© 2015 IBM Corporation

AAI-2235OpenJPA and EclipseLinkUsage Scenarios Explained

Kevin Sutter, STSM

WebSphere Java EE and JPAArchitect

Agenda

• OpenJPA and EclipseLink

• JPA version comparisons

• OpenJPA to EclipseLink Migration

• High level concepts

• OpenJPA to EclipseLink Investigation

• Specific differences

• Summary

• Q & A

2

OpenJPA andEclipseLink

OpenJPA and EclipseLink

• Apache OpenJPA

• Basis for WebSphere’s JPA solution for JPA 1.0 and JPA2.0

• EclipseLink

• JPA Reference Implementation – “gospel”

• Basis for WebSphere Liberty’s JPA solution for JPA 2.1(Beta)

• Moving forward…

• OpenJPA will continue to be supported in WAS for many,many years

• At least until JPA 2.0 is deprecated…

• Both WebSphere Full Profile and Liberty Profile

4

Multiple JPA Providers

• Do I need to switch providers?

• Absolutely not!

• If you are happy with the current OpenJPA offering,there is no need to move to EclipseLink

• If you want to use the JPA 2.1 API or features, you *may*need to change…

• Many of the JPA 2.1 features have correspondingOpenJPA proprietary solutions

• Apache OpenJPA (ie. JPA 2.0) will “play nice” with other JavaEE 7 features

• Allows for an easier, more gradual JPA migration

5

Key JPA 2.1 Features available in OpenJPA

• Schema Generation (JPA 2.1 Spec, Section 9.4)

• Generates DDL or interacts directly with database todefine table schemas based on JPA Entity definitions

• Similar functionality provided by OpenJPA’s schemamapper

• Entity Graphs (JPA 2.1 Spec, Section 3.7)

• Allows for specified fetching or processing of a graph ofEntity objects

• Similar functionality provided by OpenJPA’s FetchPlanand FetchGroup

• Stored Procedure Queries (JPA 2.1 Spec, Section 3.10.17)

• Ability to invoke database stored procedures

• Similar functionality provided by OpenJPA’s Queryinvocation

6

Additional JPA 2.1 Features available in OpenJPA

• Basic Attribute Type Conversion (JPA 2.1 Spec, Section 3.8)

• Similar functionality support provided by OpenJPA’sExternalizer feature

• @Index and @ForeignKey annotations (JPA 2.1 Spec, Sections 11.1.19 and

11.1.23)

• Similar functionality provided by OpenJPA’s annotations

• Unwrap utility methods for EntityManager, Cache, etc (JPA 2.1 Spec

Sections 3.1.1 and 7.10)

• Similar functionality provided by OpenJPA’simplementation -- EntityManagerImpl.unwrap() andOpenJPAPersistence.cast()

• Object construction when mapping results from native SQL (JPA

2.1 Spec, Section 3.10.16.2.2)

• Similar functionality provided by OpenJPA’s internalResultShape object

7

Additional JPA 2.1 Features NOT available in OpenJPA

• Criteria API Updates (JPA 2.1 Spec, Sections 6.5.15 and 6.5.7)

• Bulk update/delete

• Support for TREAT, ON, and FUNCTION operators

• Unsynchronized Persistence Contexts (JPA 2.1 Spec, Section 7.6.1)

• Provides more control over when Persistence Contextsare synchronized with a transaction. Useful for mobileapplications.

• EntityListeners and CDI (JPA 2.1 Spec, Section 3.5.1)

• EntityListeners can now use CDI for injection of objects

• JPQL Updates for JPA 2.1 (JPA 2.1 Spec, Chapter 4)

• Several extensions to JPQL in support of the otherfeatures

• OpenJPA’s JPQL will stay at the JPA 2.0 level

8

Feature Comparison Chart

9

Key JPA 2.1 Features Available in current OpenJPA solution?

