Questions d'entretiens chez Java

1. Introduction

Cet article contient des réponses à certaines des questions d'entretien d'embauche les plus importantes sur le noyau Java. Les réponses à certaines d'entre elles peuvent ne pas être évidentes, cet article vous aidera donc à clarifier les choses.

2. Questions sur le langage Core-Java pour les débutants

Q1. Les données sont-elles transmises par référence ou par valeur en Java?

Bien que la réponse à cette question soit assez simple, cette question peut être déroutante pour les débutants. Tout d'abord, clarifions le sujet de la question:

  1. Passer par valeur - cela signifie que nous passons une copie d'un objet en tant que paramètre dans une méthode.
  2. Passer par référence - cela signifie que nous passons une référence à un objet en tant que paramètre dans une méthode.

Pour répondre à la question, nous devons analyser deux cas. Ils représentent deux types de données que nous pouvons transmettre à une méthode: une primitive et un objet.

Lorsque nous transmettons des primitives à une méthode, sa valeur est copiée dans une nouvelle variable. En ce qui concerne les objets, la valeur de la référence est copiée dans une nouvelle variable. On peut donc dire que Java est un langage strictement pass-par-valeur .

Nous pouvons en apprendre plus à ce sujet dans l'un de nos articles: Pass-By-Value comme mécanisme de transmission de paramètres en Java.

Q2. Quelle est la différence entre l'importation et les importations statiques?

Nous pouvons utiliser des importations régulières pour importer une classe spécifique ou toutes les classes définies dans un package différent:

import java.util.ArrayList; //specific class import java.util.*; //all classes in util package

Nous pouvons également les utiliser pour importer des classes imbriquées publiques d'une classe englobante:

import com.baeldung.A.*

Cependant, nous devons être conscients que l'importation ci-dessus n'importe pas la classe A elle-même.

Il existe également des importations statiques qui nous permettent d'importer des membres statiques ou des classes imbriquées:

import static java.util.Collections.EMPTY_LIST;

L'effet est que nous pouvons utiliser la variable statique EMPTY_LIST sans ajouter le nom complet de la classe, c'est-à-dire comme si elle était déclarée dans la classe courante.

Q3. Quels modificateurs d'accès sont disponibles en Java et quel est leur objectif?

Il existe quatre modificateurs d'accès en Java :

  1. privé
  2. par défaut (package)
  3. protégé
  4. Publique

Le modificateur privé garantit que les membres de la classe ne seront pas accessibles en dehors de la classe. Il peut être appliqué aux méthodes, propriétés, constructeurs, classes imbriquées, mais pas aux classes de niveau supérieur elles-mêmes.

Contrairement au modificateur privé , nous pouvons appliquer le modificateur par défaut à tous les types de membres de classe et à la classe elle-même. Nous pouvons appliquer la visibilité par défaut en n'ajoutant aucun modificateur d'accès. Si nous utilisons la visibilité par défaut, notre classe ou ses membres ne seront accessibles qu'à l'intérieur du package de notre classe. Nous devons garder à l'esprit que le modificateur d'accès par défaut n'a rien de commun avec le mot-clé par défaut .

De la même manière que le modificateur par défaut , toutes les classes d'un même package peuvent accéder aux membres protégés . De plus, le modificateur protected permet aux sous-classes d'accéder aux membres protégés d'une superclasse, même s'ils ne sont pas dans le même package. Nous ne pouvons pas appliquer ce modificateur d'accès aux classes, uniquement aux membres de la classe.

Le modificateur public peut être utilisé avec le mot-clé de classe et tous les membres de la classe. Il rend les classes et les membres de classe accessibles dans tous les packages et par toutes les classes.

Nous pouvons en savoir plus dans l'article Modificateurs d'accès Java.

Q4. Quels autres modificateurs sont disponibles en Java et quel est leur objectif?

Il existe cinq autres modificateurs disponibles en Java:

  • statique
  • final
  • abstrait
  • synchronisé
  • volatil

Ceux-ci ne contrôlent pas la visibilité.

Tout d'abord, nous pouvons appliquer le mot-clé static aux champs et aux méthodes. Les champs ou méthodes statiques sont des membres de classe, tandis que les non-statiques sont des membres d'objet . Les membres de la classe n'ont besoin d'aucune instance pour être appelés. Ils sont appelés avec le nom de classe au lieu du nom de référence d'objet. Cet article donne plus de détails sur le mot clé static .

Ensuite, nous avons le mot-clé final . Nous pouvons l'utiliser avec des champs, des méthodes et des classes. Lorsque final est utilisé sur un champ, cela signifie que la référence du champ ne peut pas être modifiée. Il ne peut donc pas être réaffecté à un autre objet. Lorsque final est appliqué à une classe ou à une méthode, cela nous assure que cette classe ou méthode ne peut pas être étendue ou remplacée. Le mot-clé final est expliqué plus en détail dans cet article.

