rok commented on code in PR #14191:
URL: https://github.com/apache/arrow/pull/14191#discussion_r1025279497
##########
cpp/src/parquet/encoding.cc:
##########
@@ -2060,6 +2062,220 @@ class DictByteArrayDecoderImpl : public
DictDecoderImpl<ByteArrayType>,
}
};
+// ----------------------------------------------------------------------
+// DeltaBitPackEncoder
+
+constexpr uint32_t kValuesPerBlock = 128;
+constexpr uint32_t kMiniBlocksPerBlock = 4;
+
+template <typename DType>
+class DeltaBitPackEncoder : public EncoderImpl, virtual public
TypedEncoder<DType> {
+ public:
+ using T = typename DType::c_type;
+ using TypedEncoder<DType>::Put;
+
+ explicit DeltaBitPackEncoder(const ColumnDescriptor* descr, MemoryPool* pool,
+ const uint32_t values_per_block =
kValuesPerBlock,
+ const uint32_t mini_blocks_per_block =
kMiniBlocksPerBlock)
+ : EncoderImpl(descr, Encoding::DELTA_BINARY_PACKED, pool),
+ values_per_block_(values_per_block),
+ mini_blocks_per_block_(mini_blocks_per_block),
+ values_per_mini_block_(values_per_block / mini_blocks_per_block),
+ deltas_(values_per_block, ::arrow::stl::allocator<T>(pool)),
+ bits_buffer_(AllocateBuffer(pool, (values_per_block + 3) * sizeof(T))),
+ sink_(pool),
+ bit_writer_(bits_buffer_->mutable_data(),
+ static_cast<int>(bits_buffer_->size())) {
+ if (values_per_mini_block_ % 32 != 0) {
+ throw ParquetException(
+ "the number of values in a miniblock must be multiple of 32, but
it's " +
+ std::to_string(values_per_mini_block_));
+ }
+ }
+
+ std::shared_ptr<Buffer> FlushValues() override;
+
+ int64_t EstimatedDataEncodedSize() override { return sink_.length(); }
+
+ 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;
+
+ void FlushBlock();
+
+ private:
+ const uint32_t values_per_block_;
+ const uint32_t mini_blocks_per_block_;
+ const uint32_t values_per_mini_block_;
+ uint32_t values_current_block_{0};
+ uint32_t total_value_count_{0};
+ T first_value_{0};
+ T current_value_{0};
+ ArrowPoolVector<T> deltas_;
+ std::shared_ptr<ResizableBuffer> bits_buffer_;
+ ::arrow::BufferBuilder sink_;
+ ::arrow::bit_util::BitWriter bit_writer_;
+};
+
+template <typename DType>
+void DeltaBitPackEncoder<DType>::Put(const T* src, int num_values) {
+ if (num_values == 0) {
+ return;
+ }
+
+ int idx = 0;
+ if (total_value_count_ == 0) {
+ current_value_ = src[0];
+ first_value_ = current_value_;
+ idx = 1;
+ }
+ total_value_count_ += num_values;
+
+ while (idx < num_values) {
+ T value = src[idx];
+ deltas_[values_current_block_] = SafeSignedSubtract(value, current_value_);
+ current_value_ = value;
+ idx++;
+ values_current_block_++;
+ if (values_current_block_ == values_per_block_) {
+ FlushBlock();
+ }
+ }
+}
+
+template <typename DType>
+void DeltaBitPackEncoder<DType>::FlushBlock() {
+ if (values_current_block_ == 0) {
+ return;
+ }
+
+ const T min_delta =
+ *std::min_element(deltas_.begin(), deltas_.begin() +
values_current_block_);
+ bit_writer_.PutZigZagVlqInt(min_delta);
+
+ std::vector<uint8_t, ::arrow::stl::allocator<uint8_t>> bit_widths(
+ mini_blocks_per_block_, 0);
Review Comment:
We need to insert a bit width for each possible miniblock before at the
start of each block even if we'll have fewer miniblocks. So instead of adding
logic for that we just make a vector and populate it or not. I feel that might
be slightly more efficient since `num_miniblocks` is small.
--
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]
