We're now favoring the generic TypeInformation over trying to resolve the property type via field or PropertyDescriptor as only the former does proper generic type resolution.
Type specialization - i.e. enrichment of a raw type with a current generic context - is now only done if the current type is not yet resolved completely. This allows wildcarded target references to just fall back to the type to specialize, which will then by definition carry more generics information than the one to be specialized.
Revert the merging of a parent type's type variable map and only apply the locally available declared generics in case of parameterized types. Moved that augmentation into ParameterizedTypeInformation.
If you previously asked ProxyProjectionFactory for a proxy of an interface which the target instance already implements, it created a proxy although it could've returned the instance as is. It's now actually doing that avoiding superfluous proxy creation.
Previously, RepositoryBeanNameGenerator applied the custom bean name lookup if the BeanDefinition given was not a ScannedGenericBeanDefinition. That in turn had been the case for custom implementation classes that were obtained through classpath scanning. With Spring 5 an index file can be used by the scanner, which in turn will cause AnnotatedGenericBeanDefinition instances being returned. That caused the code path to lookup the first constructor argument to kick in (usually used to obtain the repository interface from repository factory beans) and cause a NullPointerException.
We now forward AnnotatedBeanDefinitions as is and only apply the custom lookup for everything else, i.e. the bean definitions used for the factories.
For repository query methods with a dynamic projection parameter, the ResultProcessor is recreated with the type handed to the method. This results in recreation of the ProjectingConverter, which previously recreated a DefaultConversionService instance (for fallback conversions), which is rather expensive due to the reflection lookups checking for the presence of libraries on the classpath.
This is now avoided by using copying methods that reuse the initially created DefaultConversionService.
We now override ClassPathScanningCandidateComponentProvider's getRegistry() to make sure custom conditions on repository candidates can use the currently available BeanDefinitionRegistry in their implementations. To achieve that we forward the BeanDefinitionRegistry at hand through the configuration infrastructure (both XML and annotation side of things).
ObjectInstantiator needs to be declared public as otherwise loading the implementation class fails as it doesn't have access to the (package) private interface in a different classloader.
We now use the Pageable information contained in the given Page to make sure the self link rendered by the assembler adds proper coordinates. If a base link is provided, that is used as is.
Removed custom JUnit integration as we can just create HidingClassLoader instances in the test and the integration actually causes more code being needed (additional JUnit rule, method level annotations etc.).
Tweaked ShadowingClassLoader to make obvious what has been changed over the Spring Framework variant. Created upstream ticket [0] to ask for the tweaks that would allow us to remove the class again.
Original pull request: #202.
Related ticket: SPR-15439
[0] https://jira.spring.io/browse/SPR-15439
ClassLoaderRule with @ClassLoaderConfiguration allows easy creation FilteringClassLoader. Changed usage pattern of ClassUtils.isPresent(…) in order to ease testing. Copied the ShadowingClassloader from SpringFramework to get a constructor that doesn't shadow anything by default.
Original pull request: #202.
The changes for DATACMNS-854 and DATACMNS-912 dropped the support for the simpler redeclaration of generic methods like CrudRepository.save(…) which as of the changes requires to be redeclared like <T extends Foo> T save(T entity). Previously a simple redeclaration like Foo save(Foo entity) was sufficient.
This commit reintroduces the check for a direct match of the parameter types and shortcuts the more detailed check that's necessary in case of type variables being involved.
Related tickets: DATACMNS-854, DATACMNS-912.
Oliver Gierke
Mark Paluch
Johnny Lim
Roman Rodov
Jens Schauder