Schema Generation Yes, SynchronizeMappings

Entity Graphs Yes, FetchPlans and FetchGroups

Stored Procedures Yes, Query interface

Basic Attribute Type Conversion Yes, Externalizer (and Factory)

@Index and @ForeignKey annotations Yes, proprietary @Index and @ForeignKeyannotations

Unwrap utility functions Yes, EntityManagerImpl.unwrap() andOpenJPAPersistence.cast()

Object construction when mapping results withnative SQL

Yes, ResultShape (internal OpenJPAimplementation)

Criteria API updates No

Unsynchronized PersistenceContexts No

EntityListeners and CDI No

JPQL Updates in support of JPA 2.1 No

Bottom Line

• YOU DO NOT NEED TO MIGRATE

• Our Advice

• Continue using OpenJPA for your existing applications

• Use EclipseLink for your new applications

• But… If you want to use EclipseLink with existing applications,the following slides will discuss what to look out for whenmigrating existing applications

10

OpenJPA toEclipseLinkMigration

Cache

• DataCache

• OpenJPA's L2 cache is disabled out of the box.EclipseLink's L2 cache is enabled by default.

• If you are migrating an application that isn't using theOpenJPA cache, it is highly recommended to disable theEclipseLink cache using:

<shared-cache-mode>NONE</shared-cache-mode>

• Query (Results) Cache

• If QueryCache is enabled for OpenJPA, it will cache allNamed Queries and JPQL Statement Results

• EclipseLink only caches Named Queries Results

• More information on EclipseLink QueryCache here

12

Weaving/enhancement/transformation

• OpenJPA: Entities must be enhanced

• EclipseLink: Entities do not have to be enhanced

• Documented missing features (lazy loading, performancegains, etc)

• WebSphere container hooks automatically weave/enhanceEntities for both OpenJPA and EclipseLink

• Applications packaged with entities pre-enhanced by OpenJPAmust be recompiled/repackaged

• Removes OpenJPA specific enhancement bytecode

13

OpenJPA toEclipseLinkInvestigation

Investigation

• Goal

• Find differences in behavior between OpenJPA andEclipseLink

• OpenJPA → EclipseLink migration issues

• Data

• Entities from OpenJPA's testing framework (>700 entities)

• Process

• Auto generate databases tables using both providers

• Use tables created by OpenJPA for verification

• Database: DB2

• Results

• Errors (fix EclipseLink)

• Table differences

15

Private Accessor Methods

• JPA 2.1 Spec: property accessor methods must be public orprotected

• OpenJPA: ignores “private” fields

• EclipseLink: persists “private” fields

• Migrating issue: missing database column error

• Solution: Add @Transient to the method

– Forces EclipseLink to ignore the field

16

Add @Transient

Getter/Setter accessor methods

• Annotated getter method with no corresponding setter

• OpenJPA: silently ignores the field

• EclipseLink: throws an error during entity validation

• Solution: remove annotation17

Missing setVersion method

No-arg Constructor

• JPA 2.1 Spec: an entity must have a no-arg constructor

• OpenJPA: if missing, generates a no-arg constructor.

• EclipseLink: throws an error only if the no-arg constructoris missing, but other constructors exist.

• Solution: Add an empty no-arg constructor

18

Add no-argconstructor

Unannotated Collections

• Unannotated fields of type java.util.Collection or any of it’ssubinterfaces (List<E>, Queue<E>, Set<E>, etc)

• OpenJPA: ignores fields

• EclipseLink: persists fields

• Solution: Add @javax.persistence.Transient

19

Add @Transient

@javax.persistence.ElementCollection

• @ElementCollection generates a Separate table

• Two columns: id, value

• Different value column name

• OpenJPA: 'element'

– CREATE TABLE EntityA_LISTOFSTRINGS (ENTITYA_IDINTEGER, element VARCHAR(254))

• EclipseLink: field name

– CREATE TABLE EntityA_LISTOFSTRINGS (EntityA_IDINTEGER, LISTOFSTRINGS VARCHAR(255))

