Extenject: extensions, bug fixes and updates for Zenject

This project is simply a fork of Zenject with the goal of being actively maintained. I am the primary author and was the primary maintainer until my access was removed after leaving my position at the company Modest Tree. It is called Extenject to respect Modest Tree's trademark claim on the name Zenject.

Lawsuit


Extenject has been removed from Asset Store because of a copyright claim by 
Modest Tree, so the only way to obtain it currently is through the 
https://github.com/svermeulen/Extenject/releases.

Modest Tree has also blocked me from contributing to zenject github issues and 
removed me from the zenject gitter chat, so to get my help please report 
issues https://github.com/svermeulen/Extenject/issues/new, 
and discuss in the https://gitter.im/Extenject/community instead

They have also filed a lawsuit against me and Unity tells me that they will 
not re-enable Extenject in the Asset Store until the lawsuit is complete, 
which might not occur until 2020.

I have been advised not to comment on details of the lawsuit right now 
however you can read their filing and my defense (both of which are public) 
for details if interested.  The lawsuit also relates to other MIT licensed 
open source projects I created such as Projeny and Unity3dAsyncAwaitUtil.

Support

This project is supported via donations. If you or your team have found it useful, please consider supporting further development through patreon or paypal

Introduction

Note that if you are looking for the older documentation for Zenject you can find that here: Zenject 3.x, Zenject 4.x and Zenject 5.x

Zenject is a lightweight highly performant dependency injection framework built specifically to target Unity 3D (however it can be used outside of Unity as well). It can be used to turn your application into a collection of loosely coupled parts with highly segmented responsibilities. Zenject can then glue the parts together in many different configurations to allow you to easily write, re-use, refactor and test your code in a scalable and extremely flexible way.

Tested in Unity 3D on the following platforms:

PC/Mac/Linux
iOS
Android
WebGL
PS4 (with IL2CPP backend)
Windows Store (including 8.1, Phone 8.1, Universal 8.1 and Universal 10 - both .NET and IL2CPP backend)

IL2CPP is supported, however there are some gotchas - see here for details

This project is open source.

For general troubleshooting / support, please post to stack overflow using the tag 'zenject', or post in the zenject google group

Or, if you have found a bug, you are also welcome to create an issue on the github page, or a pull request if you have a fix / extension. There is also a gitter chat that you can join for real time discussion. Finally, you can also email me directly at sfvermeulen@gmail.com or follow me on twitter at @steve_verm

Features

Injection
- Supports both normal C# classes and MonoBehaviours
- Constructor injection
- Field injection
- Property injection
- Method injection
Conditional binding (eg. by type, by name, etc.)
Optional dependencies
Support for creating objects after initialization using factories
Nested Containers aka Sub-Containers
Injection across different Unity scenes to pass information from one scene to the next
Scene parenting, to allow one scene to inherit the bindings from another
Support for global, project-wide bindings to add dependencies for all scenes
Convention based binding, based on class name, namespace, or any other criteria
Ability to validate object graphs at editor time (including dynamic object graphs created via factories)
Automatic binding on components in the scene using the ZenjectBinding component
Auto-Mocking using the Moq library
Built-in support for memory pools
Support for decorator pattern using decorator bindings
Support for automatically mapping open generic types
Built in support for unit test, integration tests, and scene tests
Just-in-time injection using the LazyInject<> construct
Support for multiple threads for resolving/instantiating
Support for 'reflection baking' to eliminate costly reflection operations completely by directly modifying the generated assemblies
Automatic injection of game objects using ZenAutoInjecter component

Installation

You can install Zenject using any of the following methods

From Releases Page Here you can choose between the following:
- Zenject-WithAsteroidsDemo.vX.X.unitypackage - This is equivalent to what you find in the Asset Store and contains both sample games "Asteroids" and "SpaceFighter" as part of the package. All the source code for Zenject is included here.
- Zenject.vX.X.unitypackage - Same as above except without the Sample projects.
- Zenject-NonUnity.vX.X.zip - Use this if you want to use Zenject outside of Unity (eg. just as a normal C# project)
From the Unity Asset Store
- Normally this should be the same as what you find in the Releases section, but may also be slightly out of date since Unity Asset Store can take a week or so to review submissions sometimes.
UPM Branch
- This option is a feature request. The package will be released when Unity is ready. Unity is not giving any insights on the development status. But the expectation is in the first or second release of 2020.
- If you can not wait. There is an alternative. But you will need the Unity extension found here. And the package found here.
From Source
- After syncing the git repo, note that you will have to build the Zenject-Usage.dll by building the solution at AssemblyBuild\Zenject-usage\Zenject-usage.sln. Or, if you prefer you can get Zenject-Usage.dll from Releases section instead
- Then you can copy the UnityProject/Assets/Plugins/Zenject directory to your own Unity3D project.

Note that when importing Zenject into your unity project, you can uncheck any folder underneath the OptionalExtras folder for cases where you don't want to include it, or if you just want the core zenject functionality, you can uncheck the entire OptionalExtras directory.

History

Unity is a fantastic game engine, however the approach that new developers are encouraged to take does not lend itself well to writing large, flexible, or scalable code bases. In particular, the default way that Unity manages dependencies between different game components can often be awkward and error prone.

This project was started because at the time there wasn't any DI frameworks for Unity, and having used DI frameworks outside of Unity (eg. Ninject) and seeing the benefits, I felt it was important to remedy that.

Finally, I will just say that if you don't have experience with DI frameworks, and are writing object oriented code, then trust me, you will thank me later! Once you learn how to write properly loosely coupled code using DI, there is simply no going back.

Documentation

The Zenject documentation is split up into the following sections. It is split up into two parts (Introduction and Advanced) so that you can get up and running as quickly as possible. I would recommend at least skimming through the Introduction section before beginning, but then feel free to jump around in the advanced section as necessary

Another great starting point is to watch this youtube series on zenject created by Infallible Code.

You might also benefit from playing with the provided sample projects (which you can find by opening Zenject/OptionalExtras/SampleGame1 or Zenject/OptionalExtras/SampleGame2).

If you are a DI veteran, then it might be worth taking a look at the cheatsheet at the bottom of this page, which should give you an idea of the syntax, which might be all you need to get started.

The tests may also be helpful to show usage for each specific feature (which you can find at Zenject/OptionalExtras/UnitTests and Zenject/OptionalExtras/IntegrationTests)

Also see further reading section for some external zenject tutorials provided elsewhere.

What is Dependency Injection?

Theory

What follows is a general overview of Dependency Injection from my perspective. However, it is kept light, so I highly recommend seeking other resources for more information on the subject, as there are many other people (often with better writing ability) that have written about the theory behind it.

When writing an individual class to achieve some functionality, it will likely need to interact with other classes in the system to achieve its goals. One way to do this is to have the class itself create its dependencies, by calling concrete constructors:

public class Foo
{
    ISomeService _service;

    public Foo()
    {
        _service = new SomeService();
    }

    public void DoSomething()
    {
        _service.PerformTask();
       … 
    }
}

This works fine for small projects, but as your project grows it starts to get unwieldy. The class Foo is tightly coupled to class 'SomeService'. If we decide later that we want to use a different concrete implementation then we have to go back into the Foo class to change it.

After thinking about this, often you come to the realization that ultimately, Foo shouldn't bother itself with the details of choosing the specific implementation of the service. All Foo should care about is fulfilling its own specific responsibilities. As long as the service fulfills the abstract interface required by Foo, Foo is happy. Our class then becomes:

public class Foo
{
    ISomeService _service;

    public Foo(ISomeService service)
    {
        _service = service;
    }

    public void DoSomething()
    {
        _service.PerformTask();
        ...
    }
}

This is better, but now whatever class is creating Foo (let's call it Bar) has the problem of filling in Foo's extra dependencies:

public class Bar
{
    public void DoSomething()
    {
        var foo = new Foo(new SomeService());
        foo.DoSomething();
        ...
    }
}

And class Bar probably also doesn't really care about what specific implementation of SomeService Foo uses. Therefore we push the dependency up again:

public class Bar
{
    ISomeService _service;

    public Bar(ISomeService service)
    {
        _service = service;
    }

    public void DoSomething()
    {
        var foo = new Foo(_service);
        foo.DoSomething();
        ...
    }
}

So we find that it is useful to push the responsibility of deciding which specific implementations of which classes to use further and further up in the 'object graph' of the application. Taking this to an extreme, we arrive at the entry point of the application, at which point all dependencies must be satisfied before things start. The dependency injection lingo for this part of the application is called the 'composition root'. It would normally look like this:

var service = new SomeService();
var foo = new Foo(service);
var bar = new Bar(service);
var qux = new Qux(bar);

.. etc.

DI frameworks such as Zenject simply help automate this process of creating and handing out all these concrete dependencies, so that you don't need to explicitly do so yourself like in the above code.

Misconceptions

There are many misconceptions about DI, due to the fact that it can be tricky to fully wrap your head around at first. It will take time and experience before it fully 'clicks'.

As shown in the above example, DI can be used to easily swap different implementations of a given interface (in the example this was ISomeService). However, this is only one of many benefits that DI offers.

More important than that is the fact that using a dependency injection framework like Zenject allows you to more easily follow the 'Single Responsibility Principle'. By letting Zenject worry about wiring up the classes, the classes themselves can just focus on fulfilling their specific responsibilities.

Another common mistake that people new to DI make is that they extract interfaces from every class, and use those interfaces everywhere instead of using the class directly. The goal is to make code more loosely coupled, so it's reasonable to think that being bound to an interface is better than being bound to a concrete class. However, in most cases the various responsibilities of an application have single, specific classes implementing them, so using interfaces in these cases just adds unnecessary maintenance overhead. Also, concrete classes already have an interface defined by their public members. A good rule of F438 thumb instead is to only create interfaces when the class has more than one implementation or in cases where you intend to have multiple implemenations in the future (this is known, by the way, as the Reused Abstraction Principle)

Other benefits include:

Refactorability - When code is loosely coupled, as is the case when using DI properly, the entire code base is much more resilient to changes. You can completely change parts of the code base without having those changes wreak havoc on other parts.
Encourages modular code - When using a DI framework you will naturally follow better design practices, because it forces you to think about the interfaces between classes.
Testability - Writing automated unit tests or user-driven tests becomes very easy, because it is just a matter of writing a different 'composition root' which wires up the dependencies in a different way. Want to only test one subsystem? Simply create a new composition root. Zenject also has some support for avoiding code duplication in the composition root itself (using Installers - described below).

Also see here and here for further discussion and justification for using a DI framework.

Introduction to Zenject API

Hello World Example

using Zenject;
using UnityEngine;
using System.Collections;

public class TestInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        Container.Bind<string>().FromInstance("Hello World!");
        Container.Bind<Greeter>().AsSingle().NonLazy();
    }
}

public class Greeter
{
    public Greeter(string message)
    {
        Debug.Log(message);
    }
}

You can run this example by doing the following:

Create a new scene in Unity
Right Click inside the Hierarchy tab and select Zenject -> Scene Context
Right Click in a folder within the Project Tab and Choose Create -> Zenject -> MonoInstaller. Name it TestInstaller.cs
Add your TestInstaller script to the scene (as its own GameObject or on the same GameObject as the SceneContext, it doesn't matter)
Add a reference to your TestInstaller to the properties of the SceneContext by adding a new row in the inspector of the "Installers" property (press the + button) and then dragging TestInstaller to it
Open up TestInstaller and paste the above code into it
Validate your scene by either selecting Edit -> Zenject -> Validate Current Scene or hitting CTRL+ALT+V. (note that this step isn't necessary but good practice to get into)
Run
Note also, that you can use the shortcut CTRL+SHIFT+R to "validate then run". Validation is usually fast enough that this does not add a noticeable overhead to running your game, and when it does detect errors it is much faster to iterate on since you avoid the startup time.
Observe unity console for output

The SceneContext MonoBehaviour is the entry point of the application, where Zenject sets up all the various dependencies before kicking off your scene. To add content to your Zenject scene, you need to write what is referred to in Zenject as an 'Installer', which declares all the dependencies and their relationships with each other. All dependencies that are marked as "NonLazy" are automatically created after the installers are run, which is why the Greeter class that we added above gets created on startup. If this doesn't make sense to you yet, keep reading!

