I have a modular application with one DAO implementation module that allows interacting with a database through JDBC. When I wrote the DAO module, I expected to get my DataSource via JNDI since the module was used until now as part of a web application. However, I need to use this same DAO module in a standalone application and therefore the datasource can't be obtained via JNDI.
Here is how I obtain my datasource:
#Bean(destroyMethod = "close")
public DataSource dataSource() throws Exception {
return new JndiDataSourceLookup().getDataSource("java:comp/env/jdbc/mydatasource");
}
I thought about the possibility to define the data source in the upper modules as follows:
For the web app module, same as previous code snippet.
For the standalong module, as follows:
#Bean(destroyMethod = "close")
public DataSource dataSource() throws Exception {
ComboPooledDataSource dataSource = new ComboPooledDataSource();
dataSource.setDriverClass(driverClassName);
dataSource.setJdbcUrl(jdbcUrl);
dataSource.setUser(username);
dataSource.setPassword(password);
return dataSource;
}
I'm looking for a better approach if there is one...
Related
I'm developing a standalone application with following technology stack:
Spring Boot version 2.1.0.RELEASE
Oracle12c System with driver ojdbc6 (11.2.0.3)
Apache Camel
JPA for the main datasource
JDBC for a secondary datasource (read only)
The JPA datasource is the primary datasource where the application itself is connected to and write data to. The JDBC is an additional datasource to read data from a database with another purpose.
During runtime I encounter the following issue:
I poll/select a JPA Entity from the primary datasource and do some processing. This processing includes running a select query on the secondary datasource via a jdbc template. Now if the execution of the query throws an exception I am able to catch it and want to update a status field on the JPA Entity and write it to the datasource.
I've already read that Oracle tries to do a rollback if a SQLException occurrs. The issue is that my JPA datasource is unable to commit the changes to the Entity I do when the JDBC query fails.
It seems to me like the two datasources/transaction managers are not completely independent of each other and that an exception in the secondary datasource causes errors in the primary datasource during commiting changes.
Is this even possible? If yes, how can I configure two independent transaction managers?
EDIT:
I have already tried to annotate the respective methods and classes with #Transactional(noRollbackFor = Exception.class) but this does not solve the problem.
Here are the two Datasource configurations:
ApplicationDatasourceConfig (JPA)
#Configuration
#EnableJpaRepositories(basePackages = "foo.bar.repository.jpa",
entityManagerFactoryRef = "applicationEntityManagerFactory",
transactionManagerRef = "applicationTransactionManager")
#ConfigurationProperties(prefix = "spring.datasource.hikari")
#EnableTransactionManagement
public class ApplicationDatasourceConfig extends HikariConfig{
#Bean("applicationDatasource")
#Primary
public DataSource applicationDataSource(){
return new HikariDataSource(this);
}
#Bean("applicationDatasourceProperties")
#Primary
public DataSourceProperties dataSourceProperties(){
return new DataSourceProperties();
}
#Bean("applicationEntityManagerFactory")
public LocalContainerEntityManagerFactoryBean applicationEntityManagerFactory(EntityManagerFactoryBuilder builder,
#Qualifier("applicationDatasource") DataSource dataSource){
return builder
.dataSource(dataSource)
.packages("foo.bar.entity")
.build();
}
#Bean("applicationTransactionManager")
#Primary
public PlatformTransactionManager applicationTransactionManager(#Qualifier("applicationEntityManagerFactory")EntityManagerFactory entityManagerFactory){
return new JpaTransactionManager(entityManagerFactory);
}
}
SecondaryDatasourceConfig (JDBC)
#Configuration
#ConfigurationProperties(prefix = "secondary.datasource")
#EnableTransactionManagement
public class SecondaryDatasourceConfig {
#Bean("secondaryDatasource")
public DataSource secondaryDataSource(){
return secondaryDataSourceProperties().initializeDataSourceBuilder().build();
}
#Bean
public DataSourceProperties secondaryDataSourceProperties(){
return new DataSourceProperties();
}
#Bean("secondaryTransactionManager")
public PlatformTransactionManager secondaryTransactionManager(#Qualifier("secondaryDatasource") DataSource secondaryDataSource){
return new DataSourceTransactionManager(secondaryDataSource);
}
}
Running under a Tomcat 9 and JDK 1.8, using Spring 5, I am trying to configure a JNDI connection to get a DataSource.
