Hi, instead of the mor advanced RowSet, I'd use the classic ResultSet ...
> I'm not familiar with that class - can you tell me what it does, from a high > level? I'm not a Spring expert, but I'll try to explain in a simple way, as i know it :-) : Given the ResultSet, the developer write an implementation of the RowMapper, where he maps the returned results rows to his objects, or better, for any row he maps the returned columns to domain object fields ... this is a sample usage as anonymous inner classes (taken from here http://static.springframework.org/spring/docs/2.5.x/reference/jdbc.html ): public Object mapRow(ResultSet rs, int rowNum) throws SQLException { Actor actor = new Actor(); actor.setFirstName(rs.getString("first_name")); actor.setSurname(rs.getString("surname")); return actor; } in this way the developer knows what are fields in the ResultSet, and what to do with any field, right ? So in this way the mapping of rows on domain objects (in the sample it's the Actor class) is leaved to the developer, without having Class mappings that could lead to problems with various Serializers (like some data types not serializable for example by the JSONSerializer, like Date, etc) ... Maybe in this way the implementation could be simpler than transform any resultset field Class to another value ... What do you think ? > Not in the way I am currently envisioning it would work. It would support > wrapping a single ResultSet only (but maybe you could use multiple instances > to solve your use case)? Ok, this can work ... maybe for the future we could think to handle also the more generic case. Sandro
