java2practice

We usually send primitive data to spring controller by using @RequestParam annotation. But how to pass whole JSONObject string or JSONArray string to spring controller directly.

For that we have to include below jar files in buildpath

1. jackson-core-asl-1.7.3.jar
2. jackson-mapper-asl-1.7.3.jar

I have created Person pojo which will be mapped with javascript JSONObject exactly, Whatever the identifiers are there in this POJO should be there in Javascript JSON.

public class Person implements Serializable{
private String id;
private String firstName;
private String lastName;
//setters and getters
}

Pojo should implement Serializable interface, as  Jackson will serialize and deserialize to send data between server and client.

Our json data is :
{"persons": [
{
"firstName": "Ramesh",
"id": "id1",
"lastName": "Kotha"
},
{
"firstName": "Sathish",
"id": "id2",
"lastName": "Kotha"
}
]
}

Below is an ajax request from jsp.

$.ajax({
type: 'POST',
dataType: 'json',
contentType:'application/json',
url: "create_persons.htm",
data:JSON.stringify(arr),
success: function(data, textStatus ){
console.log(data);
//alert("success");
},
error: function(xhr, textStatus, errorThrown){
//alert('request failed'+errorThrown);
}
});

Now will see Spring controller Code:

@RequestMapping(value="/create_persons.htm")
public @ResponseBody createPerson(@RequestBody Person[] persons){
//here you can persons array as normal
for(Person person : persons){
System.out.println(person.getId());
}
}

By seeing @RequestBody annontation json data will be converted into Java Person[] and passed to persons array.

Add this in your Configuration file :

<bean id="jacksonMessageConverter" class="org.springframework.http.converter.json.MappingJacksonHttpMessageConverter"></bean>
<bean class="org.springframework.web.servlet.mvc.annotation.AnnotationMethodHandlerAdapter">
<property name="messageConverters">
<list>
<ref bean="jacksonMessageConverter"/>
</list>
</property>

That’s it, now if you pass json string to the spring controller, it will be converted into java POJO.

Aspect Oriented Programming:

Developer has to concentrate on business logic, Cross cutting functionalities like Transactions, Logging, Data Access Logic should be taken care by container.

This principle is called AOP.

Until I have used AOP, I didn’t recognize its worth:

1.       Let’s explore the non AOP architecture

package com.java2practice.aop;
import java.math.BigDecimal;

public class Account {
BigDecimal currentAmount = new BigDecimal("5000");

public BigDecimal withdraw(BigDecimal withdrawlAmount){
if(withdrawlAmount.compareTo(currentAmount) < 0 )
     return currentAmount.subtract(withdrawlAmount);
else
     throw new FundsNotSufficientException();
 }
}

Do we have any problem with above withdraw method, if we observe closely?

First we should use logging. Otherwise resolving the errors would be a big problem.

For that we will change the above method accordingly

import java.math.BigDecimal;
import org.apache.log4j.Logger;

public class Account {
Logger logger = Logger.getLogger(Account.class);
BigDecimal currentAmount = new BigDecimal("5000");

public BigDecimal withdraw(BigDecimal withdrawlAmount){
    logger.info("withdraw method started");
    if(withdrawlAmount.compareTo(currentAmount) < 0 ){
       return currentAmount.subtract(withdrawlAmount);
   }
   else{
     throw new FundsNotSufficientException();
    }
    logger.info("withdraw method successfully completed");
  }
}

Second there is no Transaction started, As withdraw method going update the currentAmount in Database it should be within Transaction so will add transaction.

import java.math.BigDecimal;
import org.apache.log4j.Logger;
import org.hibernate.Session;
import org.hibernate.Transaction;

public class Account {
Logger logger = Logger.getLogger(Account.class);
BigDecimal currentAmount = new BigDecimal("5000");

public BigDecimal withdraw(BigDecimal withdrawlAmount){
   Transaction tx = null;
   Session session = getCurrentSession();
   try{
     logger.info("withdraw method started");
     if(withdrawlAmount.compareTo(currentAmount) < 0 ){
        tx = session.beginTransaction();
        currentAmount = currentAmount.subtract(withdrawlAmount);
        tx.commit();
        return currentAmount;
       }
     else{
       tx.rollback();
       logger.debug("FundsNotSufficientException :: withdraw method failed");
       throw new FundsNotSufficientException();
    }
   logger.info("withdraw method successfully completed");
 }catch(Exception ex){
   tx.rollback()
   logger.debug("withdraw method failed");
   }
  }
}

Now we implemented Transactions and logging successfully but what we are doing if any new requirement is coming we are changing the BankAccount class which is against Object Oriented principle.