If I configure Spring, through XML, I get my DataSource and everything seems to work fine. I configured the DataSource in my applicationContext.xml, in this way:
<jee: jndi-lookup id = "dataSource" jndi-name = "jdbc / yages"
resource-ref = "true" />
When I use the AbstractAnnotationConfigDispatcherServletInitializer class to initialize Spring, my DataSource is created but when I try to catch the connection it gives me the following error:
java.sql.SQLException: Data source is closed
at org.apache.tomcat.dbcp.dbcp2.BasicDataSource.createDataSource (BasicDataSource.java:2049)
I try to create the DataSource with this function:
#Bean (name = "dataSource")
public DataSource dataSource (Environment env) throws NamingException
{
DataSource datasource = null;
try {
JndiDataSourceLookup lookup = new JndiDataSourceLookup ();
datasource = lookup.getDataSource ("jdbc/yages");
datasource.getConnection ();
return datasource;
} catch (SQLException ex) {
ex.printStackTrace ();
}
return datasource;
}
It seems that the DataSource is created correctly, but the connection to the database seems to be closed.
However, if I use the DataSource, configure it through XML, it works well for me, which is why I assume that it is not a problem neither of the connection to the database nor of the configuration of Tomcat.
Any idea why the connection is closed?
Thank you.
I found the solution.
It is to create the Bean with this instruction
#Bean(name = "dataSource", destroyMethod = "")
The problem is that Spring to do an undeploy, destroy the Bean and close the connection. To avoid this, I have to change the default behavior of Spring.
I want to ask as currently I have my database properties like username and password inside the persistence layer in the intellij. But I want to place it somewhere outside so if someone wants to change the password or any configuration inside database he should not have to dig inside my current structure. Now my structure is persistence then main then resources and then dbconfig properties so is there any way I can do it.
You can create a file app.properties in your resources folder with all database information you need:
# Datasource details
testapp.db.driver = org.h2.Driver
testapp.db.url = jdbc:h2:mem:test
testapp.db.username = username
testapp.db.password = password
Then you can refer to it in your Java code as:
#Configuration
#PropertySource("app.properties")
public class DataConfig {
#Autowired
private Environment env;
#Bean
public DataSource dataSource() {
BasicDataSource ds = new BasicDataSource();
ds.setDriverClassName(env.getProperty("testapp.db.driver"));
ds.setUrl(env.getProperty("testapp.db.url"));
ds.setUsername(env.getProperty("testapp.db.username"));
ds.setPassword(env.getProperty("testapp.db.password"));
return ds;
}
}
I want to manage multiple DataSource using your Application Servers built-in features and access it using JNDI. I am using Spring boot with Spring JPA data.
I am able to configure the application.properties for single datasource:
spring.datasource.jndi-name=jdbc/customers
And my configuration in context.xml file as below:
<Resource name="jdbc/customer" auth="Container" type="javax.sql.DataSource"
maxTotal="100" maxIdle="30" maxWaitMillis="10000"
username="root" password="root" driverClassName="com.mysql.jdbc.Driver"
url="jdbc:mysql://localhost:3306/customer"/>
Everything works fine.
But when I am unable to configure for two datasource.
I am sure on the configuration in context.xml file:
<Resource name="jdbc/customer" auth="Container" type="javax.sql.DataSource"
maxTotal="100" maxIdle="30" maxWaitMillis="10000"
username="root" password="root" driverClassName="com.mysql.jdbc.Driver"
url="jdbc:mysql://localhost:3306/customer"/>
<Resource name="jdbc/employee" auth="Container" type="javax.sql.DataSource"
maxTotal="100" maxIdle="30" maxWaitMillis="10000"
username="root" password="root" driverClassName="com.mysql.jdbc.Driver"
url="jdbc:mysql://localhost:3306/employee"/>
I am in doubt about the application.properties file configuration.
I tried the below options with no success:
spring.datasource.jndi-name=jdbc/customers,jdbc/employee
Please let me know any details on Spring boot with JNDI for multiple data source. I was looking for this configuration for days now.
Second Trial As per Spring Boot Documentation
spring.datasource.primary.jndi-name=jdbc/customer
spring.datasource.secondary.jndi-name=jdbc/project
Configuration class.
#Bean
#Primary
#ConfigurationProperties(prefix="datasource.primary")
public DataSource primaryDataSource() {
return DataSourceBuilder.create().build();
}
#Bean
#ConfigurationProperties(prefix="datasource.secondary")
public DataSource secondaryDataSource() {
return DataSourceBuilder.create().build();
}
The application does not get started. Though the tomcat server is getting started. No errors are printed in the log.
