To support additional use cases, this commit introduces a
withAssignmentDisabled() method in the Builder for
SimpleEvaluationContext.
See gh-33319
Closes gh-33321
(cherry picked from commit e74406afd0)
SimpleEvaluationContext.forReadOnlyDataBinding() documents that it
creates a SimpleEvaluationContext for read-only access to public
properties; however, prior to this commit write access was not disabled
for indexed structures when using the assignment operator, the
increment operator, or the decrement operator.
In order to better align with the documented contract for
forReadOnlyDataBinding(), this commit makes it possible to disable
assignment in general in order to enforce read-only semantics for
SpEL's SimpleEvaluationContext when created via the
forReadOnlyDataBinding() factory method. Specifically:
- This commit introduces a new isAssignmentEnabled() "default" method
in the EvaluationContext API, which returns true by default.
- SimpleEvaluationContext overrides isAssignmentEnabled(), returning
false if the context was created via the forReadOnlyDataBinding()
factory method.
- The Assign, OpDec, and OpInc AST nodes -- representing the assignment
(=), increment (++), and decrement (--) operators, respectively --
now throw a SpelEvaluationException if assignment is disabled for the
current EvaluationContext.
See gh-33319
Closes gh-33321
(cherry picked from commit 0127de5a7a)
Prior to this commit, the Indexer in the Spring Expression Language
(SpEL) silently ignored a failure to set a property via the indexed
property syntax (['<property name>'] = <new value>) – for example, if
property write access was disabled in the EvaluationContext.
This commit addresses this issue by properly throwing a
SpelEvaluationException in PropertyIndexingValueRef.setValue(Object) if
the property could not be set.
See gh-33310
Closes gh-33311
(cherry picked from commit c57c2272a1)
Prior to this commit, the ConversionService failed to convert a primitive
array (such as int[]) to an Object[] due to an error in the logic in
ArrayToArrayConverter.
This commit addresses this by augmenting the "can bypass conversion"
check in ArrayToArrayConverter to ensure that the supplied source object
is an instance of the target type (i.e., that the source array can be
cast to the target type array without conversion).
Closes gh-33212
(cherry picked from commit cb6a5baac5)
Prior to this commit, the Spring Expression Language (SpEL) incorrectly
split single String arguments by comma for Object... varargs method and
constructor invocations.
This commit addresses this by checking if the single argument type is
already "assignable" to the varargs component type instead of "equal"
to the varargs component type.
See gh-33013
Closes gh-33189
This commit makes sure that docs artifacts have their attributes set
for staging as well. Previously they were not and deployment of Javadoc
did not occur.
Closes gh-33207
This commit harmonizes the invocation of a bean supplier with what
SimpleInstantiationStrategy does. Previously, the current factory method
was set to `null` once the invocation completes. This did not take
into account recursive scenarios where an instance supplier triggers
another instance supplier.
For consistency, the thread local is removed now if we attempt to set
the current method to null. SimpleInstantiationStrategy itself uses
the shortcut to align the code as much as possible.
Closes gh-33185
Prior to this commit, the fix for gh-32730 disabled the involvment of
the osbervation filter for async dispatches. Instead of relying on ASYNC
dispatches to close the observation for async requests, this is now
using an async listener instead: async dispatches are not guaranteed to
happen once the async request is handled.
This change caused another side-effect: because async dispatches are not
considered anymore by this filter, the observation scope is not
reinstated for async dispatches. For example, `ResponseBodyAdvice`
implementations do not have the observation scope opened during their
execution.
This commit re-enables async dispatches for this filter, but ensures
that observations are not closed during such dispatches as this will be
done by the async listener.
Fixes gh-33128
Prior to this commit, the `ServerHttpObservationFilter` would support
async dispatches and would do the following:
1. start the observation
2. call the filter chain
3. if async has started, do nothing
4. if not in async mode, stop the observation
This behavior would effectively rely on Async implementations to
complete and dispatch the request back to the container for an async
dispatch. This is what Spring web frameworks do and guarantee.
Some implementations complete the async request but do not dispatch
back; as a result, observations could leak as they are never stopped.
This commit changes the support of async requests. The filter now
opts-out of async dispatches - the filter will not be called for those
anymore. Instead, if the application started async mode during the
initial container dispatch, the filter will register an AsyncListener to
be notified of the outcome of the async handling.
Fixes gh-32986
Prior to this commit, the Spring Expression Language (SpEL) failed to
compile an expression that indexed into a Map using a primitive literal
(boolean, int, long, float, or double).
This commit adds support for compilation of such expressions by
ensuring that primitive literals are boxed into their corresponding
wrapper types in the compiled bytecode.
Closes gh-32903
(cherry picked from commit e9de426eb5)
Prior to this commit, the Spring Expression Language (SpEL) failed to
compile an expression that indexed into an array or list using an
Integer.
This commit adds support for compilation of such expressions by
ensuring that an Integer is unboxed into an int in the compiled
bytecode.
See gh-32694
Closes gh-32908
(cherry picked from commit cda577d1aa)
This covers AspectJ transaction and caching aspects when encountered by Spring AOP.
Closes gh-32882
See gh-32793
(cherry picked from commit 6d7cd9c7dc)
- Use same playbook as docs-build
- Use Env Variables to cause partial build (same as docs-build)
- Use package.json so that dependencies can be updated with dependabot
Includes JAXBContext locking revision (avoiding synchronization) and consistent treatment of DocumentBuilderFactory (in terms of caching as well as locking).
Closes gh-32851
(cherry picked from commit a4c2f291d9)
This commit avoids several instances of MockHttpServletRequest to
have a common reader for empty content as closing it will have an
unwanted side effect on the others.
Closes gh-32848
This commit fixes links from Spring Framework's Dokka HTML to Javadoc
for Spring Framework and Servlet APIs by explicitly configuring the
`element-list` page as the `package-list` in the Dokka Gradle plugin.
Closes gh-32797
(cherry picked from commit 7536980be9)
This commit fixes the artifact properties we set for "framework-docs"
artifacts. These have a different name as of 6.1.x and were backported
as is.
Closes gh-32780
A bug has existed in Spring's MergedAnnotations support since it was
introduced in Spring Framework 5.2. Specifically, if the
MergedAnnotations API is used to search for annotations with "standard
repeatable annotation" support enabled (which is the default), it's
possible to search for a repeatable annotation but not for the
repeatable annotation's container annotation.
The reason is that MergedAnnotationFinder.process(Object, int, Object,
Annotation) does not process the container annotation and instead only
processes the "contained" annotations, which prevents a container
annotation from being included in search results.
In #29685, we fixed a bug that prevented the MergedAnnotations support
from recognizing an annotation as a container if the container
annotation declares attributes other than the required `value`
attribute. As a consequence of that bug fix, since Spring Framework
5.3.25, the MergedAnnotations infrastructure considers such an
annotation a container, and due to the aforementioned bug the container
is no longer processed, which results in a regression in behavior for
annotation searches for such a container annotation.
This commit addresses the original bug as well as the regression by
processing container annotations in addition to the contained
repeatable annotations.
