Previously, CachedIntrospectionResults had three modes of caching, with the intermediate mode relying on WeakReferences in the JDK PropertyDescriptor implementation. Since the JDK is actually using SoftReferences there these days, losing information in case of a GC run with tough memory constraints, we want to allow for hard references in PropertyDescriptor objects and therefore use a full WeakReference for the CachedIntrospectionResults object itself.
Issue: SPR-10028
This change revisits the implementation of ExtendedBeanInfo, simplifying
the overall approach while also ensuring that ExtendedBeanInfo is fully
isolated from the BeanInfo instance it wraps. This includes any existing
PropertyDescriptors in the wrapped BeanInfo - along with being copied
locally into ExtendedBeanInfo, each property descriptor is now also
wrapped with our own new "simple" PropertyDescriptor variants that
bypass the soft/weak reference management that goes on in both
java.beans.PropertyDescriptor and java.beans.IndexedPropertyDescriptor,
maintaining hard references to methods and bean classes instead. This
ensures that changes we make to property descriptors, e.g. adding write
methods, do not cause subtle conflicts during garbage collection (as was
reported and reproduced in SPR-9702).
Eliminating soft/weak reference management means that we must take extra
care to ensure that we do not cause ClassLoader leaks by maintaining
hard references to methods, and therefore transitively to the
ClassLoader in which the bean class was loaded. The forthcoming
SPR-10028 addresses this aspect.
See the updated ExtendedBeanInfo Javadoc for further details.
Issue: SPR-8079, SPR-8175, SPR-8347, SPR-8432, SPR-8491, SPR-8522,
SPR-8806, SPR-8931, SPR-8937, SPR-8949, SPR-9007, SPR-9059,
SPR-9414, SPR-9453, SPR-9542, SPR-9584, SPR-9677, SPR-9702,
SPR-9723, SPR-9943, SPR-9978, SPR-10028, SPR-10029
Backport-Commit: 4a8be69099
Conflicts:
spring-beans/src/main/java/org/springframework/beans/ExtendedBeanInfoFactory.java
Allow the body of 'arg-type' XML elements to be used as an alternative to
'match' attribute when defining a 'replace-method' in XML configuration.
This change has been introduced primarily to support the samples printed
in the Apress 'Pro Spring' book.
Issue: SPR-9812
Backport-Issue: SPR-9929
Backport-Commit: 0709c033a0
Commit 6263c9abf9
@EnableSpringConfigured from beans.factory.aspectj =>
context.annotation within the spring-aspects module. This resolved a
package cycle but had the side-effect of causing a "split package" [1]
problem between spring-context and spring-aspects in OSGi-based
classloader environments because the context.annotation package now
exists in both modules.
The simplest and best solution from an OSGi perspective is to relocate
@EnableSpringConfigured and its supporting SpringConfiguredConfiguration
class into a new package. This commit moves both these types into
context.annotation.aspectj, following convention with other such
"aspectj"-qualified packages in the spring-aspects module.
As with the previous move, it is presumed this change will be low-impact
as the "spring-configured" approach to domain object injection is a
niche feature to begin with, and @EnableSpringConfigured has existed in
its current location only since 3.1.2 and this change is being made in
time for 3.1.3.
[1]: http://wiki.osgi.org/wiki/Split_Packages
Issue: SPR-9811, SPR-9441
Backport-Commit: 54db7387de
Prior to this commit, utility methods in JdbcTestUtils interpreted SQL
comments as separate statements, resulting in an exception when such a
script is executed.
This commit addresses this issue by introducing a
readScript(lineNumberReader, String) method that accepts a comment
prefix. Comment lines are therefore no longer returned in the parsed
script. Furthermore, the existing readScript(lineNumberReader) method
now delegates to this new readScript() method, supplying "--" as the
default comment prefix.
Issue: SPR-9593
Backport-Commit: 4aaf014cc6
MapToMap conversion now works when the target map does not have a
default constructor (as long as a new map copy is not required).
Issue: SPR-9284
Backport-Commit: 38c4393592
Refactor ConfigurationClassEnhancer to allow cglib caching of
generated classes. Prior to this commit each enhanced @Configuration
class would consume permgen space when created.
The CallbackFilter and Callback Types are now defined as static final
members so that they can be shared by all enhancers. Only the
callbackInstances remain specific to a @Configuration class and
these are not used by cglib as part of the cache key.
Issue: SPR-9851
Backport-Commit: c8061393fb
Update the ReflectiveMethodResolver and ReflectivePropertyAccessor
to allow methods and properties of java.lang.Class to be resolved
when the target object is a class.
Issue: SPR-9017
Backport-Commit: d28592a6c6
SpringEL300Tests's #testArray method was originally introduced with
SPR-9203, which has not yet been backported to 3.1.x. During the process
of backporting SPR-9486 functionality, this method was accidentally
introduced, causing false negative test failures. This commit removes
the offending test method.
Issue: SPR-9203, SPR-9486
This change ensures that SpEL expressions involving floats are
supported natively as opposed to the previous behavior which required
conversion to double, leading to potential downstream conversion
ambiguities.
Issue: SPR-9486
Backport-Commit: be8f23d756
Previously UriComponentsBuilder used a regular expression for parsing
query name-value pairs where both name and value were expected to not
contain neither '&', not '='. The idea is that the presence of reserved
characters makes it impossible to guess correctly how to parse the
query string (e.g. a=b&c).
This change relaxes the constraint on query param values, allowing them
to contain '='. In effect '&' is the ultimate separator of name-value
pairs, and any '=' in values is ignored. For example "q=1USD=?EUR" is
interpreted as "q equals '1USD=?EUR'".
Issue: SPR-9832
The target parameters for a FlashMap must be decoded to be able to
match them to the parameters of incoming requests given that the
HttpServletRequest returns decoded request parameters.
Issue: SPR-9657
Backport Issue: SPR-9701
The ErrorsMethodArgumentResolver expects the preceding @ModelAttribute
in the controller method signature to be the last one added in the
model -- an assumption that can break if a model attribute is added
earlier (e.g. through a @ModelAttribute method) and more attributes
are added as well. This fix ensures when an @ModelAttribute is resolved
as a controller method argument it has the highest index in the model.
Issue: SPR-9378
Backport Issue: SPR-9687
Prior to this commit spring-web's dependence on spring-aop was marked
optional, however spring-web's required dependence on spring-context
transitively forces a required dependence on spring-aop anyway.
This change marks spring-aop as a required dependency of spring-web,
which a) reflects the poms created by the 3.2 Gradle build and b) helps
solve dependency issues when using higher-level Spring projects such as
Spring Security.
Issue: SPR-9573
Reverted change for @Bean methods that declare FactoryBean as their return type: The effects of trying to create the FactoryBean to find out about its implementation type are too far-reaching. It's better to recommend declaring a specific return type in the method signature if you want the container to specifically react to your implementation type.
Issue: SPR-9857
Includes a change for factory methods that declare their return type as FactoryBean: When asked for a specific type match (e.g. LoadTimeWeaverAware), we do check early singleton instances as well (reusing the instances that we create for getObjectType checks). This is necessary in order to make @Bean method introspection as capable as XML bean definition introspection, even in case of the @Bean method using a generic FactoryBean declaration for its return type (instead of the FactoryBean impl class).
Issue: SPR-9857