• Solution

20

Add @Column

Behavior Mismatch: @GeneratedValue

• Generate values for primary keys

• Strategies: TABLE, SEQUENCE, IDENTITY, AUTO

• No Strategy or 'AUTO': JPA provider picks the strategy

– Good news: OpenJPA and EclipseLink default to'TABLE' strategy

– Bad news: different table schemas are created

• Solution

• Change entity to explicitly use 'TABLE' strategy

• Map to existing OpenJPA table

21

Java Persistence Query Language (JPQL)

• @NamedQuery: specify named query in JPQL

• OpenJPA: ignores invalid unused JPQL statements

• EclipseLink: validates all JPQL statement by default

• Solutions

• Solution 1: fix JPQL statements

• Solution 2: set eclipselink's invalid JPQL toleranceproperty to true

22

Unique Names

• Tables, NamedQuery, SQLResultSetMapping must have uniquenames

• OpenJPA behavior is undefined

– Example: tables with same name are combined

• EclipseLink throws an error during entity validation

• Solution: review entities (manual process)

23

Summary

Summary

• Do I want to utilize my existing database tables?

• Modifications to the entities may be necessary

• Database tables previously created (by OpenJPA) maynot match EclipseLink’s generated tables

• OpenJPA may have been more lenient with specviolations than Eclipselink is

• Recommendation

• Continue using OpenJPA for existing applications

• Migrate to EclipseLink only for new applications

25

Resources

Tools: Integrity Checker

• EclipseLink provides a way to verify descriptor's metadata againstdatabase metadata

• Setup the integrity checker using a session customizer (link)

• Catches missing tables or columns without having to persist

• Example: private getter methods

• setShouldCatchExceptions(true) - find all entities with missingcolumns

• Cons: doesn't catch everything or maybe anything…

• Failed to catch @ElementColumn resulting in a different columnname in the separate table, @GeneratedValue looking for adifferent table

• Doesn't compare column types

• Better than nothing, but it missed many common issues

27

Tools: Migration toolkit

• Migration toolkit

• Detect scenarios that result in different behavior betweenproviders

• Alert users when migrating from OpenJPA to EclipseLink

28

Useful Links

• EclipseLink JPA provider for Liberty profile article

https://developer.ibm.com/wasdev/blog/2014/05/28/eclipselink-jpa-provider-liberty-profile/

• EclipseLink Migration Wiki Page

https://wiki.eclipse.org/EclipseLink/Examples/JPA/Migration/OpenJPA

29

Sampling of Related Sessions…

• AAI-1713A: Introduction to Java EE 7• Monday, 2-3pm, Mandalay Bay, Reef Ballroom E

• AAI-1641A: Introduction to Web Sockets• Monday, 5-6pm, Mandalay Bay, Reef Ballroom E

• AAI-1313A: Agile Development Using Java EE 7 with WebSphere Liberty Profile (LAB)• Tuesday, 8-10am, Mandalay Bay, South Seas Ballroom D

• AAI-2236A: Using the New Java Concurrency Utilities with IBM WebSphere• Tuesday, 2-3pm, Mandalay Bay, Reef Ballroom D

• AAI-2235A: OpenJPA and EclipseLink Usage Scenarios Explained• Wednesday, 5:30-6:30pm, Mandalay Bay, Surf Ballroom A

• AAI-1610A: Configuring IBM WebSphere Application Server for Enterprise MessagingNeeds• Wednesday, 5:30-6:30pm, Mandalay Bay, Surf Ballroom E

• AAI-3085A: Don’t Wait! Develop Responsive Applications with Java EE7 Instead• Thursday, 10:30-11:30am, Mandalay Bay, Lagoon L

30

Notices and DisclaimersCopyright © 2015 by International Business Machines Corporation (IBM). No part of this document may be reproduced ortransmitted in any form without written permission from IBM.

U.S. Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract withIBM.

