Skip to content

GH-52161: [C++][Parquet] Add support for string/binary view to content defined chunking writer #52162

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.

Sign up for GitHub

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

Merged
HuaHuaY merged 2 commits into from
Oct 12, 2026
Merged

GH-52161: [C++][Parquet] Add support for string/binary view to content defined chunking writer #52162

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2aa540e
Add support for string/binary view to content defined chunking writer
robert3005 Oct 7, 2026
d69a2d9
fixes
robert3005 Oct 8, 2026
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
14 changes: 14 additions & 0 deletions cpp/src/parquet/chunker_internal.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,7 @@
#include <cstdint>
#include <iterator>
#include <string>
#include <string_view>
#include <vector>

#include "arrow/array.h"
Expand Down Expand Up @@ -349,6 +350,17 @@ class ContentDefinedChunker::Impl {
});
}

std::vector<Chunk> CalculateBinaryView(const int16_t* def_levels,
const int16_t* rep_levels, int64_t num_levels,
const ::arrow::Array& values) {
const auto& array = checked_cast<const ::arrow::BinaryViewArray&>(values);
return Calculate(def_levels, rep_levels, num_levels, [&](int64_t i) {
const std::string_view value = array.GetView(i);
Roll(reinterpret_cast<const uint8_t*>(value.data()),
static_cast<int64_t>(value.size()));
});
}

std::vector<Chunk> GetChunks(const int16_t* def_levels, const int16_t* rep_levels,
int64_t num_levels, const ::arrow::Array& values) {
auto handle_type = [&](auto&& type) -> std::vector<Chunk> {
Expand Down Expand Up @@ -380,6 +392,8 @@ class ContentDefinedChunker::Impl {
} else if constexpr (::arrow::is_large_binary_like(ArrowType::type_id)) {
return CalculateBinaryLike<::arrow::LargeBinaryArray>(def_levels, rep_levels,
num_levels, values);
} else if constexpr (::arrow::is_binary_view_like(ArrowType::type_id)) {
return CalculateBinaryView(def_levels, rep_levels, num_levels, values);
} else if constexpr (::arrow::is_dictionary(ArrowType::type_id)) {
return GetChunks(def_levels, rep_levels, num_levels,
*static_cast<const ::arrow::DictionaryArray&>(values).indices());
Expand Down
15 changes: 13 additions & 2 deletions cpp/src/parquet/chunker_internal_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,7 @@
#include "arrow/testing/generator.h"
#include "arrow/type_fwd.h"
#include "arrow/util/float16.h"
#include "arrow/visit_data_inline.h"
#include "parquet/arrow/reader.h"
#include "parquet/arrow/reader_internal.h"
#include "parquet/arrow/test_util.h"
Expand Down Expand Up @@ -170,6 +171,10 @@ Result<std::shared_ptr<Array>> GenerateArray(const std::shared_ptr<Field>& field
std::string("bin_") + std::to_string(val))
GENERATE_CASE(LARGE_BINARY, ::arrow::LargeBinaryBuilder,
std::string("bin_") + std::to_string(val))
GENERATE_CASE(STRING_VIEW, ::arrow::StringViewBuilder,
std::string("str_") + std::to_string(val))
GENERATE_CASE(BINARY_VIEW, ::arrow::BinaryViewBuilder,
std::string("bin_") + std::to_string(val))
case ::arrow::Type::FIXED_SIZE_BINARY: {
auto size =
checked_cast<const ::arrow::FixedSizeBinaryType*>(type.get())->byte_width();
Expand Down Expand Up @@ -742,6 +747,10 @@ Result<int64_t> CalculateCdcSize(const std::shared_ptr<Array>& array, bool nulla
} else if (::arrow::is_large_binary_like(type_id)) {
auto binary_array = checked_cast<const ::arrow::LargeBinaryArray*>(array.get());
result += binary_array->total_values_length();
} else if (::arrow::is_binary_view_like(type_id)) {
::arrow::VisitArraySpanInline<::arrow::BinaryViewType>(
*array->data(),
[&](std::string_view v) { result += static_cast<int64_t>(v.size()); }, [] {});
} else {
return Status::NotImplemented("CDC size calculation for type ",
array->type()->ToString(), " is not implemented");
Expand Down Expand Up @@ -784,7 +793,8 @@ void AssertContentDefinedChunkSizes(const std::shared_ptr<::arrow::ChunkedArray>
ASSERT_EQ(column_info.has_dictionary_page, expect_dictionary_page);
}

if (::arrow::is_fixed_width(type_id) || ::arrow::is_base_binary_like(type_id)) {
if (::arrow::is_fixed_width(type_id) || ::arrow::is_base_binary_like(type_id) ||
::arrow::is_binary_view_like(type_id)) {
int64_t offset = 0;

auto page_lengths = column_info.page_lengths;
Expand Down Expand Up @@ -1451,7 +1461,7 @@ INSTANTIATE_TEST_SUITE_P(
testing::Values(
CaseConfig{::arrow::boolean(), false}, CaseConfig{::arrow::int64(), true},
// Binary-like
CaseConfig{::arrow::utf8(), false},
CaseConfig{::arrow::utf8(), false}, CaseConfig{::arrow::utf8_view(), true},
CaseConfig{::arrow::fixed_size_binary(16), true},
// Nested types
CaseConfig{::arrow::list(::arrow::int32()), false},
Expand All @@ -1475,6 +1485,7 @@ INSTANTIATE_TEST_SUITE_P(
// Binary-like
CaseConfig{::arrow::utf8(), false}, CaseConfig{::arrow::binary(), true},
CaseConfig{::arrow::fixed_size_binary(16), true},
CaseConfig{::arrow::utf8_view(), false}, CaseConfig{::arrow::binary_view(), true},
// Temporal
CaseConfig{::arrow::date32(), false},
CaseConfig{::arrow::time32(::arrow::TimeUnit::MILLI), true},
Expand Down
Loading