The mapDirectory() allows you to leverage closures or lambdas to influence and filter mappings. The arguments are filter to add a filter that MUST return boolean in order to process the mapping and influence that can influence the created mapping with any custom bindings.

// influence only certain components to be singleton
mapDirectory(packagePath="coldbox.testing.testModel.ioc", influence=function(binder, path){
    if( findNoCase( "simple", arguments.path) ){
        arguments.binder.asSingleton();
    }
});

// filter some components from registration
mapDirectory(packagePath="coldbox.testing.testModel.ioc", filter=function(path){
    return ( findNoCase( "simple", arguments.path ) ? false : true );
});

The dependencies DSL methods are mostly used to define dependencies and also to activate advanced features on target objects, such as runtime mixins, virtual inheritance, etc.

Note Please note that you can concatenate more than one of these methods calls to dictate multiple constructor arguments, setter methods, cf properties, and more.

The mapping DSL is the way to configure object mappings in WireBox that will represent objects, factories or providers. All mappings DSL methods return back an instance of the binder so you can concatenate methods to create readable execution chains.

The chains are divided into three types:

1. Initiators - Start the mapping DSL process

2. Modifiers - Can modify a mapping with metadata and behavior

3. Destinations - Tells the binder to what object or behavior we should map to.

Since version 5.5.0 all mappings in WireBox will only be processed when they are requested for the very first time. This is to enhance performance and increase startup times. Processing means that the object's and its inheritance trail are inspected for metadata, which can be a very time consuming process.

process()

However, you can explicitly process a mapping right after mapping it via the binder's process() method.

That's it! If you call the process() method right after a mapping, it will be automatically processed. This means all annotations will be inspected.

mapDirectory( process=true )

If you are mapping using mapDirectory() then you can pass the process argument to true and all mappings in that directory scan will be processed automatically.

The mapping destinations tell WireBox what type of object you are mapping to. You will usually use these methods by concatenating map() or with() initiator calls:

Here are some examples:

The next step in our mapping DSL excursion is to learn about how WireBox will persist these object mappings into WireBox scopes. By default (as we have seen), all object mappings are transient objects and they belong to a scope type called NOSCOPE.

However, we need to specifically tell WireBox into what scope the declared mapped objects should be placed on in order for us to leverage caching, the singleton pattern, etc. This is accomplished by leveraging our persistence component annotations or the following methods if you prefer a non-annotation approach:

Note Please note that all WireBox configuration binders have two public properties:

These classes have on themselves several public properties that are a cool shorthand way to link to construction types or persistence scopes

So just remember that these persistence DSL methods are not mandatory. If you are an annotations kinda developer, then you can easily add these persistence annotations to your classes.

You can store a-la-carte attributes in a specific mapping so it can be retrieved at a later time by either an AOP aspect or Events. This is a great way to store custom metadata about an object so it can be read later for some meaningful purpose. Let's say you want to tag a mapping with a custom type that is not so easily determined from the object instance itself. You don't want to do all kinds of introspection in order to know what object you received in an aspect or an event.

This mapping declares that an object has some extra attributes that will be stored in the mapping, such as the location and if it belongs to a module. This is then incredibly useful when you have an attached listener to WireBox:

As you can see from this sample, the extra attributes are incredibly essential, as the listener just sends the target object. It would take lots of introspection and metadata inspections in order to determine certain metadata about an object. However, with the extra attributes, it is just a snap!

You can use our mapping DSL to register influence closures or lambdas on a per mapping basis. This will allow a developer to influence the requested instance of any object/data element and decorate objects or even return different objects.

This is similar to object providers but instead of overriding the ENTIRE creation process of the object like a provider does, the user might want to simply influence the creation of a normal mapping with some additional flair. This is accomplished via the withInfluence mapping DSL function. It receives a closure as an argument and the closure has the following signature:

Here is an example of adding some nice pizzazz to an object:

In this instance, the instance is already built and then passed into the closure for additional influence. Please note, that the object is returned from the closure. You can make this optional, but if something IS returned, it will override the instance which will allow a developer to replace or decorate the instance as they see fit.

Ok, now that we know how to configure WireBox, let's get into the fun stuff of object mapping. How do we do this? By using our DSL mapping initiators that tell WireBox how to start the object registration process. You will then concatenate the initiators with some DSL destinations methods, DI data, etc to tell WireBox all the information it might need to construct, wire and persist the object. Here are the DSL initiators: