BinaryEncoding

The BinaryEncoding library is designed to simplify encoding/decoding of native Swift types and their sequences into binary data buffers.

Supported types are: Int, UInt, Int*, UInt*, Float, Double, VarUInt, String and various collections of native numbers. Native numbers are encoded as is, using architecture-dependent representations, strings are encoded in UTF-8. VarUInt encoding is equal to w format of Perl pack function.

Requirements

Swift 4.0

Design

The library provides managed buffers of raw bytes with copy-on-write behaviour.

BinaryEncodedData

Has various read, write and append methods for all supported types:

    init(minimumCapacity: Int = 0)

    func read<T : NativeBinaryEncoding>(_: T.Type, at offset: inout Int) throws -> T
    func read(_: VarUInt.Type, at offset: inout Int) throws -> UInt
    func read(_: BinaryEncodedData.Type, withSize: Int, at offset: inout Int) throws -> BinaryEncodedData
    func read<S : LengthItem>(_: BinaryEncodedData.Type, withSizeOf: S.Type, at offset: inout Int) throws -> BinaryEncodedData
    func read(_: BinaryEncodedData.Type, withSizeOf: VarUInt.Type, at offset: inout Int) throws -> BinaryEncodedData
    func read(_: String.Type, withSize: Int, at offset: inout Int) throws -> String
    func read<S : LengthItem>(_: String.Type, withSizeOf: S.Type, at offset: inout Int) throws -> String
    func read(_: String.Type, withSizeOf: VarUInt.Type, at offset: inout Int) throws -> String
    func read<T : SequenceItem>(arrayOf: T.Type, withSize: Int, at offset: inout Int) throws -> [T]
    func read<T : SequenceItem, S : LengthItem>(arrayOf: T.Type, withSizeOf: S.Type, at offset: inout Int) throws -> [T]
    func read<T : SequenceItem>(arrayOf: T.Type, withSizeOf: VarUInt.Type, at offset: inout Int) throws -> [T]

    mutating func write<T : NativeBinaryEncoding>(_ value: T, as: T.Type, at offset: inout Int)
    mutating func write(_ value: UInt, as: VarUInt.Type, at offset: inout Int)
    mutating func write(_ value: BinaryEncodedData, at offset: inout Int)
    mutating func write<S : LengthItem>(_ value: BinaryEncodedData, withSizeOf: S.Type, at offset: inout Int)
    mutating func write(_ value: BinaryEncodedData, withSizeOf: VarUInt.Type, at offset: inout Int)
    mutating func write(_ value: String, as: String.Type, at offset: inout Int)
    mutating func write<S : LengthItem>(_ value: String, as: String.Type, withSizeOf: S.Type, at offset: inout Int)
    mutating func write(_ value: String, as: String.Type, withSizeOf: VarUInt.Type, at offset: inout Int)
    mutating func write<C : Collection>(_ value: C, asArrayOf: C.Iterator.Element.Type, at offset: inout Int)
        where C.Iterator.Element : SequenceItem
    mutating func write<C : Collection, S : LengthItem>(_ value: C, asArrayOf: C.Iterator.Element.Type, withSizeOf: S.Type, at offset: inout Int)
        where C.Iterator.Element : SequenceItem
    mutating func write<C : Collection>(_ value: C, asArrayOf: C.Iterator.Element.Type, withSizeOf: VarUInt.Type, at offset: inout Int)
        where C.Iterator.Element : SequenceItem

    mutating func append... // same as write but without parameter (at:)

Usage examples can be found in tests.

UnsafeRawBufferPointer.Reader and UnsafeMutableRawBufferPointer.Writer

These interfaces are unsafe. If you want to use them for effeciency you can find enough infromation in the source code.