In a JSF project, we wrote our own PartialViewContext to listen to some events fired by pages beans:
#RequestScoped
public class OurPartialViewContext extends PartialViewContextWrapper
{
...
// called by cdi
#SuppressWarnings("unused")
private void listenForUpdate(#Observes OurRefreshEvent event)
{
...
And we wrote the matching factory, injecting it:
public class OurPartialViewContextFactory extends PartialViewContextFactory
{
#Inject
private OurPartialViewContext customPartialViewContext;
...
Problem is that in the newest versions of JSF, the empty constructor for PartialViewContextWrapper is deprecated, asking us to use another constructor with the wrapped object in parameter.
Currently, our PartialViewContext needs to be tied to the request scope, in order to be modified during the request by the observed events and to be used by a custom PartialResponseWriter we also wrote.
So our PartialViewContext currently both:
must have an empty constructor, as it is a #RequestScoped bean;
should not have an empty constructor, as it is deprecated for PartialViewContextWrapper which it inherits from.
How could we find a solution there?
We tried to remove it from the scope and build it in the Factory with a simple new OurPartialViewContext(), but then the #Observes methods are never called.
You are required to pass the wrapped instance into the constructor and to use getWrapped() over all place in delegate methods. Otherwise your application will most probably not work when you install other JSF libraries which also ship with their own PartialViewContext implementation such as OmniFaces and PrimeFaces. You would be effectively completely skipping the functionality of their PartialViewContext implementation. This mistake was previously observed in too many custom implementations of factory-provided classes. Hence the urge to mark the default constructor as #Deprecated so that the developers are forced to use the proper design pattern.
Your specific issue can be solved by simply refactoring the listenForUpdate() method into a fullworthy request scoped CDI bean, which you then inject in the factory who in turn ultimately passes it into the constructor of your PartialViewContext implementation.
Thus, so:
#RequestScoped
public class OurEventObserver {
public void listenForUpdate(#Observes OurRefreshEvent event) {
// ...
}
}
public class OurPartialViewContextFactory extends PartialViewContextFactory {
#Inject
private OurEventObserver observer;
public OurPartialViewContextFactory(PartialViewContextFactory wrapped) {
super(wrapped);
}
#Override
public PartialViewContext getPartialViewContext(FacesContext context) {
PartialViewContext wrapped = getWrapped().getPartialViewContext(context);
return new OurPartialViewContext(wrapped, observer);
}
}
public class OurPartialViewContext extends PartialViewContextWrapper {
private OurEventObserver observer;
public OurPartialViewContext(PartialViewContext wrapped, OurEventObserver observer) {
super(wrapped);
this.observer = observer;
}
// ...
}
Inside any of the overridden methods of OurPartialViewContext you can simply access the state of the observer, provided that the listenForUpdate() modifies some instance variables representing the state.
Related
I have the following classes:
#Named
#ViewScoped
public class BaseClass {
private SomeDependency dep;
public BaseClass(){}
#Inject
public BaseClass(SomeDependency dep) {
this.dep = dep;
}
#PostConstruct
private void initialize() {
dep.doSomething(); // Point "A"
}
public String getProperty() {
return "BaseClass-Property";
}
#Specializes
public class SpecialClass extends BaseClass() {
#Override
public String getProperty() {
return "SpecialClass-Property";
}
}
Now in some .xhtml I have something like
<h:outputText value="#{baseClass.property}" />
This works fine without the SpecialClass. It breaks with a NullPointerException at Point "A" if I include the SpecialClass in the project.
Well, according to the Weld specification, this is more or less intended behavior:
When an enabled bean specializes another bean, the other bean is never
instantiated or called by the container.
Nevertheless, now I have to make sure that every #Specializes bean implements the complete constructor like
public SpecialClass() {}
#Inject
public SpecialClass(SomeDependency dep) { super(dep); }
which IMHO is kind of counter-intuitive and produces a lot of duplicated, boilerplate code, especially with something like 5-6 arguments for every constructor. Also, this is never caught when creating a new specialized bean since the project is always still compile clean.
Am I doing something wrong or is there no alternative to implementing the constructors over and over again?
BTW, I do use Constructor Injection to create easily testable classes where I can just use the constructor to "Inject" dummy implementations of the dependencies.
CDI 1.1 spec at section 4.3 says:
"The only way one bean can completely override a second bean at all
injection points is if it implements all the bean types and declares
all the qualifiers of the second bean."
Your base class is annotated with the Named qualifier and the specializing class is not. You should also mark it with Alternative and enable it in beans.xml. There's also the ViewScoped annotation. Unless it's the Omnifaces' ViewScoped, it looks like you're mixing up JSF managed beans with CDI beans.