Le mot clé suivant est abstrait . Celui-ci peut décrire des classes et des méthodes. Lorsque les classes sont abstraites , elles ne peuvent pas être instanciées. Au lieu de cela, ils sont censés être sous-classés. Lorsque les méthodes sont abstraites , elles sont laissées sans implémentation et peuvent être remplacées dans les sous-classes.

Le mot-clé synchronisé est peut-être le plus avancé. Nous pouvons l'utiliser avec l'instance ainsi qu'avec des méthodes statiques et des blocs de code. Lorsque nous utilisons ce mot-clé, nous faisons en sorte que Java utilise un verrou de moniteur pour assurer la synchronisation sur un fragment de code donné. Plus d'informations sur synchronisé peuvent être trouvées dans cet article.

Le dernier mot-clé dont nous allons discuter est volatil . Nous ne pouvons l'utiliser qu'avec des champs d'instance. Il déclare que la valeur du champ doit être lue et écrite dans la mémoire principale - en contournant le cache du processeur. Toutes les lectures et écritures pour une variable volatile sont atomiques. Le mot clé volatile est expliqué en détail dans cet article.

Q5. Quelle est la différence entre JDK, JRE et JVM?

JDK signifie Java Development Kit , qui est un ensemble d'outils nécessaires aux développeurs pour écrire des applications en Java. Il existe trois types d'environnements JDK:

  • Standard Edition - kit de développement pour créer des applications de bureau ou serveur portables
  • Enterprise Edition - une extension de l'édition Standard avec prise en charge de l'informatique distribuée ou des services Web
  • Micro Edition – development platform for embedded and mobile applications

There are plenty of tools included in the JDK which help programmers with writing, debugging or maintaining applications. The most popular ones are a compiler (javac), an interpreter (java), an archiver (jar) and a documentation generator (javadoc).

JRE is a Java Runtime Environment. It's a part of the JDK, but it contains the minimum functionality to run Java applications. It consists of a Java Virtual Machine, core classes, and supporting files. For example, it doesn't have any compiler.

JVM is the acronym for Java Virtual Machine, which is a virtual machine able to run programs compiled to bytecode. It's described by the JVM specification, as it's important to ensure interoperability between different implementations. The most important function of a JVM is to enable users to deploy the same Java application into different operating systems and environments without worrying about what lies underneath.

For more information, let's check the Difference Between JVM, JRE, and JDK article.

Q6. What Is the Difference Between Stack and Heap?

There are two parts of memory where all variables and objects are stored by the JVM. The first is the stack and the second is the heap.

The stack is a place where the JVM reserves blocks for local variables and additional data. The stack is a LIFO (last in first out) structure. It means that whenever a method is called, a new block is reserved for local variables and object references. Each new method invocation reserves the next block. When methods finish their execution, blocks are released in the reversed manner they were started.

Every new thread has its own stack.

We should be aware that the stack has much less memory space than the heap. And when a stack is full, the JVM will throw a StackOverflowError. It's likely to occur when there is a bad recursive call and the recursion goes too deep.

Every new object is created on the Java heap which is used for a dynamic allocation. There is a garbage collector which is responsible for erasing unused objects which are divided into young (nursery) and old spaces. Memory access to the heap is slower than access to the stack. The JVM throws an OutOfMemoryError when the heap is full.

We can find more details in the Stack Memory and Heap Space in Java article.

Q7. What Is the Difference Between the Comparable and Comparator Interfaces?

Sometimes when we write a new class, we would like to be able to compare objects of that class. It's especially helpful when we want to use sorted collections. There are two ways we can do this: with the Comparable interface or with the Comparator interface.

First, let's look at the Comparable interface:

public interface Comparable { int compareTo(T var1); }

We should implement that interface by the class whose objects we want to sort.

It has the compareTo() method and returns an integer. It can return three values: -1, 0, and 1 which means that this object is less than, equal to or greater than the compared object.

It's worth mentioning that the overridden compareT0() method should be consistent with the equals() method.

On the other hand, we can use the Comparator interface. It can be passed to the sort() methods of the Collection interface or when instantiating sorted collections. That's why it's mostly used to create a one-time sorting strategy.

What's more, it's also useful when we use a third-party class which doesn't implement the Comparable interface.

Like the compareTo() method, the overridden compare() methods should be consistent with the equals() method, but they may optionally allow comparison with nulls.

Let's visit the Comparator and Comparable in Java article for more information.

Q8. What Is the void Type and When Do We Use It?

Every time we write a method in Java, it must have a return type. If we want the method to return no value, we can use the void keyword.

