Implementing the Observer Pattern to get notified about every new Account - java

I am developing an application where I would like to use the observer pattern in the following way:
I have 2 classes:
public abstract class Storage<V>{
private Set<V> values;
private String filename;
protected Storage(String filename) throws ClassNotFoundException, IOException {
values = new HashSet<>();
this.filename = filename;
load();
}
...
public boolean add(V v) throws IllegalArgumentException {
if (values.contains(v))
throw new IllegalArgumentException("L'elemento è già presente");
return values.add(v);
}
...
}
Repository which is a class for saving a collection of Objects. below is a subclass that implements the singleton pattern (the others are practically the same, only the specified generic type changes)
public class AccountStorage extends Storage<Account>{
private static AccountStorage instance = null;
private AccountStorage(String filename) throws ClassNotFoundException, IOException {
super(filename);
}
public static synchronized AccountStorage getInstance() throws ClassNotFoundException, IOException {
if (instance == null) {
String savefile = "accounts.ob";
instance = new AccountStorage(savefile);
}
return instance;
}
after which I have a controller class (Controller for Spring MVC) which through a post request receives an Account in JSON format, deserializes it and adds it to the collection (Tremite the AccountStorage class) like this:
#PostMapping(value = "new/user", consumes = MediaType.APPLICATION_JSON_VALUE, produces = MediaType.APPLICATION_JSON_VALUE)
#ResponseBody
public ResponseEntity<String> newAccount(#RequestBody Account a) {
synchronized (accounts) {
try {
accounts.add(a);
// accounts.save()
} catch (IllegalArgumentException e) {
return new ResponseEntity<String>(e.getMessage(), HttpStatus.BAD_REQUEST);
} catch (IOException e) {
return new ResponseEntity<String>(e.getMessage(), HttpStatus.INTERNAL_SERVER_ERROR);
}
}
}
where accounts is: AccountStorage accounts = AccountStorage.getInstance();
I would like to make sure that, after each addition (or other methods that modify the collection) it is saved to file without calling the function affixed each time after the modification.
My idea is to use the Observer pattern. But I don't know which class must be an Observer and which Observable (assuming this approach is the correct solution).

The common practice for implementing the Observer pattern is to define an Observer interface (Listener) which will declare a general contact and each observer-implementation should provide an action which would be triggered whenever an event occurs.
A subject maintains a collection of observers (listeners), and exposes methods which allow to add and remove (subscribe/unsubscribe) an observer. Event-related behavior resides in the subject, and when a new event happens, every subscribed observer (i.e. each observer that is currently present in the collection) will be notified.
An event to which we are going to listen to is a case when a new Account gets added into an AccountStorage. And AccountStorage would be a subject. That implies that AccountStorage should hold a reference to a collection of observers, provide a functionality to subscribe/unsubscribe and override method add() of the Storage class in order to trigger all the observers when a new account will be added.
Why can't we add a collection of observers and all related functionality into the Storage class so that every implementation will inherit it? It's a valid question, the answer is that in such a scenario we can't be specific in regard to the nature of the event because we even don't know its type - method add(V) expects a mysterious V. Hence, the observer interface and its method would be faceless. It was the downside of the standard interfaces Observer and Observable that are deprecated since JDK version 9. Their names as well as the method-name update() tell nothing about an event that would be observed. It's only slightly better than define an interface MyInterface with a method myMethod() - no clue where you can use it and what actions should follow when myMethod() is fired.
It's a good practice when names of observers are descriptive, so that it's clear without looking at the code what they are meant to do. And it's not only related to the Observer pattern, it is a general practice which is called a self-documenting code.
Let's start by defining an observer interface, I'll call it listener just because AccountAddedListener sounds a bit smoothly, and it's quite common to use the terms listener and observer interchangeably.
public interface AccountAddedListener {
void onAccountAdded(Account account);
}
Now let's proceed with an implementation of the observer, let's say we need a notification manager:
public class NotificationManager implements AccountAddedListener {
#Override
public void onAccountAdded(Account account) {
// send a notification message
}
}
Now it's time to turn the AccountStorage into a subject. It should maintain a reference collection of observers, Set is a good choice because it'll not allow to add the same observer twice (which would be pointless) and is able to add and remove elements in a constant time.
Whenever a new account gets added, subject iterates over the collection of observers and invokes onAccountAdded() method on each of them.
We need to define a method to add a new observer, and it's also good practice to add another one to be able to unregister the observer when it's no longer needed.
public class AccountStorage extends Storage<Account> {
private Set<AccountAddedListener> listeners = new HashSet<>(); // collection of observers
#Override
public boolean add(Account account) throws IllegalArgumentException {
listeners.forEach(listener -> listener.onAccountAdded(account)); // notifying observers
return super.add(account);
}
public boolean registerAccountAddedListener(AccountAddedListener listener) {
return listeners.add(listener);
}
public boolean unregisterAccountAddedListener(AccountAddedListener listener) {
return listeners.remove(listener);
}
// all other functionality of the AccountStorage
}

