To resolve XSD files properly from the classpath, their HTTPS reference must be present in the spring.schemas to avoid internet interaction for resolving an XSD file.
Both the XML and annotation based configuration sources now support a bootstrap mode configuration property that allow to configure whether repositories are eagerly initialized unless declared as lazy, initialized in a deferred way (just before the application context finishs bootstrapping) or entirely lazy (upon first usage, i.e. a method invocation).
We now register a custom AutowireCandidateResolver that will consider all injection points of lazy repositories lazy too. This will prevent them to accidentally trigger downstream infrastructure initialization.
Both the configuration code looking up Spring Data specific Jackson modules as well as the code detecting whether we have multiple Spring Data modules on the classpath used component scanning. That can have quite significant impact on startup times.
This commit replaces the classpath scanning with the use of SpringFactoriesLoader that now requires modules that want to extend the behavior of Spring Data's web configuration or indicate a repository implementation being present by shipping a file called META-INF/spring.factories. Spring Data relies on the following keys:
- org.springframework.data.web.config.SpringDataJacksonModules - list the Spring configuration classes that will expose additional Jackson modules that are supposed to be registered for Spring Data's Jackson web support
- org.springframework.data.repository.core.support.RepositoryFactorySupport - list the Spring Data repository factory implementation class that implements repository support for your module. The general detection and configuration mechanism for user repositories is not affected by this. Currently Spring Data only uses the pure number of different entries for that key to switch into strict configuration mode in case we find more than one entry.
RepositoryFactorySupport now exposes a setRepositoryBaseClass(…) to take a custom type that instances will be created reflectively for. This removes the need for a boilerplate repository factory and FactoryBean implementation to hook custom repository base class implementations into the infrastructure.
RepositoryFactorySupport now exposes a getTargetRepositoryViaReflection(…) method to allow sub-classes to create repository instances and consider customized repository base classes nonetheless. getTargetRepository(…) needs a tiny signature change which unfortunately requires imple
Fixed schema registration for version 1.8 schema. Added new XSD containing a base-class attribute on the <repositories /> element to customize the repository base class via XML.
Removed outdated section of how to create a custom base repository class from the reference documentation and replaced it with new simplified instructions.
Deprecated usage of factory-class attribute in configuration. Point users to newly introduced way of configuring a base class directly to avoid the need for a boilerplate repository factory and factory bean implementation.
DefaultRepositoryInformation now makes target class methods assignable before caching and returning them, so that they can always be invoked reflectively. Made this aspect part of the contract of RepositoryInformation.getTargetClassMethod(…).
Extracted KeySpaceUtils from KeyValueTemplate and pulled in the copy of MetaAnnotationUtils to hide it from the outside. Separated related test cases into dedicated test class.
Renamed BasicMappingContext to KeyValueMappingContext and BasicPersistentProperty to KeyValuePersistentProperty. Fixed spelling in SpelQueryCreatorUnitTests.
Moved MappingContext into separate package to satisfy architectural guidelines. Created dedicated KeyValueRepositoryNamespaceHandler to avoid a circular reference between commons and the key-value subsystem.
Made the query creator type an annotation on @EnableKeyValueRepositories so that it's not exposed to the user to be configured. Added the necessary ImportBeanDefinitionRegistrars to the Hazelcast and EhCache implementation. Removed @since tags from those modules at they're not going to make it into Spring Data Commons. Renamed EnableEhCacheRepsotoriesUnitTests to EnableEhCacheRepositoriesIntegrationTests as well as the same for the Hazelcast specific test.
The KeyValueRepositoryConfigurationExtension now explicitly registers a KeyValueMappingContext unless one is already registered. AnnotationRepositoryConfigurationSource now explicitly exposes an AnnotationMetadata instance for the annotation that triggered the configuration.
Turned SpelCriteriaAccessor and SpelPropertyComparator into classes to be able to pass them a SpelExpressionParser to be able to get rid of SpelExpressionFactory eventually. Made both classes as well as SpelQueryEngine package protected for now.
Moved repository implementations into support package. Updated architcture description JavaDoc polish, spacing, blank lines. Added TODOs.
Original pull request: #95.
We now optionally consider nested repository interfaces defined as inner-classes. This can be configured by setting the considerNestedRepositories property on EnableXXXRepository annotations or the consider-nested-repositories attribute to true - it defaults to false.
Original pull request: #50.
