跳至主要内容

Java EE Security API 1.0: IdentityStore


There are two built-in IdentityStore implementations provided in Glassfish v5, Database or Ldap.
An example of using built-in @DatabaseIdentityStoreDefinition to setup database based IdentityStore.
@DatabaseIdentityStoreDefinition(
    dataSourceLookup = "${'java:global/MyDS'}",
    callerQuery = "#{'select password from caller where name = ?'}",
    groupsQuery = "select group_name from caller_groups where caller_name = ?",
    hashAlgorithm = Pbkdf2PasswordHash.class,
    priorityExpression = "#{100}",
    hashAlgorithmParameters = {
        "Pbkdf2PasswordHash.Iterations=3072",
        "${applicationConfig.dyna}"
    } // just for test / example
)
@ApplicationScoped
@Named
public class ApplicationConfig {

    public String[] getDyna() {
        return new String[]{"Pbkdf2PasswordHash.Algorithm=PBKDF2WithHmacSHA512", "Pbkdf2PasswordHash.SaltSizeBytes=64"};
    }

}
Initializes database with the initial users.
@DataSourceDefinition(
        // global to circumvent https://java.net/jira/browse/GLASSFISH-21447
        name = "java:global/MyDS",
        className = "org.h2.jdbcx.JdbcDataSource",
        // :mem:test would be better, but TomEE insists on this being a file
        url = "jdbc:h2:mem:test;DB_CLOSE_ON_EXIT=FALSE"
)
@Singleton
@Startup
public class DatabaseSetup {

    @Resource(lookup = "java:global/MyDS")
    private DataSource dataSource;

    @Inject
    private Pbkdf2PasswordHash passwordHash;

    @Inject
    Logger LOG;

    @PostConstruct
    public void init() {

        LOG.info("initializing database...");

        Map<String, String> parameters = new HashMap<>();
        parameters.put("Pbkdf2PasswordHash.Iterations", "3072");
        parameters.put("Pbkdf2PasswordHash.Algorithm", "PBKDF2WithHmacSHA512");
        parameters.put("Pbkdf2PasswordHash.SaltSizeBytes", "64");
        passwordHash.initialize(parameters);

        executeUpdate(dataSource, "DROP TABLE IF EXISTS caller");
        executeUpdate(dataSource, "DROP TABLE IF EXISTS caller_groups");

        executeUpdate(dataSource, "CREATE TABLE IF NOT EXISTS caller(name VARCHAR(64) PRIMARY KEY, password VARCHAR(255))");
        executeUpdate(dataSource, "CREATE TABLE IF NOT EXISTS caller_groups(caller_name VARCHAR(64), group_name VARCHAR(64))");

        executeUpdate(dataSource, "INSERT INTO caller VALUES('user', '" + passwordHash.generate("password".toCharArray()) + "')");
        executeUpdate(dataSource, "INSERT INTO caller VALUES('reza', '" + passwordHash.generate("secret1".toCharArray()) + "')");
        executeUpdate(dataSource, "INSERT INTO caller VALUES('alex', '" + passwordHash.generate("secret2".toCharArray()) + "')");
        executeUpdate(dataSource, "INSERT INTO caller VALUES('arjan', '" + passwordHash.generate("secret2".toCharArray()) + "')");
        executeUpdate(dataSource, "INSERT INTO caller VALUES('werner', '" + passwordHash.generate("secret2".toCharArray()) + "')");

        executeUpdate(dataSource, "INSERT INTO caller_groups VALUES('user', 'foo')");
        executeUpdate(dataSource, "INSERT INTO caller_groups VALUES('user', 'bar')");

        executeUpdate(dataSource, "INSERT INTO caller_groups VALUES('reza', 'foo')");
        executeUpdate(dataSource, "INSERT INTO caller_groups VALUES('reza', 'bar')");

        executeUpdate(dataSource, "INSERT INTO caller_groups VALUES('alex', 'foo')");
        executeUpdate(dataSource, "INSERT INTO caller_groups VALUES('alex', 'bar')");

        executeUpdate(dataSource, "INSERT INTO caller_groups VALUES('arjan', 'foo')");
        executeUpdate(dataSource, "INSERT INTO caller_groups VALUES('werner', 'foo')");
    }