Injection

There are many different ways of declaring dependencies on the container, which are documented in the next section. There are also several ways of having these dependencies injected into your classes. These are:

1 - Constructor Injection

public class Foo
{
    IBar _bar;

    public Foo(IBar bar)
    {
        _bar = bar;
    }
}

2 - Field Injection

public class Foo
{
    [Inject]
    IBar _bar;
}

Field injection occurs immediately after the constructor is called. All fields that are marked with the [Inject] attribute are looked up in the container and given a value. Note that these fields can be private or public and injection will still occur.

3 - Property Injection

public class Foo
{
    [Inject]
    public IBar Bar
    {
        get;
        private set;
    }
}

Property injection works the same as field injection except is applied to C# properties. Just like fields, the setter can be private or public in this case.

4 - Method Injection

public class Foo
{
    IBar _bar;
    Qux _qux;

    [Inject]
    public void Init(IBar bar, Qux qux)
    {
        _bar = bar;
        _qux = qux;
    }
}

Method Inject injection works very similarly to constructor injection.

Note the following:

Inject methods are the recommended approach for MonoBehaviours, since MonoBehaviours cannot have constructors.
There can be any number of inject methods. In this case, they are called in the order of Base class to Derived class. This can be useful to avoid the need to forward many dependencies from derived classes to the base class via constructor parameters, while also guaranteeing that the base class inject methods complete first, just like how constructors work.
Inject methods are called after all other injection types. It is designed this way so that these methods can be used to execute initialization logic which might make use of injected fields or properties. Note also that you can leave the parameter list empty if you just want to do some initialization logic only.
You can safely assume that the dependencies that you receive via inject methods will themselves already have been injected (the only exception to this is in the case where you have circular dependencies). This can be important if you use inject methods to perform some basic initialization, since in that case you may need the given dependencies to be initialized as well
Note however that it is usually not a good idea to use inject methods for initialization logic. Often it is better to use IInitializable.Initialize or Start() methods instead, since this will allow the entire initial object graph to be created first.

Recommendations

Best practice is to prefer constructor/method injection compared to field/property injection.

Constructor injection forces the dependency to only be resolved once, at class creation, which is usually what you want. In most cases you don't want to expose a public property for your initial dependencies because this suggests that it's open to changing.
Constructor injection guarantees no circular dependencies between classes, which is generally a bad thing to do. Zenject does allow circular dependencies when using other injections types however such as method/field/property injection
Constructor/Method injection is more portable for cases where you decide to re-use the code without a DI framework such as Zenject. You can do the same with public properties but it's more error prone (it's easier to forget to initialize one field and leave the object in an invalid state)
Finally, Constructor/Method injection makes it clear what all the dependencies of a class are when another programmer is reading the code. They can simply look at the parameter list of the method. This is also good because it will be more obvious when a class has too many dependencies and should therefore be split up (since its constructor parameter list will start to seem long)

Register Mappings to the DI Container

The core of a dependency injection framework is the DI container. In it's simplest form it's an object which contains a dictionary that holds all the registrations. In this section we are going to cover the 'register a new mapping' part. In Zenject it's called binding. As it creates a binding between an abstraction to a concrete type.

Binding

Every dependency injection framework is ultimately just a framework to bind types to instances.

In Zenject, dependency mapping is done by adding bindings to something called a container. The container should then 'know' how to create all the object instances in your application, by recursively resolving all dependencies for a given object.

When the container is asked to construct an instance of a given type, it uses C# reflection to find the list of constructor arguments, and all fields/properties that are marked with an [Inject] attribute. It then attempts to resolve each of these required dependencies, which it uses to call the constructor and create the new instance.

Each Zenject application therefore must tell the container how to resolve each of these dependencies, which is done via Bind commands. For example, given the following class:

public class Foo
{
    IBar _bar;

    public Foo(IBar bar)
    {
        _bar = bar;
    }
}

You can wire up the dependencies for this class with the following:

Container.Bind<Foo>().AsSingle();
Container.Bind<IBar>().To<Bar>().AsSingle();

This tells Zenject that every class that requires a dependency of type Foo should use the same instance, which it will automatically create when needed. And similarly, any class that requires the IBar interface (like Foo) will be given the same instance of type Bar.

The full format for the bind command is the following. Note that in most cases you will not use all of these methods and that they all have logical defaults when unspecified

Container.Bind<ContractType>()
    .WithId(Identifier)
    .To<ResultType>()
    .FromConstructionMethod()
    .AsScope()
    .WithArguments(Arguments)
    .OnInstantiated(InstantiatedCallback)
    .When(Condition)
    .(Copy|Move)Into(All|Direct)SubContainers()
    .NonLazy()
    .IfNotBound();

Where:

ContractType = The type that you are creating a binding for.
- This value will correspond to the type of the field/parameter that is being injected.
ResultType = The type to bind to.
- Default: ContractType
- This type must either to equal to ContractType or derive from ContractType. If unspecified, it assumes ToSelf(), which means that the ResultType will be the same as the ContractType. This value will be used by whatever is given as the ConstructionMethod to retrieve an instance of this type
Identifier = The value to use to uniquely identify the binding. This can be ignored in most cases, but can be quite useful in cases where you need to distinguish between multiple bindings with the same contract type. See here for details.
ConstructionMethod = The method by which an instance of ResultType is created/retrieved. See this section for more details on the various construction methods.
- Default: FromNew()
- Examples: eg. FromGetter, FromMethod, FromResolve, FromComponentInNewPrefab, FromSubContainerResolve, FromInstance, etc.
Scope = This value determines how often (or if at all) the generated instance is re-used across multiple injections.
- Default: AsTransient. Note however that not all bindings have a default, so an exception will be thrown if not supplied. The bindings that do not require the scope to be set explicitly are any binding with a construction method that is a search rather than creating a new object from scratch (eg. FromMethod, FromComponentX, FromResolve, etc.)
- It can be one of the following:
  1. AsTransient - Will not re-use the instance at all. Every time ContractType is requested, the DiContainer will execute the given construction method again
  2. AsCached - Will re-use the same instance of ResultType every time ContractType is requested, which it will lazily generate upon first use
  3. AsSingle - Exactly the same as AsCached, except that it will sometimes throw exceptions if there already exists a binding for ResultType. It is simply a way to ensure that the given ResultType is unique within the container. Note however that it will only guarantee that there is only one instance across the given container, which means that using AsSingle with the same binding in a sub-container could generate a second instance.
- In most cases, you will likely want to just use AsSingle, however AsTransient and AsCached have their uses too.
Arguments = A list of objects to use when constructing the new instance of type ResultType. This can be useful as an alternative to adding other bindings for the arguments in the form Container.BindInstance(arg).WhenInjectedInto<ResultType>()
InstantiatedCallback = In some cases it is useful to be able customize an object after it is instantiated. In particular, if using a third party library, it might be necessary to change a few fields on one of its types. For these cases you can pass a method to OnInstantiated that can customize the newly created instance. For example:
```
Container.Bind<Foo>().AsSingle().OnInstantiated<Foo>(OnFooInstantiated);

void OnFooInstantiated(InjectContext context, Foo foo)
{
    foo.Qux = "asdf";
}
```
Or, equivalently:
```
Container.Bind<Foo>().AsSingle().OnInstantiated<Foo>((ctx, foo) => foo.Bar = "qux");
```
Note that you can also bind a custom factory using FromFactory that directly calls Container.InstantiateX before customizing it for the same effect, but OnInstantiated can be easier in some cases
Condition = The condition that must be true for this binding to be chosen. See here for more details.
(Copy|Move)Into(All|Direct)SubContainers = This value can be ignored for 99% of users. It can be used to automatically have the binding inherited by subcontainers. For example, if you have a class Foo and you want a unique instance of Foo to be automatically placed in the container and every subcontainer, then you could add the following binding:
```
Container.Bind<Foo>().AsSingle().CopyIntoAllSubContainers()
```
In other words, the result will be equivalent to copying and pasting the Container.Bind<Foo>().AsSingle() statement into the installer for every sub-container.

Or, if you only wanted Foo in the subcontainers and not the current container:
```
Container.Bind<Foo>().AsSingle().MoveIntoAllSubContainers()
```
Or, if you only wanted Foo to be in the immediate child subcontainer, and not the subcontainers of these subcontainers:
```
Container.Bind<Foo>().AsSingle().MoveIntoDirectSubContainers()
```
NonLazy = Normally, the ResultType is only ever instantiated when the binding is first used (aka "lazily"). However, when NonLazy is used, ResultType will immediately be created on startup.
IfNotBound = When this is added to a binding and there is already a binding with the given contract type + identifier, then this binding will be skipped.

Construction Methods

FromNew - Create via the C# new operator. This is the default if no construction method is specified.
```
// These are both the same
Container.Bind<Foo>();
Container.Bind<Foo>().FromNew();
```

FromInstance - Add a given instance to the container. Note that the given instance will not be injected in this case. If you also want your instance to be injected at startup, see QueueForInject

Container.Bind<Foo>().FromInstance(new Foo());

// You can also use this short hand which just takes ContractType from the parameter type
Container.BindInstance(new Foo());

// This is also what you would typically use for primitive types
Container.BindInstance(5.13f);
Container.BindInstance("foo");

// Or, if you have many instances, you can use BindInstances
Container.BindInstances(5.13f, "foo", new Foo());

FromMethod - Create via a custom method

Container.Bind<Foo>().FromMethod(SomeMethod);

Foo SomeMethod(InjectContext context)
{
    ...
    return new Foo();
}

FromMethodMultiple - Same as FromMethod except allows returning multiple instances at once (or zero).

Container.Bind<Foo>().FromMethodMultiple(GetFoos);

IEnumerable<Foo> GetFoos(InjectContext context)
{
    ...
    return new Foo[]
    {
        new Foo(),
        new Foo(),
        new Foo(),
    }
}

FromFactory - Create instance using a custom factory class. This construction method is similar to FromMethod except can be cleaner in cases where the logic is more complicated or requires dependencies (since the factory itself can have dependencies injected)
```
class FooFactory : IFactory<Foo>
{
    public Foo Create()
    {
        // ...
        return new Foo();
    }
}

Container.Bind<Foo>().FromFactory<FooFactory>()
```

FromIFactory - Create instance using a custom factory class. This is a more generic and more powerful alternative to FromFactory, because you can use any kind of construction method you want for your custom factory (unlike FromFactory which assumes FromNew().AsCached())

For example, you could use a factory that is a scriptable object like this:

class FooFactory : ScriptableObject, IFactory<Foo>
{
    public Foo Create()
    {
        // ...
        return new Foo();
    }
}

Container.Bind<Foo>().FromIFactory(x => x.To<FooFactory>().FromScriptableObjectResource("FooFactory")).AsSingle();

Or, you might want to have your custom factory be placed in a subcontainer like this:

public class FooFactory : IFactory<Foo>
{
    public Foo Create()
    {
        return new Foo();
    }
}

public override void InstallBindings()
{
    Container.Bind<Foo>().FromIFactory(x => x.To<FooFactory>().FromSubContainerResolve().ByMethod(InstallFooFactory)).AsSingle();
}

void InstallFooFactory(DiContainer subContainer)
{
    subContainer.Bind<FooFactory>().AsSingle();
}

Also note that the following two lines are equivalent:

Container.Bind<Foo>().FromFactory<FooFactory>().AsSingle();
Container.Bind<Foo>().FromIFactory(x => x.To<FooFactory>().AsCached()).AsSingle();

FromComponentInNewPrefab - Instantiate the given prefab as a new game object, inject any MonoBehaviour's on it, and then search the result for type ResultType in a similar way that GetComponentInChildren works
```
Container.Bind<Foo>().FromComponentInNewPrefab(somePrefab);
```
ResultType must either be an interface or derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case

Note that if there are multiple matches for ResultType on the prefab it will only match the first one encountered just like how GetComponentInChildren works. So if you are binding a prefab and there isn't a specific MonoBehaviour/Component that you want to bind to, you can just choose Transform and it will match the root of the prefab.
FromComponentsInNewPrefab - Same as FromComponentInNewPrefab except will match multiple values or zero values. You might use it for example and then inject List<ContractType> somewhere. Can be thought of as similar to GetComponentsInChildren
FromComponentInNewPrefabResource - Instantiate the given prefab (found at the given resource path) as a new game object, inject any MonoBehaviour's on it, and then search the result for type ResultType in a similar way that GetComponentInChildren works (in that it will return the first matching value found)
```
Container.Bind<Foo>().FromComponentInNewPrefabResource("Some/Path/Foo");
```
ResultType must either be an interface or derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case
FromComponentsInNewPrefabResource - Same as FromComponentInNewPrefab except will match multiple values or zero values. You might use it for example and then inject List<ContractType> somewhere. Can be thought of as similar to GetComponentsInChildren
FromNewComponentOnNewGameObject - Create a new empty game object and then instantiate a new component of the given type on it.
```
Container.Bind<Foo>().FromNewComponentOnNewGameObject();
```
ResultType must derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case
FromNewComponentOnNewPrefab - Instantiate the given prefab as a new game object and also instantiate a new instance of the given component on the root of the new game object. NOTE: It is not necessary that the prefab contains a copy of the given component.
```
Container.Bind<Foo>().FromNewComponentOnNewPrefab(somePrefab);
```
ResultType must derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case
FromNewComponentOnNewPrefabResource - Instantiate the given prefab (found at the given resource path) and also instantiate a new instance of the given component on the root of the new game object. NOTE: It is not necessary that the prefab contains a copy of the given component.
```
Container.Bind<Foo>().FromNewComponentOnNewPrefabResource("Some/Path/Foo");
```
ResultType must derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case
FromNewComponentOn - Instantiate a new component of the given type on the given game object
```
Container.Bind<Foo>().FromNewComponentOn(someGameObject);
```
ResultType must derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case
FromNewComponentSibling - Instantiate a new component of the given on the current transform. The current transform here is taken from the object being injected, which must therefore be a MonoBehaviour derived type.
```
Container.Bind<Foo>().FromNewComponentSibling();
```
Note that if the given component type is already attached to the current transform that this will just return that instead of creating a new component. As a result, this bind statement functions similar to Unity's [RequireComponent] attribute.

