Berin Loritsch wrote:
It seems to work. I replaced excalibur-datasource-1.0.jar in the newest Phoenix release candidate with the three jars from your version and my app worked. BTW, what are excalibur-datasource-cluster-1.1a.jar and excalibur-datasource-ids-1.1a.jar for?
Not sure, they were added by Leif to handle some aspects of database
manipulation. They are additional functionality that is optional which
is why they are in separate JARs.
Datasource-cluster:
This package was built to make it possible to work with very large data sets by distributing
the data across multiple database servers. Rather than paying for a super expensive Oracle
license, this datasource implementation will correctly route queries to the correct data base
making the fact that the database is clustered fairly transparent. See the Java docs of the
DefaultHashedDataSourceCluster, DefaultIndexedDataSourceCluster, and
DefaultRoundRobinDataSourceCluster. They have reasonable documentation in the javadocs.
Datasource-ids:
This is a package which makes it possible for applications to manage ids in a database
independent way. Check out the java docs in the TableIdGenerator and SequenceIdGenerator
classes. Once again, they are fairly well documented. When a component needs an id to be
able to insert a record into a database, the IdGenerator is called to get that id. The Table
implementation works by using a table of ids where each table has its own rows. It will then
allocate blocks of Ids to use at will. This system works great where multiple servers all need
to be inserting into the same database without id conflicts. This method can also be used
with the clustering above.
I know these need some real documentation <:-}
Cheers, Leif
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