    @PreDestroy
    public void destroy() {
        try {
            executeUpdate(dataSource, "DROP TABLE IF EXISTS caller");
            executeUpdate(dataSource, "DROP TABLE IF EXISTS caller_groups");
        } catch (Exception e) {
            LOG.log(Level.SEVERE, "error drop tables:." + e.getMessage());

        }
    }

    private void executeUpdate(DataSource dataSource, String query) {
        try (Connection connection = dataSource.getConnection()) {
            try (PreparedStatement statement = connection.prepareStatement(query)) {
                statement.executeUpdate();
            }
        } catch (SQLException e) {
            LOG.log(Level.SEVERE, "error executed query:." + e.getMessage());
        }
    }

}
Note, we configure a Pbkdf2PasswordHash bean which is used to hash password.
Similiar with @DatabaseIdentityStoreDefinition, there is a @LdapIdentityStoreDefinition for configuring users and groups in Ldap servers, such Microsoft Active Directory, Apache Directory.
You can also customsize IdentityStore by implementate the IdentityStore interface.
@ApplicationScoped
public class TestIdentityStore implements IdentityStore {

    public CredentialValidationResult validate(UsernamePasswordCredential usernamePasswordCredential) {

        if (usernamePasswordCredential.compareTo("user", "password")) {
            return new CredentialValidationResult("user", new HashSet<>(asList("foo", "bar")));
        }

        return INVALID_RESULT;
    }

}
Grab the source codes from my github account, and have a try.

评论

此博客中的热门博文

Build a Reactive application with Angular 5 and Spring Boot 2.0

I have created a post to describe Reactive programming supports in Spring 5 and its subprojects, all codes of this article are updated the latest Spring 5 RELEASE, check spring-reactive-sample under my Github account. In this post, I will create a simple blog system, including: A user can sign in and sign out. An authenticated user can create a post. An authenticated user can update a post. Only the user who has ADMIN role can delete a post. All users(including anonymous users) can view post list and post details. An authenticated user can add his comments to a certain post. The backend will be built with the latest Spring 5 reactive stack, including: Spring Boot 2.0, at the moment the latest version is 2.0.0.M7 Spring Data MongoDB supports reactive operations for MongoDB Spring Session adds reactive support for WebSession Spring Security 5 aligns with Spring 5 reactive stack The frontend is an Angular based SPA and it will be generated by Angular CLI. The so...

Create a restful application with AngularJS and Zend 2 framework

Create a restful application with AngularJS and Zend 2 framework This example application uses AngularJS/Bootstrap as frontend and Zend2 Framework as REST API producer. The backend code This backend code reuses the database scheme and codes of the official Zend Tutorial, and REST API support is also from the Zend community. Getting Started with Zend Framework 2 Getting Started with REST and Zend Framework 2 Zend2 provides a   AbstractRestfulController   for RESR API producing. class AlbumController extends AbstractRestfulController { public function getList() { $results = $this->getAlbumTable()->fetchAll(); $data = array(); foreach ($results as $result) { $data[] = $result; } return new JsonModel(array( 'data' => $data) ); } public function get($id) { $album = $this->getAlbumTable()->getAlbum($id); return new JsonModel(array("data" =...

Auditing with Hibernate Envers

Auditing with Hibernate Envers The approaches provided in JPA lifecyle hook and Spring Data auditing only track the creation and last modification info of an Entity, but all the modification history are not tracked. Hibernate Envers fills the blank table. Since Hibernate 3.5, Envers is part of Hibernate core project. Configuration Configure Hibernate Envers in your project is very simple, just need to add   hibernate-envers   as project dependency. <dependency> <groupId>org.hibernate</groupId> <artifactId>hibernate-envers</artifactId> </dependency> Done. No need extra Event listeners configuration as the early version. Basic Usage Hibernate Envers provides a simple   @Audited   annotation, you can place it on an Entity class or property of an Entity. @Audited private String description; If   @Audited   annotation is placed on a property, this property can be tracked. @Entity @Audited public clas...