Related

How to use non-keyed state with Kafka Consumer in Flink?

I'm trying to implement (just starting work with Java and Flink) a non-keyed state in KafkaConsumer object, since in this stage no keyBy() in called. This object is the front end and the first module to handle messages from Kafka.
SourceOutput is a proto file representing the message.
I have the KafkaConsumer object :
public class KafkaSourceFunction extends ProcessFunction<byte[], SourceOutput> implements Serializable
{
#Override
public void processElement(byte[] bytes, ProcessFunction<byte[], SourceOutput>.Context
context, Collector<SourceOutput> collector) throws Exception
{
// Here, I want to call to sorting method
collector.collect(output);
}
}
I have an object (KafkaSourceSort) that do all the sorting and should keep the unordered message in priorityQ in the state and also responsible to deliver the message if it comes in the right order thru the collector.
class SessionInfo
{
public PriorityQueue<SourceOutput> orderedMessages = null;
public void putMessage(SourceOutput Msg)
{
if(orderedMessages == null)
orderedMessages = new PriorityQueue<SourceOutput>(new SequenceComparator());
orderedMessages.add(Msg);
}
}
public class KafkaSourceState implements Serializable
{
public TreeMap<String, SessionInfo> Sessions = new TreeMap<>();
}
I read that I need to use a non-keyed state (ListState) which should contain a map of sessions while each session contains a priorityQ holding all messages related to this session.
I found an example so I implement this:
public class KafkaSourceSort implements SinkFunction<KafkaSourceSort>,
CheckpointedFunction
{
private transient ListState<KafkaSourceState> checkpointedState;
private KafkaSourceState state;
#Override
public void snapshotState(FunctionSnapshotContext functionSnapshotContext) throws Exception
{
checkpointedState.clear();
checkpointedState.add(state);
}
#Override
public void initializeState(FunctionInitializationContext context) throws Exception
{
ListStateDescriptor<KafkaSourceState> descriptor =
new ListStateDescriptor<KafkaSourceState>(
"KafkaSourceState",
TypeInformation.of(new TypeHint<KafkaSourceState>() {}));
checkpointedState = context.getOperatorStateStore().getListState(descriptor);
if (context.isRestored())
{
state = (KafkaSourceState) checkpointedState.get();
}
}
#Override
public void invoke(KafkaSourceState value, SinkFunction.Context contex) throws Exception
{
state = value;
// ...
}
}
I see that I need to implement an invoke message which probably will be called from processElement() but the signature of invoke() doesn't contain the collector and I don't understand how to do so or even if I did OK till now.
Please, a help will be appreciated.
Thanks.
A SinkFunction is a terminal node in the DAG that is your job graph. It doesn't have a Collector in its interface because it cannot emit anything downstream. It is expected to connect to an external service or data store and send data there.
If you share more about what you are trying to accomplish perhaps we can offer more assistance. There may be an easier way to go about this.

How to process Websocket messages from client in Java?