See gh-29685
Closes gh-32731
(cherry picked from commit 4baad16437)
This commit changes the guard against multiple subscriptions, as the
previously used doOnSubscribe hook could not function as guard in
certain scenarios.
See gh-32727
Closes gh-32732
This commit refines the preDetermineBeanTypes algorithm that AOT uses
to approximate the order of preInstantiateSingletons more closely.
Previously, the algorithm assumed that all beans where non-lazy
singletons in terms of initialization order, which led to inconsistent
order in CGLIB proxy generation.
We now explicitly park lazy beans so that their types are determined
during a second phase, matching the order of regular initialization
order.
Closes gh-32701
Co-authored-by: Juergen Hoeller <juergen.hoeller@broadcom.com>
Before this change temporary files would not consistently be deleted
when the connection which uploads the multipart files closes naturally.
This change uses the usingWhen Reactor operator to ensure that the
termination of the connection doesn't prevent individual file parts
from being deleted due to a cancellation signal.
See gh-31217
Closes gh-32638
This commit refines the expressions for the scheme, user info, host and
port parts of the URL in UriComponentsBuilder to better conform to
RFC 3986.
Fixes gh-32617
Prior to this commit, MethodIntrospector failed to properly detect
bridge methods for subsequent invocations of selectMethods() with the
same targetType and MetadataLookup, if such subsequent invocations
occurred after the ApplicationContext had been refreshed.
The reason this occurs is due to the following.
- Class#getDeclaredMethods() always returns "child copies" of the
underlying Method instances -- which means that `equals()` should be
used instead of `==` whenever the compared Method instances can come
from different sources (such as the static caches mentioned below).
- BridgeMethodResolver caches resolved bridge methods in a static cache
-- which is never cleared.
- ReflectionUtils caches declared methods in a static cache
-- which gets cleared when an ApplicationContext is refreshed.
Consequently, if you attempt to load an ApplicationContext twice in the
same ClassLoader, the second attempt uses the existing, populated cache
for bridged methods but a cleared, empty cache for declared methods.
This results in new invocations of Class#getDeclaredMethods(), and
identity checks with `==` then fail to detect equivalent bridge methods.
This commit addresses this by additionally comparing bridge methods
using `equals()` in MethodIntrospector.selectMethods().
Note that the `==` checks remain in place as an optimization for when
`equals()` is unnecessary.
Closes gh-32586
(cherry picked from commit e702733c7b)
The fix for #31254 resulted in an InvalidMimeTypeException being thrown
by MimeTypeUtils.sortBySpecificity() instead of an
IllegalArgumentException. However, InvalidMimeTypeException extends
IllegalArgumentException. Consequently, the change from
IllegalArgumentException to InvalidMimeTypeException did not result in
the desired effect in HeaderContentNegotiationStrategy.
HeaderContentNegotiationStrategy.resolveMediaTypes() still allows the
InvalidMimeTypeException to propagate as-is without wrapping it in an
HttpMediaTypeNotAcceptableException.
To address this issue, this commit catches InvalidMediaTypeException
and InvalidMimeTypeException in HeaderContentNegotiationStrategy and
wraps the exception in an HttpMediaTypeNotAcceptableException.
See gh-31254
See gh-31769
Closes gh-32483
(cherry picked from commit ef02f0bad8)
Prior to this commit, error signals flowing from the client response
publisher in `WebClient` would be set on the `Observation.Context`. This
is enough for the observation convention to collect data about the error
but observation handlers are not notified of this error.
This commit sets the error instead on the observation directly to fix
this issue.
Fixes gh-32399
webmvc.fn now also uses the StandardServletAsyncWebRequest wrapped response
to enforce lifecycle rules from Servlet spec (section 2.3.3.4).
See gh-32341
The wrapped response prevents use after AsyncListener onError or completion
to ensure compliance with Servlet Spec 2.3.3.4.
The wrapped response is applied in RequestMappingHandlerAdapter.
The wrapped response raises AsyncRequestNotUsableException that is now
handled in DefaultHandlerExceptionResolver.
See gh-32341
Based on feedback from several members of the community, we have
decided to revert the caching of the Content-Type header that was
introduced in ContentCachingResponseWrapper in 375e0e6827.
This commit therefore completely removes Content-Type caching in
ContentCachingResponseWrapper and updates the existing tests
accordingly.
To provide guards against future regressions in this area, this commit
also introduces explicit tests for the 6 ways to set the content length
in ContentCachingResponseWrapper and modifies a test in
ShallowEtagHeaderFilterTests to ensure that a Content-Type header set
directly on ContentCachingResponseWrapper is propagated to the
underlying response even if content caching is disabled for the
ShallowEtagHeaderFilter.
See gh-32039
See gh-32317
Closes gh-32321
Commit 375e0e6827 introduced a regression in
ContentCachingResponseWrapper (CCRW). Specifically, CCRW no longer
honors Content-Type and Content-Length headers that have been set in
the wrapped response and now incorrectly returns null for those header
values if they have not been set directly in the CCRW.
This commit fixes this regression as follows.
- The Content-Type and Content-Length headers set in the wrapped
response are honored in getContentType(), containsHeader(),
getHeader(), and getHeaders() unless those headers have been set
directly in the CCRW.
- In copyBodyToResponse(), the Content-Type in the wrapped response is
only overridden if the Content-Type has been set directly in the CCRW.
Furthermore, prior to this commit, getHeaderNames() returned duplicates
for the Content-Type and Content-Length headers if they were set in the
wrapped response as well as in CCRW.
This commit fixes that by returning a unique set from getHeaderNames().
This commit also updates ContentCachingResponseWrapperTests to verify
the expected behavior for Content-Type and Content-Length headers that
are set in the wrapped response as well as in CCRW.
See gh-32039
See gh-32317
Closes gh-32321
This commit adapts Hibernate native support to handle
the changes performed as part of HHH-17643 which impacts
Hibernate versions 6.4.3+ and 6.2.21+.
It ignores the BytecodeProvider services loaded by the
service loader feature in order to default to the
"no-op" provider with native.
gh-32314 is expected to remove the need for such
substitutions which are not great for maintainability
by design.
Closes gh-32312
This notably enables Jackson to reflectively call a user-provided
builder class and invoke its declared methods (setters and build) in
a native app.
See gh-32238
Closes gh-32257
ServletAttributesMap inherited default implementations of the size
and isEmpty methods from AbstractMap which delegates to the Set returned
by entrySet. ServletAttributesMap's entrySet method made this fairly
expensive, since it would copy the attributes to a List, then use a
Stream to build the Set. To avoid the cost, add implementations of
isEmpty / size that don't need to call entrySet at all.
Additionally, change entrySet to return a Set view that simply lazily
delegates to the underlying servlet request for iteration.
Closes gh-32197
This commit improves PersistenceManagedTypesBeanRegistrationAotProcessor
so that it does not attempt to load JPA classes when checking for the
presence of a PersistenceManagedTypes bean. To make it more clear a
check on the presence for JPA has been added to prevent the nested
classes to be loaded regardless of the presence of the bean.
Closes gh-32160
Before this commit, the JDK and Jetty connectors do not have any
safeguards against multiple body subscriptions. Such as check has now
been added.
