Skip to content


Switch branches/tags


Failed to load latest commit information.
Latest commit message
Commit time


CI Status CI Status Gem Version

ActiveRecord attributes stored serialized in a json column, super smooth. For Rails 5.0 through 6.1. Ruby 2.4+.

Typed and cast like Active Record. Supporting nested models, dirty tracking, some querying (with postgres jsonb contains), and working smoothy with form builders.

Use your database as a typed object store via ActiveRecord, in the same models right next to ordinary ActiveRecord column-backed attributes and associations. Your json-serialized attr_json attributes use as much of the existing ActiveRecord architecture as we can.

Why might you want or not want this?

Developed for postgres, but most features should work with MySQL json columns too, although has not yet been tested with MySQL.

Basic Use

# migration
class CreatMyModels < ActiveRecord::Migration[5.0]
  def change
    create_table :my_models do |t|
      t.jsonb :json_attributes

    # If you plan to do any querying with jsonb_contains below..
    add_index :my_models, :json_attributes, using: :gin

class MyModel < ActiveRecord::Base
   include AttrJson::Record

   # use any ActiveModel::Type types: string, integer, decimal (BigDecimal),
   # float, datetime, boolean.
   attr_json :my_string, :string
   attr_json :my_integer, :integer
   attr_json :my_datetime, :datetime

   # You can have an _array_ of those things too.
   attr_json :int_array, :integer, array: true

   #and/or defaults
   attr_json :int_with_default, :integer, default: 100

These attributes have type-casting behavior very much like ordinary ActiveRecord values.

model =
model.my_integer = "12"
model.my_integer # => 12
model.int_array = "12"
model.int_array # => [12]
model.my_datetime = "2016-01-01 17:45"
model.my_datetime # => a Time object representing that, just like AR would cast

You can use ordinary ActiveRecord validation methods with attr_json attributes.

All the attr_json attributes are serialized to json as keys in a hash, in a database jsonb/json column. By default, in a column json_attributes. If you look at model.json_attributes, you'll see values already cast to their ruby representations.

But one way to see something like what it's really like in the db is to save the record and then use the standard Rails *_before_type_cast method.!
# => string containing: {"my_integer":12,"int_array":[12],"my_datetime":"2016-01-01T17:45:00.000Z"}

Specifying db column to use

While the default is to assume you want to serialize in a column called json_attributes, no worries, of course you can pick whatever named jsonb column you like, class-wide or per-attribute.

class OtherModel < ActiveRecord::Base
  include AttrJson::Record

  # as a default for the model
  attr_json_config(default_container_attribute: :some_other_column_name)

  # now this is going to serialize to column 'some_other_column_name'
  attr_json :my_int, :integer

  # Or on a per-attribute basis
  attr_json :my_int, :integer, container_attribute: "yet_another_column_name"

Store key different than attribute name/methods

You can also specify that the serialized JSON key should be different than the attribute name/methods, by using the store_key argument.

class MyModel < ActiveRecord::Base
  include AttrJson::Record

  attr_json :special_string, :string, store_key: "__my_string"

model =
model.special_string = "foo"
model.json_attributes # => {"__my_string"=>"foo"}!
model.json_attributes_before_type_cast # => string containing: {"__my_string":"foo"}

You can of course combine array, default, store_key, and container_attribute params however you like, with whatever types you like: symbols resolvable with ActiveRecord::Type.lookup, or any ActiveModel::Type::Value subclass, built-in or custom.

You can register your custom ActiveModel::Type::Value in a Rails initializer or early on in your app boot sequence:

ActiveRecord::Type.register(:my_type, MyActiveModelTypeSubclass)


There is some built-in support for querying using postgres jsonb containment (@>) operator. (or see here or here). For now you need to additionally include AttrJson::Record::QueryScopes to get this behavior.

model = MyModel.create(my_string: "foo", my_integer: 100)

MyModel.jsonb_contains(my_string: "foo", my_integer: 100).to_sql
# SELECT "products".* FROM "products" WHERE (products.json_attributes @> ('{"my_string":"foo","my_integer":100}')::jsonb)
MyModel.jsonb_contains(my_string: "foo", my_integer: 100).first
# Implemented with scopes, this is an ordinary relation, you can
# combine it with whatever, just like ordinary `where`.

