Hi, I agree with your description - columns should be dropped. Looks like a bug or unimplemented feature. Don't you mind creating a JIRA [1] issue?
[1] http://issues.apache.org/jira/browse/OPENJPA Greetings, Milosz > > Steps to reproduce: > 1. Create the class: > @Entity > @Table(catalog = "testmysql", name = "city") > public class CityEntity { > public CityEntity() { > } > > private String m_cityname; > > @Basic > @Column(name = "CityName", length = 15) > public String getCityname() { > return m_cityname; > } > > public void setCityname(String cityname) { > m_cityname = cityname; > } > > private long m_Id; > > @Id > @GeneratedValue(strategy = GenerationType.AUTO) > public long getId() { > return m_Id; > } > > public void setId(long id) { > m_Id = id; > } > } > > 2. Create the persistance.xml: > <?xml version="1.0" encoding="UTF-8"?> > <persistence xmlns="http://java.sun.com/xml/ns/persistence"; version="1.0"> > <persistence-unit name="NewPeristenceUnit"> > > <provider>org.apache.openjpa.persistence.PersistenceProviderImpl</provider> > <class>testmysql.server.entity.CityEntity</class> > <properties> > <property name="openjpa.jdbc.SynchronizeMappings" > value="refresh(primaryKeys=true, ForeignKeys=true)"/> > <property name="openjpa.ConnectionURL" > value="jdbc:mysql://localhost:33306/testmysql2"/> > <property name="openjpa.ConnectionDriverName" > value="com.mysql.jdbc.Driver"/> > <property name="openjpa.ConnectionUserName" value="root"/> > <property name="openjpa.ConnectionPassword" value="12345"/> > <property name="openjpa.Log" value="DefaultLevel=TRACE, > Tool=INFO"/> > </properties> > </persistence-unit> > </persistence> > 3. Create the test method: > private String JPATest() { > EntityManagerFactory factory = null; > EntityManager manager = null; > try { > factory = > Persistence.createEntityManagerFactory("NewPeristenceUnit"); > manager = factory.createEntityManager(); > > CityEntity city = new CityEntity(); > > manager.getTransaction().begin(); > manager.persist(city); > manager.getTransaction().commit(); > } > catch{...} > > 4. Run the test. In specified table in database are now 2 columns id and > CityName. > 5. Now delete all strings in CityEntity.java so, that entity CityEntity > hasn't field CityName more: > @Entity > @Table(catalog = "testmysql", name = "city") > public class CityEntity { > public CityEntity() { > } > > private long m_Id; > > @Id > @GeneratedValue(strategy = GenerationType.AUTO) > public long getId() { > return m_Id; > } > > public void setId(long id) { > m_Id = id; > } > } > > 6. Run test again. > 7. Column CityName is still present in database table. But documentation for > MappingTool about refresh schemaAction says: > "refresh: Equivalent to retain, then add. > retain: Keep all schema components in the given XML definition, but drop the > rest from the database. This action never adds any schema components. " > Unnecessary columns should be deleted in database since the entity doesn't > contain mapped fields for them, when schemaAction is refresh. > -- > View this message in context: > http://n2.nabble.com/Reverse-Mapping-by-MappingTool-does-n-delete-unnecessary-columns-tp3748079p3748079.html > Sent from the OpenJPA Developers mailing list archive at Nabble.com. >
