char(n) (BPCHAR) length fix

[tanner0101]

The char(n) (BPCHAR) data type now correctly supports values longer than one byte when converting to fixed-width integers or strings (fixes #71, fixes #79, #72).

[tkrajacic]

This now crashes in the PostgresDataDecoder.swift on

return convertible.init(postgresData: data)! as! T

with

NIO-ELT-0-#0 (3): Fatal error: Unexpectedly found nil while unwrapping an Optional value

(lldb) po data
▿ 0x0020 (BPCHAR)
  ▿ type : BPCHAR
    - rawValue : 1042
  ▿ typeModifier : Optional<Int32>
    - some : 5
  - formatCode : binary
  ▿ value : Optional<ByteBuffer>
    ▿ some : ByteBuffer { readerIndex: 0, writerIndex: 2, readableBytes: 2, capacity: 2, slice: _ByteBufferSlice { 1453..<1455 }, storage: 0x00007b7000011000 (2048 bytes) }
      ▿ _storage : <_Storage: 0x7b100006d7c0>
      - _readerIndex : 0
      - _writerIndex : 2
      ▿ _slice : _ByteBufferSlice { 1453..<1455 }
        - upperBound : 1455
        ▿ _begin : 1453
          - b12 : 1453
          - b3 : 0

The actual type it crashes on is now

public enum Subscription: Int8, Equatable, Hashable, Codable, CaseIterable {
    case unknown = 0
    case subscribed = 1
    case unsubscribed = 2
}

with

extension Subscription: DatabaseTypeRepresentable {
    public static var databaseType: String { "CHARACTER" }
}

public protocol DatabaseTypeRepresentable {
    static var databaseType: String { get }
}

extension Subscription: PostgresDataConvertible { }

/cc @calebkleveter

[tanner0101]

@tkrajacic can you try CHAR instead of CHARACTER?

From https://dba.stackexchange.com/questions/159090/how-to-store-one-byte-integer-in-postgresql

"char" is an "internal" type intended for simple and cheap enumeration.

Either way, the force cast should be fixed obviously. That's in postgres-kit, right?

[tkrajacic]

As it turns out, Postgres has no 1-byte type really, and even a character(1) uses 2 bytes.

See https://www.postgresql.org/docs/current/datatype-character.html

The storage requirement for a short string (up to 126 bytes) is 1 byte plus the actual string, which includes the space padding in the case of character

There would be an internal type of "CHAR" (including the quotes) but this is not adhering to any standard

The type "char" (note the quotes) is different from char(1) in that it only uses one byte of storage. It is internally used in the system catalogs as a simplistic enumeration type.

[tanner0101]

Merging this since it at least fixes #79. Solutions for Postgres' lack of a good "single byte" type can be discussed further.

[tanner0101]

These changes are now available in 1.0.0-beta.2.8