MyModel.not_jsonb_contains(my:string: "foo", my_integer: 100).to_sql
# SELECT "products".* FROM "products" WHERE NOT (products.json_attributes @> ('{"my_string":"foo","my_integer":100}')::jsonb)

# typecasts much like ActiveRecord on query too:
MyModel.jsonb_contains(my_string: "foo", my_integer: "100")
# no problem

# works for arrays too
model = MyModel.create(int_array: [10, 20, 30])
MyModel.jsonb_contains(int_array: 10) # finds it
MyModel.jsonb_contains(int_array: [10]) # still finds it
MyModel.jsonb_contains(int_array: [10, 20]) # it contains both, so still finds it
MyModel.jsonb_contains(int_array: [10, 1000]) # nope, returns nil, has to contain ALL listed in query for array args

jsonb_contains will handle any store_key you have set -- you should specify attribute name, it'll actually query on store_key. And properly handles any container_attribute -- it'll look in the proper jsonb column.

Anything you can do with jsonb_contains should be handled by a postgres USING GIN index (I think! can anyone help confirm/deny?). To be sure, I recommend you investigate: Check out to_sql on any query to see what jsonb SQL it generates, and explore if you have the indexes you need.

Nested models -- Structured/compound data

The AttrJson::Model mix-in lets you make ActiveModel::Model objects that can be round-trip serialized to a json hash, and they can be used as types for your top-level AttrJson::Record. AttrJson::Models can contain other AJ::Models, singly or as arrays, nested as many levels as you like.

That is, you can serialize complex object-oriented graphs of models into a single jsonb column, and get them back as they went in.

AttrJson::Model has an identical attr_json api to AttrJson::Record, with the exception that container_attribute is not supported.

class LangAndValue
  include AttrJson::Model

  attr_json :lang, :string, default: "en"
  attr_json :value, :string

  # Validations work fine, and will post up to parent record
  validates :lang, inclusion_in: I18n.config.available_locales.collect(&:to_s)

class MyModel < ActiveRecord::Base
  include AttrJson::Record
  include AttrJson::Record::QueryScopes

  attr_json :lang_and_value, LangAndValue.to_type

  # YES, you can even have an array of them
  attr_json :lang_and_value_array, LangAndValue.to_type, array: true

# Set with a model object, in initializer or writer
m = "fr", value: "S'il vous plaît"))
m.lang_and_value = "es", value: "hola")
# => #<LangAndValue:0x007fb64f12bb70 @attributes={"lang"=>"es", "value"=>"hola"}>!
# => string containing: {"lang_and_value":{"lang":"es","value":"hola"}}

# Or with a hash, no problem.

m = { lang: 'fr', value: "S'il vous plaît"})
m.lang_and_value = { lang: 'en', value: "Hey there" }!
# => string containing: {"lang_and_value":{"lang":"en","value":"Hey there"}}
found = MyModel.find(
# => #<LangAndValue:0x007fb64eb78e58 @attributes={"lang"=>"en", "value"=>"Hey there"}>

# Arrays too, yup

m = [{ lang: 'fr', value: "S'il vous plaît"}, { lang: 'en', value: "Hey there" }])
# => [#<LangAndValue:0x007f89b4f08f30 @attributes={"lang"=>"fr", "value"=>"S'il vous plaît"}>, #<LangAndValue:0x007f89b4f086e8 @attributes={"lang"=>"en", "value"=>"Hey there"}>]!
# => string containing: {"lang_and_value_array":[{"lang":"fr","value":"S'il vous plaît"},{"lang":"en","value":"Hey there"}]}

You can nest AttrJson::Model objects inside each other, as deeply as you like.