ResultType must derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case

Also note that if a non-MonoBehaviour requests the given type, an exception will be thrown, since there is no current transform in that case.
FromComponentInHierarchy - Look up the component within the scene hierarchy associated with the current context, as well as the hierarchy associated with any parent contexts. Works similar to GetComponentInChildren in that it will return the first matching value found
```
Container.Bind<Foo>().FromComponentInHierarchy().AsSingle();
```
ResultType must either be an interface or derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case

In the most common case where the context is a SceneContext, this will search the entire scene hierarchy (except any sub-contexts such as GameObjectContext). In other words, when the current context is a scene context, it will behave similar to GameObject.FindObjectsOfType. Note that since this could b 10000 e a big search, it should be used with caution, just like GameObject.FindObjectsOfType should be used with caution.

In the case where the context is GameObjectContext, it will only search within and underneath the game object root (and any parent contexts).

In the case where the context is ProjectContext, it will only search within the project context prefab
FromComponentsInHierarchy - Same as FromComponentInHierarchy except will match multiple values or zero values. You might use it for example and then inject List<ContractType> somewhere. Can be thought of as similar to GetComponentsInChildren
FromComponentSibling - Look up the given component type by searching over the components that are attached to the current transform. The current transform here is taken from the object being injected, which must therefore be a MonoBehaviour derived type.
```
Container.Bind<Foo>().FromComponentSibling();
```
ResultType must either be an interface or derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case

Note that if a non-MonoBehaviour requests the given type, an exception will be thrown, since there is no current transform in that case.
FromComponentsSibling - Same as FromComponentSibling except will match multiple values or zero values.
FromComponentInParents - Look up the component by searching the current transform or any parent for the given component type. The current transform here is taken from the object being injected, which must therefore be a MonoBehaviour derived type.
```
Container.Bind<Foo>().FromComponentInParents();
```
ResultType must either be an interface or derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case

Note that if a non-MonoBehaviour requests the given type, an exception will be thrown, since there is no current transform in that case.
FromComponentsInParents - Same as FromComponentInParents except will match multiple values or zero values. You might use it for example and then inject List<ContractType> somewhere
FromComponentInChildren - Look up the component by searching the current transform or any child transform for the given component type. The current transform here is taken from the object being injected, which must therefore be a MonoBehaviour derived type. Similar to GetComponentInChildren in that it will return the first matching value found
```
Container.Bind<Foo>().FromComponentInChildren();
```
ResultType must either be an interface or derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case

Note that if a non-MonoBehaviour requests the given type, an exception will be thrown, since there is no current transform in that case.
FromComponentsInChildren - Same as FromComponentInChildren except will match multiple values or zero values. You might use it for example and then inject List<ContractType> somewhere. Can be thought of as similar to GetComponentsInChildren
FromNewComponentOnRoot - Instantiate the given component on the root of the current context. This is most often used with GameObjectContext.
```
Container.Bind<Foo>().FromNewComponentOnRoot();
```
ResultType must derive from UnityEngine.MonoBehaviour / UnityEngine.Component in this case
FromResource - Create by calling the Unity3d function Resources.Load for ResultType. This can be used to load any type that Resources.Load can load, such as textures, sounds, prefabs, etc.
```
Container.Bind<Texture>().WithId("Glass").FromResource("Some/Path/Glass");
```
FromResources - Same as FromResource except will match multiple values or zero values. You might use it for example and then inject List<ContractType> somewhere
FromScriptableObjectResource - Bind directly to the scriptable object instance at the given resource path. NOTE: Changes to this value while inside unity editor will be saved persistently. If this is undesirable, use FromNewScriptableObjectResource.
```
public class Foo : ScriptableObject
{
}

Container.Bind<Foo>().FromScriptableObjectResource("Some/Path/Foo");
```
FromNewScriptableObjectResource - Same as FromScriptableObjectResource except it will instantiate a new copy of the given scriptable object resource. This can be useful if you want to have multiple distinct instances of the given scriptable object resource, or if you want to ensure that the saved values for the scriptable object are not affected by changing at runtime.
FromResolve - Get instance by doing another lookup on the container (in other words, calling DiContainer.Resolve<ResultType>()). Note that for this to work, ResultType must be bound in a separate bind statement. This construction method can be especially useful when you want to bind an interface to another interface, as shown in the below example
```
public interface IFoo
{
}

public interface IBar : IFoo
{
}

public class Foo : IBar
{
}

Container.Bind<IFoo>().To<IBar>().FromResolve();
Container.Bind<IBar>().To<Foo>();
```
FromResolveAll - Same as FromResolve except will match multiple values or zero values.
FromResolveGetter<ObjectType> - Get instance from the property of another dependency which is obtained by doing another lookup on the container (in other words, calling DiContainer.Resolve<ObjectType>() and then accessing a value on the returned instance of type ResultType). Note that for this to work, ObjectType must be bound in a separate bind statement.
```
public class Bar
{
}

public class Foo
{
    public Bar GetBar()
    {
        return new Bar();
    }
}

Container.Bind<Foo>();
Container.Bind<Bar>().FromResolveGetter<Foo>(x => x.GetBar());
```
FromResolveAllGetter<ObjectType> - Same as FromResolveGetter except will match multiple values or zero values.
FromSubContainerResolve - Get ResultType by doing a lookup on a subcontainer. Note that for this to work, the sub-container must have a binding for ResultType. This approach can be very powerful, because it allows you to group related dependencies together inside a mini-container, and then expose only certain classes (aka "Facades") to operate on this group of dependencies at a higher level. For more details on using sub-containers, see this section. There are several ways to define the subcontainer:
1. ByNewPrefabMethod - Initialize subcontainer by instantiating a new prefab. Note that unlike ByNewContextPrefab, this bind method does not require that there be a GameObjectContext attached to the prefab. In this case the GameObjectContext is added dynamically and then run with the given installer method.
```
Container.Bind<Foo>().FromSubContainerResolve().ByNewPrefabMethod(MyPrefab, InstallFoo);

void InstallFoo(DiContainer subContainer)
{
    subContainer.Bind<Foo>();
}
```
  Note that instead of passing in a prefab directly, you can also pass in a getter method. For example:
```
Container.Bind<Foo>().FromSubContainerResolve().ByNewPrefabMethod(ChooseFooPrefab, InstallFoo);

UnityEngine.Object ChooseFooPrefab(InjectContext context) {
    return FooPrefabs[Random.Range(0, FooPrefabs.Length)];
}
```
2. ByNewPrefabInstaller - Initialize subcontainer by instantiating a new prefab. Same as ByNewPrefabMethod, except it initializes the dynamically created GameObjectContext with the given installer rather than a method.
```
Container.Bind<Foo>().FromSubContainerResolve().ByNewPrefabInstaller<FooInstaller>(MyPrefab);

class FooInstaller : Installer
{
    public override void InstallBindings()
    {
        Container.Bind<Foo>();
    }
}
```
  Note that instead of passing in a prefab directly, you can also pass in a getter method. For example:
```
Container.Bind<Foo>().FromSubContainerResolve().ByNewPrefabInstaller<FooInstaller>(ChooseFooPrefab);

UnityEngine.Object ChooseFooPrefab(InjectContext context) {
    return FooPrefabs[Random.Range(0, FooPrefabs.Length)];
}
```
3. ByNewPrefabResourceMethod - Initialize subcontainer instantiating a new prefab obtained via Resources.Load. Note that unlike ByNewPrefabResource, this bind method does not require that there be a GameObjectContext attached to the prefab. In this case the GameObjectContext is added dynamically and then run with the given installer method.
```
Container.Bind<Foo>().FromSubContainerResolve().ByNewPrefabResourceMethod("Path/To/MyPrefab", InstallFoo);

void InstallFoo(DiContainer subContainer)
{
    subContainer.Bind<Foo>();
}
```
4. ByNewPrefabResourceInstaller - Initialize subcontainer instantiating a new prefab obtained via Resources.Load. Same as ByNewPrefabResourceMethod, except it initializes the dynamically created GameObjectContext with the given installer rather than a method.
```
Container.Bind<Foo>().FromSubContainerResolve().ByNewPrefabResourceInstaller<FooInstaller>("Path/To/MyPrefab");

class FooInstaller : Installer<FooInstaller>
{
    public override void InstallBindings()
    {
        Container.Bind<Foo>();
    }
}
```
5. ByNewGameObjectInstaller - Initialize subcontainer by instantiating a empty game object, attaching GameObjectContext, and then installing using the given installer. This approach is very similar to ByInstaller except has the following advantages:
  - Any ITickable, IInitializable, IDisposable, etc. bindings will be called properly
  - Any new game objects that are instantiated inside the subcontainer will be parented underneath the game object context object
  - You can destroy the subcontainer by just destroying the game object context game object
6. ByNewGameObjectMethod - Same as ByNewGameObjectInstaller except the subcontainer is initialized based on the given method rather than an installer type.
7. ByMethod - Initialize the subcontainer by using a method.
```
Container.Bind<Foo>().FromSubContainerResolve().ByMethod(InstallFooFacade);

void InstallFooFacade(DiContainer subContainer)
{
    subContainer.Bind<Foo>();
}
```
  Note that when using ByMethod, if you want to use zenject interfaces such as ITickable, IInitializable, IDisposable inside your subcontainer then you have to also use the WithKernel bind method like this:
```
Container.Bind<Foo>().FromSubContainerResolve().ByMethod(InstallFooFacade).WithKernel();

void InstallFooFacade(DiContainer subContainer)
{
    subContainer.Bind<Foo>();
    subContainer.Bind<ITickable>().To<Bar>();
}
```
8. ByInstaller - Initialize the subcontainer by using a class derived from Installer. This can be a cleaner and less error-prone alternative than ByMethod, especially if you need to inject data into the installer itself. Less error prone because when using ByMethod it is common to accidentally use Container instead of subContainer in your method.
```
Container.Bind<Foo>().FromSubContainerResolve().ByInstaller<FooFacadeInstaller>();

class FooFacadeInstaller : Installer
{
    public override void InstallBindings()
    {
        Container.Bind<Foo>();
    }
}
```
  Note that when using ByInstaller, if you want to use zenject interfaces such as ITickable, IInitializable, IDisposable inside your subcontainer then you have to also use the WithKernel bind method like this:
```
Container.Bind<Foo>().FromSubContainerResolve().ByInstaller<FooFacadeInstaller>().WithKernel();
```
9. ByNewContextPrefab - Initialize subcontainer by instantiating a new prefab. Note that the prefab must contain a GameObjectContext component attached to the root game object. For details on GameObjectContext see this section.
```
Container.Bind<Foo>().FromSubContainerResolve().ByNewContextPrefab(MyPrefab);

// Assuming here that this installer is added to the GameObjectContext at the root
// of the prefab.  You could also use a ZenjectBinding in the case where Foo is a MonoBehaviour
class FooFacadeInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        Container.Bind<Foo>();
    }
}
```
10. ByNewContextPrefabResource - Initialize subcontainer instantiating a new prefab obtained via Resources.Load. Note that the prefab must contain a GameObjectContext component attached to the root game object.
```
Container.Bind<Foo>().FromSubContainerResolve().ByNewContextPrefabResource("Path/To/MyPrefab");
```
11. ByInstance - Initialize the subcontainer by directly using a given instance of DiContainer that you provide yourself. This is only useful in some rare edge cases.
FromSubContainerResolveAll - Same as FromSubContainerResolve except will match multiple values or zero values.