Third Trial: With JndiObjectFactoryBean
I have the below application.properties
spring.datasource.primary.expected-type=javax.sql.DataSource
spring.datasource.primary.jndi-name=jdbc/customer
spring.datasource.primary.jpa.database-platform=org.hibernate.dialect.MySQL5Dialect
spring.datasource.primary.jpa.show-sql=false
spring.datasource.primary.jpa.hibernate.ddl-auto=validate
spring.datasource.secondary.jndi-name=jdbc/employee
spring.datasource.secondary.expected-type=javax.sql.DataSource
spring.datasource.secondary.jpa.database-platform=org.hibernate.dialect.MySQL5Dialect
spring.datasource.secondary.jpa.show-sql=false
spring.datasource.secondary.jpa.hibernate.ddl-auto=validate
And the below java configuration:
#Bean(destroyMethod="")
#Primary
#ConfigurationProperties(prefix="spring.datasource.primary")
public FactoryBean primaryDataSource() {
return new JndiObjectFactoryBean();
}
#Bean(destroyMethod="")
#ConfigurationProperties(prefix="spring.datasource.secondary")
public FactoryBean secondaryDataSource() {
return new JndiObjectFactoryBean();
}
But still getting error:
Related cause: org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'primaryDataSource' defined in class path resource [com/web/initializer/MvcConfig.class]: Invocation of init method failed; nested exception is javax.naming.NameNotFoundException: Name [jdbc/customer] is not bound in this Context. Unable to find [jdbc].
Related cause: org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'secondaryDataSource' defined in class path resource [com/web/initializer/MvcConfig.class]: Invocation of init method failed; nested exception is javax.naming.NameNotFoundException: Name [jdbc/employee] is not bound in this Context. Unable to find [jdbc].
at org.springframework.boot.context.embedded.EmbeddedWebApplicationContext.onRefresh(EmbeddedWebApplicationContext.java:133)
at org.springframework.context.support.AbstractApplicationContext.refresh(AbstractApplicationContext.java:474)
at org.springframework.boot.context.embedded.EmbeddedWebApplicationContext.refresh(EmbeddedWebApplicationContext.java:118)
at org.springframework.boot.SpringApplication.refresh(SpringApplication.java:686)
at org.springframework.boot.SpringApplication.run(SpringApplication.java:320)
at org.springframework.boot.context.web.SpringBootServletInitializer.run(SpringBootServletInitializer.java:117)
at org.springframework.boot.context.web.SpringBootServletInitializer.createRootApplicationContext(SpringBootServletInitializer.java:108)
at org.springframework.boot.context.web.SpringBootServletInitializer.onStartup(SpringBootServletInitializer.java:68)
at org.springframework.web.SpringServletContainerInitializer.onStartup(SpringServletContainerInitializer.java:175)
Update:
Trial using the below properties file:
spring.datasource.primary.expected-type=javax.sql.DataSource
spring.datasource.primary.jndi-name=java:comp/env/jdbc/customer
spring.datasource.secondary.jndi-name=java:comp/env/jdbc/employee
spring.datasource.secondary.expected-type=javax.sql.DataSource
spring.jpa.database-platform=org.hibernate.dialect.MySQL5Dialect
spring.jpa.show-sql=false
spring.jpa.hibernate.ddl-auto=validate
It creates all the tables in customer schema, but fails trying to find the other tables also.(from the second schema)
This is the solution for your third trial a little bit modified.