We should also know that there is a Void class. It's a placeholder class that may be used, for example, when working with generics. The Void class can neither be instantiated nor extended.

Q9. What Are the Methods of the Object Class and What Do They Do?

It's important to know what methods the Object class contains and how they work. It's also very helpful when we want to override those methods:

  • clone() – returns a copy of this object
  • equals() – returns true when this object is equal to the object passed as a parameter
  • finalize() – the garbage collector calls this method while it's cleaning the memory
  • getClass() – returns the runtime class of this object
  • hashCode() – returns a hash code of this object. We should be aware that it should be consistent with the equals() method
  • notify() – sends a notification to a single thread waiting for the object's monitor
  • notifyAll() – sends a notification to all threads waiting for the object's monitor
  • toString() – returns a string representation of this object
  • wait() – there are three overloaded versions of this method. It forces the current thread to wait the specified amount of time until another thread calls notify() or notifyAll() on this object.

Q10. What Is an Enum and How We Can Use It?

Enum is a type of class that allows developers to specify a set of predefined constant values. To create such a class we have to use the enum keyword. Let's imagine an enum of days of the week:

public enum Day { SUNDAY, MONDAY, TUESDAY, WEDNESDAY, THURSDAY, FRIDAY, SATURDAY }

To iterate over all constants we can use the static values() method. What's more, enums enable us to define members such as properties and methods like regular classes.

Although it's a special type of class, we can't subclass it. An enum can, however, implement an interface.

Another interesting advantage of Enums is that they are thread-safe and so they are popularly used as singletons.

We can find more about Enums in one of our guides.

Q11. What Is a JAR?

JAR is a shortcut for Java archive. It's an archive file packaged using the ZIP file format. We can use it to include the class files and auxiliary resources that are necessary for applications. It has many features:

  • Security – we can digitally sign JAR files
  • Compression – while using a JAR, we can compress files for efficient storage
  • Portability – we can use the same JAR file across multiple platforms
  • Versioning – JAR files can hold metadata about the files they contain
  • Sealing – we can seal a package within a JAR file. This means that all classes from one package must be included in the same JAR file
  • Extensions – we can use the JAR file format to package modules or extensions for existing software

Q12. What Is aNullPointerException?

The NullPointerException is probably the most common exception in the Java world. It's an unchecked exception and thus extends RuntimeException. We shouldn't try to handle it.

This exception is thrown when we try to access a variable or call a method of a null reference, like when:

  • invoking a method of a null reference
  • setting or getting a field of a null reference
  • checking the length of a null array reference
  • setting or getting an item of a null array reference
  • throwing null

Q13. What Are Two Types of Casting in Java? Which Exception May Be Thrown While Casting? How Can We Avoid It?

We can distinguish two types of casting in Java. We can do upcasting which is casting an object to a supertype or downcasting which is casting an object to a subtype.

Upcasting is very simple, as we always can do that. For example, we can upcast a String instance to the Object type:

Object str = "string";

Alternatively, we can downcast a variable. It's not as safe as upcasting as it involves a type check. If we incorrectly cast an object, the JVM will throw a ClassCastExcpetion at runtime. Fortunately, we can use the instanceof keyword to prevent invalid casting:

Object o = "string"; String str = (String) o; // it's ok Object o2 = new Object(); String str2 = (String) o2; // ClassCastException will be thrown if (o2 instanceof String) { // returns false String str3 = (String) o2; }

We can learn more about type casting in this article.

3. Core-Java Language Questions for Advanced Programmers

Q1. Why Is String an Immutable Class?

We should know that String objects are treated differently than other objects by the JVM. One difference is that String objects are immutable. It means that we can't change them once we have created them. There are several reasons why they behave that way:

  1. They are stored in the string pool which is a special part of the heap memory. It's responsible for saving a lot of space.
  2. The immutability of the String class guarantees that its hash code won't change. Due to that fact, Strings can be effectively used as keys in hashing collections. We can be sure that we won't overwrite any data because of a change in hash codes.
  3. They can be used safely across several threads. No thread can change the value of a String object, so we get thread safety for free.
  4. Strings are immutable to avoid serious security issues. Sensitive data such as passwords could be changed by an unreliable source or another thread.

We can learn more about the immutability of Strings in this article.

Q2. What Is the Difference Between Dynamic Binding and Static Binding?

Binding in Java is a process of associating a method call with the proper method body. We can distinguish two types of binding in Java: static and dynamic.

The main difference between static binding and dynamic binding is that static binding occurs at compile time and dynamic binding at runtime.

Static binding uses class information for binding. It's responsible for resolving class members that are private or static and final methods and variables. Also, static binding binds overloaded methods.

Dynamic binding, on the other hand, uses object information to resolve bindings. That's why it's responsible for resolving virtual and overridden methods.

