SQLServer double-byte enablement

Question

We are using MS SQL Server and are trying to be fully double-byte enabled.
We have subclassed SQLServerDictionary to modify the SQLTypeMap to use the
column type NVARCHAR (rather than VARCHAR) for all appropriate string
column definitions. Now it seems that in order to actually insert/update a
string column, the string literal in the SQL statement must be preceeded
by an 'N' character (before the quotes). Is there a way I can use the
SQLServerDictionary subclass to modify the INSERT/UPDATE statements that
are generated in order to get this 'N' character in there?
Thanks,
David A. King

Answer 1

David-
You should be able to override the method toSQL (type, obj):
(see: http://www.solarmetric.com/Software/Documentation/2.2.4/javadoc/com/solarmetric/kodo/impl/jdbc/schema/dict/GenericDictionary.html#toSQL(int,%20java.lang.Object)
To read something like:
public Object toSQL (int type, Object val)
     switch (type)
          case SQLTypes.STRING:
          case SQLTypes.LENSTRING:
          case SQLTypes.CLOB:
               return "N" + super.toSQL (type, val);
          default:
               return super.toSQL (type, val);
David A. King <[email protected]> wrote:
We are using MS SQL Server and are trying to be fully double-byte enabled.
We have subclassed SQLServerDictionary to modify the SQLTypeMap to use the
column type NVARCHAR (rather than VARCHAR) for all appropriate string
column definitions. Now it seems that in order to actually insert/update a
string column, the string literal in the SQL statement must be preceeded
by an 'N' character (before the quotes). Is there a way I can use the
SQLServerDictionary subclass to modify the INSERT/UPDATE statements that
are generated in order to get this 'N' character in there?
Thanks,
David A. King--
Marc Prud'hommeaux [email protected]
SolarMetric Inc. http://www.solarmetric.com
Kodo Java Data Objects Full featured JDO: eliminate the SQL from your code