arrowio

Documentation

Index

• Variables
• type Reader
  • func NewReader(zctx *zed.Context, r io.Reader) (*Reader, error)
  • func NewReaderFromRecordReader(zctx *zed.Context, rr pqarrow.RecordReader) (*Reader, error)
  • func (r *Reader) Close() error
  • func (r *Reader) Read() (*zed.Value, error)
• type WriteCloser
• type Writer
  • func NewWriter(w io.WriteCloser) *Writer
  • func (w *Writer) Close() error
  • func (w *Writer) Write(val zed.Value) error

Variables

var (
	ErrMultipleTypes   = errors.New("arrowio: encountered multiple types (consider 'fuse')")
	ErrNotRecord       = errors.New("arrowio: not a record")
	ErrUnsupportedType = errors.New("arrowio: unsupported type")
)

Types

type Reader

type Reader struct {
	// contains filtered or unexported fields
}

Reader is a zio.Reader for the Arrow IPC stream format.

func NewReader

func NewReader(zctx *zed.Context, r io.Reader) (*Reader, error)

func NewReaderFromRecordReader

func NewReaderFromRecordReader(zctx *zed.Context, rr pqarrow.RecordReader) (*Reader, error)

func (*Reader) Close

func (r *Reader) Close() error

func (*Reader) Read

func (r *Reader) Read() (*zed.Value, error)

type WriteCloser

type WriteCloser interface {
	Write(rec arrow.Record) error
	Close() error
}

type Writer

type Writer struct {
	NewWriterFunc func(io.Writer, *arrow.Schema) (WriteCloser, error)
	// contains filtered or unexported fields
}

Writer is a zio.Writer for the Arrow IPC stream format. Given Zed values with appropriately named types (see the newArrowDataType implementation), it can write all Arrow types except dictionaries and sparse unions. (Although dictionaries are not part of the Zed data model, write support could be added using a named type.)

func NewWriter

func NewWriter(w io.WriteCloser) *Writer

func (*Writer) Close

func (w *Writer) Close() error

func (*Writer) Write

func (w *Writer) Write(val zed.Value) error