[CChunkedArray]] Make(vector[shared_ptr[CArray]] chunks,
+                                                shared_ptr[CDataType] type)
         int64_t length()
         int64_t null_count()
         int num_chunks()

diff --git a/python/pyarrow/table.pxi b/python/pyarrow/table.pxi
@@ -1303,7 +1303,8 @@ def chunked_array(arrays, type=None):
     cdef:
         Array arr
         vector[shared_ptr[CArray]] c_arrays
-        shared_ptr[CChunkedArray] sp_chunked_array
+        shared_ptr[CChunkedArray] c_result
+        shared_ptr[CDataType] c_type
 
     type = ensure_type(type, allow_none=True)
 
@@ -1318,25 +1319,13 @@ def chunked_array(arrays, type=None):
             # subsequent arrays to the firstly inferred array type
             # it also spares the inference overhead after the first chunk
             type = arr.type
-        else:
-            if arr.type != type:
-                raise TypeError(
-                    "All array chunks must have type {}".format(type)
-                )
 
         c_arrays.push_back(arr.sp_array)
 
-    if c_arrays.size() == 0 and type is None:
-        raise ValueError("When passing an empty collection of arrays "
-                         "you must also pass the data type")
-
-    sp_chunked_array.reset(
-        new CChunkedArray(c_arrays, pyarrow_unwrap_data_type(type))
-    )
+    c_type = pyarrow_unwrap_data_type(type)
     with nogil:
-        check_status(sp_chunked_array.get().Validate())
-
-    return pyarrow_wrap_chunked_array(sp_chunked_array)
+        c_result = GetResultValue(CChunkedArray.Make(c_arrays, c_type))
+    return pyarrow_wrap_chunked_array(c_result)
 
 
 cdef _schema_from_arrays(arrays, names, metadata, shared_ptr[CSchema]* schema):

diff --git a/python/pyarrow/tests/test_compute.py b/python/pyarrow/tests/test_compute.py
@@ -394,6 +394,14 @@ def test_mode_chunked_array():
     assert len(pc.mode(arr)) == 0
 
 
+def test_empty_chunked_array():
+    msg = "cannot construct ChunkedArray from empty vector and omitted type"
+    with pytest.raises(pa.ArrowInvalid, match=msg):
+        pa.chunked_array([])
+
+    pa.chunked_array([], type=pa.int8())
+
+
 def test_variance():
     data = [1, 2, 3, 4, 5, 6, 7, 8]
     assert pc.variance(data).as_py() == 5.25

diff --git a/python/pyarrow/tests/test_table.py b/python/pyarrow/tests/test_table.py
@@ -97,10 +97,7 @@ def test_chunked_array_construction():
     assert len(arr) == 3
     assert len(arr.chunks) == 2
 
-    msg = (
-        "When passing an empty collection of arrays you must also pass the "
-        "data type"
-    )
+    msg = "cannot construct ChunkedArray from empty vector and omitted type"
     with pytest.raises(ValueError, match=msg):
         assert pa.chunked_array([])
 
@@ -143,14 +140,15 @@ def test_chunked_array_to_numpy():
 
 
 def test_chunked_array_mismatch_types():
-    with pytest.raises(TypeError):
+    msg = "chunks must all be same type"
+    with pytest.raises(TypeError, match=msg):
         # Given array types are different
         pa.chunked_array([
             pa.array([1, 2, 3]),
             pa.array([1., 2., 3.])
         ])
 
-    with pytest.raises(TypeError):
+    with pytest.raises(TypeError, match=msg):
         # Given array type is different from explicit type argument
         pa.chunked_array([pa.array([1, 2, 3])], type=pa.float64())