See gh-32100
Closes gh-32102
Prior to this commit, the `RestTemplate` observation instrumentation
would only record `RestClientException` and `IOException` as errors in
the observation. Other types of errors can be thrown by custom
components, such as `ResponseErrorHandler` and in this case they aren't
recorded with the observation.
Also, the current instrumentation does not create any observation scope
around the execution. While this would have a limited benefit as no
application code is executed there, developers could set up custom
components (such as, again, `ResponseErrorHandler`) that could use
contextual logging with trace ids.
This commit ensures that all `Throwable` are recorded as errors with the
observations and that an observation `Scope` is created around the
execution of the client exchange.
Fixes gh-32063
This commit ensures that setting the Content-Length through
setHeader("Content-Length", x") has the same effect as calling
setContentLength in the ShallowEtagHeaderFilter. It also filters out
Content-Type headers similarly to Content-Length.
See gh-32039
Closes gh-32050
ResponseEntityExceptionHandler should not set the exception attribute
when there is a response body, and the response is fully handled.
Closes gh-31541
This commit updates ContentRequestMatchers#multipartData
Javadoc to mention Tomcat fork of Commons FileUpload library
instead of the original variant.
It also adds a similar note to
ContentRequestMatchers#multipartDataContains.
Closes gh-31989
(Backport of 598c972a78)
This commit adds CORS support for Private Network Access
by adding an Access-Control-Allow-Private-Network response
header when the preflight request is sent with an
Access-Control-Request-Private-Network header and that
Private Network Access has been enabled in the CORS
configuration.
See https://developer.chrome.com/blog/private-network-access-preflight/
for more details.
Closes gh-31975
(cherry picked from commit 318d460256)
Includes TypeVariable bypass for reflection-free annotation retrieval.
Includes info log message for annotation attribute retrieval failure.
Closes gh-27182
(cherry picked from commit 70247c4a94)
Prior to this commit, the `HandlerMappingIntrospector` would comparea
request with a cached request by using `String#matches` on their String
URI. This could lead to `PatternSyntaxException` exceptions at runtime
if the request URI contained pattern characters.
This commit fixes this typo to use `String#equals` instead.
Fixes gh-31946
Prior to this commit, the `ForwardedHeaderFilter` and the forwarded
header utils would throw `IllegalArgumentException` and
`IllegalStateException` when request headers are invalid and cannot be
parsed for Forwarded handling.
This commit aligns the behavior with the WebFlux counterpart by
rejecting such requests with HTTP 400 responses directly.
Fixes gh-31894
This commit ensures that the capacity of the remainder buffer after a
split operation is set directly on the field. Calling capacity(int)
caused a new buffer to be allocated.
See gh-31848
Closes gh-31859
This commit introduces an integration test for the regression fixed in
the previous commit (76bc9cf325).
See gh-31826
Closes gh-31828
(cherry picked from commit 8d4deca2a6)
Required by Spring Security to complete work on
https://github.com/spring-projects/spring-security/issues/14128
The setCache and resetCache methods used from createCacheFilter are now
public. Generally they don't need to be used outside of the Filter if
only making checks against the current request. Spring Security, however,
makes additional checks against requests with alternative paths.
Prior to this commit, the AnnotationsScanner used in the
MergedAnnotations infrastructure found duplicate annotations on methods
within multi-level interface hierarchies.
This commit addresses this issue by scanning methods at a given level
in the interface hierarchy using ReflectionUtils#getDeclaredMethods
instead of Class#getMethods, since the latter includes public methods
declared in super-interfaces which will anyway be scanned when
processing super-interfaces recursively.
Closes gh-31803
(cherry picked from commit 75da9c3c47)
This commit fixes a bug in DefaultPartHttpMessageReader's
MultipartParser, due to which the last token in a part window was not
properly indicated.
See gh-30953
Closes gh-31766
Prior to this commit, regressions were introduced with gh-31417:
1. the observation keyvalues would be inconsistent with the HTTP response
2. the observation scope would not cover all controller handlers, causing
traceIds to be missing
The first issue is caused by the fact that in case of error signals, the
observation was stopped before the response was fully committed, which
means further processing could happen and update the response status.
This commit delays the stop event until the response is committed in
case of errors.
The second problem is caused by the change from a `contextWrite`
operator to using the `tap` operator with a `SignalListener`. The
observation was started in the `doOnSubscription` callback, which is too
late in some cases. If the WebFlux controller handler is synchronous
non-blocking, the execution of the handler is performed before the
subscription happens. This means that for those handlers, the
observation was not started, even if the current observation was
present in the reactor context. This commit changes the
`doOnSubscription` to `doFirst` to ensure that the observation is
started at the right time.
Fixes gh-31715
Fixes gh-31716
This commit updates MetadataNamingStrategy to quote an ObjectName
attribute value if necessary. For now, only the name attribute is
handled as it is usually a bean name, and we have no control over
its structure.
Closes gh-31708
This commit supports the scroll() and scroll(ScrollMode) methods from
Hibernate's Query API in SharedEntityManagerCreator's query-terminating
methods set.
See gh-31682
Closes gh-31683
(cherry picked from commit a15f472898)
Prior to this commit, `ExchangeFilterFunction` could only get the
current observation from the reactor context. This is particularly
useful when such filters want to add KeyValues to the observation
context.
This commit makes this use case easier by adding the context of the
current observation as a request attribute. This also aligns the
behavior with other instrumentations.
Fixes gh-31646
Prior to this commit, the Jackson 2.x encoders, in case of encoding a
stream of data, would first release the `ByteArrayBuilder` and then the
`JsonGenerator`. This order is inconsistent with the single value
variant (see `o.s.h.codec.json.AbstractJackson2Encoder#encodeValue`) and
invalid since the `JsonGenerator` uses internally the
`ByteArrayBuilder`.
In case of a CSV Encoder, the codec can buffer data to write the column
names of the CSV file. Writing an empty Flux with this Encoder would not
fail but still log a NullPointerException ignored by the reactive
pipeline.
This commit fixes the order and avoid such issues at runtime.
Fixes gh-31656
Legacy EJB attributes are ignored since 6.0 due to being bound to a plain JndiObjectFactoryBean - but can still be declared now, e.g. when validating against the common versions of spring-jee.xsd out there.
Closes gh-31627
(cherry picked from commit 695559879e)
Adding generated code in the default package is not supported as we
intend to import it, most probably from another package, and that is
not supported. While this situation is hard to replicate with Java,
Kotlin is unfortunately more lenient and users can end up in that
situation if they forget to add a package statement.
This commit checks for the presence of a valid package, and throws
a dedicated exception if necessary.
Closes gh-31629
Prior to this commit, the getResource() methods in PathResourceResolver
implementations allowed an exception thrown from Resource#getURL() to
propagate instead of logging a warning about the missing resource as
intended.
This commit modifies the getResource() methods in PathResourceResolver
implementations so that the log messages include the output of the
toString() implementations of the underlying resources instead of their
getURL() implementations, which may throw an exception.
Furthermore, logging the toString() output of resources aligns with the
existing output for "allowed locations" in the same log message.