Installers

Often, there is some collections of related bindings for each sub-system and so it makes sense to group those bindings into a re-usable object. In Zenject this re-usable object is called an 'installer'. You can define a new installer as follows:

public class FooInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        Container.Bind<Bar>().AsSingle();
        Container.BindInterfacesTo<Foo>().AsSingle();
        // etc...
    }
}

You add bindings by overriding the InstallBindings method, which is called by whatever Context the installer has been added to (usually this is SceneContext). MonoInstaller is a MonoBehaviour so you can add FooInstaller by attaching it to a GameObject. Since it is a GameObject you can also add public members to it to configure your installer from the Unity inspector. This allows you to add references within the scene, references to assets, or simply tuning data (see here for more information on tuning data).

Note that in order for your installer to be triggered it must be attached to the Installers property of the SceneContext object. Installers are installed in the order given to SceneContext (with scriptable object installers first, then mono installers, then prefab installers) however this order should not usually matter (since nothing should be instantiated during the install process).

In many cases you want to have your installer derive from MonoInstaller, so that you can have inspector settings. There is also another base class called simply Installer which you can use in cases where you do not need it to be a MonoBehaviour.

You can also call an installer from another installer. For example:

public class BarInstaller : Installer<BarInstaller>
{
    public override void InstallBindings()
    {
        ...
    }
}

public class FooInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        BarInstaller.Install(Container);
    }
}

Note that in this case BarInstaller is of type Installer<> (note the generic arguments) and not MonoInstaller, which is why we can simply call BarInstaller.Install(Container) and don't require that BarInstaller be added to our scene already. Any calls to BarInstaller.Install will immediately instantiate a temporary instance of BarInstaller and then call InstallBindings on it. This will repeat for any installers that this installer installs. Note also that when using the Installer<> base class, we always must pass in ourself as the generic argument to Installer<>. This is necessary so that the Installer<> base class can define the static method BarInstaller.Install. It is also designed this way to support runtime parameters (described below).

One of the main reasons we use installers as opposed to just having all our bindings declared all at once for each scene, is to make them re-usable. This is not a problem for installers of type Installer<> because you can simply call FooInstaller.Install as described above for every scene you wish to use it in, but then how would we re-use a MonoInstaller in multiple scenes?

There are three ways to do this.

Prefab instances within the scene. After attaching your MonoInstaller to a gameobject in your scene, you can then create a prefab out of it. This is nice because it allows you to share any configuration that you've done in the inspector on the MonoInstaller across scenes (and also have per-scene overrides if you want). After adding it in your scene you can then drag and drop it on to the Installers property of a Context
Prefabs. You can also directly drag your installer prefab from the Project tab into the InstallerPrefabs property of SceneContext. Note that in this case you cannot have per-scene overrides like you can when having the prefab instantiated in your scene, but can be nice to avoid clutter in the scene.
Prefabs within Resources folder. You can also place your installer prefabs underneath a Resoures folder and install them directly from code by using the Resources path. For details on usage see here.

Another option in addition to MonoInstaller and Installer<> is to use ScriptableObjectInstaller which has some unique advantages (especially for settings) - for details see here.

When calling installers from other installers it is common to want to pass parameters into it. See here for details on how that is done.

Using Non-MonoBehaviour Classes

ITickable

In some cases it is preferable to avoid the extra weight of MonoBehaviours in favour of just normal C# classes. Zenject allows you to do this much more easily by providing interfaces that mirror functionality that you would normally need to use a MonoBehaviour for.

For example, if you have code that needs to run per frame, then you can implement the ITickable interface:

public class Ship : ITickable
{
    public void Tick()
    {
        // Perform per frame tasks
    }
}

Then, to hook it up in an installer:

Container.Bind<ITickable>().To<Ship>().AsSingle();

Or if you don't want to have to always remember which interfaces your class implements, you can use the shortcut described here

Note that the order that the Tick() is called in for all ITickables is also configurable, as outlined here.

Also note that there are interfaces ILateTickable and IFixedTickable which mirror Unity's LateUpdate and FixedUpdated methods

IInitializable

If you have some initialization that needs to occur on a given object, you could include this code in the constructor. However, this means that the initialization logic would occur in the middle of the object graph being constructed, so it may not be ideal.

A better alternative is to implement IInitializable, and then perform initialization logic in an Initialize() method.

Then, to hook it up in an installer:

Container.Bind<IInitializable>().To<Foo>().AsSingle();

Or if you don't want to have to always remember which interfaces your class implements, you can use the shortcut described here

The Foo.Initialize method would then be called after the entire object graph is constructed and all constructors have been called.

Note that the constructors for the initial object graph are called during Unity's Awake event, and that the IInitializable.Initialize methods are called immediately on Unity's Start event. Using IInitializable as opposed to a constructor is therefore more in line with Unity's own recommendations, which suggest that the Awake phase be used to set up object references, and the Start phase be used for more involved initialization logic.

This can also be better than using constructors or [Inject] methods because the initialization order is customizable in a similar way to ITickable, as explained here.

public class Ship : IInitializable
{
    public void Initialize()
    {
        // Initialize your object here
    }
}

IInitializable works well for start-up initialization, but what about for objects that are created dynamically via factories? (see this section for what I'm referring to here). For these cases you will most likely want to either use an [Inject] method or an explicit Initialize method that is called after the object is created. For example:

public class Foo
{
    [Inject]
    IBar _bar;

    [Inject]
    public void Initialize()
    {
        ...
        _bar.DoStuff();
        ...
    }
}

IDisposable

If you have external resources that you want to clean up when the app closes, the scene changes, or for whatever reason the context object is destroyed, you can declare your class as IDisposable like below:

public class Logger : IInitializable, IDisposable
{
    FileStream _outStream;

    public void Initialize()
    {
        _outStream = File.Open("log.txt", FileMode.Open);
    }

    public void Log(string msg)
    {
        _outStream.WriteLine(msg);
    }

    public void Dispose()
    {
        _outStream.Close();
    }
}

Then in your installer you can include:

Container.Bind(typeof(Logger), typeof(IInitializable), typeof(IDisposable)).To<Logger>().AsSingle();

Or you can use the BindInterfaces shortcut:

Container.BindInterfacesAndSelfTo<Logger>().AsSingle();

This works because when the scene changes or your unity application is closed, the unity event OnDestroy() is called on all MonoBehaviours, including the SceneContext class, which then triggers Dispose() on all objects that are bound to IDisposable.

You can also implement the ILateDisposable interface which works similar to ILateTickable in that it will be called after all IDisposable objects have been triggered. However, for most cases you're probably better off setting 10000 an explicit execution order instead if the order is an issue.

BindInterfacesTo and BindInterfacesAndSelfTo

If you end up using the ITickable, IInitializable, and IDisposable interfaces as described above, you will often end up with code like this:

Container.Bind(typeof(Foo), typeof(IInitializable), typeof(IDisposable)).To<Logger>().AsSingle();

This can be a bit verbose sometimes. Also, it is not ideal because if I later on decide that Foo doesn't need a Tick() or a Dispose() then I have to keep the installer in sync.

A better idea might be to just always use the interfaces like this:

Container.Bind(new[] { typeof(Foo) }.Concat(typeof(Foo).GetInterfaces())).To<Foo>().AsSingle();

This pattern is useful enough that Zenject includes a custom bind method for it. The above code is equivalent to:

Container.BindInterfacesAndSelfTo<Foo>().AsSingle();

Now, we can add and remove interfaces to/from Foo and the installer remains the same.

In some cases you might only want to bind the interfaces, and keep Foo hidden from other classes. In that case you would use the BindInterfacesTo method instead:

Container.BindInterfacesTo<Foo>().AsSingle()

Which, in this case, would expand to:

Container.Bind(typeof(IInitializable), typeof(IDisposable)).To<Foo>().AsSingle();

Using the Unity Inspector To Configure Settings

One implication of writing most of your code as normal C# classes instead of MonoBehaviour's is that you lose the ability to configure data on them using the inspector. You can however still take advantage of this in Zenject by using the following pattern:

public class Foo : ITickable
{
    readonly Settings _settings;

    public Foo(Settings settings)
    {
        _settings = settings;
    }

    public void Tick()
    {
        Debug.Log("Speed: " + _settings.Speed);
    }

    [Serializable]
    public class Settings
    {
        public float Speed;
    }
}

Then, in an installer:

public class TestInstaller : MonoInstaller<TestInstaller>
{
    public Foo.Settings FooSettings;

    public override void InstallBindings()
    {
        Container.BindInstance(FooSettings);
        Container.BindInterfacesTo<Foo>().AsSingle();
    }
}

Or, equivalently:

public class TestInstaller : MonoInstaller<TestInstaller>
{
    public Foo.Settings FooSettings;

    public override void InstallBindings()
    {
        Container.BindInterfacesTo<Foo>().AsSingle().WithArguments(FooSettings);
    }
}

Now, if we run our scene we can change the speed value to tune the Foo class in real time.

Another (arguably better) way to do this is to use ScriptableObjectInstaller instead of MonoInstaller, which have the added advantage that you can change your settings at runtime and have those changes automatically persist when play mode is stopped. See here for details.

Object Graph Validation

Overview

The usual workflow when setting up bindings using a DI framework is something like this:

Add some number of bindings in code
Execute your app
Observe a bunch of DI related exceptions
Modify your bindings to address problem
Repeat

This works ok for small projects, but as the complexity of your project grows it is often a tedious process. The problem gets worse if the startup time of your application is particularly bad, or when the exceptions only occur from factories at various points at runtime. What would be great is some tool to analyze your object graph and tell you exactly where all the missing bindings are, without requiring the cost of firing up your whole app.

You can do this in Zenject out-of-the-box by executing the menu item Edit -> Zenject -> Validate Current Scene or simply hitting SHIFT+ALT+V with the scenes open that you want to validate. This will execute all installers for the current scene, with the result being a fully bound container. It will then iterate through the object graphs and verify that all bindings can be found (without actually instantiating any of them). In other words, it executes a 'dry run' of the normal startup procedure. Under the hood, this works by storing dummy objects in the container in place of actually instantiating your classes.

Alternatively, you can execute the menu item Edit -> Zenject -> Validate Then Run or simply hitting CTRL+ALT+R. This will validate the scenes you have open and then if validation succeeds, it will start play mode. Validation is usually pretty fast so this can be a good alternative to always just hitting play, especially if your game has a costly startup time.

Note that this will also include factories and memory pools, which is especially helpful because those errors might not be caught until sometime after startup.

There are a few things to be aware of:

No actual logic code is executed - only install bindings is called. This means that if you have logic inside installers other than bind commands that these will be executed as well and may cause issues when running validation (if that logic requires that the container return actual values)
null values are injected into the dependencies that are actually instantiated such as installers (regardles of what was binded)

You might want to inject some classes even in validation mode. In that case you can mark them with [ZenjectAllowDuringValidation].

Also note that some validation behaviour is configurable in zenjectsettings

Custom validatables

If you want to add your own validation logic, you can do this simply by having one of your classes inherit from IValidatable. After doing this, as long as your class is bound in an installer somewhere, it will be instantiated during validation and then its Validate() method will be called. Note however that any dependencies it has will be injected as null (unless marked with [ZenjectAllowDuringValidation] attribute).

Inside the Validate method you can throw exceptions if you want validation to fail, or you can just log information to the console. One common thing that occurs in custom validatables is to instantiate types that would not otherwise be validated. By instantiating them during validation it will ensure that all their dependencies can be resolved.