I am developing a client-server application in Java using Websocket. Currently, all the client messages are processed using switch-case as shown below.
#OnMessage
public String onMessage(String unscrambledWord, Session session) {
switch (unscrambledWord) {
case "start":
logger.info("Starting the game by sending first word");
String scrambledWord = WordRepository.getInstance().getRandomWord().getScrambledWord();
session.getUserProperties().put("scrambledWord", scrambledWord);
return scrambledWord;
case "quit":
logger.info("Quitting the game");
try {
session.close(new CloseReason(CloseCodes.NORMAL_CLOSURE, "Game finished"));
} catch (IOException e) {
throw new RuntimeException(e);
}
}
String scrambledWord = (String) session.getUserProperties().get("scrambledWord");
return checkLastWordAndSendANewWord(scrambledWord, unscrambledWord, session);
}
The server has to process more than 50 different requests from client and that results in more than 50 case statements. And in future, I expect it to grow. Is there any better way to process Websocket messages from client? Or, is this how it is usually done?
I read somewhere about the use of hashtable to avoid long switch-case scenario by mapping to function pointers. Is this possible in Java? Or, is there any better solutions?
Thanks.
After a bit of testing and study, I found two alternatives to avoid long switch case scenario.
Anonymous class method (Strategy pattern)
Reflection with Annotations
Using Anonymous Class
Anonymous class method is the norm and following code shows how to implement it. I used Runnable in this example. If more control is required, create a custom interface.
public class ClientMessageHandler {
private final HashMap<String, Runnable> taskList = new HashMap<>();
ClientMessageHandler() {
this.populateTaskList();
}
private void populateTaskList() {
// Populate the map with client request as key
// and the task performing objects as value
taskList.put("action1", new Runnable() {
#Override
public void run() {
// define the action to perform.
}
});
//Populate map with all the tasks
}
public void onMessageReceived(JSONObject clientRequest) throws JSONException {
Runnable taskToExecute = taskList.get(clientRequest.getString("task"));
if (taskToExecute == null)
return;
taskToExecute.run();
}
}
Major drawback of this method is object creation. Say, we have 100 different tasks to perform. This Anonymous class approach will result in creating 100 objects for a single client. Too much object creation is not affordable for my application, where there will be more than 5,000 active concurrent connections. Have a look at this article http://blogs.microsoft.co.il/gilf/2009/11/22/applying-strategy-pattern-instead-of-using-switch-statements/
Reflection with Annotation
I really like this approach. I created a custom annotation to represent the tasks performed by methods. There is no overhead of object creation, like in Strategy pattern method, as tasks are performed by a single class.
Annotation
#Retention(RetentionPolicy.RUNTIME)
#Target(ElementType.METHOD)
public #interface TaskAnnotation {
public String value();
}
The code given below maps the client request keys to the methods which process the task. Here, map is instantiated and populated only once.
public static final HashMap<String, Method> taskList = new HashMap<>();
public static void main(String[] args) throws Exception {
// Retrieves declared methods from ClientMessageHandler class
Method[] classMethods = ClientMessageHandler.class.getDeclaredMethods();
for (Method method : classMethods) {
// We will iterate through the declared methods and look for
// the methods annotated with our TaskAnnotation
TaskAnnotation annot = method.getAnnotation(TaskAnnotation.class);
if (annot != null) {
// if a method with TaskAnnotation is found, its annotation
// value is mapped to that method.
taskList.put(annot.value(), method);
}
}
// Start server
}
Now finally, our ClientMessageHandler class looks like the following
public class ClientMessageHandler {
public void onMessageReceived(JSONObject clientRequest) throws JSONException {
// Retrieve the Method corresponding to the task from map
