MySQL Bugs: #2111: add a rs.copy(busObject) method

Bug #2111	add a rs.copy(busObject) method
Submitted:	14 Dec 2003 5:44	Modified:	15 Dec 2003 7:06
Reporter:	Ralf Hauser	Email Updates:
Status:	Closed	Impact on me:	None
Category:	Connector / J	Severity:	S4 (Feature request)
Version:	0312 nightly	OS:	Any (any)
Assigned to:	Mark Matthews	CPU Architecture:	Any

Description:
Recently, I learned to appreciate the existence of 
org.apache.commons.beanutils.PropertyUtils or rather its
PropertyUtils.copyProperties (http://jakarta.apache.org/commons/beanutils/api/org/apache/commons/beanutils/PropertyUtils...))

Similarly, it would be great to have the equivalent to have the content from a ResultSet copied into a  business Object.
Additional parameter my be hints on the naming scheme used between Java and MySQL:
For example, a Java "private boolean isActive" may be a mysql "tinyint isactive" but also "is_active"

I guess this is a little bit less efficient in execution time, but it may save the programmer so many (, possibly error-prone) single copy statements that at least for a prototype implementation, it is very worthwhile to have such a method.

What do you think?

How to repeat:
.

Suggested fix:
.

There are many fine frameworks that already do this, like Hibernate, OJB, Torque, , JDO, etc. There is also talk in the JDBC expert group for JDBC-4.0 to have some rudimentary object-relational-mapping support. Until there is a standard for JDBC, I recommend you investigate one of the existing O-R frameworks out there, because it is unlikely we will add a non-standard method of doing this to the JDBC driver.

thx for the hint.
But don't these all add much more overhead than just the copy-related one?
Wouldn't one loose the control over the exact sql statement when using those?

Object-relational mapping is a complex topic, and even a 'simple' mapper like you propose has a lot of caveats once you start talking about making the data 'flow' in both directions (in and out of the database), as well as managing state correctly. 

If you take a look at many of the tools I mentioned, many of them can in some cases be _faster_ than straight JDBC, because of the caching they can do, as well as writing more efficient queries than most developers can (Hibernate is one example of this).

Mark, just stumbled over 
http://jakarta.apache.org/commons/dbutils/apidocs/org/apache/commons/dbutils/handlers/Bean...
and http://jakarta.apache.org/commons/dbutils/apidocs/org/apache/commons/dbutils/handlers/Bean...
Do you have any experience with those?
Could they do what I want?
Rgds r.