Model-type defaults

If you want to set a default for an AttrJson::Model type, you should use a proc argument for the default, to avoid accidentally re-using a shared global default value, similar to issues people have with ruby Hash default.

  attr_json :lang_and_value, LangAndValue.to_type, default: -> { "en", value: "default") }

You can also use a Hash value that will be cast to your model, no need for proc argument in this case.

  attr_json :lang_and_value, LangAndValue.to_type, default: { lang: "en", value: "default" }

Polymorphic model types

There is some support for "polymorphic" attributes that can hetereogenously contain instances of different AttrJson::Model classes, see comment docs at AttrJson::Type::PolymorphicModel.

class SomeLabels
  include AttrJson::Model

  attr_json :hello, LangAndValue.to_type, array: true
  attr_json :goodbye, LangAndValue.to_type, array: true
class MyModel < ActiveRecord::Base
  include AttrJson::Record
  include AttrJson::Record::QueryScopes

  attr_json :my_labels, SomeLabels.to_type

m =
m.my_labels = {}
# => #<SomeLabels:0x007fed2a3b1a18>
m.my_labels.hello = [{lang: 'en', value: 'hello'}, {lang: 'es', value: 'hola'}]
# => #<SomeLabels:0x007fed2a3b1a18 @attributes={"hello"=>[#<LangAndValue:0x007fed2a0eafc8 @attributes={"lang"=>"en", "value"=>"hello"}>, #<LangAndValue:0x007fed2a0bb4d0 @attributes={"lang"=>"es", "value"=>"hola"}>]}>
m.my_labels.hello.find { |l| l.lang == "en" }.value = "Howdy"!
# => {"my_labels"=>#<SomeLabels:0x007fed2a714e80 @attributes={"hello"=>[#<LangAndValue:0x007fed2a714cf0 @attributes={"lang"=>"en", "value"=>"Howdy"}>, #<LangAndValue:0x007fed2a714ac0 @attributes={"lang"=>"es", "value"=>"hola"}>]}>}
# => string containing: {"my_labels":{"hello":[{"lang":"en","value":"Howdy"},{"lang":"es","value":"hola"}]}}

GUESS WHAT? You can QUERY nested structures with jsonb_contains, using a dot-keypath notation, even through arrays as in this case. Your specific defined attr_json types determine the query and type-casting.

MyModel.jsonb_contains("my_labels.hello.lang" => "en").to_sql
# => SELECT "products".* FROM "products" WHERE (products.json_attributes @> ('{"my_labels":{"hello":[{"lang":"en"}]}}')::jsonb)
MyModel.jsonb_contains("my_labels.hello.lang" => "en").first

# also can give hashes, at any level, or models themselves. They will
# be cast. Trying to make everything super consistent with no surprises.

MyModel.jsonb_contains("my_labels.hello" => 'en')).to_sql
# => SELECT "products".* FROM "products" WHERE (products.json_attributes @> ('{"my_labels":{"hello":[{"lang":"en"}]}}')::jsonb)

MyModel.jsonb_contains("my_labels.hello" => {"lang" => "en"}).to_sql
# => SELECT "products".* FROM "products" WHERE (products.json_attributes @> ('{"my_labels":{"hello":[{"lang":"en"}]}}')::jsonb)

Remember, we're using a postgres containment (@>) operator, so queries always mean 'contains' -- the previous query needs a my_labels.hello which is a hash that includes the key/value, lang: en, it can have other key/values in it too. String values will need to match exactly.

Single AttrJson::Model serialized to an entire json column

The main use case of the gem is set up to let you combine multiple primitives and nested models under different keys combined in a single json or jsonb column.

But you may also want to have one AttrJson::Model class that serializes to map one model class, as a hash, to an entire json column on it's own.

AttrJson::Model can supply a simple coder for the ActiveRecord serialization feature to easily do that.

class MyModel
  include AttrJson::Model

  attr_json :some_string, :string
  attr_json :some_int, :int

class MyTable < ApplicationRecord
  serialize :some_json_column, MyModel.to_serialization_coder

MyTable.create(some_json_column: "string"))

# will cast from hash for you
MyTable.create(some_json_column: { some_int: 12 })

# etc

To avoid errors raised at inconvenient times, we recommend you set these settings to make 'bad' data turn into nil, consistent with most ActiveRecord types:

class MyModel
  include AttrJson::Model

  attr_json_config(bad_cast: :as_nil, unknown_key: :strip)
  # ...

And/or define a setter method to cast, and raise early on data problems:

class MyTable < ApplicationRecord
  serialize :some_json_column, MyModel.to_serialization_coder

  def some_json_column=(val)
    super(   )

Serializing a model to an entire json column is a relatively recent feature, please let us know how it's working for you.