Method method = taskList.get(clientRequest.getString("task"));
if (method == null)
return;
try {
// Invoke the Method for this object, if Method corresponding
// to client request is found
method.invoke(this);
} catch (IllegalAccessException | IllegalArgumentException
| InvocationTargetException e) {
logger.error(e);
}
}
#TaskAnnotation("task1")
public void processTaskOne() {
}
#TaskAnnotation("task2")
public void processTaskTwo() {
}
// Methods for different tasks, annotated with the corresponding
// clientRequest code
}
Major drawback of this approach is the performance hit. This approach is slow compared to Direct Method calling approach. Moreover, many articles are suggesting to stay away from Reflection, unless we are dealing with dynamic programming.
Read these answers to know more about reflection What is reflection and why is it useful?
Reflection performance related articles
Faster alternatives to Java's reflection
https://dzone.com/articles/the-performance-cost-of-reflection
FINAL RESULT
I continue to use switch statements in my application to avoid any performance hit.
As mentioned in the comments, one of websockets drawback is that you'll to specify the communication protocol yourself. AFAIK, the huge switch is the best option. To improve code readability and maintenance, I'll suggest to use encoders and decoders. Then, your problem becomes: how should I design my messages?
Your game looks like Scrabble. I don't know how to play Scrabble so let's take the example of card game with money. Let's assume you have three types of actions:
Global action (join table, leave table ...)
Money action (place bet, split bet, ...)
Card action (draw card, etc)
Then your messages can look like
public class AbstractAction{
// not relevant for global action but let's put that aside for the example
public abstract void endTurn();
}
public class GlobalAction{
// ...
}
public class MoneyAction{
enum Action{
PLACE_BET, PLACE_MAX_BET, SPLIT_BET, ...;
}
private MoneyAction.Action action;
// ...
}
public class CardAction{
// ...
}
Once your decoder and encoders are properly defined, your switch would be easier to read and easier to maintain. In my project, the code would look like this:
#ServerEndPoint(value = ..., encoders = {...}, decoders = {...})
public class ServerEndPoint{
#OnOpen
public void onOpen(Session session){
// ...
}
#OnClose
public void onClose(Session session){
// ...
}
#OnMessage
public void onMessage(Session session, AbstractAction action){
// I'm checking the class here but you
// can use different check such as a
// specific attribute
if(action instanceof GlobalAction){
// do some stuff
}
else if (action instanceof CardAction){
// do some stuff
}
else if (action instance of MoneyAction){
MoneyAction moneyAction = (MoneyAction) action;
switch(moneyAction.getAction()){
case PLACE_BET:
double betValue = moneyAction.getValue();
// do some stuff here
break;
case SPLIT_BET:
doSomeVeryComplexStuff(moneyAction);
break;
}
}
}
private void doSomeVeryComplexStuff(MoneyAction moneyAction){
// ... do something very complex ...
}
}
I prefer this approach because:
The messages design can leverage your entities design (if you are using JPA behind)
As messages are not plain text anymore but objects, enumerations can be used and enumerations are very powerful in this kind of switch-case situation. With the same logic but in a lesser extend, class abstraction can be useful as well
The ServerEndPoint class only handles communication. The business logic is handled out of this class, either directly in Messages classes or in some EJB. Because of this split, code maintenance is much easier
Bonus: #OnMessage method can be read as a summary of the protocol but details should not be displayed here. Each case must contain few lines only.
I prefer avoid using Reflection: it'll ruin your code readability, in the specific scenario of websocket
To go further beyond code readability, maintenance and efficiency, you can use a SessionHandler to intercept some CDI event if this can improve your code. I gave an example in this answer. If you need a more advanced example, Oracle provides a great tutorial about it. It might help you to improve your code.

