sparrow-ipc 0.2.0
Loading...
Searching...
No Matches
deserialize_primitive_array.hpp
Go to the documentation of this file.
1#pragma once
2
3#include <optional>
4#include <unordered_set>
5#include <vector>
6
7#include <sparrow/arrow_interface/arrow_array_schema_proxy.hpp>
8#include <sparrow/primitive_array.hpp>
9
10#include "Message_generated.h"
14
15namespace sparrow_ipc
16{
17 template <typename T>
18 [[nodiscard]] sparrow::primitive_array<T> deserialize_non_owning_primitive_array(
19 const org::apache::arrow::flatbuf::RecordBatch& record_batch,
20 std::span<const uint8_t> body,
21 std::string_view name,
22 const std::optional<std::vector<sparrow::metadata_pair>>& metadata,
23 bool nullable,
24 size_t& buffer_index
25 )
26 {
27 const std::string_view format = data_type_to_format(
28 sparrow::detail::get_data_type_from_array<sparrow::primitive_array<T>>::get()
29 );
30
31 // Set up flags based on nullable
32 std::optional<std::unordered_set<sparrow::ArrowFlag>> flags;
33 if (nullable)
34 {
35 flags = std::unordered_set<sparrow::ArrowFlag>{sparrow::ArrowFlag::NULLABLE};
36 }
37
38 ArrowSchema schema = make_non_owning_arrow_schema(
39 format,
40 name.data(),
41 metadata,
42 flags,
43 0,
44 nullptr,
45 nullptr
46 );
47
48 const auto compression = record_batch.compression();
49 std::vector<arrow_array_private_data::optionally_owned_buffer> buffers;
50
51 auto validity_buffer_span = utils::get_buffer(record_batch, body, buffer_index);
52 auto data_buffer_span = utils::get_buffer(record_batch, body, buffer_index);
53
54 if (compression)
55 {
56 buffers.push_back(utils::get_decompressed_buffer(validity_buffer_span, compression));
57 buffers.push_back(utils::get_decompressed_buffer(data_buffer_span, compression));
58 }
59 else
60 {
61 buffers.push_back(validity_buffer_span);
62 buffers.push_back(data_buffer_span);
63 }
64
65 // TODO bitmap_ptr is not used anymore... Leave it for now, and remove later if no need confirmed
66 const auto [bitmap_ptr, null_count] = utils::get_bitmap_pointer_and_null_count(validity_buffer_span, record_batch.length());
67
69 record_batch.length(),
70 null_count,
71 0,
72 0,
73 nullptr,
74 nullptr,
75 std::move(buffers)
76 );
77
78 sparrow::arrow_proxy ap{std::move(array), std::move(schema)};
79 return sparrow::primitive_array<T>{std::move(ap)};
80 }
81}
std::span< const uint8_t > get_buffer(const org::apache::arrow::flatbuf::RecordBatch &record_batch, std::span< const uint8_t > body, size_t &buffer_index)
Extracts a buffer from a RecordBatch's body.
std::pair< std::uint8_t *, int64_t > get_bitmap_pointer_and_null_count(std::span< const uint8_t > validity_buffer_span, const int64_t length)
Extracts bitmap pointer and null count from a validity buffer span.
std::variant< std::vector< std::uint8_t >, std::span< const std::uint8_t > > get_decompressed_buffer(std::span< const uint8_t > buffer_span, const org::apache::arrow::flatbuf::BodyCompression *compression)
Retrieves a decompressed buffer or a view of the original buffer.
ArrowSchema make_non_owning_arrow_schema(std::string_view format, const char *name, std::optional< M > metadata, std::optional< std::unordered_set< sparrow::ArrowFlag > > flags, size_t children_count, ArrowSchema **children, ArrowSchema *dictionary)
sparrow::primitive_array< T > deserialize_non_owning_primitive_array(const org::apache::arrow::flatbuf::RecordBatch &record_batch, std::span< const uint8_t > body, std::string_view name, const std::optional< std::vector< sparrow::metadata_pair > > &metadata, bool nullable, size_t &buffer_index)
ArrowArray make_arrow_array(int64_t length, int64_t null_count, int64_t offset, size_t children_count, ArrowArray **children, ArrowArray *dictionary, Arg &&private_data_arg)