mapleFU commented on code in PR #14341:
URL: https://github.com/apache/arrow/pull/14341#discussion_r1149423665
##########
cpp/src/parquet/encoding.cc:
##########
@@ -3037,11 +3073,180 @@ class RleBooleanDecoder : public DecoderImpl, virtual
public BooleanDecoder {
// ----------------------------------------------------------------------
// DELTA_BYTE_ARRAY
-class DeltaByteArrayDecoder : public DecoderImpl,
- virtual public TypedDecoder<ByteArrayType> {
+/// Delta Byte Array encoding also known as incremental encoding or front
compression:
+/// for each element in a sequence of strings, store the prefix length of the
previous
+/// entry plus the suffix.
+///
+/// This is stored as a sequence of delta-encoded prefix lengths
(DELTA_BINARY_PACKED),
+/// followed by the suffixes encoded as delta length byte arrays
+/// (DELTA_LENGTH_BYTE_ARRAY).
+
+// ----------------------------------------------------------------------
+// DeltaByteArrayEncoder
+
+template <typename DType>
+class DeltaByteArrayEncoder : public EncoderImpl, virtual public
TypedEncoder<DType> {
public:
- explicit DeltaByteArrayDecoder(const ColumnDescriptor* descr,
+ using T = typename DType::c_type;
+
+ explicit DeltaByteArrayEncoder(const ColumnDescriptor* descr,
MemoryPool* pool =
::arrow::default_memory_pool())
+ : EncoderImpl(descr, Encoding::DELTA_BYTE_ARRAY, pool),
+ sink_(pool),
+ prefix_length_encoder_(nullptr, pool),
+ suffix_encoder_(nullptr, pool),
+ last_value_("") {}
+
+ std::shared_ptr<Buffer> FlushValues() override;
+
+ int64_t EstimatedDataEncodedSize() override {
+ return prefix_length_encoder_.EstimatedDataEncodedSize() +
+ suffix_encoder_.EstimatedDataEncodedSize();
+ }
+
+ using TypedEncoder<DType>::Put;
+
+ void Put(const ::arrow::Array& values) override;
+
+ void Put(const T* buffer, int num_values) override;
+
+ void PutSpaced(const T* src, int num_values, const uint8_t* valid_bits,
+ int64_t valid_bits_offset) override {
+ if (valid_bits != NULLPTR) {
+ PARQUET_ASSIGN_OR_THROW(auto buffer, ::arrow::AllocateBuffer(num_values
* sizeof(T),
+
this->memory_pool()));
+ T* data = reinterpret_cast<T*>(buffer->mutable_data());
+ int num_valid_values = ::arrow::util::internal::SpacedCompress<T>(
+ src, num_values, valid_bits, valid_bits_offset, data);
+ Put(data, num_valid_values);
+ } else {
+ Put(src, num_values);
+ }
+ }
+
+ protected:
+ template <typename ArrayType>
+ void PutBinaryArray(const ArrayType& array) {
+ uint32_t previous_len = 0;
+ std::string_view last_value_view = last_value_;
+
+ PARQUET_THROW_NOT_OK(::arrow::VisitArraySpanInline<typename
ArrayType::TypeClass>(
+ *array.data(),
+ [&](::std::string_view view) {
+ if (ARROW_PREDICT_FALSE(view.size() >= kMaxByteArraySize)) {
+ return Status::Invalid("Parquet cannot store strings with size 2GB
or more");
+ }
+ // Convert view to ByteArray so it can be passed to the
suffix_encoder_.
+ const ByteArray src{view};
+
+ uint32_t j = 0;
+ while (j < std::min(previous_len, src.len)) {
+ if (last_value_view[j] != view[j]) {
+ break;
+ }
+ j++;
+ }
+ previous_len = j;
+ prefix_length_encoder_.Put({static_cast<int32_t>(j)}, 1);
+
+ const uint8_t* suffix_ptr = src.ptr + j;
+ const uint32_t suffix_length = static_cast<uint32_t>(src.len - j);
+ last_value_view =
+ string_view{reinterpret_cast<const char*>(suffix_ptr),
suffix_length};
Review Comment:
```suggestion
last_value_view = view;
```
##########
cpp/src/parquet/encoding.cc:
##########
@@ -3196,6 +3397,30 @@ class DeltaByteArrayDecoder : public DecoderImpl,
std::shared_ptr<ResizableBuffer> buffered_data_;
};
+class DeltaByteArrayDecoder : public DeltaByteArrayDecoderImpl<ByteArrayType> {
+ public:
+ using Base = DeltaByteArrayDecoderImpl<ByteArrayType>;
+ using Base::DeltaByteArrayDecoderImpl;
+
+ int Decode(ByteArray* buffer, int max_values) override {
+ return GetInternal(buffer, max_values);
+ }
+};
+
+class DeltaByteArrayFLBADecoder : public DeltaByteArrayDecoderImpl<FLBAType>,
+ virtual public FLBADecoder {
+ public:
+ using Base = DeltaByteArrayDecoderImpl<FLBAType>;
+ using Base::DeltaByteArrayDecoderImpl;
+
+ int Decode(ByteArray* buffer, int max_values) {
+ return GetInternal(buffer, max_values);
+ }
+ int Decode(FixedLenByteArray* buffer, int max_values) override {
Review Comment:
Yes, I think maybe check-and-cast is ok here, because
`DeltaByteArrayFLBADecoder` cannot decode FLBA is really weird.
##########
cpp/src/parquet/encoding.cc:
##########
@@ -3037,11 +3073,180 @@ class RleBooleanDecoder : public DecoderImpl, virtual
public BooleanDecoder {
// ----------------------------------------------------------------------
// DELTA_BYTE_ARRAY
-class DeltaByteArrayDecoder : public DecoderImpl,
- virtual public TypedDecoder<ByteArrayType> {
+/// Delta Byte Array encoding also known as incremental encoding or front
compression:
+/// for each element in a sequence of strings, store the prefix length of the
previous
+/// entry plus the suffix.