How to create some sort of event framework in java?

I don't have a GUI (my classes are part of a Minecraft Mod). I wanted to be able to mimic C# event framework: A class declares events and lets others subscribe to them.
My first approach was to create a class called EventArgs and then do something like this:
public class EventArgs
{
public boolean handled;
}
#FunctionalInterface
public interface IEventHandler<TEvtArgs extends EventArgs>
{
public void handle(Object source, TEvtArgs args);
}
public class Event<TEvtArgs extends EventArgs>
{
private final Object owner;
private final LinkedList<IEventHandler<TEvtArgs>> handlers = new LinkedList<>();
public Event(Object owner)
{
this.owner = owner;
}
public void subscribe(IEventHandler<TEvtArgs> handler)
{
handlers.add(handler);
}
public void unsubscribe(IEventHandler<TEvtArgs> handler)
{
while(handlers.remove(handler));
}
public void raise(TEvtArgs args)
{
for(IEventHandler<TEvtArgs> handler : handlers)
{
handler.handle(owner, args);
if(args.handled)
break;
}
}
}
Then a class would do something like this:
public class PropertyChangedEvtArgs extends EventArgs
{
public final Object oldValue;
public final Object newValue;
public PropertyChangedEvtArgs(final Object oldValue, final Object newValue)
{
this.oldValue = oldValue;
this.newValue = newValue;
}
}
public class SomeEventPublisher
{
private int property = 0;
private final Random rnd = new Random();
public final Event<PropertyChangedEvtArgs> PropertyChanged = new Event<>(this);
public void raiseEventOrNot(int value)
{
if(rnd.nextBoolean())//just to represent the fact that the event is not always raised
{
int old = property;
property = value;
PropertyChanged.raise(new PropertyChangedEvtArgs("old(" + old + ")", "new(" + value + ")"));
}
}
}
public class SomeSubscriber
{
private final SomeEventPublisher eventPublisher = new SomeEventPublisher();
public SomeSubscriber()
{
eventPublisher.PropertyChanged.subscribe(this::handlePropertyAChanges);
}
private void handlePropertyAChanges(Object source, PropertyChangedEvtArgs args)
{
System.out.println("old:" + args.oldValue);
System.out.println("new:" + args.newValue + "\n");
}
public void someMethod(int i)
{
eventPublisher.raiseEventOrNot(i);
}
}
public class Main
{
private static final SomeSubscriber subscriber = new SomeSubscriber();
public static void main(String[] args)
{
for(int i = 0; i < 10; ++i)
{
subscriber.someMethod(i);
}
}
}
The biggest problem with this naïve approach is that it breaks proper encapsullation by exposing raise as public. I can't see a way around it, and maybe my whole pattern is wrong. I would like some ideas.
There's also a related problem: I would like the events to be raised immediately after the method raising them returns. Is there a way to synchronize this using threads or some other construct? The caller code, of course, can't be involved in the task of synchronization. It has to be completely transparent to it.
The best thing to do here is to avoid implementing your own event framework in the first place, and instead rely on some existing library. Out of the box Java provides EventListener, and at a minimum you can follow the patterns documented there. Even for non-GUI applications most of this advice applies.
Going beyond the JDK Guava provides several possible options, depending on your exact use case.
The most likely candidate is EventBus, which:
allows publish-subscribe-style communication between components without requiring the components to explicitly register with one another (and thus be aware of each other).
Or ListenableFuture (and ListeningExecutorService) which:
allows you to register callbacks to be executed once [a task submitted to an Executor] is complete, or if the computation is already complete, immediately. This simple addition makes it possible to efficiently support many operations that the basic Future interface cannot support.
Or the Service API which:
represents an object with an operational state, with methods to start and stop. For example, webservers, RPC servers, and timers can implement the Service interface. Managing the state of services like these, which require proper startup and shutdown management, can be nontrivial, especially if multiple threads or scheduling is involved.
This API similarly lets you register listeners to respond to state changes in your services.
Even if none of these options directly work for your use case, take a look at Guava's source code for examples of event-driven behavior and listeners you can try to emulate.

Java send event automatically on setter

I am building an Java App, with Eclipse RCP.
When my model classes are modified (thanks to my setter methods) I would like to send events (IEventBroker) corresponding to each differents setters, that my differents UI elements can catch. I thought about this pattern :
Model {
setValueForA() {...}
setValueForB() {...}
setValueForC() {...}
}
ServiceLayer {
setValueForA() {
model.setValueForA();
sendEventAUpdated();
}
setValueForB() {
model.setValueForB();
sendEventBUpdated();
}
setValueForC() {
model.setValueForC();
sendEventCUpdated();
}
}
Is there a way with a framework, library, annotations,... whatever who can helps me to implement less code, and to avoid this copying code? I am on 1.8 JDK
This sounds like a candidate for Observer pattern. Java language provides some basic support through Observable and Observer classes. The downside is that you will have to extend these in your corresponding classes. This frees you from the overhead of adding observer/obervable code though. Broadly, you just add notify code in your setter methods of model object. Define Observers, then tie them together from application(Service) code. Whenever a set happens, if it has notify coded, it will notify all observers associated with the model object.
Following is a rough example based on code in question :
Model extends Observable { //various observers can be notified when property changes.
setValueForA(Object value) {...
setChanged();
notifyObservers(value);
}
//Example setter above. Similar code needed in all required properties.
}
Now create observers
public class SomePropertyObserver implements Observer
{
private ObservableValue ov = null;
public SomePropertyObserver(ObservableValue ov)
{
this.ov = ov;
}
//This method is called whenever the observed object is changed.
public void update(Observable obs, Object obj)
{
//obj is the value sent from notifyObservers called from observable
}
}
Tie them together in your application code :
Model model = new Model();
SomePropertyObserver observer = new SomePropertyObserver(model);
model.addObserver(observer); // now observer will get notification whenever a property changes in model.
There are some downsides, as you have to extend Observable and hence you are very tightly coupled and restricted of further object extension.

