Introduction

dry-struct is a gem built on top of dry-types which provides virtus-like DSL for defining typed struct classes.

Basic Usage

You can define struct objects which will have readers for specified attributes using a simple dsl:

require 'dry-struct'

module Types
  include Dry.Types()
end

class User < Dry::Struct
  attribute :name, Types::String.optional
  attribute :age, Types::Coercible::Integer
end

user = User.new(name: nil, age: '21')

user.name # nil
user.age # 21

user = User.new(name: 'Jane', age: '21')

user.name # => "Jane"
user.age # => 21

Value

You can define value objects which will behave like structs but will be deeply frozen:

class Location < Dry::Struct::Value
  attribute :lat, Types::Float
  attribute :lng, Types::Float
end

loc1 = Location.new(lat: 1.23, lng: 4.56)
loc2 = Location.new(lat: 1.23, lng: 4.56)

loc1.frozen? # true
loc2.frozen? # true

loc1 == loc2
# true

Hash Schemas

Dry::Struct out of the box uses hash schemas from dry-types for processing input hashes. with_type_transform and with_key_transform are exposed as transform_types and transform_keys:

class User < Dry::Struct
  transform_keys(&:to_sym)

  attribute :name, Types::String.optional
  attribute :age, Types::Coercible::Integer
end

User.new('name' => 'Jane', 'age' => '21')
# => #<User name="Jane" age=21>

This plays nicely with inheritance, you can define a base struct for symbolizing input and then reuse it:

class SymbolizeStruct < Dry::Struct
  transform_keys(&:to_sym)
end

class User < SymbolizeStruct
  attribute :name, Types::String.optional
  attribute :age, Types::Coercible::Integer
end

Validating data with dry-struct

Please don't. Structs are meant to work with valid input, it cannot generate error messages good enough for displaying them for a user etc. Use dry-validation for validating incoming data and then pass its output to structs.

Differences between dry-struct and virtus

dry-struct look somewhat similar to Virtus but there are few significant differences:

  • Structs don't provide attribute writers and are meant to be used as "data objects" exclusively
  • Handling of attribute values is provided by standalone type objects from dry-types, which gives you way more powerful features
  • Handling of attribute hashes is provided by standalone hash schemas from dry-types, which means there are different types of constructors in dry-struct
  • Structs are not designed as swiss-army knifes, specific constructor types are used depending on the use case
  • Struct classes quack like dry-types, which means you can use them in hash schemas, as array members or sum them

octocatEdit on GitHub