Note that the toString() implementations could potentially also throw
exceptions, but that is considered less likely.
See gh-31623
Closes gh-31624
(cherry picked from commit 7d2ea7e7e1)
Expose methods to set and reset cache to use from a Filter instead
of a method to create such a Filter. Also use cached results only
if they match by dispatcher type and requestURI.
See gh-31588
This commit updates PathMatchingResourcePatternResolver to avoid
returning duplicate resources on MS Windows when searching using the
`classpath*:` prefix and a wildcard pattern that matches resources
which are directly present in a JAR as well as present via classpath
manifest entries.
Closes gh-31598
Prior to this commit, the `ConcurrentReferenceHashMap#Segment` would
recalculate its element count during the restructure phase by
subtracting the number of polled elements from the queue to the current
count.
Later, the newly restructured `references` array would only contain
references that 1) are not in the set of elements to purge and
2) that hold a non-null value.
The issues with this approach are multiple. The newly calculated count
is only an estimate, as some situations can make it invalid, even
temporarily.
* since we initially collected all elements to be purged from the queue,
the GC might have collected new values. This means that we might
filter out more references that we initially intended to
* because the restructure operation re-creates new references for all
elements in the original array, we might later get references from the
queue that are not in the array anymore. This could lead to
"duplicate" removals for the same value
Because several methods in the Segment class have special no-op behavior
when `count == 0`, an invalid count can lead to keys appearing missing
when they are actually still present. In some scenarios, this can
decrease the performance of the cache since values need to be
recalculated.
This commit fixes this inconsistency count issue by first using an
estimate in order to decide whether the array needs a resize and then by
counting the actual numbers of elements inserted in the restructured
array as the new count.
Fixes gh-31373
This commit adds a warning in the reference guide to address the
use cases where users might be tempted to use several property
placeholder configurers.
Closes gh-14623
Prior to this commit, HTTP server observations for Spring WebFlux could
be recorded twice for a single request in some cases. The "COMPLETE" and
"CANCEL" signals would race in the reactive pipeline and would trigger
both the `doOnComplete()` and ` `doOnCancel()` operators, each calling
`observation.stop()` on the current observation.
This would in fact publish two different observations for the same
request.
This commit ensures that the instrumentation uses the `Mono#tap`
operator to guard against this case and only call `Observation#stop`
once for each request.
Fixes gh-31417
Prior to this commit, a cancelled exchange would always result in an
`"status":"UNKNOWN"` KeyValue. This only applied to reactive variants,
as cancelled exchanges are not currently detected for Servlet
implementations.
In some cases, exchanges can be cancelled by clients before they are
completed, but the response was actually received by the client. The
response status information has been set by the application and the
response has been committed. For those cases, we shouldn't assume an
"UNKNOWN" value.
This commit assumes that committed responses have a response status set
by the application and that the observations should reflect that. From
now on, we only assume an "UNKNOWN" status if the response has not been
commited.
Fixes gh-31388
Due to the changes in gh-31341, if the repeat count in a SpEL
expression (using the repeat operator '*') is negative, we throw a
SpelEvaluationException with the MAX_REPEATED_TEXT_SIZE_EXCEEDED
message which is incorrect and misleading.
Prior to gh-31341, a negative repeat count resulted in an
IllegalArgumentException being thrown by String#repeat(), which was
acceptable in terms of diagnostics, but that did not make it
immediately clear to the user what the underlying cause was.
In light of the above, this commit improves diagnostics for a negative
repeated text count in SpEL expressions by throwing a
SpelEvaluationException with a new NEGATIVE_REPEATED_TEXT_COUNT error
message.
Closes gh-31342
If the resulting size of repeated text in a SpEL expression (using the
repeat operator '*') would exceed MAX_REPEATED_TEXT_SIZE, we currently
throw a SpelEvaluationException with the
MAX_REPEATED_TEXT_SIZE_EXCEEDED message.
However, if the calculation of the repeated text size results in
integer overflow, our max size check fails to detect that, and
String#repeat(int) throws a preemptive OutOfMemoryError from which the
application immediately recovers.
To improve diagnostics for users, this commit ensures that we
consistently throw a SpelEvaluationException with the
MAX_REPEATED_TEXT_SIZE_EXCEEDED message when integer overflow occurs.
Closes gh-31341
Prior to this commit the Spring Expression Language (SpEL) was able to
properly parse an expression that uses the safe navigation operator
(?.) with a method that has a `void` return type (for example,
"myObject?.doSomething()"); however, SpEL was not able to evaluate or
compile such expressions.
This commit addresses the evaluation issue by selectively not boxing
the exit type descriptor (for inclusion in the generated bytecode) when
the method's return type is `void`.
This commit addresses the compilation issue by pushing a null object
reference onto the stack in the generated byte code when the method's
return type is `void`.
Closes gh-27421
Prior to this commit, if a Spring Expression Language (SpEL) expression
contained property, field, or method references using the null-safe
navigation operator (?.), the generated AST String representation
incorrectly omitted the '?' characters.
For example, 'myProperty?.myMethod()' had a generated AST string
representation of 'myProperty.myMethod()'.
This commit addresses this by introducing isNullSafe() in
MethodReference and reworking the logic in
CompoundExpression.toStringAST().
Closes gh-31326
Since the rewrite of ConcurrentLruCache in Spring Framework 6.0, an
attempt to create a ConcurrentLruCache with zero capacity results in an
IllegalArgumentException even though the documentation states that zero
capacity indicates "no caching, always generating a new value".
This commit restores the ability to configure a ConcurrentLruCache with
zero capacity and introduces corresponding tests (which were first
verified against the 5.3.x branch to ensure backward compatibility).
See gh-26320
Closes gh-31317
Previously, MessagingMessageListenerAdapter or any adapter relying on
the default MessagingMessageConverter would log an incoming message
with a toString of the Message that does not provide any extra
information. This is due to the default implementation providing a
lazy resolution message that only attempts to extract the payload
when necessary.
This commit implements a toString method that uses the raw JMS message
if the payload is not available. If it is, the payload is used instead.
Closes gh-21265
Given a @Configuration class named org.example.AppConfig which
contains @Bean methods, in Spring Framework 5.3.x and previous
versions, the following classes were created when generating the CGLIB
proxy.
org.example.AppConfig$$EnhancerBySpringCGLIB$$fd7e9baa
org.example.AppConfig$$FastClassBySpringCGLIB$$3fec86e
org.example.AppConfig$$EnhancerBySpringCGLIB$$fd7e9baa$$FastClassBySpringCGLIB$$82534900
Those class names indicate that 1 class was generated for the proxy for
the @Configuration class itself and that 2 additional FastClass
classes were generated to support proxying of @Bean methods in
superclasses.
However, since Spring Framework 6.0, the following classes are created
when generating the CGLIB proxy.
org.example.AppConfig$$SpringCGLIB$$0
org.example.AppConfig$$SpringCGLIB$$1
org.example.AppConfig$$SpringCGLIB$$2
The above class names make it appear that 3 proxy classes are generated
for each @Configuration class, which is misleading.