Q3. What Is JIT?

JIT stands for “just in time”. It's a component of the JRE that runs in the runtime and increases the performance of the application. Specifically, it's a compiler which runs just after the program's start.

This is different from the regular Java compiler which compiles the code long before the application is started. JIT can speed up the application in different ways.

For example, the JIT compiler is responsible for compiling bytecode into native instructions on the fly to improve performance. Also, it can optimize the code to the targeted CPU and operating system.

Additionally, it has access to many runtime statistics which may be used for recompilation for optimal performance. With this, it can also do some global code optimizations or rearrange code for better cache utilization.

Q4. What Is Reflection in Java?

Reflection is a very powerful mechanism in Java. Reflection is a mechanism of Java language which enables programmers to examine or modify the internal state of the program (properties, methods, classes etc.) at runtime. The java.lang.reflect package provides all required components for using reflection.

When using this feature, we can access all possible fields, methods, constructors that are included within a class definition. We can access them irrespective of their access modifier. It means that for example, we are able to access private members. To do that, we don't have to know their names. All we have to do is to use some static methods of Class.

It's worth to know that there is a possibility to restrict access via reflection. To do that we can use the Java security manager and the Java security policy file. They allow us to grant permissions to classes.

When working with modules since Java 9, we should know that by default, we aren't able to use reflection on classes imported from another module. To allow other classes to use reflection to access the private members of a package we have to grant the “Reflection” Permission.

This article goes into more depth about Java Reflection.

Q5. What Is a Classloader?

The classloader is one of the most important components in Java. It's a part of the JRE.

Simply put, the classloader is responsible for loading classes into the JVM. We can distinguish three types of classloaders:

  • Bootstrap classloader – it loads the core Java classes. They are located in the /jre/lib directory
  • Extension classloader – it loads classes located in /jre/lib/ext or in the path defined by the java.ext.dirs property
  • System classloader – it loads classes on the classpath of our application

A classloader loads classes “on demand”. It means that classes are loaded after they are called by the program. What's more, a classloader can load a class with a given name only once. However, if the same class is loaded by two different class loaders, then those classes fail in an equality check.

There is more information about classloaders in the Class Loaders in Java article.

Q6. What Is the Difference Between Static and Dynamic Class Loading?

Static class loading takes place when we have source classes available at compile time. We can make use of it by creating object instances with the new keyword.

Dynamic class loading refers to a situation when we can't provide a class definition at the compile time. Yet, we can do that at runtime. To create an instance of a class, we have to use the Class.forName() method:

Class.forName("oracle.jdbc.driver.OracleDriver") 

Q7. What Is the Purpose of the Serializable Interface?

We can use the Serializable interface to enable serializability of a class, using Java's Serialization API. Serialization is a mechanism for saving the state of an object as a sequence of bytes while deserialization is a mechanism for restoring the state of an object from a sequence of bytes. The serialized output holds the object's state and some metadata about the object's type and types of its fields.

We should know that subtypes of serializable classes are also serializable. However, if we want to make a class serializable, but its supertype is non-serializable we have to do two things:

  • implement the Serializable interface
  • assure that a no argument constructor is present in the superclass

We can read more about Serialization in one of our articles.

Q8. Is There a Destructor in Java?

In Java, the garbage collector automatically deletes the unused objects to free up the memory. Developers have no need to mark the objects for deletion, which is error-prone. So it's sensible Java has no destructors available.

In case the objects hold open sockets, open files, or database connections, the garbage collector is not able to reclaim those resources. We can release the resources in close method and use try-finally syntax to call the method afterward before Java 7, such as the I/O classes FileInputStreamand FileOutputStream. As of Java 7, we can implement interface AutoCloseable and use try-with-resources statement to write shorter and cleaner code. But it's possible the API users forget to call the close method, so the finalize method and Cleaner class come into existence to act as the safety net. But please be cautioned they are not equivalent to the destructor.

It's not assured both the finalize method and Cleaner class will run promptly. They even get no chance to run before the JVM exits. Although we could call System.runFinalization to suggest that JVM run the finalize methods of any objects pending for finalization, it's still non-deterministic.

Moreover, the finalize method can cause performance issues, deadlocks etc. We can find more information by looking at one of our articles: A Guide to the finalize Method in Java.

À partir de Java 9 , la classe Cleaner est ajoutée pour remplacer la méthode finalize en raison des inconvénients qu'elle présente. En conséquence, nous avons un meilleur contrôle sur le thread qui effectue les actions de nettoyage.

Mais la spécification Java souligne que le comportement des nettoyeurs pendant System.exit est spécifique à l'implémentation et Java ne fournit aucune garantie si les actions de nettoyage seront appelées ou non.