- How is the JDBCModelReader supposed to handle default values? This is a problem espacially regarding the often used default date. I recognized that this seems to be handled as pure string.
I've noticed this also. I submitted a bug about this a while back. The default values in the the schema definition file aren't used if you don't set a value for that column. So, even if I define a default date, when I'm creating my data using DynaBeans I have to set "NOW" for the dates. It's annoying, and I'd like to fix it.
- The individual SQL types of a database are not handled correctly because the type mapping is done using the java.sql.Types definitions. I recognized this when reading from a MSSql database I got complains about the types "NVARCHAR" and "NTEXT". Is it a target of Sql to handle these types correct as well or do you suppose to support only pure JDBC features?
I'm not sure about this one. It really is up to the users to decide I guess. I would prefer an architecture which would allow customized type mappings to be plugged in for situations like these. But also, perhaps the project could be split into components, where the core code would sit in one area, and the database-specific things would sit in another.
For my Postgres db, I needed to find a way to alter the start value of my sequences, and also set permissions on my tables. There's no way to do this with [sql]. I didn't want to manually write a SQL script to handle this, so I wrote an XSL style sheet which generates it from the schema definition. It's pretty cool.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