Consider this solution (Spring Boot 1.3.2):
application.properties file:
spring.datasource.primary.jndi-name=java:/comp/env/jdbc/SecurityDS
spring.datasource.primary.driver-class-name=org.postgresql.Driver
spring.datasource.secondary.jndi-name=java:/comp/env/jdbc/TmsDS
spring.datasource.secondary.driver-class-name=org.postgresql.Driver
spring.jpa.properties.hibernate.dialect=org.hibernate.dialect.PostgreSQL9Dialect
spring.jpa.show-sql=false
Configuration:
#Configuration# EnableConfigurationProperties
public class AppConfig {
#Bean# ConfigurationProperties(prefix = "spring.datasource.primary")
public JndiPropertyHolder primary() {
return new JndiPropertyHolder();
}
#Bean# Primary
public DataSource primaryDataSource() {
JndiDataSourceLookup dataSourceLookup = new JndiDataSourceLookup();
DataSource dataSource = dataSourceLookup.getDataSource(primary().getJndiName());
return dataSource;
}
#Bean# ConfigurationProperties(prefix = "spring.datasource.secondary")
public JndiPropertyHolder secondary() {
return new JndiPropertyHolder();
}
#Bean
public DataSource secondaryDataSource() {
JndiDataSourceLookup dataSourceLookup = new JndiDataSourceLookup();
DataSource dataSource = dataSourceLookup.getDataSource(secondary().getJndiName());
return dataSource;
}
private static class JndiPropertyHolder {
private String jndiName;
public String getJndiName() {
return jndiName;
}
public void setJndiName(String jndiName) {
this.jndiName = jndiName;
}
}
}
And then you can follow guide http://docs.spring.io/spring-data/jpa/docs/1.3.0.RELEASE/reference/html/jpa.repositories.html to use your datasources with jpa repositories.
You could use a plain JndiObjectFactoryBean for this. Simply replace the DataSourceBuilder with a JndiObjectFactoryBean should do the trick.
Java configuration
#Bean(destroyMethod="")
#Primary
#ConfigurationProperties(prefix="datasource.primary")
public FactoryBean primaryDataSource() {
return new JndiObjectFactoryBean();
}
#Bean(destroyMethod="")
#ConfigurationProperties(prefix="datasource.secondary")
public FactoryBean secondaryDataSource() {
return new JndiObjectFactoryBean();
}
Properties
datasource.primary.jndi-name=jdbc/customer
datasource.primary.expected-type=javax.sql.DataSource
datasource.secondary.jndi-name=jdbc/project
datasource.secondary.expected-type=javax.sql.DataSource
You can set every property of the JndiObjectFactoryBean using the #ConfigurationProperties annotation. (See the expected-type I added, but you could also set cache or lookup-on-startup etc.).
Note: when doing a JNDI lookup set the destroyMethod to an "" else you might get the situation that when the application is shutdown your JNDI resource is getting closed/shutdown as well. This is not something you want in a shared environment.
It works for me and contains less code
#Configuration
public class Config {
#Value("${spring.datasource.primary.jndi-name}")
private String primaryJndiName;
#Value("${spring.datasource.secondary.jndi-name}")
private String secondaryJndiName;
private JndiDataSourceLookup lookup = new JndiDataSourceLookup();
#Primary
#Bean(destroyMethod = "") // destroy method is disabled for Weblogic update app ability
public DataSource primaryDs() {
return lookup.getDataSource(primaryJndiName);
}
#Bean(destroyMethod = "") // destroy method is disabled for Weblogic update app ability
public DataSource secondaryDs() {
return lookup.getDataSource(secondaryJndiName);
}
}
In my case, when i start my application using Spring Boot App, the database configurations are read on application-dev.properties, when I publish on tomcat, using datasources, was necessary add a validation to check if my profile is prod, in this case, i do a jndi lookup
#Bean(name = "dsName")
#ConfigurationProperties("ds.datasource.configuration")
public DataSource dataSource(#Qualifier("dsProperties") DataSourceProperties db1DataSourceProperties)
{
if(Arrays.asList(environment.getActiveProfiles()).contains("prod"))
{
final JndiDataSourceLookup dataSourceLookup = new JndiDataSourceLookup();
return dataSourceLookup.getDataSource("java:comp/env/jdbc/DS1");
}
else
{
return db1DataSourceProperties.initializeDataSourceBuilder().type(HikariDataSource.class).build();
}
}
The concise way I get success and explore more
set up many jndi resources in external tomcat , that you can start/stop in eclipse.Note- double click the tomcat in eclipse and select use workspace metedata , means dont deploy the app to tomcat webapp folder. Add jndi resources in respective eclipse server files( context.xml - ResourceLink, server.xml - Resource , web.xml - resource-ref).
no need to set spring.datasource.* in application.properties. since jndi-contest which is a datasource type( i.e. type="javax.sql.DataSource") is exported to external server.
in SpringBootApplication annotated class , create the datasource beans from all the jndi resources(those setup as per #1) through jndi lookup
#Bean(name = "abcDataSource")
public DataSource getAbcDataSource() {
JndiDataSourceLookup dataSourceLookup = new JndiDataSourceLookup();
DataSource dataSource = dataSourceLookup.getDataSource("java:comp/env/jdbc/abcDataSource");
return dataSource;
}
if spring jdbc used in your project then provide the above datasource to create a jdbcTemplate bean
#Bean(name = "jdbcAbcTemplate")
public JdbcTemplate abcJdbcTemplate(#Lazy #Qualifier("abcDataSource")
DataSource refDS)
{
return new JdbcTemplate(refDS);
}
just autowire a property of DataSource type and get systemout its details to explore more.