I have some legacy code that put objects as http session attributes using code like this:
MyObject object = new MyObject();
Map<String, Object> sessionMap = FacesContext.getCurrentInstance().getExternalContext().getSessionMap();
sessionMap.put("attrname", object);
The old facelets accessed the code using
#ManagedProperty("#{attrname}")
private MyObject object;
Is there any way using CDI (#Inject) to inject this session attribute to a Bean?
In new code that uses CDI what's the better way to create and inject objects that need to be created in a controlled way.
Get hold of it in a session scoped managed bean with a #Produces#Named on the getter.
#SessionScoped
public class MyObjectProducer implements Serializable {
private MyObject myObject;
#Produces
#Named("attrname")
public MyObject getMyObject() {
return myObject;
}
public void setMyObject(MyObject myObject) {
this.myObject = myObject;
}
}
When you set it somehow via e.g. myObjectProducer.setMyObject(myObject) elsewhere (or perhaps a CDI #Observes event), then you can inject it anywhere using #Inject #Named.
#Inject
#Named("attrname")
private MyObject myObject;
And yes, it's still available via #{attrname} in EL the usual way. And no, it won't be auto-created when not set, it'll remain null until you actually set it as a property of the producer class.
Alternatively, if you really intend to keep the legacy way of setting the instance via ExternalContext#getSessionMap() (e.g. because it's third party and you can thus not change it), then you can alternatively also let the producer return it directly from the session map:
#SessionScoped
public class MyObjectProducer implements Serializable {
#Produces
#Named("attrname")
public MyObject getMyObject() {
return (MyObject) FacesContext.getCurrentInstance().getExternalContext().getSessionMap().get("attrname");
}
}
This however isn't guaranteed to work when injected in a non-JSF artifact, such as an arbitrary #WebServlet, as the FacesContext#getCurrentInstance() would obviously return null.
This is the way to go, for both non-jsf and jsf artifacts.
#Qualifier
#Retention(RUNTIME)
#Target({TYPE,METHOD,FIELD,PARAMETER});
public #interface SessionAttribute {
#NonBinding
String value() default "";
}
#ApplicationScope
public class SessionAttributeService {
//Dont worry, this is a proxy, and CDI will ensure that the right one called at the right time.
#Inject
private HttpServletRequest servletRequest;
#Produces
#SessionAttribute
#RequestScope
public String sessionAttribute(final InjectionPoint ip){
final SessionAttribute sa = ip.getAnnotated().getAnnotation(SessionAttribute.class);
final HttpSession session = servletRequest.getSession();
return session.getAttribute(sa.value());
}
}
And use case:
#RequestScope
public class MyServiceBean {
#Inject
#SessionAttribute("theAttribute")
private String attribute;
}
I'm looking for a minimal example on how to pass a method expression to a custom component. I tried the following, but the responsible setter of my component is never called.
Bean for my view:
public String bar(){
//do stuff
return "";
}
My view:
<my:comp foo="#{bean.bar}" />
Bean for my component:
private static final String FOO = "foo";
public void setFoo(MethodExpression me){
//never called
getStateHelper().put(FOO, me);
}
public MethodExpression getFoo(){
//actually gets called
return (MethodExpression) getStateHelper().get(FOO);
}
Inside my component renderer, I call component.getFoo() and get a NPE.
You need to implement the ActionSource2 interface:
#FacesComponent(MyComponent.COMPONENT_TYPE)
public class Mycomponent extends UIComponentBase implements ActionSource2 {
// ...
}
Or, easier, to extend from UICommand class (which are also used by <h:commandXxx> components) which has all of those ActionSource2 methods already implemented the right way so that you don't need to repeat the job:
#FacesComponent(MyComponent.COMPONENT_TYPE)
public class Mycomponent extends UICommand {
// ...
}
Either way, you can use action="#{some method expression}" and even attach actionlisteners on it. Note that you can't rename the attribute name action. It really has to be action.
An open source example of such a custom component is the OmniFaces <o:commandScript>. The source code is available here.
I want to access the property of a #SessionScoped bean in another bean using #ManagedProperty. In short, I want to access the name property of firstBean in secondBean.
#ManagedBean
#SessionScoped
public class FirstBean implements Serializable{
private String name;
//...other attributes
//...constructor
public String getSelectedModel() {
return selectedModel;
}
public void setSelectedModel(String selectedModel) {
this.selectedModel = selectedModel;
}
//other getters&setters
}
And second bean:
#ManagedBean
#SessionScoped
public class SecondBean implements Serializable{
#ManagedProperty(value="#{firstBean}")
private FirstBean firstBean
public SecondBean() {
System.out.println(firstBean.getName());
}
public IndexBean getFirstBean() {
return firstBean;
}
public void setFirstBean(FirstBean firstBean) {
this.firstBean = firstBean;
}
When I run this, I always get NullPointerException on System.out.println(firstBean.getName()); in the constructor of second bean, which seems to mean that I need to create a new instance of firstBean.
