Struct parquet::decoding::PlainDecoder

pub struct PlainDecoder<T: DataType> { /* fields omitted */ }

Plain decoding that supports all types. Values are encoded back to back. For native types, data is encoded as little endian. Floating point types are encoded in IEEE. See PlainEncoder for more information.

Implementations

impl<T: DataType> PlainDecoder<T>

pub fn new(type_length: i32) -> Self

Creates new plain decoder.

Trait Implementations

impl<T: DataType> Decoder<T> for PlainDecoder<T>

fn set_data(&mut self, data: ByteBufferPtr, num_values: usize) -> Result<()>

Sets the data to decode to be data, which should contain num_values of values to decode.

fn values_left(&self) -> usize

Returns the number of values left in this decoder stream.

fn encoding(&self) -> Encoding

Returns the encoding for this decoder.

fn get(&mut self, buffer: &mut [T::T]) -> Result<usize>

Consumes values from this decoder and write the results to buffer. This will try to fill up buffer.

fn get_spaced(
    &mut self,
    buffer: &mut [T::T],
    null_count: usize,
    valid_bits: &[u8]
) -> Result<usize>

Consume values from this decoder and write the results to buffer, leaving “spaces” for null values.