For example, if you create a custom factory that directly instantiates a type using Container.Instantiate<Foo>(), then Foo will not be validated, so you will not find out if it is missing some dependencies until runtime. However you can fix this by having your factory implement IValidatable and then calling Container.Instantiate<Foo>() inside the Validate() method.

Scene Bindings

In many cases, you have a number of MonoBehaviours that have been added to the scene within the Unity editor (ie. at editor time not runtime) and you want to also have these MonoBehaviours added to the Zenject Container so that they can be injected into other classes.

The usual way this is done is to add public references to these objects within your installer like this:

public class Foo : MonoBehaviour
{
}

public class GameInstaller : MonoInstaller
{
    public Foo foo;

    public override void InstallBindings()
    {
        Container.BindInstance(foo);
        Container.Bind<IInitializable>().To<GameRunner>().AsSingle();
    }
}

public class GameRunner : IInitializable
{
    readonly Foo _foo;

    public GameRunner(Foo foo)
    {
        _foo = foo;
    }

    public void Initialize()
    {
        ...
    }
}

This works fine however in some cases this can get cumbersome. For example, if you want to allow an artist to add any number of Enemy objects to the scene, and you also want all those Enemy objects added to the Zenject Container. In this case, you would have to manually drag each one to the inspector of one of your installers. This is very error prone since its easy to forget one, or to delete the Enemy game object but forget to delete the null reference in the inspector for your installer, etc.

Another way to do this would be to use the FromComponentInHierarchy bind method like this:

public class GameInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        Container.Bind<Foo>().FromComponentInHierarchy().AsTransient();
        Container.Bind<IInitializable>().To<GameRunner>().AsSingle();
    }
}

Now, whenever a dependency of type Foo is required, zenject will search the entire scene for any MonoBehaviours of type Foo. This will function very similarly to use Unity's FindObjectsOfType method every time you want to look up a certain dependency. Note that because this method can be a very heavy operation, you probably want to mark it AsCached or AsSingle like this instead:

public class GameInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        Container.Bind<Foo>().FromComponentInHierarchy().AsCached();
        Container.Bind<IInitializable>().To<GameRunner>().AsSingle();
    }
}

This way, you only incur the performance hit for the search once the first time it is needed instead of every time it is injected to any class. Note also that we can FromComponentsInHierarchy (note the plural) in cases where we expect there to be multiple Foos.

Yet another way to do this is to use the ZenjectBinding component. You can do this by adding a ZenjectBinding MonoBehaviour to the same game object that you want to be automatically added to the Zenject container.

For example, if you have a MonoBehaviour of type Foo in your scene, you can just add ZenjectBinding alongside it, and then drag the Foo component into the Component property of the ZenjectBinding component.

Then our installer becomes:

public class GameInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        Container.Bind<IInitializable>().To<GameRunner>().AsSingle();
    }
}

The ZenjectBinding component has the following properties:

Bind Type - This will determine what 'contract type' to use. It can be set to any of the following values:
1. Self
This is equivalent to the first example where we did this:
```
Container.Bind<Foo>().FromInstance(_foo);
```
Or, equivalently:
```
Container.BindInstance(_foo);
```
So if we duplicate this game object to have multiple game objects with Foo on them (as well as the ZenjectBinding), they will all be bound to the Container this way. So after doing this, we would have to change GameRunner above to take a List<Foo> otherwise we would get Zenject exceptions (see here for info on list bindings).
1. AllInterfaces
This bind type is equivalent to the following:
```
Container.BindInterfacesTo(_foo.GetType()).FromInstance(_foo);
```
Note however, in this case, that GameRunner must ask for type IFoo in its constructor. If we left GameRunner asking for type Foo then Zenject would throw exceptions, since the BindInterfaces method only binds the interfaces, not the concrete type. If you want the concrete type as well then you can use:
1. AllInterfacesAndSelf
This bind type is equivalent to the following:
```
Container.BindInterfacesAndSelfTo(_foo.GetType()).FromInstance(_foo);
```
This is the same as AllInterfaces except we can directly access Foo using type Foo instead of needing an interface.
1. BaseType
This bind type is equivalent to the following:
```
Container.Bind(_foo.GetType().BaseType()).FromInstance(_foo)
```
Identifier - This value can be left empty most of the time. It will determine what is used as the identifier for the binding. For example, when set to "Foo1", it is equivalent to doing the following:
```
Container.BindInstance(_foo).WithId("Foo1");
```
Use Scene Context - This is optional but useful in cases where you want to bind a dependency that is inside a GameObjectContext to the SceneContext. You could also drag the SceneContext to the Context properly but this flag is a bit easier.
Context - This is completely optional and in most cases should be left unset. This will determine which Context to apply the binding to. If left unset, it will use whatever context the GameObject is in. In most cases this will be SceneContext, but if it's inside a GameObjectContext it will be bound into the GameObjectContext container instead. One important use case for this field is to allow dragging the SceneContext into this field, for cases where the component is inside a GameObjectContext. This allows you to treat this MonoBehaviour as a Facade for the entire sub-container given by the GameObjectContext.

General Guidelines / Recommendations / Gotchas / Tips and Tricks

Do not use GameObject.Instantiate if you want your objects to have their dependencies injected
- If you want to instantiate a prefab at runtime and have any MonoBehaviour's automatically injected, we recommend using a factory. You can also instantiate a prefab by directly using the DiContainer by calling any of the InstantiatePrefab methods. Using these ways as opposed to GameObject.Instantiate will ensure any fields that are marked with the [Inject] attribute are filled in properly, and all [Inject] methods within the prefab are called.
Best practice with DI is to only reference the container in the composition root "layer"
- Note that factories are part of this layer and the container can be referenced there (which is necessary to create objects at runtime). See here for more details on this.
Do not use IInitializable, ITickable and IDisposable for dynamically created objects
- Objects that are of type IInitializable are only initialized once - at startup during Unity's Start phase. If you create an object through a factory, and it derives from IInitializable, the Initialize() method will not be called. You should use [Inject] methods in this case or call Initialize() explicitly yourself after calling Create.
- The same applies to ITickable and IDisposable. Deriving from these will do nothing unless they are part of the original object graph created at startup.
- If you have dynamically created objects that have an Update() method, it is usually best to call Update() on those manually, and often there is a higher level manager-like class in which it makes sense to do this from. If however you prefer to use ITickable for dynamically objects you can declare a dependency to TickableManager and add/remove it explicitly as well.
Using multiple constructors
- You can have multiple constructors however you must mark one of them with the [Inject] attribute so Zenject knows which one to use. If you have multiple constructors and none of them are marked with [Inject] then Zenject will guess that the intended constructor is the one with the least amount of arguments.
Lazily instantiated objects and the object graph
- Zenject does not immediately instantiate every object defined by the bindings that you've set up in your installers. It will only instantiate those bindings that are marked NonLazy. All other bindings are only instantiated when they are needed. All the NonLazy objects as well as all their dependencies form the 'initial object graph' of the application. Note that this automatically includes all types that implement IInitializable, ITickable, IDisposable, etc. So if you have a binding that is not being created because nothing in the initial object graph references it, then you can make this explicit by adding NonLazy to your binding
Restrict the use of bind commands to the 'composition root' only. In other words, do not make calls to Container.Bind, Container.Rebind, or Container.Unbind after the install phase is completed. This important because immediately after install completes the initial object graph of your application is constructed, and needs access to the full set of bindings.
The order that things occur in is wrong, like injection is occurring too late, or Initialize() event is not called at the right time, etc.
- It may be because the 'script execution order' of the Zenject classes ProjectKernel or SceneKernel or SceneContext is incorrect. These classes should always have the earliest or near earliest execution order. This should already be set by default (since this setting is included in the cs.meta files for these classes). However if you are compiling Zenject yourself or have a unique configuration, you may want to make sure, which you can do by going to "Edit -> Project Settings -> Script Execution Order" and confirming that these classes are at the top, before the default time.

Advanced

Binding

Game Object Bind Methods

For bindings that create new game objects (eg. FromComponentInNewPrefab, FromNewComponentOnNewGameObject, etc.) there are also two extra bind methods.

WithGameObjectName = The name to give the new Game Object associated with this binding.

Container.Bind<Foo>().FromComponentInNewPrefabResource("Some/Path/Foo").WithGameObjectName("Foo1");
Container.Bind<Foo>().FromNewComponentOnNewGameObject().WithGameObjectName("Foo1");

UnderTransformGroup(string) = The name of the transform group to place the new game object under. This is especially useful for factories, which can be used to create many copies of a prefab, so it can be nice to have them automatically grouped together within the scene hierarchy.
```
Container.BindFactory<Bullet, Bullet.Factory>()
    .FromComponentInNewPrefab(BulletPrefab)
    .UnderTransformGroup("Bullets");
```

UnderTransform(Transform) = The actual transform to place the new game object under.

Container.BindFactory<Bullet, Bullet.Factory>()
    .FromComponentInNewPrefab(BulletPrefab)
    .UnderTransform(BulletTransform);

UnderTransform(Method) = A method to provide the transform to use.

Container.BindFactory<Foo, Foo.Factory>()
    .FromComponentInNewGameObject()
    .UnderTransform(GetParent);

Transform GetParent(InjectContext context)
{
    if (context.ObjectInstance is Component)
    {
        return ((Component)context.ObjectInstance).transform;
    }

    return null;
}

This example will automatically parent the Foo GameObject underneath the game object that it is being injected into, unless the injected object is not a MonoBehaviour in which case it will leave Foo at the root of the scene hierarchy.

Optional Binding

You can declare some dependencies as optional as follows:

public class Bar
{
    public Bar(
        [InjectOptional]
        IFoo foo)
    {
        ...
    }
}
...

// You can comment this out and it will still work
Container.Bind<IFoo>().AsSingle();

Or, if you have an identifier as well:

public class Bar
{
    public Bar(
        [Inject(Optional = true, Id = "foo1")]
        IFoo foo)
    {
        ...
    }
}

If an optional dependency is not bound in any installers, then it will be injected as null.

If the dependency is a primitive type (eg. int, float, struct) then it will be injected with its default value (eg. 0 for ints).

You may also assign an explicit default using the standard C# way such as:

public class Bar
{
    public Bar(int foo = 5)
    {
        ...
    }
}
...

// Can comment this out and 5 will be used instead
Container.BindInstance(1);

Note also that the [InjectOptional] is not necessary in this case, since it's already implied by the default value.

Alternatively, you can define the primitive parameter as nullable, and perform logic depending on whether it is supplied or not, such as:

public class Bar
{
    int _foo;

    public Bar(
        [InjectOptional]
        int? foo)
    {
        if (foo == null)
        {
            // Use 5 if unspecified
            _foo = 5;
        }
        else
        {
            _foo = foo.Value;
        }
    }
}

...

// Can comment this out and it will use 5 instead
Container.BindInstance(1);

Conditional Bindings

< 10000 svg class="octicon octicon-link" viewBox="0 0 16 16" version="1.1" width="16" height="16" aria-hidden="true">

In many cases you will want to restrict where a given dependency is injected. You can do this using the following syntax:

Container.Bind<IFoo>().To<Foo1>().AsSingle().WhenInjectedInto<Bar1>();
Container.Bind<IFoo>().To<Foo2>().AsSingle().WhenInjectedInto<Bar2>();

Note that WhenInjectedInto is simple shorthand for the following, which uses the more general When() method:

Container.Bind<IFoo>().To<Foo>().AsSingle().When(context => context.ObjectType == typeof(Bar));

The InjectContext class (which is passed as the context parameter above) contains the following information that you can use in your conditional:

Type ObjectType - The type of the newly instantiated object, which we are injecting dependencies into. Note that this is null for root calls to Resolve<> or Instantiate<>
object ObjectInstance - The newly instantiated instance that is having its dependencies filled. Note that this is only available when injecting fields or into [Inject] methods and null for constructor parameters
string Identifier - This will be null in most cases and set to whatever is given as a parameter to the [Inject] attribute. For example, [Inject(Id = "foo")] _foo will result in Identifier being equal to the string "foo".
object ConcreteIdentifier - This will be null in most cases and set to whatever is given as the identifier in the WithConcreteIdentifier bind method
string MemberName - The name of the field or parameter that we are injecting into. This can be used, for example, in the case where you have multiple constructor parameters that are strings. However, using the parameter or field name can be error prone since other programmers may refactor it to use a different name. In many cases it's better to use an explicit identifier
Type MemberType - The type of the field or parameter that we are injecting into.
InjectContext ParentContext - This contains information on the entire object graph that precedes the current class being created. For example, dependency A might be created, which requires an instance of B, which requires an instance of C. You could use this field to inject different values into C, based on some condition about A. This can be used to create very complex conditions using any combination of parent context information. Note also that ParentContext.MemberType is not necessarily the same as ObjectType, since the ObjectType could be a derived type from ParentContext.MemberType
bool Optional - True if the [InjectOptional] parameter is declared on the field being injected

List Bindings

When Zenject finds multiple bindings for the same type, it interprets that to be a list. So, in the example code below, Bar would get a list containing a new instance of Foo1, Foo2, and Foo3:

// In an installer somewhere
Container.Bind<IFoo>().To<Foo1>().AsSingle();
Container.Bind<IFoo>().To<Foo2>().AsSingle();
Container.Bind<IFoo>().To<Foo3>().AsSingle();

...

public class Bar
{
    public Bar(List<IFoo> foos)
    {
    }
}

The order of the list will be the same as the order in which they were added with a Bind method. The only exception is when you use subcontainers, since in that case the list will be ordered first by the associated subcontainer, with the first set of instances taken from the bottom most subcontainer, and then the parent, then the grandparent, etc.