If we think again there should be data access logic for the Bank Account class then we have to change once again the class to like this:

package com.java2practice.aop;
import java.math.BigDecimal;
import org.apache.log4j.Logger;
import org.hibernate.Session;
import org.hibernate.Transaction;

public class Account {
Logger logger = Logger.getLogger(Account.class);
BigDecimal currentAmount = new BigDecimal("5000");

public BigDecimal withdraw(BigDecimal withdrawlAmount){
    logger.info("withdraw method started");
    Transaction tx = null;
    Session session = getCurrentSession();
    try{
      if(currentUser.isWIthdrawEnabled()){
        if(withdrawlAmount.compareTo(currentAmount) < 0 ){
          tx = session.beginTransaction();
          currentAmount = currentAmount.subtract(withdrawlAmount);
          tx.commit();
          return currentAmount;
        }
      else{
        tx.rollback();
        logger.debug("FundsNotSufficientException :: withdraw method failed");
        throw new FundsNotSufficientException();
       }
     logger.info("withdraw method successfully completed");
     }else {
     logger.info("withdraw method Failed withdrawal is disabled for this user");
   }
  }catch(Exception ex){
    tx.rollback();
    logger.debug("withdraw method failed");
   }
  }
}

Now it looks fine, but in future any new cross cutting functionality comes, we need to change the Bank Account class once again. What if we separate the cross cutting functionalities like Logging, Security, Transaction to outside of the business logic then we can add new cross cutting functionality without touching BankAccount class.

Imagine if the Data access logic changes, we may need to change lot of files. Same with the entire cross cutting functionalities, They should not be coupled with the business logic and should be defined separately. This we can achieve through Spring AOP.

2. Will see the AOP architecture:

For Cross cutting functionality Logging will see the AOP architecture

1. CustomerDAO.java

package com.java2practice.dao;

public interface CustomerDAO {
   String addCustomer();
   String addCustomerReturnValue()
   void addCustomerThrowException() throws Exception;
   void addCustomerAround(String name);
}

1. CustomerDAOImpl.java

package com.java2practice.dao.impl;
import com.java2practice.dao.CustomerDAO;

public class CustomerDAOImpl implements CustomerDAO {
  public String addCustomer(){
     System.out.println("addCustomer() is running ");
     //throw new ArithmeticException();
     return "ramesh";
   }
  public String addCustomerReturnValue(){
     System.out.println("addCustomerReturnValue() is running ");
     return "abc";
  }
  public void addCustomerThrowException() throws Exception {
     System.out.println("addCustomerThrowException() is running ");
     throw new Exception("Generic Error");
   }
  public void addCustomerAround(String name){
    System.out.println("addCustomerAround() is running, args : " + name);
  }
}

Here logging will be implemented by AOP, for that we need to write an Aspect.

1. LoggingAspect.java

package com.java2practice.aspect;
import org.aspectj.lang.JoinPoint;
import org.aspectj.lang.ProceedingJoinPoint;
import org.aspectj.lang.annotation.After;
import org.aspectj.lang.annotation.AfterReturning;
import org.aspectj.lang.annotation.AfterThrowing;
import org.aspectj.lang.annotation.Around;
import org.aspectj.lang.annotation.Aspect;
import org.aspectj.lang.annotation.Before;

@Aspect
public class LoggingAspect {
  @Before("execution(* com.java2practice.dao.*.*(..))")
  public void logBefore(JoinPoint joinPoint) {
     System.out.println(joinPoint.getSignature().getName()+" is started! from logBefore");
  }
  @After("execution(* com.java2practice.dao.*.*(..))")
  public void logAfter(JoinPoint joinPoint){
    System.out.println(joinPoint.getSignature().getName()+" is completed! from logAfter");
  }
  @AfterThrowing(pointcut = "execution(* com.java2practice.dao.CustomerDAO.*(..))", throwing= "error")
  public void logAround(JoinPoint joinPoint, Object error){
    System.out.println(joinPoint.getSignature().getName()+" after throwing called! from after throwing "+error);
  }
  @AfterReturning(  pointcut = "execution(* com.java2practice.dao.CustomerDAO.addCustomerReturnValue(..))", returning= "result")
  public void afterReturning(JoinPoint joinPoint, Object result){
    System.out.println("after returning called result: "+result);
  }
  @Around("execution(* com.java2practice.dao.CustomerDAO.addCustomerAround(..))")
  public void logAround(ProceedingJoinPoint joinPoint) throws Throwable {
    System.out.println("Around before is running!");
    if(joinPoint.getArgs()[0].equals("ramesh")){
       joinPoint.proceed(); //continue on the intercepted method
       System.out.println("Around after is running!");
     }
    else{
      System.out.println("you dont have permission to call this method");
    }
  }
}

