Index: examples.xml =================================================================== RCS file: /home/cvspublic/jakarta-commons/dbutils/xdocs/examples.xml,v retrieving revision 1.6 diff -u -r1.6 examples.xml --- examples.xml 19 Mar 2004 00:25:39 -0000 1.6 +++ examples.xml 27 Nov 2004 20:22:43 -0000 @@ -149,6 +149,39 @@ your application. The provided implementation delegates datatype conversion to the JDBC driver.

+ +

Using the BeanHandler potentially adds a new requirement - it +requires that the names of the column in the database match the +property names defined of the JavaBean. In the above example, that +means that if the columns in the database table "Person" were +"PERSON_ID", "NAME", "FIRST_NAME", and "LAST_NAME", then the +properties of the JavaBean Person.java must be "person_id", +"name", "first_name", and "last_name". (Note that the case does +not have to match.) Such a situation is a real world reality in +that many DBAs, er, encourage a naming schema that contains +only capital letters and where each word is separated by the +underscore character.

+ +

There are at least three potential solutions to this issue: +

    +
  1. Change the database column names to match the JavaBean property names.
    2. +
  2. Change the JavaBean property names to match the database column names.
    3. +
  3. Modify the SELECT statement to use the AS SQL keyword
    4. +
+

+ +

The first two options presented are usually not feasible. The +third option has you change the SQL statement. So, given a JavaBean +with properties "person_id", "name", "first_name", and "last_name", +the SELECT statement would need to be changed to SELECT +PERSON_ID AS personId, NAME, FIRST_NAME AS firstName, LAST_NAME +AS lastName FROM Person WHERE name=?. (Note that the AS +keyword is not used with the column named NAME since it matches the +JavaBean property name.)

+ +

A fourth option, as mentioned above, would involve implementing +a customized RowProcessor

+