Global Bindings Using Project Context

This all works great for each individual scene, but what if you have dependencies that you wish to persist permanently across all scenes? In Zenject you can do this by adding installers to a ProjectContext object.

To do this, first you need to create a prefab for the ProjectContext, and then you can add installers to it. You can do this most easily by selecting the menu item Edit -> Zenject -> Create Project Context. You should then see a new asset in the folder Assets/Resources called 'ProjectContext'. Alternatively, you can right click somewhere in Projects tab and select Create -> Zenject -> ProjectContext.

If you click on this it will appear nearly identically to the inspector for SceneContext. The easiest way to configure this prefab is to temporarily add it to your scene, add Installers to it, then click "Apply" to save it back to the prefab before deleting it from your scene. In addition to installers, you can also add your own custom MonoBehaviour classes to the ProjectContext object directly.

Then, when you start any scene that contains a SceneContext, your ProjectContext object will always be initialized first. All the installers you add here will be executed and the bindings that you add within them will be available for use in all scenes within your project. The ProjectContext game object is set as DontDestroyOnLoad so it will not be destroyed when changing scenes.

Note also that this only occurs once. If you load another scene from the first scene, your ProjectContext will not be called again and the bindings that it added previously will persist into the new scene. You can declare ITickable / IInitializable / IDisposable objects in your project context installers in the same way you do for your scene installers with the result being that IInitializable.Initialize is called only once across each play session and IDisposable.Dispose is only called once the application is fully stopped.

The reason that all the bindings you add to a global installer are available for any classes within each individual scene, is because the Container in each scene uses the ProjectContext Container as it's "parent". For more information on nested containers see here.

ProjectContext is a very convenient place to put objects that you want to persist across scenes. However, the fact that it's completely global to every scene can lead to some unintended behaviour. For example, this means that even if you write a simple test scene that uses Zenject, it will load the ProjectContext, which may not be what you want. To address these problems it is often better to use Scene Parenting instead, since that approach allows you to be selective in terms of which scenes inherit the same common bindings. See here for more details on that approach.

Note also that by default, any game objects that are instantiated inside ProjectContext will be parented underneath it by default. If you'd prefer that each newly instantiated object is instead placed at the root of the scene hierarchy (but still marked DontDestroyOnLoad) then you can change this by unchecking the flag 'Parent New Objects Under Context' in the inspector of ProjectContext.

Identifiers

You can also give an ID to your binding if you need to have distinct bindings for the same type, and you don't want it to just result in a List<>. For example:

Container.Bind<IFoo>().WithId("foo").To<Foo1>().AsSingle();
Container.Bind<IFoo>().To<Foo2>().AsSingle();

...

public class Bar1
{
    [Inject(Id = "foo")]
    IFoo _foo;
}

public class Bar2
{
    [Inject]
    IFoo _foo;
}

In this example, the Bar1 class will be given an instance of Foo1, and the Bar2 class will use the default version of IFoo which is bound to Foo2.

Note also that you can do the same thing for constructor/inject-method arguments as well:

public class Bar
{
    Foo _foo;

    public Bar(
        [Inject(Id = "foo")]
        Foo foo)
    {
    }
}

In many cases, the ID is created as a string, however you can actually use any type you like for this. For example, it's sometimes useful to use an enum instead:

enum Cameras
{
    Main,
    Player,
}

Container.Bind<Camera>().WithId(Cameras.Main).FromInstance(MyMainCamera);
Container.Bind<Camera>().WithId(Cameras.Player).FromInstance(MyPlayerCamera);

You can also use custom types, as long as they implement the Equals operator.

Non Generic Bindings

In some cases you may not know the exact type you want to bind at compile time. In these cases you can use the overload of the Bind method which takes a System.Type value instead of a generic parameter.

// These two lines will result in the same behaviour
Container.Bind(typeof(Foo));
Container.Bind<Foo>();

Note also that when using non generic bindings, you can pass multiple arguments:

Container.Bind(typeof(Foo), typeof(Bar), typeof(Qux)).AsSingle();

// The above line is equivalent to these three:
Container.Bind<Foo>().AsSingle();
Container.Bind<Bar>().AsSingle();
Container.Bind<Qux>().AsSingle();

The same goes for the To method:

Container.Bind<IFoo>().To(typeof(Foo), typeof(Bar)).AsSingle();

// The above line is equivalent to these two:
Container.Bind<IFoo>().To<Foo>().AsSingle();
Container.Bind<IFoo>().To<Bar>().AsSingle();

You can also do both:

Container.Bind(typeof(IFoo), typeof(IBar)).To(typeof(Foo1), typeof(Foo2)).AsSingle();

// The above line is equivalent to these:
Container.Bind<IFoo>().To<Foo>().AsSingle();
Container.Bind<IFoo>().To<Bar>().AsSingle();
Container.Bind<IBar>().To<Foo>().AsSingle();
Container.Bind<IBar>().To<Bar>().AsSingle();

This can be especially useful when you have a class that implements multiple interfaces:

Container.Bind(typeof(ITickable), typeof(IInitializable), typeof(IDisposable)).To<Foo>().AsSingle();

Though in this particular example there is already a built-in shortcut method for this:

Container.BindInterfacesTo<Foo>().AsSingle();

Convention Based Binding

Convention based binding can come in handy in any of the following scenarios:

You want to define a naming convention that determines how classes are bound to the container (eg. using a prefix, suffix, or regex)
You want to use custom attributes to determine how classes are bound to the container
You want to automatically bind all classes that implement a given interface within a given namespace or assembly

Using "convention over configuration" can allow you to define a framework that other programmers can use to quickly and easily get things done, instead of having to explicitly add every binding within installers. This is the philosophy that is followed by frameworks like Ruby on Rails, ASP.NET MVC, etc. Of course, there are both advantages and disadvantages to this approach.

They are specified in a similar way to Non Generic bindings, except instead of giving a list of types to the Bind() and To() methods, you describe the convention using a Fluent API. For example, to bind IFoo to every class that implements it in the entire codebase:

Container.Bind<IFoo>().To(x => x.AllTypes().DerivingFrom<IFoo>());

Note that you can use the same Fluent API in the Bind() method as well, and you can also use it in both Bind() and To() at the same time.

For more examples see the examples section below. The full format is as follows:

x.InitialList().Conditional().AssemblySources()

Where:

InitialList = The initial list of types to use for our binding. This list will be filtered by the given Conditionals. It can be one of the following (fairly self explanatory) methods:
1. AllTypes
2. AllNonAbstractClasses
3. AllAbstractClasses
4. AllInterfaces
5. AllClasses
Conditional = The filter to apply to the list of types given by InitialList. Note that you can chain as many of these together as you want, and they will all be applied to the initial list in sequence. It can be one of the following:
1. DerivingFrom - Only match types deriving from T
2. DerivingFromOrEqual - Only match types deriving from or equal to T
3. WithPrefix(value) - Only match types with names that start with value
4. WithSuffix(value) - Only match types with names that end with value
5. WithAttribute - Only match types that have the attribute [T] above their class declaration
6. WithoutAttribute - Only match types that do not have the attribute [T] above their class declaration
7. WithAttributeWhere(predicate) - Only match types that have the attribute [T] above their class declaration AND in which the given predicate returns true when passed the attribute. This is useful so you can use data given to the attribute to create bindings
8. InNamespace(value) - Only match types that are in the given namespace
9. InNamespaces(value1, value2, etc.) - Only match types that are in any of the given namespaces
10. MatchingRegex(pattern) - Only match types that match the given regular expression
11. Where(predicate) - Finally, you can also add any kind of conditional logic you want by passing in a predicate that takes a Type parameter
AssemblySources = The list of assemblies to search for types when populating InitialList. It can be one of the following:
1. FromAllAssemblies - Look up types in all loaded assemblies. This is the default when unspecified.
2. FromAssemblyContaining - Look up types in whatever assembly the type T is in
3. FromAssembliesContaining(type1, type2, ..) - Look up types in all assemblies that contains any of the given types
4. FromThisAssembly - Look up types only in the assembly in which you are calling this method
5. FromAssembly(assembly) - Look up types only in the given assembly
6. FromAssemblies(assembly1, assembly2, ...) - Look up types only in the given assemblies
7. FromAssembliesWhere(predicate) - Look up types in all assemblies that match the given predicate

Examples:

Note that you can chain together any combination of the below conditionals in the same binding. Also note that since we aren't specifying an assembly here, Zenject will search within all loaded assemblies.

Bind IFoo to every class that implements it in the entire codebase:
```
Container.Bind<IFoo>().To(x => x.AllTypes().DerivingFrom<IFoo>());
```
Note that this will also have the same result:
```
Container.Bind<IFoo>().To(x => x.AllNonAbstractTypes());
```
This is because Zenject will skip any bindings in which the concrete type does not actually derive from the base type. Also note that in this case we have to make sure we use AllNonAbstractTypes instead of just AllTypes, to ensure that we don't bind IFoo to itself

Bind an interface to all classes implementing it within a given namespace

Container.Bind<IFoo>().To(x => x.AllTypes().DerivingFrom<IFoo>().InNamespace("MyGame.Foos"));

Auto-bind IController every class that has the suffix "Controller" (as is done in ASP.NET MVC):

Container.Bind<IController>().To(x => x.AllNonAbstractTypes().WithSuffix("Controller"));

You could also do this using MatchingRegex:

Container.Bind<IController>().To(x => x.AllNonAbstractTypes().MatchingRegex("Controller$"));

Bind all types with the prefix "Widget" and inject into Foo

Container.Bind<object>().To(x => x.AllNonAbstractTypes().WithPrefix("Widget")).WhenInjectedInto<Foo>();

Auto-bind the interfaces that are used by every type in a given namespace
```
Container.Bind(x => x.AllInterfaces())
    .To(x => x.AllNonAbstractClasses().InNamespace("MyGame.Things"));
```
This is equivalent to calling Container.BindInterfacesTo<T>() for every type in the namespace "MyGame.Things". This works because, as touched on above, Zenject will skip any bindings in which the concrete type does not actually derive from the base type. So even though we are using AllInterfaces which matches every single interface in every single loaded assembly, this is ok because it will not try and bind an interface to a type that doesn't implement this interface.

Decorator Bindings

Decorator Bindings has it's own documentation here.

Scriptable Object Installer

Another alternative to deriving from MonoInstaller or Installer when implementing your own installers, is to derive from the ScriptableObjectInstaller class. This is most commonly used to store game settings. This approach has the following advantages:

Any changes you make to the properties of the installer will persist after you stop play mode. This can be very useful when tweaking runtime parameters. For other installer types as well as any MonoBehaviour's in your scene, any changes to the inspector properties at runtime will be undone when play mode is stopped. However, there is a 'gotcha' to be aware of: Any changes to these settings in code will also be saved persistently (unlike with settings on MonoInstaller's). So if you go this route you should treat all settings objects as read-only to avoid this from happening.
You can very easily swap out multiple instances of the same installer. For example, using the below example, you might have an instance of GameSettingsInstaller called GameSettingsEasy, and another one called GameSettingsHard, etc.

