Subject | New DataSource implementation |
---|---|
Author | rrokytskyy |
Post date | 2002-08-11T23:40:34Z |
Hi,
I have committed a FBSimpleDataSource and made FBWrappingDataSource a
subclass of FBSimpleDataSource.
This new DataSource implementation is really simple and does not
provide any connection pooling. The reason for this class is the
dependency of FBWrappingDataSource on EDU.oswego.* classes
(concurrent.jar). In my opinion, when people use this driver in
standalone applications, they use connection pooling already. So,
dependency on a third-party jar does not bring too much, only might
cause deployment problems.
This simple DataSource implementation is supposed to be included in a
standalone jar version (comming soon, David is investigating issue of
re-distributing needed javax.* classes). There still will be a version
of a jar file including all classes that are in firebirdsql.jar now.
I would be very grateful if those using FBWrappingDataSource could
check out new version of a driver and check if I haven't broken
anything during refactoring (even all test cases succeeded).
Best regards,
Roman Rokytskyy
I have committed a FBSimpleDataSource and made FBWrappingDataSource a
subclass of FBSimpleDataSource.
This new DataSource implementation is really simple and does not
provide any connection pooling. The reason for this class is the
dependency of FBWrappingDataSource on EDU.oswego.* classes
(concurrent.jar). In my opinion, when people use this driver in
standalone applications, they use connection pooling already. So,
dependency on a third-party jar does not bring too much, only might
cause deployment problems.
This simple DataSource implementation is supposed to be included in a
standalone jar version (comming soon, David is investigating issue of
re-distributing needed javax.* classes). There still will be a version
of a jar file including all classes that are in firebirdsql.jar now.
I would be very grateful if those using FBWrappingDataSource could
check out new version of a driver and check if I haven't broken
anything during refactoring (even all test cases succeeded).
Best regards,
Roman Rokytskyy