Hi Tobias, You might want to experiment with @Strategy or @Externalizer. See the links below.
http://openjpa.apache.org/builds/latest/docs/manual/manual.html#ref_guide_pc_extern http://openjpa.apache.org/builds/latest/docs/manual/manual.html#ref_guide_mapping_custom_field https://issues.apache.org/jira/browse/OPENJPA-1687 Regards, Milosz > > Hi, > > I have to access a legacy DB2 schema where text is stored in fixed length > CHAR columns. We are using OpenJPA 2.0 on WebSphere 7. I face two problems: > > a) when loading an entity, I always read a fixed length String of size N > with trailing spaces (that's how a CHAR column works). I trim the spaces > before they are display and edited in a web frontend. > > b) when updating the entity, the trimmed string values are updated because > of the missing trailing spaces, even when the string itself didn't change. > > This seems to me like a common problem when dealing with legacy schemas with > CHAR columns. Is there a common solution for this problem? Can I register > some kind of TypeMapper for CHAR columns? > > If a have a column like > > @Column(name = "PROD_VAR", length = 5) > private String produktVariante; > > this mapper should be able to honour the length attribute. After reading the > value from the ResultSet, the mapper just trims the string. When writing the > value, the string has to be padded up to length. Is this possible with > OpenJPA? > > TIA, > Tobias >
