User-defined metadata.

User-defined metadata.

Optional compression of the message body.

indicesBuffer stores the location and size of the array that contains the column indices of the corresponding non-zero values.

indicesBuffer stores the location and size of the array that contains the column indices of the corresponding non-zero values.

The location and size of the tensor's data.

Name of the dimension, optional.

The location and size of the tensor's data.

The timezone is an optional string indicating the name of a timezone, one of: * As used in the Olson timezone database (the "tz database" or "tzdata"), such as "America/New_York".

Each constituent buffer is first compressed with the indicated compressor, and then written with the uncompressed length in the first 8 bytes as a 64-bit little-endian signed integer followed by the compressed buffer bytes (and then padding as required by the protocol).

The stream makes use of compressed bodies as described in Message.fbs.

The stream makes use of multiple full dictionaries with the same ID and assumes clients implement dictionary replacement correctly.

Needed to make flatbuffers happy.

0.1.0 (October 2016).

0.2.0 (February 2017).

0.3.0 -> 0.7.1 (May - December 2017).

>= 0.8.0 (December 2017).

>= 1.0.0 (July 2020.

Opaque binary data.

Optional compression for the memory buffers constituting IPC message bodies.

---------------------------------------------------------------------- A Buffer represents a single contiguous memory segment.

Date is either a 32-bit or 64-bit signed integer type representing an elapsed time since UNIX epoch (1970-01-01), stored in either of two units: * Milliseconds (64 bits) indicating UNIX time elapsed since the epoch (no leap seconds), where the values are evenly divisible by 86400000 * Days (32 bits) since the UNIX epoch.

Exact decimal value represented as an integer value in two's complement.

For sending dictionary encoding information.

---------------------------------------------------------------------- A field represents a named column in a record / row batch or child of a nested type.

---------------------------------------------------------------------- Data structures for describing a table row batch (a collection of equal-length Arrow arrays) Metadata about a field at some level of a nested type tree (but not its children).

---------------------------------------------------------------------- Arrow File metadata .

---------------------------------------------------------------------- user defined key value pairs to add custom metadata to arrow key namespacing is the responsibility of the user.

Same as Binary, but with 64-bit offsets, allowing to represent extremely large data values.

Same as List, but with 64-bit offsets, allowing to represent extremely large data values.

Same as Utf8, but with 64-bit offsets, allowing to represent extremely large data values.

A Map is a logical nested type that is represented as List<entries: Struct<key: K, value: V>> In this layout, the keys and values are each respectively contiguous.

These are stored in the flatbuffer in the Type union below.

A data header describing the shared memory layout of a "record" or "row" batch.

---------------------------------------------------------------------- A Schema describes the columns in a row batch.

Compressed Sparse Row format, that is matrix-specific.

Compressed Sparse format, that is matrix-specific.

---------------------------------------------------------------------- EXPERIMENTAL: Data structures for sparse tensors Coordinate (COO) format of sparse tensor index.

Compressed Sparse Fiber (CSF) sparse tensor index.

A Struct_ in the flatbuffer metadata is the same as an Arrow Struct (according to the physical memory layout).

---------------------------------------------------------------------- Data structures for dense tensors Shape data for a single axis in a tensor.

Time is either a 32-bit or 64-bit signed integer type representing an elapsed time since midnight, stored in either of four units: seconds, milliseconds, microseconds or nanoseconds.

Timestamp is a 64-bit signed integer representing an elapsed time since a fixed epoch, stored in either of four units: seconds, milliseconds, microseconds or nanoseconds, and is optionally annotated with a timezone.

A union is a complex type with children in Field By default ids in the type vector refer to the offsets in the children optionally typeIds provides an indirection between the child offset and the type id for each child `typeIds[offset]` is the id used in the type vector.

Unicode with UTF-8 encoding.

Provided for forward compatibility in case we need to support different strategies for compressing the IPC message body (like whole-body compression rather than buffer-level) in the future.

---------------------------------------------------------------------- Dictionary encoding metadata Maintained for forwards compatibility, in the future Dictionaries might be explicit maps between integers and values allowing for non-contiguous index values.

---------------------------------------------------------------------- Endianness of the platform producing the data.

Represents Arrow Features that might not have full support within implementations.

---------------------------------------------------------------------- The root Message type This union enables us to easily send different message types without redundant storage, and in the future we can easily add new message types.

---------------------------------------------------------------------- Top-level Type value, enabling extensible type-specific metadata.