* Add tenant-partitioned L2 caches (#2956)
When `tenantPartitionedCache` is enabled, each tenant gets its own
cache namespace (keys include the tenant id), improving cache-hit
ratio by preventing cross-tenant key collisions.
Refactor BeanDescriptorCacheHelp to abstract base class with two
concrete subclasses:
- BeanDescriptorCacheHelpFixed - static cache refs (original behaviour)
- BeanDescriptorCacheHelpPartitioned - per-request Supplier<> lookup
so the correct tenant-scoped cache is resolved on each access
Fix background-thread invalidation: clear(name) in partitioned mode
now scans all matching cache entries by key prefix rather than calling
tenantProvider.currentId() which is null/wrong on background executor
threads.
Add SpiCacheManager.clearTenant(tenantId) to allow removal of all
cache entries for a deactivated tenant, preventing unbounded memory
growth in long-running multi-tenant deployments.
Also load cache settings (cacheMaxSize, cacheMaxIdleTime, etc.) from
application properties - these were previously missing from loadSettings().
* Fix versions in test-java16
---------
Co-authored-by: robin.bygrave <robin.bygrave@eroad.com>
* Add support for Java 21 SequencedSet and SequencedMap
Such that these can be used in place of Set and Map if desired.
* Refactor BeanSet, BeanList, BeanMap replacing setActualSet|List|Map
Replace with collectionAdd() and refresh() methods.
* Refactor rename method getBeanCollectionAdd() -> collectionAdd()
* Build needs to use Java 21 to support the multi-release jar
* Update SequencedSet etc from recent changes
* Update build, needs package to use MR-JAR
---------
Co-authored-by: robin.bygrave <robin.bygrave@eroad.com>
* Change [implicit] Lazy loading to be by TYPE rather than by PATH
We have some graph models where a common type (in the tests it is Label)
is used in many different *paths* of the graph. When we are loading via
*path* then all the loading of the Labels isn't in a single batch / load context
but instead split into different load contexts PER PATH.
This change, means that lazy loading operates by TYPE instead of by PATH.
In the tests, all the Labels are read via a single lazy loading query rather
that one lazy loading query per PATH, and this is more efficient.
* ImmutableBeanCache part 1
* ImmutableBeanCache part 2
* ImmutableBeanCache part 3 - add ImmutableBeanCaches
* ImmutableBeanCache - add propagation of caches to secondary queries
* ImmutableBeanCache - use and improve BeanDescriptor.merge() to positional
Using positional avoids the lookup by property name.
* ImmutableBeanCache - add unit testing with BeanDescriptorMergeTest
* ImmutableBeanCache - add testing for the further lazy loading on mutable beans
Also add doc / guides
* Add ImmutableCacheBuilder with underlying DefaultServerCache implementation
This provides a cache with maxSize, maxIdleSeconds, maxSecondsToLive
* Improve javadoc and doc / guides for ImmutableBeanCache
* Add cache invalidation for ImmutableBeanCache
* Use immutable cache direct hits for unmodifiable assoc-one refs
- add ImmutableBeanCache.getIfPresent() as a non-loading probe
- use immutable cache hits in AssocOneHelp / AssocOneHelpRefInherit
- avoid creating ref beans and merge/copy for unmodifiable cache hits
- keep existing mutable-query and miss/backfill behavior unchanged
* creates support for generics in mapped superclasses
* Changes DeployCreateProperties to allow processing of inheritance hierarchies with generics
* adds tests for ebean-querybean
* bumps querybean-generator version
* fixes tests by changing table names for ProductWithGenericLong and ProductWithGenericString
* Restore format, this reduces the diff
* Restore format, this reduces the diff
---------
Co-authored-by: Rob Bygrave <robin.bygrave@gmail.com>
Currently, the SELF response type of QueryBuilder returns as Object when really we want it to return the SELF generic type of QueryBuilder, QueryBean, IQueryBean etc
Note that this change requires an updated ebean-agent
Rob Bygrave