But strangely, when I commented out this line, I can do something like this with no errors, which means that firstBean is actually a property of secondBean.
<h:outputText value="#{secondBean.firstBean.name}" />
What's the problem here?
It's not possible to access an injected dependency in the constructor. You're basically expecting that Java is able to do something like this:
SecondBean secondBean; // Declare.
secondBean.firstBean = new FirstBean(); // Inject.
secondBean = new SecondBean(); // Construct.
It's absolutely not possible to set an instance variable if the instance is not constructed yet. Instead, it works as follows:
SecondBean secondBean; // Declare.
secondBean = new SecondBean(); // Construct.
secondBean.firstBean = new FirstBean(); // Inject.
Then, in order to perform business actions based on injected dependencies, use a method annotated with #PostConstruct. It will be invoked by the dependency injection manager directly after construction and dependency injection.
So, just replace
public SecondBean() {
System.out.println(firstBean.getName());
}
by
#PostConstruct
public void init() { // Note: method name is fully to your choice.
System.out.println(firstBean.getName());
}
I' have a question about initialization of List in the POJO as it follows the next code:
public class Person {
//other fields...
private List<String> friends=new ArrayList<>();
public List<String> getFriends() {
return friends;
}
public void setFriends(List<String> friends) {
this.friends = friends;
}
}
OR is it better like this and have initalization in other class(like for example Bean(JSF))
public class Person {
//other fields...
private List<String> friends;
public List<String> getFriends() {
return friends;
}
public void setFriends(List<String> friends) {
this.friends = friends;
}
}
So my question is what approach is better?
If it's a managed bean as you say, you should do this in a method annotated with #PostConstruct
public class Person {
private List<String> friends;
#PostConstruct
public void init(){
friends = new ArrayList<String>();
}
//getter and setter...
}
The practice of doing any initialization in the getter and setter is generally frowned upon within the context of JSF. See Why JSF calls getters multiple times
Also, per the API for #PostConstruct, the contract specifies safety features and guarantees that if an exception is thrown in a method annotated as such, the bean should not be put into service. There are no such guarantees on a plain constructor.
In a managed bean, injection happens immediately after construction. This means that any operations you're carrying out in the constructor cannot depend on any injected resources (via #ManagedProperty). Whereas in a #PostConstruct method, you'll have access to all the resources declared on the managed bean
EDIT: It's important to note that there can be only one #PostConstruct for any #ManagedBean, so all important initializations should happen in there.
It's also worthwhile to note that, while the #PostConstruct method is the ideal place to initialize a backing bean variable/List, there are implications regarding the scope of the managed bean
#RequestScoped: In a managed bean with this annotation, the method will be called per submit of the JSF view concerned. A #RequestScoped bean is destroyed and recreated with every request, The implication of this is that depending on your setup, the list initialized in the #PostConstruct may be reset to empty or default values during each request. Under certain circumstances, conversion errors may occur as a result of the re-initialization of the list mid-JSF request.
#ViewScoped: In a managed bean with this annotation, you're guaranteed to have the #PostConstruct method run once, if and only if you're dealing with the same instance of the #ViewScoped bean. If the viewscoped bean is destroyed and recreated, the #PostConstruct method will run again.
#SessionScoped: A bean with this annotation is created once and stays alive until the user's HTTP session ends. In this scenario, the #PostConstruct method is guaranteed to run once and only once until the bean is destroyed
See also
https://stackoverflow.com/a/3406631/1530938
I would suggest this:
public class Person {
//other fields...
private List<String> friends=new ArrayList<>();
// returns a copy to protect original list
public List<String> getFriends() {
Collections.unmodifiableList(new ArrayList<>(friends));
}
public void addFriend(String> friend) {
this.friends.add(friend);
}
public void addFriends(List<String> friends) {
this.friends.addAll(friends);
}
}
In my opinion it would be best to handle that in the constructors. If a default constructor is used, initialize the list in the constructor.
public Person() {
friends = new ArrayList<>();
}
If a constructor which accepts parameters is used, let the calling class pass in a list.
public Person(ArrayList<> friends) {
this.friends = friends;//friends
}
My suggestion, add a null check in the getter:
public class Person {
//other fields...
private List<String> friends;
public List<String> getFriends() {
if (this.friends == null) friends = new ArrayList<String>();
return friends;
}
}
But also notice I have omitted the setter. Instead, in any client code, call like this:
personInstance.getFriends().add("Some Item");
Or if you have a full list to add:
personInstance.getFriends().addAll(someStringCollection);
It depends. Usually first way preferable because you may want to add something to collection later. If you won't know was your collection initialized or not you must check it every time.