github-actions[bot] commented on code in PR #39022: URL: https://github.com/apache/doris/pull/39022#discussion_r1706453662
########## be/src/vec/columns/column_object.cpp: ########## @@ -415,6 +456,71 @@ void ColumnObject::Subcolumn::insert(Field field, FieldInfo info) { data.back()->insert(field); } +static DataTypePtr create_array(TypeIndex type, size_t num_dimensions) { + DataTypePtr result_type; + auto nested_type = make_nullable(DataTypeFactory::instance().create_data_type(type)); + for (size_t i = 0; i < num_dimensions; ++i) { + result_type = std::make_shared<DataTypeArray>(nested_type); + } + return make_nullable(result_type); +} + +Array create_empty_array_field(size_t num_dimensions) { + if (num_dimensions == 0) { + throw doris::Exception(ErrorCode::INVALID_ARGUMENT, + "Cannot create array field with 0 dimensions"); + } + + Array array; + Array* current_array = &array; + for (size_t i = 1; i < num_dimensions; ++i) { + current_array->push_back(Array()); + current_array = ¤t_array->back().get<Array&>(); + } + + return array; +} + +// Recreates column with default scalar values and keeps sizes of arrays. +static ColumnPtr recreate_column_with_default_values(const ColumnPtr& column, TypeIndex scalar_type, + size_t num_dimensions) { + const auto* column_array = check_and_get_column<ColumnArray>(column.get()); + if (column_array && num_dimensions) { + return make_nullable(ColumnArray::create( + recreate_column_with_default_values(column_array->get_data_ptr(), scalar_type, + num_dimensions - 1), + IColumn::mutate(column_array->get_offsets_ptr()))); + } + + return create_array(scalar_type, num_dimensions) + ->create_column() + ->clone_resized(column->size()); +} + +ColumnObject::Subcolumn ColumnObject::Subcolumn::recreateWithDefaultValues( + const FieldInfo& field_info) const { + Subcolumn new_subcolumn(*this); + new_subcolumn.least_common_type = + LeastCommonType {create_array(field_info.scalar_type_id, field_info.num_dimensions)}; + + for (auto& part : new_subcolumn.data) { + part = recreate_column_with_default_values(part, field_info.scalar_type_id, + field_info.num_dimensions); + } + + return new_subcolumn; +} + +Field ColumnObject::Subcolumn::get_last_field() const { + if (data.empty()) { + return Field(); Review Comment: warning: avoid repeating the return type from the declaration; use a braced initializer list instead [modernize-return-braced-init-list] ```suggestion return {}; ``` ########## be/src/vec/json/path_in_data.h: ########## @@ -29,15 +29,18 @@ #include <vector> #include "gen_cpp/segment_v2.pb.h" Review Comment: warning: 'gen_cpp/segment_v2.pb.h' file not found [clang-diagnostic-error] ```cpp #include "gen_cpp/segment_v2.pb.h" ^ ``` ########## be/src/olap/rowset/segment_v2/column_reader.cpp: ########## @@ -1663,4 +1663,53 @@ Status VariantRootColumnIterator::read_by_rowids(const rowid_t* rowids, const si return Status::OK(); } +Status DefaultNestedColumnIterator::next_batch(size_t* n, vectorized::MutableColumnPtr& dst) { + bool has_null = false; + return next_batch(n, dst, &has_null); +} + +static void fill_nested_with_defaults(vectorized::MutableColumnPtr& dst, + vectorized::MutableColumnPtr& sibling_column, size_t nrows) { + const auto* sibling_array = vectorized::check_and_get_column<vectorized::ColumnArray>( + remove_nullable(sibling_column->get_ptr())); + CHECK(sibling_array) << "Expected array column, but mmet " << sibling_column->get_name(); + ; + const auto* dst_array = vectorized::check_and_get_column<vectorized::ColumnArray>( + remove_nullable(dst->get_ptr())); + if (!dst_array || !sibling_array) { + throw doris::Exception(ErrorCode::INTERNAL_ERROR, + "Expected array column, but met %s and %s", dst->get_name(), + sibling_column->get_name()); + } + auto new_nested = + dst_array->get_data_ptr()->clone_resized(sibling_array->get_data_ptr()->size()); + auto new_array = make_nullable(vectorized::ColumnArray::create( + new_nested->assume_mutable(), sibling_array->get_offsets_ptr()->assume_mutable())); + dst->insert_range_from(*new_array, 0, new_array->size()); +} + +Status DefaultNestedColumnIterator::next_batch(size_t* n, vectorized::MutableColumnPtr& dst, Review Comment: warning: pointer parameter 'n' can be pointer to const [readability-non-const-parameter] ```suggestion Status DefaultNestedColumnIterator::next_batch(const size_t* n, vectorized::MutableColumnPtr& dst, ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org For additional commands, e-mail: commits-h...@doris.apache.org