| [eclipselink-users] Understanding the CacheInterceptor |
|
Our application
require some ultra-fast lookups on a unique index other than the primary key.
There are some posts and notes on the web indicating that cache indexes might
be coming at some point, but in the meantime we actually have to go with a home
grown solution. So the current thinking has two pieces: 1) To manage the
index we create a CacheInterceptor, and intercept cache operations. Whenever an
object (of the required class) is inserted or updated into the cache, we
extract the other index and create a simple Map of {index,object}. 2) To read objects,
we create a Redirect Query which reads the map. So, two questions,
really: - Does this make
sense? - Is there any sample
code out there using the CacheInterceptor? Eirik |