Storing Arbitrary JSON data

Arbitrary JSON data (hashes, arrays, primitives of any depth) can be stored within attributes by using the rails built in ActiveModel::Type::Value as the attribute type. This is basically a "no-op" value type -- JSON alone will be used to serialize/deserialize whatever values you put there, because of the json type on the container field.

class MyModel < ActiveRecord::Base
  include AttrJson::Record

  attr_json :arbitrary_hash,

Forms and Form Builders

Use with Rails form builders is supported pretty painlessly. Including with simple_form and cocoon (integration-tested in CI).

If you have nested AttrJson::Models you'd like to use in your forms much like Rails associated records: Where you would use Rails accepts_nested_attributes_for, instead include AttrJson::NestedAttributes and use attr_json_accepts_nested_attributes_for. Multiple levels of nesting are supported.

To get simple_form to properly detect your attribute types, define your attributes with rails_attribute: true. You can default rails_attribute to true with attr_json_config(default_rails_attribute: true)

For more info, see doc page on Use with Forms and Form Builders.

Dirty tracking

Full change-tracking, ActiveRecord::Attributes::Dirty-style, is available in Rails 5.1+ on attr_jsons on your ActiveRecord classes that include AttrJson::Record, by including AttrJson::Record::Dirty. Change-tracking methods are available off the attr_json_changes method.

class MyModel < ActiveRecord::Base
   include AttrJson::Record
   include AttrJson::Record::Dirty

   attr_json :str, :string

model =
model.str = "old"
model.str = "new"

# All and only "new" style dirty tracking methods (Raisl 5.1+)
# are available:

# etc

More options are available, including merging changes from 'ordinary' ActiveRecord attributes in. See docs on Dirty Tracking

Do you want this?

Why might you want this?

  • You have complicated data, which you want to access in object-oriented fashion, but want to avoid very complicated normalized rdbms schema -- and are willing to trade the powerful complex querying support normalized rdbms schema gives you.

  • Single-Table Inheritance, with sub-classes that have non-shared data fields. You rather not make all those columns, some of which will then also appear to inapplicable sub-classes.

  • A "content management system" type project, where you need complex structured data of various types, maybe needs to be vary depending on plugins or configuration, or for different article types -- but doesn't need to be very queryable generally -- or you have means of querying other than a normalized rdbms schema.

  • You want to version your models, which is tricky with associations between models. Minimize associations by inlining the complex data into one table row.

  • Generally, we're turning postgres into a simple object-oriented document store. That can be mixed with an rdbms. The very same row in a table in your db can have document-oriented json data and foreign keys and real rdbms associations to other rows. And it all just feels like ActiveRecord, mostly.

Why might you not want this?

  • An rdbms and SQL is a wonderful thing, if you need sophisticated querying and reporting with reasonable performance, complex data in a single jsonb probably isn't gonna be the best.

  • This is pretty well-designed code that mostly only uses fairly stable and public Rails API, but there is still some risk of tying your boat to it, it's not Rails itself, and there is some risk it won't keep up with Rails in the future.

Note on Optimistic Locking

When you save a record with any changes to any attr_jsons, it will overwrite the whole json structure in the relevant column for that row. Unlike ordinary AR attributes where updates just touch changed attributes.

Becuase of this, you probably want to seriously consider using ActiveRecord Optimistic Locking to prevent overwriting other updates from processes.

State of Code, and To Be Done

This code is solid and stable and is being used in production by at least a handful of people, including the primary maintainer, jrochkind.

The project is currently getting very little maintainance -- and is still working reliably through Rails releases. It is tested on edge rails and ruby (and has needed very few if any changes with subsequent releases), and I endeavor to keep it working as Rails keeps releasing.

In order to keep the low-maintenace scenario sustainable, I am very cautious accepting new features, especially if they increase code complexity at all. Even if you have a working PR, I may be reluctant to accept it. I'm prioritizing sustainability and stability over new features, and so far this is working out well. However, discussion is always welcome! Especially when paired with code (failing tests for the bugfix or feature you want are super helpful on their own!).

We are committed to semantic versioning and will endeavor to release no backwards breaking changes without a major version. We are also serious about minimizing backwards incompat releases altogether (ie minimiing major version releases).

Feedback of any kind of very welcome, please feel free to use the issue tracker. It is hard to get a sense of how many people are actually using this, which is helpful both for my own sense of reward and for anyone to get a sense of the size of the userbase -- feel free to say hi and let us know how you are using it!