1. @Aspect annotation will tells the container it is an Aspect
2. @Before(“execution(* com.java2practice.dao.*.*(..))”) — it will be executed before any method executes  in com.java2.practice.dao package
3. @After(“execution(* com.java2practice.dao.*.*(..))”) — it will be executed after any method execution in com.java2.practice.dao package
4. @AfterThrowing(pointcut = “execution(* com.java2practice.dao.CustomerDAO.*(..))”, throwing= “error”) —  It will be executed if any method throws any exception in com.java2practice.dao.CustomerDAO class and the error will be captured in “error”.
5. @AfterReturning(  pointcut = “execution(* com.java2practice.dao.CustomerDAO.addCustomerReturnValue(..))”, returning= “result”) – it will be executed if com.java2practice.dao.CustomerDAO.addCustomerReturnValue() method returns any value and the returned value will be stored in “result”.
6. @Around(“execution(* com.java2practice.dao.CustomerDAO.addCustomerAround(..))”) – This is very powerful l aspect, it will be executed before the method execution and after the method execution.  Even we can control the method execution  by joinPoint.proceed()

So now will write spring-config.xml

<beans xmlns="http://www.springframework.org/schema/beans"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns:aop="http://www.springframework.org/schema/aop"
xsi:schemaLocation="http://www.springframework.org/schema/beans
http://www.springframework.org/schema/beans/spring-beans-3.0.xsd
http://www.springframework.org/schema/aop
http://www.springframework.org/schema/aop/spring-aop-3.0.xsd ">
    <aop:aspectj-autoproxy />
    <bean id="customerDAO" />
    <!-- Aspect -->
    <bean id="logAspect" />
</beans>

Finally will write  Main program to see the power of AOP.

package com.java2practice.aspect;
import org.springframework.context.ApplicationContext;
import org.springframework.context.support.ClassPathXmlApplicationContext;
import com.java2practice.dao.CustomerDAO;

public class Main {
  public static void main(String[] args)  {
  ApplicationContext context = new ClassPathXmlApplicationContext("spring-config.xml");
  CustomerDAO customer = (CustomerDAO) context.getBean("customerDAO");
  try {
    //customer.addCustomer();
      customer.addCustomerAround("ramesh");
    //customer.addCustomerReturnValue();
    //customer.addCustomerThrowException();
     } catch (Exception e) {
    //e.printStackTrace();
   }
 }
}

Have fun with  AOP.

Spring MVC in simple terms:

It is a presentation framework, you may have doubt “How MVC belongs to only presentation layer”.

MVC Framework:

View and Controller belongs to Presentation Layer as Controller takes the URL and directs to appropriate View based on the given URL.

 MVC Flow:

1. Request comes to the controller, controller decides is there any business logic to be implemented for this request or not.
2.  If no business logic required it will redirect to the appropriate View. Ex: if we are requesting a login form no business logic required here so controller directly redirect login.jsp.
3. If business logic required to the particular request it will call the appropriate Service class to process the request. Based on the service object response, controller redirects to the appropriate view.
4. In the view if there is any data is to be shown, then the data will be taken from appropriate model.

With the above diagram View and Controller directly interacting with Presentation Layer.

And model is used by View to show the data.

So MVC belongs to Presentation Layer

MVC in Three tier architecture:

In three tier architecture we have Presentation Tier, Middle Tier, Data Access Tier, simply we can call Presentation layer, Service layer  and DAO Layer.

Presentation layer interacts with the Service layer and Service layer interacts with DAO layer.

The Flow will be like this:

1. Request  comes to controller and controller decides whether this request needs any business logic to be implemented or not
2. If no it simply returns a view.
3. If yes it interacts with Service Layer and Service Layer interacts with DAO Layer
4. Later the based on the response from Service Layer appropriate View will be shown by the Controller.

Spring recommends the same model can be used for presentation layer, In Struts we have form beans as presentation layer model. But in Spring we can use the same model across all the Three Layers.