While the above answers are good I am going to add one more to illustrate a deadly gotcha if mixing jndi and full data connection configuration. In a typical development environment you may fully qualify the database connection in your local dev environment then use jndi as you push to qa, etc.. Your
application .properties looks like so:
spring.datasource.url=jdbc:sqlserver://server.domain.org:1433;databaseName=dbxx
spring.datasource.username=userxxyyzz
spring.datasource.password=passxxyyzz
spring.datasource.platform=mssql
spring.datasource.driverClassName=com.microsoft.sqlserver.jdbc.SQLServerDriver
and application-qa.properties like so:
spring.datasource.jndi-name=java:jboss/datasources/dbxx
The problem arises when you have to define your own beans to have multiple datasources. If you use the default Spring managed datasource then it automatically detects jndi vs fully qualified connection and returns a datasource with no change needed in the application code. If you define your own datasource it no longer does this. If you have application.properties like so:
spring.custom.datasource.url=jdbc:sqlserver://server.domain.org:1433;databaseName=dbxx
spring.custom.datasource.username=userxxyyzz
spring.custom.datasource.password=passxxyyzz
spring.custom.datasource.platform=mssql
spring.custom.datasource.driverClassName=com.microsoft.sqlserver.jdbc.SQLServerDriver
and application-qa.properties like so:
spring.custom.datasource.jndi-name=java:jboss/datasources/dbxx
with a datasource bean like so, as suggested in Spring docs https://docs.spring.io/spring-boot/docs/2.1.11.RELEASE/reference/html/howto-data-access.html
#Primary
#Bean(name="customDataSourcePropertiesBean")
#ConfigurationProperties("spring.custom.datasource")
public DataSourceProperties customDataSourceProperties() {
return new DataSourceProperties();
}
#Primary
#Bean(name="customDataSourceBean")
#ConfigurationProperties("spring.custom.datasource")
public HiakriDataSource customDataSource(#Qualifier("customDataSourcePropertiesBean") DataSourceProperties properties) {
return properties.initializeDataSourceBuilder().type(HikariDataSource.class).build();
}
This datasource builder does not attempt to read the jndi config in application-qa.properties and silently fails back to application.properties returning the WRONG database connection. Resolution is fairly simple - test which environment you are in and customize the type of database connection created. Debugging this was a bear, as the symptom was that the app appeared to be ignoring application-qa.properties. I share to spare others the pain. Add spring.profiles.active=qa etc. to your properties files to know which environment you are in then:
#Value("${spring.profiles.active}")
String profile;
#Value("${spring.custom.jndi-name}")
String jndi;
#Primary
#Bean(name="customDataSourcePropertiesBean")
#ConfigurationProperties("spring.custom.datasource")
public DataSourceProperties customDataSourceProperties() {
return new DataSourceProperties();
}
#Primary
#Bean(name="customDataSourceBean")
#ConfigurationProperties("spring.custom.datasource")
public DataSource customDataSource(#Qualifier("customDataSourcePropertiesBean") DataSourceProperties properties) {
if(profile.equals("localhost")) {
return DataSourceBuilder
.create()
.username(properties.getDataUsername())
.password(properties.getPassword())
.url(properties.getUrl())
.driverClassName(properties.getDriverClassName())
.build();
}else {
JndiDataSourceLookup dataSourceLookup = new JndiDataSourceLookup();
return dataSourceLookup.getDataSource(jndi);
}
}
I'm using Oracle Spatial, and I have a table with an SDO_GEOMETRY field.
The table is mapped to a JPA entity. I want to have the SDO_GEOMETRY field mapped to a java oracle.spatial.geometry.JGeometry type.
I figured I should use a JPA Converter and to convert to and from java.sql.Struct (or maybe oracle.sql.STRUCT).
The problem is the JGeometry method that converts to Struct, JGeometry.storeJS(Connection conn, JGeometry geom), wants the jdbc connection as a parameter.