Example:

Open Unity
Right click somewhere in the Project tab and select Create -> Zenject -> ScriptableObjectInstaller
Name it GameSettingsInstaller
Right click again in the same location
Select the newly added menu item Create -> Installers -> GameSettingsInstaller
Following the approach to settings outlined here, you might then replace it with the following:

public class GameSettings : ScriptableObjectInstaller
{
    public Player.Settings Player;
    public SomethingElse.Settings SomethingElse;
    // ... etc.

    public override void InstallBindings()
    {
        Container.BindInstances(Player, SomethingElse, etc.);
    }
}

public class Player : ITickable
{
    readonly Settings _settings;
    Vector3 _position;

    public Player(Settings settings)
    {
        _settings = settings;
    }

    public void Tick()
    {
        _position += Vector3.forward * _settings.Speed;
    }

    [Serializable]
    public class Settings
    {
        public float Speed;
    }
}

Now, you should be able to run your game and adjust the Speed value that is on the GameSettingsInstaller asset at runtime, and have that change saved permanently

Runtime Parameters For Installers

Often when calling installers from other installers it is desirable to be able to pass parameters. You can do this by adding generic arguments to whichever installer base class you are using with the types for the runtime parameters. For example, when using a non-MonoBehaviour Installer:

public class FooInstaller : Installer<string, FooInstaller>
{
    string _value;

    public FooInstaller(string value)
    {
        _value = value;
    }

    public override void InstallBindings()
    {
        ...

        Container.BindInstance(_value).WhenInjectedInto<F
10000
oo>();
    }
}

public class MainInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        FooInstaller.Install(Container, "asdf");
    }
}

Or when using a MonoInstaller prefab:

public class FooInstaller : MonoInstaller<string, FooInstaller>
{
    string _value;

    // Note that in this case we can't use a constructor
    [Inject]
    public void Construct(string value)
    {
        _value = value;
    }

    public override void InstallBindings()
    {
        ...

        Container.BindInstance(_value).WhenInjectedInto<Foo>();
    }
}

public class MainInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        // For this to work, there must be a prefab with FooInstaller attached to it at
        // Resources/My/Custom/ResourcePath.prefab
        FooInstaller.InstallFromResource("My/Custom/ResourcePath", Container, "asdf")

        // If a resource path is not provided then it is assumed to exist at resource path
        // 'Resources/Installers/FooInstaller'
        // For example:
        // FooInstaller.InstallFromResource(Container, "asdf");
    }
}

ScriptableObjectInstaller works the same as MonoInstaller in this regard.

Using Zenject Outside Unity Or For DLLs

If you are building some code as DLLs and then including them in Unity, you can still add bindings for those classes inside your installers, with the only limitation being that you have to use constructor injection. If you want to use the other inject approaches such as member injection or method injection, then you can do that too, however in that case you will need to add a reference for your project to Zenject-Usage.dll which can be found in the Zenject\Source\Usage directory. This DLL also includes the standard interfaces such as ITickable, IInitializable, etc. so you can use those as well.

You can also use Zenject for non-unity projects by downloading Zenject-NonUnity.zip from the releases section

Finally, if you are attempting to run unit tests outside of Unity using the Unity generated solution, you might encounter run time errors in the Zenject code when it attempts to access the Unity API. You can disable this behaviour by adding a define for ZEN_TESTS_OUTSIDE_UNITY in the generated solution.

Zenject Settings

A lot of the default behaviour in Zenject can be customized via a settings property on the ProjectContext. This includes the following:

Validation Error Response - This value controls the behaviour that is triggered when zenject encounters a validation error. It can be set to either 'Log' or 'Throw'. The difference here is that when set to 'Log' there will be multiple validation errors printed every time validation is run, whereas if set to 'Throw' only the first validation error will be output to the console. When unset the default value is 'Log'. 'Throw' is also sometimes useful if running validation inside unit tests.
Validation Root Resolve Method - When validation is triggered for a given scene, the DiContainer will do a 'dry run' and pretend to instantiate the entire object graph as defined by the installers in the scene. However, by default it will only validate the 'roots' of the object graph - that is, the 'NonLazy' bindings or the bindings which are injected into the 'NonLazy' bindings. As an option, you can change this behaviour to 'All' which will validate all bindings, even those that are not currently used.
Display Warning When Resolving During Install - This value will control whether a warning is issued to the console when either a Resolve or an Instantiate is triggered during the install phase which looks like this:

Zenject Warning: It is bad practice to call Inject/Resolve/Instantiate before all the Installers have completed!  This is important to ensure that all bindings have properly been installed in case they are needed when injecting/instantiating/resolving.  Detected when operating on type 'Foo'.

So if you often encounter this warning and are aware of the implications of what you're doing then you might set this value to false to suppress it.

Ensure Deterministic Destruction Order On Application Quit - When set to true, this will ensure that all GameObject's and IDisposables are destroyed in a predictable order when the application is closed. By default it is set to false, because there are some undesirable implications to enabling this feature as described in this section.

These settings can also be configured on a per DiContainer basis by changing the DiContainer.Settings property. Changing this property will affect the given container as well as any subcontainers.

There are also settings for the signals system which are documented here.

Signals

See here.

Factories: Creating Objects Dynamically

See here.

Memory Pools

See here.

Update / Initialization Order

In many cases, especially for small projects, the order that classes update or initialize in does not matter. However, in larger projects update or initialization order can become an issue. This can especially be an issue in Unity, since it is often difficult to predict in what order the Start(), Awake(), or Update() methods will be called in. Unfortunately, Unity does not have an easy way to control this (besides in Edit -> Project Settings -> Script Execution Order, though that can be awkward to use)

In Zenject, by default, ITickables and IInitializables are called in the order that they are added, however for cases where the update or initialization order does matter, there is another way that is sometimes better: By specifying their priorities explicitly in the installer. For example, in the sample project you can find this code in the scene installer:

public class AsteroidsInstaller : MonoInstaller
{
    ...

    void InitExecutionOrder()
    {
        // In many cases you don't need to worry about execution order,
        // however sometimes it can be important
        // If for example we wanted to ensure that AsteroidManager.Initialize
        // always gets called before GameController.Initialize (and similarly for Tick)
        // Then we could do the following:
        Container.BindExecutionOrder<AsteroidManager>(-10);
        Container.BindExecutionOrder<GameController>(-20);

        // Note that they will be disposed of in the reverse order given here
    }

    ...

    public override void InstallBindings()
    {
        ...
        InitExecutionOrder();
        ...
    }

}

This way, you won't hit a wall at the end of the project due to some unforeseen order-dependency.

Note here that the value given to BindExecutionOrder will apply to ITickable / IInitializable and IDisposable (with the order reversed for IDisposable's).

You can also assign priorities for each specific interface separately like this:

Container.BindInitializableExecutionOrder<Foo>(-10);
Container.BindInitializableExecutionOrder<Bar>(-20);

Container.BindTickableExecutionOrder<Foo>(10);
Container.BindTickableExecutionOrder<Bar>(-80);

Any ITickables, IInitializables, or IDisposables that are not assigned a priority are automatically given the priority of zero. This allows you to have classes with explicit priorities executed either before or after the unspecified classes. For example, the above code would result in Foo.Initialize being called before Bar.Initialize.

Zenject Order Of Operations

What follows below is a more detailed view of what happens when running a scene that uses Zenject. This can be useful to know to fully understand exactly how Zenject works.

Unity Awake() phase begins
- SceneContext.Awake() method is called. This should always be the first thing executed in your scene. It should work this way by default (see here if you are noticing otherwise).
- Project Context is initialized. Note that this only happens once per play session. If a previous scene already initialized the ProjectContext, then this step is skipped
  - All injectable MonoBehaviour's on the ProjectContext prefab are passed to the container via DiContainer.QueueForInject
  - ProjectContext iterates through all the Installers that have been added to its prefab via the Unity Inspector, runs injects on them, then calls InstallBindings() on each. Each Installer calls some number of Bind methods on the DiContainer.
  - ProjectContext then constructs all the non-lazy root objects, which includes any classes that derive from ITickable / IInitializable or IDisposable, as well as those classes that are added with a NonLazy() binding.
  - All instances that were added via DiContainer.QueueForInject are injected
- Scene Context is initialized.
  - All injectable MonoBehaviour's in the entire scene are passed to the SceneContext container via DiContainer.QueueForInject
  - SceneContext iterates through all the Installers that have been added to it via the Unity Inspector, runs injects on them, then calls InstallBindings() on each. Each Installer calls some number of Bind<> methods on the DiContainer.
  - SceneContext then constructs all the non-lazy root objects, which includes any classes that derive from ITickable / IInitializable or IDisposable, as well as those classes that are added with a NonLazy() binding.
  - All instances that were added via DiContainer.QueueForInject are injected
- If any required dependencies cannot be resolved, a ZenjectResolveException is thrown
- All other MonoBehaviour's in the scene have their Awake() method called
Unity Start() phase begins
- ProjectKernel.Start() method is called. This will trigger the Initialize() method on all IInitializable objects in the order specified in the ProjectContext installers.
- SceneKernel.Start() method is called. This will trigger the Initialize() method on all IInitializable objects in the order specified in the SceneContext installers.
- All other MonoBehaviour's in your scene have their Start() method called
Unity Update() phase begins
- ProjectKernel.Update() is called, which results in Tick() being called for all ITickable objects (in the order specified in the ProjectContext installers)
- SceneKernel.Update() is called, which results in Tick() being called for all ITickable objects (in the order specified in the SceneContext installers)
- All other MonoBehaviour's in your scene have their Update() method called
These same steps repeated for LateUpdate and ILateTickable
At the same time, These same steps are repeated for FixedUpdate according to the physics timestep
Unity scene is unloaded
- Dispose() is called on all objects mapped to IDisposable within all the GameObjectContext's (see here for details)
- Dispose() is called on all objects mapped to IDisposable within the SceneContext installers (see here for details)
App is exitted
- Dispose() is called on all objects mapped to IDisposable within the ProjectContext installers (see here for details)

Injecting data across scenes

In some cases it's useful to pass arguments from one scene to another. The way Unity allows us to do this by default is fairly awkward. Your options are to create a persistent GameObject and call DontDestroyOnLoad() to keep it alive when changing scenes, or use global static classes to temporarily store the data.

Let's pretend you want to specify a 'level' string to the next scene. You have the following class that requires the input:

public class LevelHandler : IInitializable
{
    readonly string _startLevel;

    public LevelHandler(
        [InjectOptional]
        string startLevel)
    {
        if (startLevel == null)
        {
            _startLevel = "default_level";
        }
        else
        {
            _startLevel = startLevel;
        }
    }

    public void Initialize()
    {
        ...
        [Load level]
        ...
    }
}

You can load the scene containing LevelHandler and specify a particular level by using the following syntax:

public class Foo
{
    readonly ZenjectSceneLoader _sceneLoader;

    public Foo(ZenjectSceneLoader sceneLoader)
    {
        _sceneLoader = sceneLoader;
    }

    public void AdvanceScene()
    {
        _sceneLoader.LoadScene("NameOfSceneToLoad", LoadSceneMode.Single, (container) =>
            {
                container.BindInstance("custom_level").WhenInjectedInto<LevelHandler>();
            });
    }
}

The bindings that we add here inside the lambda will be added to the container as if they were inside an installer in the new scene.

Note that you can still run the scene directly, in which case it will default to using "default_level". This is possible because we are using the InjectOptional flag.

An alternative and arguably cleaner way to do this would be to customize the installer itself rather than the LevelHandler class. In this case we can write our LevelHandler class like this (without the [InjectOptional] flag).

public class LevelHandler : IInitializable
{
    readonly string _startLevel;

    public LevelHandler(string startLevel)
    {
        _startLevel = startLevel;
    }

    public void Initialize()
    {
        ...
        [Load level]
        ...
    }
}

Then, in the installer for our scene we can include the following:

public class GameInstaller : Installer
{
    [InjectOptional]
    public string LevelName = "default_level";

    ...

    public override void InstallBindings()
    {
        ...
        Container.BindInstance(LevelName).WhenInjectedInto<LevelHandler>();
        ...
    }
}

Then, instead of injecting directly into the LevelHandler we can inject into the installer instead.

public class Foo
{
    readonly ZenjectSceneLoader _sceneLoader;

    public Foo(ZenjectSceneLoader sceneLoader)
    {
        _sceneLoader = sceneLoader;
    }

    public void AdvanceScene()
    {
        _sceneLoader.LoadScene("NameOfSceneToLoad", (container) =>
            {
                container.BindInstance("custom_level").WhenInjectedInto<GameInstaller>();
            });
    }
}

The ZenjectSceneLoader class also allows for more complex scenarios, such as loading a scene as a "child" of the current scene, which would cause the new scene to inherit all the dependencies in the current scene. However, it is often better to use Scene Contract Names for this instead.

