On 8/06/2005 3:58 PM, Matt Hogstrom wrote:
Regarding the it turns out that I was incorrect, the problem was not
the Date field but was a java.math.BigDecimal field. The database
column is a DOUBLE and the Java type is java.math.BigDecimal. I
worked with Jeremy earlier today on this and am working out a patch.
This brings up a question. In TranQL is there a framework to provide
database specific mappings and transforms? For instance DB2 and
Oracle both provide support for converting a DOUBLE to long, int,
short, byte, float, java.math.BigDecimal, boolean or
java.lang.String. I'm not sure about Derby. The point is that there
are extensions to common drivers that provide additional conversion
that simplifies life. From what I can see TranQL may be artificially
limiting by requiring more TypeConverters than needed based on the
JDBC driver that is being used.
I do not yet see your point: the implementation uses these type
conversions, no?
For instance, if a CMP field having the Java type BigDecimal is mapped
to a column having the SQL type DOUBLE, then:
* PrepareStatement.setBigDecimal() is used to set its value. As the
standard mapping from the Java type BigDecimal to the JDBC type is
NUMERIC, the driver will have to do the conversion if it supports it
(with lost of precision I think); and
* Resultset.getBigDecimal() is used to get its value. This is not the
recommended method for retrieving a DOUBLE JDBC type; yet it is still a
supported type conversion.
Also, with regard to doing JOINS and providing support for SELECT FOR
UPDATE different databases have different issues. For instance,
Oracle allows a SELECT FOR UPDATE in a JOIN and DB2 does not.
Although, DB2 has specialized predicated that would allow for an
UPDATE in a JOIN that became available in DB2 8.2. I believe sybase
does not support a SELECT FOR UPDATE semantic but can use a null
UPDATE command to obtain a database lock. I guess my question is does
TranQL allow for database specific query and update generators to be
plugged inthat will allow for maximal SQL optimization ?
I think that the response is yes. When a org.tranql.ql.Query instance is
being traversed to generate a SQL query, it is possible to apply minor
syntactic transformations.
For instance, the CONCAT method does not need to be escaped with Oracle
and does in the case of Derby. This means that this query:
SELECT field1 FROM myTable WHERE field2 = CONCAT('test', 'test')
needs to be written:
SELECT field1 FROM myTable WHERE field2 = {fn CONCAT('test', 'test')}
to be understood by the Derby JDBC driver. If you have a look to
DerbyGenerator, you will see that it is in charge of doring this minor
syntactic transformation.
If a more complex "transformation" needs to be applied, then the best
approach is to implement a specific EJBQLCompilerFactory.EJBQLCompiler.
This guy is used to transform a Query from the EJB domain to the SQL one.
For instance, it alllows us to generate two SQL queries (semantically
equivalent) for the same EJB-QL query. One of these two queries may be
specific to Oracle and the other one to DB2. You can have a look to
DerbyEJBQLCompilerFactory to get an idea of what is actually required.
Thanks,
Gianny
I'l let you know how my testing goes.
Cheers.
Matt