The spring EntityManagerFactory is configured with the persistence unit name, the persistence unit contains the data source jndi name, and the data source is defined in tomcat, as a connection pool.
Any idea on how I can get the Connection in the converter ?
This what I want to achieve:
#Converter(autoApply = true)
public class GeometryConverter implements AttributeConverter<JGeometry, Struct> {
#Override
public Struct convertToDatabaseColumn(JGeometry geometry) {
// How to get this connection ?
return JGeometry.storeJS(connection, geometry);
}
#Override
public JGeometry convertToEntityAttribute(Struct struct) {
try {
return JGeometry.loadJS(struct);
} catch (SQLException e) {
throw new RuntimeException("Failed to convert geometry", e);
}
}
}
I am using Spring 4, spring-data-jpa 1.6, Hibernate 4, Tomcat 8, Oracle 12c.
Updated with more info:
Spring configuration:
#Configuration
#EnableJpaRepositories("com.package.repository")
#EnableTransactionManagement
#ComponentScan("com.package")
public class SpringConfig {
#Bean(name = "entityManagerFactory", destroyMethod = "destroy")
public LocalContainerEntityManagerFactoryBean entityManagerFactory() {
LocalContainerEntityManagerFactoryBean emf = new LocalContainerEntityManagerFactoryBean();
emf.setPersistenceUnitName("persistence-unit");
return emf;
}
#Bean(name = "transactionManager")
public JpaTransactionManager getTransactionManager() {
return new JpaTransactionManager();
}
}
If you use spring, and you need to use both JPA and JDBC, you should :
construct a datasource bean and make connection pooling there (or get if from jndi(*))
inject that datasource in one on the spring helpers for building the EntityManagerFactory (such as LocalContainerEntityManagerFactoryBean)
inject that datasource in any bean where you want to do direct JDBC
That way you can use JPA for your normal DAO, and still have access to JDBC in special parts - without a too strong dependance of the internals of your JPA provider.
EDIT:
(*) If your datasource is defined by a jndi name, all is fine. Expose it as a bean (ref)
If using Spring's XML schema based configuration, setup in the Spring context like this:
<xmlns:jee="http://www.springframework.org/schema/jee"
xsi:schemaLocation="http://www.springframework.org/schema/jee
http://www.springframework.org/schema/jee/spring-jee-3.2.xsd">
...
<jee:jndi-lookup id="dbDataSource"
jndi-name="jdbc/DatabaseName"
expected-type="javax.sql.DataSource" />
Alternatively, setup using simple bean configuration like this:
<bean id="dbDataSource" class="org.springframework.jndi.JndiObjectFactoryBean">
<property name="jndiName" value="java:comp/env/jdbc/DatabaseName"/>
</bean>
As you are using a JpaTransactionManager, there will not be any problem because as specified in spring javadoc This transaction manager also supports direct DataSource access within a transaction (i.e. plain JDBC code working with the same DataSource). This allows for mixing services which access JPA and services which use plain JDBC (without being aware of JPA)! provided you get your Connection through DataSourceUtils.getConnection(javax.sql.DataSource)
EDIT2 :
Ok now the only problem is how to access a singleton bean from a non bean object. A simple way to solve it is to create a holder singleton bean with a static method.
#Bean
public class DataSourceHolder implements InitializingBean {
private DataSource dataSource;
private static DataSourceHolder instance;
public static DataSource getDataSource() {
return instance.dataSource;
}
#Autowired
public void setDataSource(DataSource dataSource) {
this.dataSource = dataSource;
}
#Override
public void afterPropertiesSet() throws Exception {
DataSourceHolder.instance = this;
}
}
Then in any object, be it a bean or not, you can use
DataSource ds = DataSourceHolder.getDataSource();
Connection con = DataSourceUtils.getConnection(ds);
That would be tricky. Purely from the jpa api. You will have to dig into the specific provider implementation and get hold of the DataSource object or the PersistenceUnitInfo object.
From here you can get hold of the Connection object.
Now depending on which environment you are working. If you are on an JavaEE environment, and you inject EntityManager or its EntityManagerFactory, there is no guarantee that the return instance is an instance of the provider own implementation as this may just be a proxy that implements the interface, and hence no relation to the provider's own implementation.
On JSE environment, since you are the one creating the EntityManagerFactory from Persistence.createEntityManagerFactory(), you could tweak the provider in order to get the Connection.