Avoiding instanceof when checking a message type

I have the following situation where a client class executes different behavior based on the type of message it receives. I'm wondering if there is a better way of doing this since I don't like the instanceof and the if statements.
One thing I thought of doing was pulling the methods out of the client class and putting them into the messages. I would put a method like process() in the IMessage interface and then put the message specific behavior in each of the concrete message types. This would make the client simple because it would just call message.process() rather than checking types. However, the only problem with this is that the behavior contained in the conditionals has to do with operations on data contained within the Client class. Thus, if I did implement a process method in the concrete message classes I would have to pass it the client and I don't know if this really makes sense either.
public class Client {
messageReceived(IMessage message) {
if(message instanceof concreteMessageA) {
concreteMessageA msg = (concreteMessageA)message;
//do concreteMessageA operations
}
}
if (message instanceof concreteMessageB) {
concreteMessageb msg = (concreteMessageB)message;
//do concreteMessageB operations
}
}
The simple way to avoid instanceof testing is to dispatch polymorphicly; e.g.
public class Client {
void messageReceived(IMessage message) {
message.doOperations(this);
}
}
where each message class defines an appropriate doOperations(Client client) method.
EDIT: second solution which better matches the requirements.
An alternative that replaces a sequence of 'instanceof' tests with a switch statement is:
public class Client {
void messageReceived(IMessage message) {
switch (message.getMessageType()) {
case TYPE_A:
// process type A
break;
case TYPE_B:
...
}
}
}
Each IMessage class needs to define an int getMessageType() method to return the appropriate code. Enums work just as well ints, and are more more elegant, IMO.
One option here is a handler chain. You have a chain of handlers, each of which can handle a message (if applicable) and then consume it, meaning it won't be passed further down the chain. First you define the Handler interface:
public interface Handler {
void handle(IMessage msg);
}
And then the handler chain logic looks like:
List<Handler> handlers = //...
for (Handler h : handlers) {
if (!e.isConsumed()) h.handle(e);
}
Then each handler can decide to handle / consume an event:
public class MessageAHandler implements Handler {
public void handle(IMessage msg) {
if (msg instanceof MessageA) {
//process message
//consume event
msg.consume();
}
}
}
Of course, this doesn't get rid of the instanceofs - but it does mean you don't have a huge if-elseif-else-if-instanceof block, which can be unreadable
What type of message system are you using?
Many have options to add a filter to the handlers based on message header or content. If this is supported, you simply create a handler with a filter based on message type, then your code is nice and clean without the need for instanceof or checking type (since the messaging system already checked it for you).
I know you can do this in JMS or the OSGi event service.
Since you are using JMS, you can basically do the following to register your listeners. This will create a listener for each unique message type.
String filterMsg1 = "JMSType='messageType1'";
String filterMsg2 = "JMSType='messageType2'";
// Create a receiver using this filter
Receiver receiverType1 = session.createReceiver(queue, filterMsg1);
Receiver receiverType2 = session.createReceiver(queue, filterMsg2);
receiverType1.setMessageHandler(messageType1Handler);
receiverType2.setMessageHandler(messageType2Handler);
Now each handler will receive the specific message type only (no instanceof or if-then), assuming of course that the sender sets the type via calls to setJMSType() on the outgoing message.
This method is built into message, but you can of course create your own header property and filter on that instead as well.
//Message.java
abstract class Message{
public abstract void doOperations();
}
//MessageA.java
class MessageA extends Message{
public void doOperations(){
//do concreteMessageA operations ;
}
}
//MessageB.java
class MessageB extends Message {
public void doOperations(){
//do concreteMessageB operations
}
}
//MessageExample.java
class MessageExample{
public static void main(String[] args) {
doSmth(new MessageA());
}
public static void doSmth(Message message) {
message.doOperations() ;
}
}
A Java 8 solution that uses double dispatch. Doesn't get rid of instanceof completely but does only require one check per message instead of an if-elseif chain.
public interface Message extends Consumer<Consumer<Message>> {};
public interface MessageA extends Message {
#Override
default void accept(Consumer<Message> consumer) {
if(consumer instanceof MessageAReceiver){
((MessageAReceiver)consumer).accept(this);
} else {
Message.super.accept(this);
}
}
}
public interface MessageAReceiver extends Consumer<Message>{
void accept(MessageA message);
}
With JMS 2.0 you can use:
consumer.receiveBody(String.class)
For more information you can refer here:

Categories