Except for the jsonb_contains stuff using postgres jsonb contains operator, I don't believe any postgres-specific features are used. It ought to work with MySQL, testing and feedback welcome. (Or a PR to test on MySQL?). My own interest is postgres.

This is still mostly a single-maintainer operation, so has all the sustainability risks of that. Although there are other people using and contributing to it, check out the Github Issues and Pull Request tabs yourself to get a sense.

Possible future features:

  • Make AttrJson::Model lean more heavily on ActiveModel::Attributes API that did not fully exist in first version of attr_json [hope for a future attr_json 2.0 release]

  • Make AttrJson::Record insist on creating rails cover attributes (no longer optional) and integrating more fully into rails, including rails dirty tricking, eliminating need for custom dirty tracking implementation. Overall decrease in lines of code. Can use jsonb_accessor as guide for some aspects. [hope for inclusion in future attr_json 2.0 release]

  • partial updates for json hashes would be really nice: Using postgres jsonb merge operators to only overwrite what changed. In my initial attempts, AR doesn't make it easy to customize this. [update: this is hard, probably not coming soon]

  • seamless compatibility with ransack [update: not necessarily prioritized]

  • Should we give AttrJson::Model a before_serialize hook that you might want to use similar to AR before_save? Should AttrJson::Models raise on trying to serialize an invalid model? [update: eh, hasn't really come up]

  • There are limits to what you can do with just jsonb_contains queries. We could support operations like >, <, <> as jsonb_accessor, even accross keypaths. (At present, you could use a before_savee to denormalize/renormalize copy your data into ordinary AR columns/associations for searching. Or perhaps a postgres ts_vector for text searching. Needs to be worked out.) [update: interested, but not necessarily prioritized. This one would be interesting for a third-party PR draft!]

  • We could/should probably support jsonb_order clauses, even accross key paths, like jsonb_accessor. [update: interested but not necessarily prioritized]

  • Could we make these attributes work in ordinary AR where, same as they do in jsonb_contains? Maybe. [update: probably not]


While attr_json depends only on active_record, we run integration tests in the context of a full Rails app, in order to test working with simple_form and cocoon, among other things. (Via combustion, with app skeleton at ./spec/internal).

At present this does mean that all our automated tests are run in a full Rails environment, which is not great (any suggestions or PR's to fix this while still running integration tests under CI with full Rails app).

Tests are in rspec, run tests simply with ./bin/rspec.

We use appraisal to test with multiple rails versions, including on travis. Locally you can run bundle exec appraisal rspec to run tests multiple times for each rails version, or eg bundle exec appraisal rails-5-1 rspec. If the Gemfile or Appraisal file changes, you may need to re-run bundle exec appraisal install and commit changes. (Try to put dev dependencies in gemspec instead of Gemfile, but sometimes it gets weird.)

  • If you've been switching between rails versions and you get integration test failures, try rm -rf spec/internal/tmp/cache. Rails 6 does some things in there apparently not compatible with Rails 5, at least in our setup, and vice versa.

There is a ./bin/console that will give you a console in the context of attr_json and all it's dependencies, including the combustion rails app, and the models defined there.

Acknowledements and Prior Art

  • The excellent work Sean Griffin did on ActiveModel::Type really lays the groundwork and makes this possible. Plus many other Rails developers. Rails has a reputation for being composed of messy or poorly designed code, but it's some really nice design in Rails that allows us to do some pretty powerful stuff here, in surprisingly few lines of code.

  • The existing jsonb_accessor was an inspiration, and provided some good examples of how to do some things with AR and ActiveModel::Types. I started out trying to figure out how to fit in nested hashes to jsonb_accessor... but ended up pretty much rewriting it entirely, to lean on object-oriented polymorphism and ActiveModel::Type a lot heavier and have the API and internals I wanted/imagined.

  • Took a look at existing active_model_attributes too.

  • Didn't actually notice existing json_attributes until I was well on my way here. I think it's not updated for Rails5 or type-aware, haven't looked at it too much.

  • store_model was created after attr_json, and has some overlapping functionality.


Serialized json-hash-backed ActiveRecord attributes, super smooth







No packages published