To address that and to align more closely with how such generated
classes were named in previous versions of the framework, this commit
modifies SpringNamingPolicy so that generated class names once again
include "FastClass" when the generated class is for a CGLIB FastClass
as opposed to the actual proxy for the @Configuration class.
Consequently, with this commit the following classes are created when
generating the CGLIB proxy.
org.example.AppConfig$$SpringCGLIB$$0
org.example.AppConfig$$SpringCGLIB$$FastClass$$0
org.example.AppConfig$$SpringCGLIB$$FastClass$$1
Closes gh-31272
For equivalence, we only need to compare the preInstantiationPointcut
fields since they include the declaredPointcut fields. In addition, we
should not compare the aspectInstanceFactory fields since
LazySingletonAspectInstanceFactoryDecorator does not implement equals().
See gh-31238
This commit expands the scope of equality checks in the implementation
of equals() for PerTargetInstantiationModelPointcut to include all
fields instead of just the pointcut expression for the declared
pointcut.
See gh-31238
The introduction of AdvisedSupport.AdvisorKeyEntry in Spring Framework
6.0.10 resulted in a regression regarding caching of CGLIB generated
proxy classes. Specifically, equality checks for the proxy class cache
became based partially on identity rather than equivalence. For
example, if an ApplicationContext was configured to create a
class-based @Transactional proxy, a second attempt to create the
ApplicationContext resulted in a duplicate proxy class for the same
@Transactional component.
On the JVM this went unnoticed; however, when running Spring
integration tests within a native image, if a test made use of
@DirtiesContext, a second attempt to create the test
ApplicationContext resulted in an exception stating, "CGLIB runtime
enhancement not supported on native image." This is because Test AOT
processing only refreshes a test ApplicationContext once, and the
duplicate CGLIB proxy classes are only requested in subsequent
refreshes of the same ApplicationContext which means that duplicate
proxy classes are not tracked during AOT processing and consequently
not included in a native image.
This commit addresses this regression as follows.
- AdvisedSupport.AdvisorKeyEntry is now based on the toString()
representations of the ClassFilter and MethodMatcher in the
corresponding Pointcut instead of the filter's and matcher's
identities.
- Due to the above changes to AdvisorKeyEntry, ClassFilter and
MethodMatcher implementations are now required to implement equals(),
hashCode(), AND toString().
- Consequently, the following now include proper equals(), hashCode(),
and toString() implementations.
- CacheOperationSourcePointcut
- TransactionAttributeSourcePointcut
- PerTargetInstantiationModelPointcut
Closes gh-31238
This commit reviews when an AOT-generated bean definition defines a
beanClass or targetType. Previously, a beanClass was not consistently
set which could lead to issues.
Closes gh-31242
Prior to this commit, the `RuntimeHintsPredicates` would assume that
registering introspection or invocation hints for "all declared methods"
on a type would also include "all public methods". This is not true, as
the Java reflection API itself behaves differently.
`getDeclaredMethods()` does not return a superset of `getMethods()`, as
the latter can return inherited methods, but not the former.
Same reasoning applies to fields.
This commit fixes the hints predicates to only match if the correct hint
has been registered.
Fixes gh-31224
This commit refines CORS wildcard processing Javadoc to
provides more details on how wildcards are handled for
Access-Control-Allow-Methods, Access-Control-Allow-Headers
and Access-Control-Expose-Headers CORS headers.
For Access-Control-Expose-Headers, it is not possible to copy
the response headers which are not available at the point
when the CorsProcessor is invoked. Since all the major browsers
seem to support wildcard including on requests with credentials,
and since this is ultimately the user-agent responsibility to
check on client-side what is authorized or not, Spring Framework
continues to support this use case.
See gh-31143
Prior to this commit, it was unclear to users and third parties that it
is necessary to manually configure a StandardTypeLocator with a
specific ClassLoader to ensure that the SpEL expression parser is able
to reliably locate user types.
For example, the StandardBeanExpressionResolver in the spring-context
module configures a StandardTypeLocator using the bean ClassLoader of
the corresponding BeanFactory.
This commit improves the documentation to raise awareness of this fact.
Closes gh-26253
gh-30810 introduced explicit support for collections and maps in
ObjectUtils.nullSafeConciseToString() by invoking isEmpty() on a Map or
Collection to determine which concise string representation should be
used. However, this caused a regression in which an exception was
thrown if the Map or Collection was a proxy generated by
AbstractFactoryBean to support <util:set />, <util:list />, and
<util:map /> in XML configuration.
This commit addresses this set of regressions by always returning
"[...]" or "{...}" for a Collection or Map, respectively, disregarding
whether the map is empty or not.
Closes gh-31138
This commit prints a log message at debug level without
a stacktrace for TypeNotPresentException and uses
warn level instead of error level for other exceptions
since the processing of such bean will just be skipped.
Closes gh-31147
This commit revises the FilePatternResourceHintsRegistrar API and
introduces List<String> overrides of various var-args methods used with
the new builder API.
Closes gh-29161
Due to the introduction of the `framework-api` project in the main branch,
we now have 3 `framework-*` projects that should be ignored. I have
therefore removed `/framework-bom/build` and `/framework-docs/build` and
added `/framework-*/build` to ignore all 3 framework projects.
See gh-31049
This commit refines Reactor field precomputing on native
to only compute at build-time fields in the reactor.core
package, since doing so in reactor.netty has unwanted side
effects like Epoll always disabled.
Closes gh-31141
* pr/29753:
Polish "Make sure NoUniqueBeanDefinitionException to be serializable"
Make sure NoUniqueBeanDefinitionException to be serializable
Closes gh-29753
The tests were ignored due to "Fails against TestNG 6.11"; however,
these tests pass against the current version of TestNG that we build
against (7.8.0).
Prior to this commit, when PathMatchingResourcePatternResolver
processed a `file:` pattern (for example, `file:/app-config/**`) for a
`rootPath` that did not exist in the filesystem, the resolver attempted
to search the directory and logged a WARNING message similar to the
following when it failed to do so.
Failed to search in directory [/app-config/] for files matching pattern
[**]: java.nio.file.NoSuchFileException: /app-config/
To avoid unnecessary attempts to search a nonexistent directory,
PathMatchingResourcePatternResolver now skips searching of a nonexistent
directory and preemptively logs an INFO message similar to the
following.
Skipping search for files matching pattern [**]: directory [/app-config]
does not exist
Closes gh-31111
This commit ensures that PathMatchingResourcePatternResolver's
doFindPathMatchingFileResources() method returns a mutable Set in order
to comply with the documented contract.
This commit optimizes ClassUtils#getMostSpecificMethod which is
a method frequently invoked in typical Spring applications.
It refines ClassUtils#isOverridable by considering static and
final modifiers as non overridable and optimizes its implementation.
Closes gh-30272
This commit adds @Nullable to the argValue parameters in the
doSetValue() methods in ArgumentPreparedStatementSetter and
ArgumentTypePreparedStatementSetter.
Closes gh-31086
Prior to this commit, a dynamic property registered via a
@DynamicPropertySource method in a @Nested test class was not able to
override a property registered via a @DynamicPropertySource method in
the enclosing class.