Information in these presentations (including information relating to products that have not yet been announced by IBM) has beenreviewed for accuracy as of the date of initial publication and could include unintentional technical or typographical errors. IBMshall have no responsibility to update this information. THIS DOCUMENT IS DISTRIBUTED "AS IS" WITHOUT ANY WARRANTY,EITHER EXPRESS OR IMPLIED. IN NO EVENT SHALL IBM BE LIABLE FOR ANY DAMAGE ARISING FROM THE USE OFTHIS INFORMATION, INCLUDING BUT NOT LIMITED TO, LOSS OF DATA, BUSINESS INTERRUPTION, LOSS OF PROFITOR LOSS OF OPPORTUNITY. IBM products and services are warranted according to the terms and conditions of theagreements under which they are provided.

Any statements regarding IBM's future direction, intent or product plans are subject to change or withdrawal withoutnotice.

Performance data contained herein was generally obtained in a controlled, isolated environments. Customer examples arepresented as illustrations of how those customers have used IBM products and the results they may have achieved. Actualperformance, cost, savings or other results in other operating environments may vary.

References in this document to IBM products, programs, or services does not imply that IBM intends to make such products,programs or services available in all countries in which IBM operates or does business.

Workshops, sessions and associated materials may have been prepared by independent session speakers, and do notnecessarily reflect the views of IBM. All materials and discussions are provided for informational purposes only, and are neitherintended to, nor shall constitute legal or other guidance or advice to any individual participant or their specific situation.

It is the customer’s responsibility to insure its own compliance with legal requirements and to obtain advice of competent legalcounsel as to the identification and interpretation of any relevant laws and regulatory requirements that may affect the customer’sbusiness and any actions the customer may need to take to comply with such laws. IBM does not provide legal advice orrepresent or warrant that its services or products will ensure that the customer is in compliance with any law.

Notices and Disclaimers (con’t)

Information concerning non-IBM products was obtained from the suppliers of those products, their publishedannouncements or other publicly available sources. IBM has not tested those products in connection with thispublication and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBMproducts. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products.IBM does not warrant the quality of any third-party products, or the ability of any such third-party products tointeroperate with IBM’s products. IBM EXPRESSLY DISCLAIMS ALL WARRANTIES, EXPRESSED OR IMPLIED,INCLUDING BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR APARTICULAR PURPOSE.

The provision of the information contained herein is not intended to, and does not, grant any right or license under anyIBM patents, copyrights, trademarks or other intellectual property right.

•IBM, the IBM logo, ibm.com, Bluemix, Blueworks Live, CICS, Clearcase, DOORS®, Enterprise DocumentManagement System™, Global Business Services ®, Global Technology Services ®, Information on Demand, ILOG,Maximo®, MQIntegrator®, MQSeries®, Netcool®, OMEGAMON, OpenPower, PureAnalytics™, PureApplication®,pureCluster™, PureCoverage®, PureData®, PureExperience®, PureFlex®, pureQuery®, pureScale®, PureSystems®,QRadar®, Rational®, Rhapsody®, SoDA, SPSS, StoredIQ, Tivoli®, Trusteer®, urban{code}®, Watson, WebSphere®,Worklight®, X-Force® and System z® Z/OS, are trademarks of International Business Machines Corporation,registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or othercompanies. A current list of IBM trademarks is available on the Web at "Copyright and trademark information" at:www.ibm.com/legal/copytrade.shtml.

Thank YouYour Feedback is

Important!

Access the InterConnect 2015Conference CONNECT AttendeePortal to complete your sessionsurveys from your smartphone,

laptop or conference kiosk.

Backup

Differences in Column Types

• Mapping fields to columns

• Example: java.lang.String fields

– OpenJPA: VARCHAR(254)

– EclipseLink: VARCHAR(255)

• Mostly non-problematic

• Exception: java.util.Locale fields

– OpenJPA: VARCHAR(254)

– EclipseLink: BLOB(64000)

– Solution

– Use a type converter (introduced in JPA 2.1)

– BLOB (object) ↔ VARCHAR (database)

35