Scene Parenting Using Contract Names

Putting bindings inside ProjectContext is a fast and easy way to add common long-lasting dependencies that are shared across scenes. However, in many cases you have bindings that you only want to be shared between specific scenes, so using ProjectContext doesn't work since in that case, the bindings we add there are global to every single scene in our entire project.

As an example, let's pretend that we are working on a spaceship game, and we want to create one scene to serve as the environment (involving planets, asteroids, stars, etc.) and we want to create another scene to represent the ship that the player is in. We also want all the classes in the ship scene to be able to reference bindings declared in the environment scene. Also, we want to be able to define multiple different versions of both the ship scene and the environment scene. To achieve all this, we will use a Zenject feature called 'Scene Contract Names'.

We will start by using Unity's support for multi-scene editting, and dragging both our environment scene and our ship scene into the Scene Hierarchy tab. Then we will select the SceneContext in the environment scene and add a 'Contract Name'. Let's call it 'Environment'. Then all we have to do now is select the SceneContext inside the ship scene and set its 'Parent Contract Name' to the same value ('Environment'). Now if we press play, all the classes in the ship scene can access the declared bindings in the environment scene.

The reason we use a name field here instead of explicitly using the scene name is to support swapping out the various environment scenes for different implementations. In this example, we might define several different environments, all using the same Contract Name 'Environment', so that we can easily mix and match them with different ship scenes just by dragging the scenes we want into the scene hierarchy then hitting play.

It is called 'Contract Name' because all the environment scenes will be expected to follow a certain 'contract' by the ship scenes. For example, the ship scenes might require that regardless of which environment scene was loaded, there is a binding for 'AsteroidManager' containing the list of asteroids that the ship must avoid.

Note that you do not need to load the environment scene and the ship scene at the same time for this to work. For example, you might want to have a menu embedded inside the environment to allow the user to choose their ship before starting. So you could create a menu scene and load that after the environment scene. Then once the user chooses their ship, you could load the associated ship scene by calling the unity method SceneManager.LoadScene (making sure to use LoadSceneMode.Additive).

Also note that the Validate command can be used to quickly verify the different multi-scene setups. If you find that scenes are unloaded when you do this see here.

Also, I should mention that Unity currently doesn't have a built-in way to save and restore multi-scene setups. We use a simple editor script for this that you can find here if interested.

Default Scene Parents

One drawback to using scene parent contract names instead of ProjectContext to add bindings shared across scenes, is that you always have to remember to configure the scene hierarchy inside Unity to contain the correct scenes before running it. You can't simply open different scenes and hit Play like you can with ProjectContext. So to address this, zenject allows specifying a default value for the different contract names in cases where an existing value is not already loaded.

To take our example from above, we'd like to be able to open Ship scene and immediately hit play without needing to place an environment scene above that first. To do this, right click in your Projects tab and select Create -> Zenject -> Default Scene Contract Config. Note that you will have to do this by right clicking on a folder named Resources for this to work.

After adding this you can click on the ZenjectDefaultSceneContractConfig object and add any number of defaults by typing in contract names and then dragging in scene files from the Project tab into the Scene property. After doing this, you should be able to directly run the Ship scene and the default environment scene will automatically be loaded above that.

Note that the default scene for a given contract will also be used when using scene decorators

Note that this is an editor only feature. The default contract names will not be used in builds. In those cases you will have to explicitly load the correct parent scenes yourself in code.

ZenAutoInjecter

As explained in the section on factories, any object that you create dynamically needs to be created through zenject in order for it to be injected. You cannot simply execute GameObject.Instantiate(prefab), or call new Foo().

However, this is sometimes problematic especially when using other third party libraries. For example, some networking libraries work by automatically instantiating prefabs to sync state across multiple clients. And it is still desirable in these cases to execute zenject injection.

So to solve these cases, we've added a helper MonoBehaviour that you can add to these kinds of objects called ZenAutoInjecter. If you add this MonoBehaviour to your prefab, then you should be able to call GameObject.Instantiate and injection should happen automatically.

After adding this component you might notice that there is a field on it called 'Container Source'. This value will determine how it calculates which container to use for the injection and has the following options:

SearchInHiearchy - This will find the container to use by searching the scene hierarchy where the prefab is instantiated. So if the prefab is instantiated underneath a GameObjectContext, it will use the container associated with the GameObjectContext. If it is instantiated underneath a DontDestroyOnLoad object, then it will use the ProjectContext container. And otherwise will use the SceneContext container for whatever scene it is instantiated in.
SceneContext - Don't bother searching the hierarchy and instead just always use the current SceneContext container.
ProjectContext - Don't bother searching the hierarchy and instead just always use the ProjectContext container.

Scene Decorators

Scene Decorators offer another approach to using multiple scenes together with zenject in addition to scene parenting described above. The difference is that with scene decorators, the multiple scenes in question will all share the same Container and therefore all scenes can access bindings in all other scenes (unlike with scene parenting where only the child can access the parent bindings and not vice versa).

Another way to think about scene decorators is that it is a more advanced way doing the process described for injecting data across scenes. That is, they can be used to add behaviour to another scene without actually changing the installers in that scene.

Usually, when you want to customize different behaviour for a given scene depending on some conditions, you would use boolean or enum properties on MonoInstallers, which would then be used to add different bindings depending on the values set. However, the scene decorator approach can be cleaner sometimes because it doesn't involve changing the main scene.

For example, let's say we want to add some special keyboard shortcuts to your main production scene for testing purposes. In order to do this using decorators, you would do the following:

Open the main production scene
Right click on the far right menu beside the scene name within the scene hierarchy and select Add New Scene
Drag the scene so it's above the main scene
Right Click inside the new scene and select Zenject -> Decorator Context
Select the Decorator Context and set the 'Decorated Contract Name' field to 'Main'
Select the SceneContext in the main scene and add a contract name with the same value ('Main')
Create a new C# script with the following contents, then add this MonoBehaviour to your decorator scene as a gameObject, then drag it to the Installers property of SceneDecoratorContext

public class ExampleDecoratorInstaller : MonoInstaller
{
    public override void InstallBindings()
    {
        Container.Bind<ITickable>().To<TestHotKeysAdder>().AsSingle();
    }
}

public class TestHotKeysAdder : ITickable
{
    public void Tick()
    {
        if (Input.GetKeyDown(KeyCode.Space))
        {
            Debug.Log("Hotkey triggered!");
        }
    }
}

Note the following:

If you run your scene it should now behave exactly like the main scene except with the added functionality in your decorator installer. Also note that while not shown here, both scenes can access each other's bindings as if everything was in the same scene.
The Validate command (CTRL+ALT+V) can be used to quickly verify the different multi-scene setups. If you find that scenes are unloaded when you do this see here.
Decorator scenes must be loaded before the scenes that they are decorating.
Unity currently doesn't have a built-in way to save and restore multi-scene setups. We use a simple editor script for this that you can find here if interested.
Finally, if you want to save yourself some time you could add a default scene for the contract name that you are using above

Sub-Containers And Facades

See here.

Writing Automated Unit Tests / Integration Tests

See here.

Philosophy Of Zenject

One thing that is helpful to be aware of in terms of understanding the design of Zenject is that unlike many other frameworks it is not opinionated. Many frameworks, such as ROR, ECS, ASP.NET MVC, etc. make rigid design choices for you that you have to follow. The only assumption that Zenject makes is that you are writing object oriented code and otherwise, how you design your code is entirely up to you.

In my view, dependency injection is pretty fundamental to object oriented programming. And without a dependency injection framework, the composition root quickly becomes a headache to maintain. Therefore dependency injection frameworks are fairly fundamental as well.

And that's all Zenject strives to be - a dependency injection framework that targets Unity. There are certainly quite a few features in Zenject, but they are all optional. If you want, you can follow traditional unity development and use MonoBehaviours for every class, with the one exception that you use [Inject] instead of [SerializeField]. Or you can abandon MonoBehaviours's completely and use the included interfaces such as ITickable and IInitializable. It is up to you which design you want to use.

Of course, using a DI framework has some disadvantages compared to more rigid frameworks. The main drawback is that it can be more challenging for new developers to get up and running quickly in a code base, because they need to understand the specific architecture chosen by the previous developers. Whereas with rigid frameworks developers are given a very clear pathway to follow and so can be productive more quickly. It is more difficult to make huge design mistakes when using a rigid framework. However, with this rigidity also comes limitations, because whatever design decisions that are enforced by the framework might not necessarily be ideal for every single problem.

Just-In-Time Resolving Using LazyInject<>

In some cases it can be useful to delay the creation of certain dependencies until after startup. You can use the LazyInject<> construct for this.

For example, let's imagine a scenario like this:

< D7AE div class="highlight highlight-source-cs notranslate position-relative overflow-auto" dir="auto" data-snippet-clipboard-copy-content="public class Foo { public void Run() { ... } } public class Bar { Foo _foo; public Bar(Foo foo) { _foo = foo; } public void Run() { _foo.Run(); } } public class TestInstaller : MonoInstaller<TestInstaller> { public override void InstallBindings() { Container.Bind<Foo>().AsSingle(); Container.Bind<Bar>().AsSingle(); } }">

public class Foo
{
    public void Run()
    {
        ...
    }
}

public class Bar
{
    Foo _foo;

    public Bar(Foo foo)
    {
        _foo = foo;
    }

    public void Run()
    {
        _foo.Run();
    }
}

public class TestInstaller : MonoInstaller<TestInstaller>
{
    public override void InstallBindings()
    {
        Container.Bind<Foo>().AsSingle();
        Container.Bind<Bar>().AsSingle();
    }
}

Name		Name	Last commit message	Last commit date
Latest commit History 1,921 Commits
.github		.github
AssemblyBuild/Zenject-usage		AssemblyBuild/Zenject-usage
Build		Build
Documentation		Documentation
NonUnityBuild		NonUnityBuild
UnityProject		UnityProject
.gitignore		.gitignore
.releaserc.yml		.releaserc.yml
CONTRIBUTING.md		CONTRIBUTING.md
License.md		License.md
README.md		README.md

License

erinloy/Extenject

Folders and files

Latest commit

History

Repository files navigation

Extenject: extensions, bug fixes and updates for Zenject

Table Of Contents

Lawsuit

Support

Introduction

Features

Installation

History

Documentation

What is Dependency Injection?

Theory

Misconceptions

Introduction to Zenject API

Hello World Example

Injection

Register Mappings to the DI Container

Binding

Construction Methods

Installers

Using Non-MonoBehaviour Classes

ITickable

IInitializable

IDisposable

BindInterfacesTo and BindInterfacesAndSelfTo

Using the Unity Inspector To Configure Settings

Object Graph Validation

Scene Bindings

General Guidelines / Recommendations / Gotchas / Tips and Tricks

Further Reading

Advanced

Binding

Game Object Bind Methods

Optional Binding

Conditional Bindings

List Bindings

Global Bindings Using Project Context

Identifiers

Non Generic Bindings

Convention Based Binding

Where:

Examples:

Decorator Bindings

Scriptable Object Installer

Runtime Parameters For Installers

Using Zenject Outside Unity Or For DLLs

Zenject Settings

Signals

Factories: Creating Objects Dynamically

Memory Pools

Update / Initialization Order

Zenject Order Of Operations

Injecting data across scenes

Scene Parenting Using Contract Names

Default Scene Parents

ZenAutoInjecter

Scene Decorators

Sub-Containers And Facades

Writing Automated Unit Tests / Integration Tests

Philosophy Of Zenject

Just-In-Time Resolving Using LazyInject<>

Open Generic Types

Notes About Destruction/Dispose Order

UniRx Integration

Auto-Mocking using Moq

Creating Unity EditorWindow's with Zenject

Optimization Recommendations/Notes

Reflection Baking

Baking External DLLs

Under the hood

Coverage Settings

Upgrade Guide for Zenject 6

DiContainer Methods

DiContainer.Instantiate

DiContainer.Bind

Packages