See gh-26091
Closes gh-31083
This commit contributes proxy hints needed by
SharedEntityManagerCreator for
org.hibernate.query.sql.internal.NativeQueryImpl interfaces.
A related smoke test has been contributed via
spring-projects/spring-aot-smoke-tests#188.
Closes gh-29603
This commit turns TypeMismatchException thrown in
ModelAttributeMethodArgumentResolver#createAttribute into
proper ServerWebInputException in order get HTTP response
with 400 Bad Request status code instead of 500 Internal error.
Closes gh-31045
This commit contributes proxy hints needed by
SharedEntityManagerCreator for
org.hibernate.query.sqm.internal.QuerySqmImpl interfaces.
Until Hibernate 6.1, those hints were erroneously provided
at GraalVM reachability metadata level. As of Hibernate 6.2,
they are not, hence the need to contribute them at Spring
Framework level.
A related smoke test has been contributed via
spring-projects/spring-aot-smoke-tests#188.
Closes gh-31050
Reuses ValidationAnnotationUtils which is slightly optimized for the detection of Spring's Validated annotation now, also to the benefit of common web scenarios.
Closes gh-21852
This commit changes Any to Any? in ProceedingJoinPoint
Kotlin examples in order to be consistent with Java
and avoid a "NullPointerException: pjp.proceed() must
not be null" error.
Closes gh-31015
This commit changes Any to Any? in ProceedingJoinPoint
Kotlin examples in order to be consistent with Java
and avoid a "NullPointerException: pjp.proceed() must
not be null" error.
See gh-31015
Prior to this commit, the ignoreResourceNotFound flag in
@PropertySource was ignored by PropertySourceProcessor if a
PropertySourceFactory threw an exception which wrapped an exception
that would otherwise be ignored -- for example, a FileNotFoundException.
To address this issue, this commit updates PropertySourceFactory so
that it catches RuntimeException and IOException and then checks if the
exception or its cause is an "ignorable" exception in terms of
ignoreResourceNotFound semantics.
Closes gh-22276
When the content length is known, use readNBytes on the InputStream in
StringHttpMessageConverter, which avoids some extra copying and allocations.
Closes gh-30942
This commit updates WebMVC converters and WebFlux
encoders/decoders to support custom serializers
with Kotlin Serialization when specified via
a custom SerialFormat.
It also turns the serializers cache to a non-static
field in order to allow per converter/encoder/decoder
configuration.
Closes gh-30870
Prior to this commit, the `SseEventBuilder` would be used to create SSE
events and write them to the connection using the `ResponseBodyEmitter`.
This would send each data item one by one, effectively writing and
flushing to the network for each. Since multiple data lines are prepared
by the `SseEventBuilder`, a typical write of an SSE event performs
multiple flushes operations.
This commit adds a method on `ResponseBodyEmitter` to perform batch
writes (given a `Set<DataWithMediaType>`) and only flush once all
elements of the set have been written.
This also applies in case of early writes, where now all buffered
elements are written then flushed altogether.
Fixes gh-30912
In order to reduce the surface area of published APIs in the affected
classes, this commit:
- Reverts the changes made to GeneratedClasses in c354b1014d.
- Reverts the changes made to DefaultGenerationContext in a28ec3a0a8.
- Makes the DefaultGenerationContext(DefaultGenerationContext, String)
constructor protected.
- Reworks the internals of TestContextGenerationContext to align with
the above changes.
See gh-30861
Closes gh-30895
Closes gh-30897
Aligned with shortcut handling in AutowiredAnnotationBeanPostProcessor.
Includes minor MethodInvoker optimization for pre-resolved targetClass.
Closes gh-30883
Prior to this commit, test AOT processing failed if a feature name for
generated class names was used for more than one ApplicationContext.
For example, when generating code for org.example.MessageService with a
"Management" feature name, the "BeanDefinitions" class was named as
follows (without a uniquely identifying TestContext###_ feature name
prefix).
org/example/MessageService__ManagementBeanDefinitions.java
When another attempt was made to generate code for the MessageService
using the same "Management" feature name, a FileAlreadyExistsException
was thrown denoting that the class/file name was already in use.
To avoid such naming collisions, this commit introduces a
TestContextGenerationContext which provides a custom implementation of
withName(String) that prepends an existing feature name (if present) to
a new feature name, thereby treating any existing feature name as a
prefix to a new, nested feature name.
Consequently, code generation for the above example now results in
unique class/file names like the following (which retain the uniquely
identifying TestContext###_ prefixes).
org/example/MessageService__TestContext002_ManagementBeanDefinitions.java
org/example/MessageService__TestContext003_ManagementBeanDefinitions.java
Closes gh-30861
This commit introduces a `failOnError` flag in TestContextAotGenerator.
When set to `true`, any error encountered during AOT processing will
result in an exception that fails the overall process. When set to
`false` (the default), the previous behavior remains unchanged: a DEBUG
or WARN message will be logged, and processing will continue.
This feature is currently only used for internal testing.
See gh-30861
Closes gh-30898
Previously, BeanInstanceSupplier had three variants of the
`withGenerator` callback, one with a bi function, one with a function,
and with a supplier. This could lead to compilation failure when the
target type has a method with the same name and a number of arguments
that match another variant.
It turns out the supplier-based variant is only used a shortcut. This
commit deprecates it and update ghe code generation to use the function
instead.
Closes gh-29278
After further consideration, the team has decided to remove the
getAutodetectMode() method since its return type conflicts with the
parameter type in setAutodetectMode(int), making it an invalid bean
property.
See gh-30855
This commit changes the documentation to advise using
https://github.com/stomp-js/stompjs library on client
side and to configure by default WebSocket without
SockJS as browsers and proxies now have a pretty good
WebSocket support.
Closes gh-30857
Prior to this commit, MBeanExporter used
org.springframework.core.Constants which used reflection to find
constant fields in the MBeanExporter class. Consequently, one had to
register reflection hints in order to use MBeanExporter in a GraalVM
native image.
This commit addresses this by replacing the use of the `Constants`
class with a simple java.util.Map which maps constant names to constant
values for the autodetect constants defined in MBeanExporter.
See gh-30851
Closes gh-30846
This merges the existing support for the legacy JSR-250 PostConstruct/PreDestroy annotations into CommonAnnotationBeanPostProcessor itself, opening up the InitDestroyAnnotationBeanPostProcessor base class for multiple init/destroy methods in a single post-processor. This removes the need for a separate JSR-250 InitDestroyAnnotationBeanPostProcessor in AnnotationConfigUtils.
Closes gh-30695
This commit documents the fact that the Servlet Filter based
observations for MVC applications is limited by the Servlet Filter
contract in the first place. All processing and logging that happens
outside of the scope of the filter is not observed.
Log statements from the catalina engine (in the case of Tomcat), or any
container-specific infrastructure, is not covered by the
instrumentation.
Closes gh-29398
This commit improves the documentation for the
`ShallowEtagHeaderFilter`, stating that it is only meant to support a
subset of conditional HTTP requests: GET requests with "If-None-Match"
headers. Other headers and state changing HTTP methods are not supported
here, as the filter only operates on the content of the response and has
no knowledge of the resource being served.
