- Integrate OData-JPA-Adapter into project
- Start OData-JPA-Adapter as standalone servlet
- Extend the OData-JPA-Adapter
- Use DTO's instead of JPA mapped entities
- OpenTypes and Map's
- Call Java methods as OData action
- Upload files via Multipart/form-data
- Detailed informations
- More hints
- Use Excel export
- EclipseLink, Hibernate and weaving (byte code enhancement)
- Migration guide
- Annotation overview
- Example code
Integration with JPA providers
Recommend is Eclipselink, because no major issues are known!
Hibernate
- Hibernate has a broken meta model, so a few workarounds are required:
- @IdClass with multiple key attributes are not supported
- Switching from one persistence unit to another in a dynamic way will result in a invalid meta model
- It cannot compare a java.lang.Short as java.lang.Number, so OData filter conditions like 'population lt 6000' will not work
- Hibernate build a default join column name (as in @JoinColumn#name() described) always with quotes as part of the column name
- The handling for columns declared without quoting and the default join column name derivation is not working (adaption of SQL table model by developer required as workaround)
- Weaving (byte code enhancement) seems to work only in very specific scenarios and must assumed as not usable.