across the package. Moved marshaller creation and management into the
PersistenceContext because marshallers should be shared across a persistence
context.
Also moved the actual "get a connection and do the query" code into the
persistence context which is sort of a nice separation. The repository and
friends now just put together Query and Modifier objects and pass them to the
context for execution. The context will eventually handle interaction with the
cache as well (exposing manual manipulation mechanisms as needed to the
DepotRepository).
files which explains why Ray's "there are no languages but English" system
resulted in complaints and my "named after the African word for community"
system did not. Fortunately we can just tell the Java compiler what to expect.
I'm pretty sure source files should be ISO-8859-1.
If you're more comfortable with entity tags appearing in a /* */ style
comment, then maybe that's what we should use for the header text.
(The Java Coding Guidlines say you should use a /* */ comment at the top
of the file.)
properly cache the results of our insert based on the newly assigned primary
key. Also avoid freakout if an attempt is made to store() a non-primary-key
using object.
where we will need to cache the results of a modifier (like an insert or an
update). Added hasPrimaryKey() to DepotMarshaller which we will soon use
externally.
provide one. Separate out logging for depot so that fine grained logging can be
enabled if needed without turning on such logging for all of samskivert.
various fancy things like embedded objects and constraints and kitchen sinks
are not yet implemented, but the basic bits are functional (including IDENTITY
column primary key handling).
Michael Bayne