Closes gh-30517
Prior to this commit, there was no explicit support for arrays,
collections, and maps in nullSafeConciseToString(). This lead to string
representations such as the following, regardless of whether the array,
collection, or map was empty.
- char[]@1623b78d
- java.util.ImmutableCollections$List12@74fe5c40
- java.util.ImmutableCollections$MapN@10e31a9a
This commit introduces explicit support for arrays, collections, and
maps in nullSafeConciseToString(), which results in the following
empty/non-empty string representations.
- array: {} / {...}
- collection: [] / [...]
- map: {} / {...}
The reason a string representation of an array uses "{}" instead of
"[]" (like in Arrays.toString(...)) is that
ObjectUtils.nullSafeToString(<array>) already follows that convention,
and the implementation of nullSafeConciseToString() aligns with that
for the sake of consistency.
Closes gh-30810
We would preferably use ObjectUtils.nullSafeConciseToString(rejectedValue) here but revert to the full nullSafeToString representation for strict backwards compatibility (programmatic toString calls as well as exception messages).
Closes gh-30799
In addition to the previously addressed removal of bean definitions, this is able to deal with prototype factory methods returning non-null after null or also null after non-null. Stale cached values are getting refreshed rather than bypassed.
Closes gh-30794
This commit extends the list of explicitly supported types in
ObjectUtils.nullSafeConciseToString() with the following.
- Optional
- File
- Path
- InetAddress
- Charset
- Currency
- TimeZone
- ZoneId
- Pattern
Closes gh-30805
This commit also reverts to using ReflectionUtils.findMethod in order
to make the check more robust in case the Micrometer team refactors the
code base and declares the `getObservationRegistry()` method in a super
type.
Prior to this commit, the required runtime dependencies were checked
via reflection each time an attempt was made to instantiate
MicrometerObservationRegistryTestExecutionListener.
Since it's sufficient to check for the presence of required runtime
dependencies only once, this commit caches the results of the
dependency checks in a static field.
This commit also introduces automated tests for the runtime dependency
checks in MicrometerObservationRegistryTestExecutionListener.
See gh-30747
Prior to this commit, dependency checks in the static initialization
block for MicrometerObservationRegistryTestExecutionListener resulted
in an ExceptionInInitializerError which led to verbose logging in
TestContextFailureHandler.
This commit improves the logging for missing dependencies in
MicrometerObservationRegistryTestExecutionListener by moving the
dependency checks to the constructor and by throwing a
NoClassDefFoundError instead of an IllegalStateException. This allows
TestContextFailureHandler to log a concise DEBUG message denoting that
the listener is being skipped due to missing dependencies.
This commit also now checks for the presence of
io.micrometer.context.ThreadLocalAccessor in addition to
io.micrometer.observation.contextpropagation.ObservationThreadLocalAccessor.
Furthermore, this commit now explicitly mentions the need for
io.micrometer:context-propagation in the error message.
The following demonstrate the generated DEBUB message when
ObservationThreadLocalAccessor and ThreadLocalAccessor are missing,
respectively.
Skipping candidate TestExecutionListener [org.springframework.test.context.observation.MicrometerObservationRegistryTestExecutionListener] due to a missing dependency. Specify custom TestExecutionListener classes or make the default TestExecutionListener classes and their required dependencies available. Offending class: io.micrometer.observation.contextpropagation.ObservationThreadLocalAccessor. MicrometerObservationRegistryTestExecutionListener requires io.micrometer:micrometer-observation:1.10.8 or higher and io.micrometer:context-propagation:1.0.3 or higher.
Skipping candidate TestExecutionListener [org.springframework.test.context.observation.MicrometerObservationRegistryTestExecutionListener] due to a missing dependency. Specify custom TestExecutionListener classes or make the default TestExecutionListener classes and their required dependencies available. Offending class: io.micrometer.context.ThreadLocalAccessor. MicrometerObservationRegistryTestExecutionListener requires io.micrometer:micrometer-observation:1.10.8 or higher and io.micrometer:context-propagation:1.0.3 or higher.
Closes gh-30747
Prior to this commit, private (and non-visible package-private)
init/destroy methods were not supported in AOT mode. The reason is that
such methods are tracked using their fully-qualified method names, and
the AOT support for init/destroy methods previously did not take
fully-qualified method names into account. In addition, the invocation
order of init/destroy methods differed vastly between standard JVM mode
and AOT mode.
This commit addresses these issues in the following ways.
- AbstractAutowireCapableBeanFactory.invokeCustomInitMethod(),
DisposableBeanAdapter.determineDestroyMethod(), and
BeanDefinitionPropertiesCodeGenerator.addInitDestroyHint() now parse
fully-qualified method names to locate the correct init/destroy
methods.
- AbstractAutowireCapableBeanFactory and DisposableBeanAdapter delegate
to a new MethodDescriptor record which encapsulates the parsing of
fully-qualified method names; however,
BeanDefinitionPropertiesCodeGenerator duplicates this logic since it
resides in a different package, and we do not currently want to make
MethodDescriptor public.
- Init/destroy methods detected via annotations (such as @PostConstruct
and @PreDestroy) are now invoked prior to init/destroy methods that
are explicitly configured by name or convention. This aligns with the
invocation order in standard JVM mode; however,
InitializingBean#afterPropertiesSet() and DisposableBean#destroy()
are still invoked before annotated init/destroy methods in AOT mode
which differs from standard JVM mode.
- Unit and integration tests have been updated to test the revised
behavior.
Closes gh-30692
This commits changes the return type from Publisher<?> to
Object in order to avoid potential compatibility issues when
the Reactive Streams dependency is not in the classpath.
Closes gh-30716
This commit updates AbstractMessageListenerContainer's Javadoc
regarding the log level used in invokeErrorHandler() so that the
documentation aligns with the implementation, namely that errors will
logged at WARN level if no ErrorHandler has been registered.
Closes gh-30730
In the original implementation of
MicrometerObservationRegistryTestExecutionListener I accidentally
imported JUnit 5's org.junit.platform.launcher.TestExecutionListener
instead Spring's
org.springframework.test.context.TestExecutionListener. The code
therefore attempts to use the ClassLoader for the JUnit Platform's
TestExecutionListener which may fail to see the required types. In
addition, if the JUnit Platform's TestExecutionListener is not on the
classpath, the attempt to access its ClassLoader will fail.
This commit addresses this by properly using the ClassLoader for
Spring's TestExecutionListener to detect dependencies of the
MicrometerObservationRegistryTestExecutionListener.
Closes gh-30726
Prior to this commit, if an init/destroy method was package-private and
declared in a superclass in a package different from the package in
which the registered bean resided, a local init/destroy method with the
same name would effectively "shadow" the method from the different
package, resulting in only the local init/destroy method being invoked.
This commit addresses this issue by tracking package-private init
methods from different packages using their fully-qualified method
names, analogous to the existing support for private init/destroy
methods.