+///
+/// This is stored as a sequence of delta-encoded prefix lengths
(DELTA_BINARY_PACKED),
+/// followed by the suffixes encoded as delta length byte arrays
+/// (DELTA_LENGTH_BYTE_ARRAY).
+
+// ----------------------------------------------------------------------
+// DeltaByteArrayEncoder
+
+template <typename DType>
+class DeltaByteArrayEncoder : public EncoderImpl, virtual public
TypedEncoder<DType> {
public:
- explicit DeltaByteArrayDecoder(const ColumnDescriptor* descr,
+ using T = typename DType::c_type;
+
+ explicit DeltaByteArrayEncoder(const ColumnDescriptor* descr,
MemoryPool* pool =
::arrow::default_memory_pool())
+ : EncoderImpl(descr, Encoding::DELTA_BYTE_ARRAY, pool),
+ sink_(pool),
+ prefix_length_encoder_(nullptr, pool),
+ suffix_encoder_(nullptr, pool),
+ last_value_("") {}
+
+ std::shared_ptr<Buffer> FlushValues() override;
+
+ int64_t EstimatedDataEncodedSize() override {
+ return prefix_length_encoder_.EstimatedDataEncodedSize() +
+ suffix_encoder_.EstimatedDataEncodedSize();
+ }
+
+ using TypedEncoder<DType>::Put;
+
+ void Put(const ::arrow::Array& values) override;
+
+ void Put(const T* buffer, int num_values) override;
+
+ void PutSpaced(const T* src, int num_values, const uint8_t* valid_bits,
+ int64_t valid_bits_offset) override {
+ if (valid_bits != NULLPTR) {
+ PARQUET_ASSIGN_OR_THROW(auto buffer, ::arrow::AllocateBuffer(num_values
* sizeof(T),
+
this->memory_pool()));
+ T* data = reinterpret_cast<T*>(buffer->mutable_data());
+ int num_valid_values = ::arrow::util::internal::SpacedCompress<T>(
+ src, num_values, valid_bits, valid_bits_offset, data);
+ Put(data, num_valid_values);
+ } else {
+ Put(src, num_values);
+ }
+ }
+
+ protected:
+ template <typename ArrayType>
+ void PutBinaryArray(const ArrayType& array) {
+ uint32_t previous_len = 0;
+ std::string_view last_value_view = last_value_;
+
+ PARQUET_THROW_NOT_OK(::arrow::VisitArraySpanInline<typename
ArrayType::TypeClass>(
+ *array.data(),
+ [&](::std::string_view view) {
+ if (ARROW_PREDICT_FALSE(view.size() >= kMaxByteArraySize)) {
+ return Status::Invalid("Parquet cannot store strings with size 2GB
or more");
+ }
+ // Convert view to ByteArray so it can be passed to the
suffix_encoder_.
+ const ByteArray src{view};
+
+ uint32_t j = 0;
+ while (j < std::min(previous_len, src.len)) {
+ if (last_value_view[j] != view[j]) {
+ break;
+ }
+ j++;
+ }
+ previous_len = j;
+ prefix_length_encoder_.Put({static_cast<int32_t>(j)}, 1);
+
+ const uint8_t* suffix_ptr = src.ptr + j;
+ const uint32_t suffix_length = static_cast<uint32_t>(src.len - j);
+ last_value_view =
+ string_view{reinterpret_cast<const char*>(suffix_ptr),
suffix_length};
+ // Convert suffix to ByteArray so it can be passed to the
suffix_encoder_.
+ const ByteArray suffix(suffix_length, suffix_ptr);
+ suffix_encoder_.Put(&suffix, 1);
+
+ return Status::OK();
+ },
+ []() { return Status::OK(); }));
+ last_value_ = last_value_view;
+ }
+
+ ::arrow::BufferBuilder sink_;
+ DeltaBitPackEncoder<Int32Type> prefix_length_encoder_;
+ DeltaLengthByteArrayEncoder<ByteArrayType> suffix_encoder_;
+ std::string last_value_;
+};
+
+template <typename DType>
+void DeltaByteArrayEncoder<DType>::Put(const T* src, int num_values) {
+ if (num_values == 0) {
+ return;
+ }
+ ArrowPoolVector<int32_t> prefix_lengths(num_values,
+
::arrow::stl::allocator<int32_t>(pool_));
+ std::string_view last_value_view = last_value_;
+
+ int i = 0;
+ while (i < num_values) {
+ // Convert to ByteArray so we can pass to the suffix_encoder_.
+ auto value = reinterpret_cast<const ByteArray*>(&src[i]);
+ if (ARROW_PREDICT_FALSE(value->len >= kMaxByteArraySize)) {
+ throw Status::Invalid("Parquet cannot store strings with size 2GB or
more");
+ }
+
+ auto view = string_view{reinterpret_cast<const char*>(value->ptr),
value->len};
+ uint32_t j = 0;
+ while (j < std::min(value->len,
static_cast<uint32_t>(last_value_view.length()))) {
+ if (last_value_view[j] != view[j]) {
+ break;
+ }
+ j++;
+ }
+
+ prefix_lengths[i] = j;
+ const uint8_t* suffix_ptr = value->ptr + j;
+ const uint32_t suffix_length = static_cast<uint32_t>(value->len - j);
+ last_value_view =
+ string_view{reinterpret_cast<const char*>(suffix_ptr), suffix_length};
Review Comment:
```suggestion
last_value_view = view;
```
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