Closes gh-30718
Previously, a bean definition that is optimized AOT could have
different metadata based on whether its resolved type had a generic or
not. This is due to RootBeanDefinition taking either a Class or a
ResolvableType doing fundamentally different things. While the former
sets the bean class which is to little use with an instance supplier,
the latter specifies the target type of the bean.
This commit sets the target type of the bean, using the existing
setter methods that take either a class or a ResolvableType and set the
same attribute consistently.
Closes gh-30689
This commit only tests which init/destroy methods are "registered" in
AOT mode.
This commit does NOT test that the registered methods can actually be
invoked in AOT mode: that will be addressed in a separate commit.
See gh-30692
This commit overhauls the TestExecutionListener for Micrometer's
ObservationRegistry that was introduced in the previous commit.
Specifically, this commit:
- Renames the listener to MicrometerObservationRegistryTestExecutionListener
since the use of a ThreadLocal is an implementation detail that may
change over time.
- Makes the listener package-private instead of public in order to
allow the team greater flexibility in evolving this feature.
- Eagerly loads the ObservationThreadLocalAccessor class and verifies
that it has a getObservationRegistry() method to ensure that the
listener is properly skipped when SpringFactoriesLoader attempts to
load it, if Micrometer 1.10.8+ is not on the classpath.
- Switches the listener's automatic registration order to 2500 in order
to register it after the DependencyInjectionTestExecutionListener.
- Only tracks the previous ObservationRegistry in beforeTestMethod() if
the test's ApplicationContext contains an ObservationRegistry bean.
- Properly removes the TestContext attribute for the previous
ObservationRegistry in afterTestMethod().
- Introduces DEBUG logging for diagnostics.
- Adds an entry in the Javadoc for TestExecutionListener as well as in
the Testing chapter in the reference manual.
Closes gh-30658
Prior to this commit, there was no way to specify the
ObservationRegistry that is registered in the given test's
ApplicationContext as the one that should be used by Micrometer's
ObservationThreadLocalAccessor for context propagation.
This commit introduces a TestExecutionListener for Micrometer's
ObservationRegistry in the Spring TestContext Framework. Specifically,
this listener obtains the ObservationRegistry registered in the test's
ApplicationContext, stores it in ObservationThreadLocalAccessor for the
duration of each test method execution, and restores the original
ObservationRegistry in ObservationThreadLocalAccessor after each test.
Co-authored-by: Sam Brannen <sam@sambrannen.com>
See gh-30658
The dependency on spring-web from spring-beans makes it impossible to
import the projects in Eclipse IDE due to cycles between projects.
This commit therefore moves the web-related test for
BeanUtilsRuntimeHints to spring-web.
See gh-30491
Lazy and read-only database transactions remain exposed to potential isolation level mismatches in case of concurrent transactions with custom isolation levels.
Closes gh-29997
This commit updates JmsAccessor to handle custom JMS acknowledgment
modes as client acknowledge, which is useful when working with JMS
providers that provide non-standard variations of CLIENT_ACKNOWLEDGE,
such as AWS SQS and its UNORDERED_ACKNOWLEDGE mode.
This commit ensures that any storage used for multipart handling only
gets cleaned up if multipart data is actually retrieved via
ServerWebExchange::getMultipartData.
Closes gh-30590
Prior to this commit, if a Spring Expression Language (SpEL) expression
contained indexed access to an object, the generated AST String
representation incorrectly included a dot ('.') before the index access.
For example, 'property[0]' had a generated AST string representation of
'property.[0]'.
This commit addresses this by reworking the logic in
CompoundExpression.toStringAST().
Closes gh-30610
Prior to this commit, when an unsupported character such as "ü" was
encountered in a SpEL expression, the error message was:
Cannot handle (252) 'ü'
With this commit, the error message is now similar to:
Unsupported character 'ü' (252) encountered at position 5 in expression.
See gh-30580
Closes gh-30602
Since EclipseLink applies a custom transaction isolation value to its shared DatabasePlatform instance, we need to immediately restore the original value after the current value got picked up for JDBC Connection access inside of EclipseLink. In order to not interfere with concurrent transactions, we need to use synchronization around the transaction begin sequence in such a case.
Closes gh-29997
Includes hashCode optimization in AbstractBeanFactoryBasedTargetSource.
Includes ThreadLocal naming fix in ThreadLocalTargetSource.
Closes gh-30576
Closes gh-30581
Prior to this commit, the tests we had in place for SpEL 'selector'
support did not assert what happens when a selector expression does not
match or when a selector header is not present.
See gh-30550
gh-23846 introduced a new
MethodArgumentNotValidException(Executable, BindingResult)
constructor that can be advantageously replaced by using
MethodArgumentNotValidException(MethodParameter, BindingResult)
in ModelAttributeMethodProcessor.
This commit updates ModelAttributeMethodProcessor accordingly,
and deprecates MethodArgumentNotValidException(Executable,
BindingResult) in favor of
MethodArgumentNotValidException(MethodParameter, BindingResult).
Closes gh-30558
This commit removes the build configuration for the repository
"libs-spring-framework-build", as it is not needed anymore for the
dependencies it contains.
This commit refines how GraalVM tracing agent detection works
for both test and application executions.
It rolls back the introduction of TestAotDetector done in 111309605c
and instead updates AotDetector.useGeneratedArtifacts()
to only detect "buildtime" and "runtime" imagecode system
property values by leveraging a new method
NativeDetector.inNativeImage(NativeDetector.Context...).
This commit also adds a workaround for
https://github.com/oracle/graal/issues/6691.
Closes gh-30511
This commit adds the "antora" task to the check-project script in order
to check for documentation failures when pushing to the current branch.
The actual documentation is build is performed on the docs-build branch.
Closes gh-30528
Prior to this commit, if a non-annotated [1] class was registered
directly with an ApplicationContext -- for example, via
AnnotatedBeanDefinitionReader, AnnotationConfigApplicationContext, or
@ContextConfiguration -- it was not considered a configuration class in
'lite' mode unless @Bean methods were declared locally. In other words,
if the registered class didn't declare local @Bean methods but rather
extended a class that declared @Bean methods, then the registered class
was not parsed as a @Configuration class in 'lite' mode, and the @Bean
methods were ignored.
Whereas, a non-annotated class registered via @Import is always
considered to be a configuration class candidate.
To address this discrepancy between @Import'ed classes and classes
registered directly with an ApplicationContext, this commit treats any
class registered via AnnotatedBeanDefinitionReader as a @Configuration
class candidate with @Bean 'lite' mode semantics.
[1] In this context, "non-annotated" means a class not annotated (or
meta-annotated) with @Component, @ComponentScan, @Import, or
@ImportResource.
Closes gh-30449
This commit removes the JBoss Application Server external javadoc link,
which covers too many packages and can thus problematically take
precedence for said packages and lead to dead links.
It is replaced by Hibernate javadocs for `org.hibernate.*` packages,
JakartaEE and jsr305 for annotations. We lose a limited number of links
around org.xnio classes and Hibernate-and-JTA-related classes, as a
trade-off.
See gh-30455
See gh-30428
This commit improves how the build deals with javadoc invalid references
in two ways.
Link/see references that are temporarily invalid during javadoc
generation of individual modules are better masked by using the option
`Xdoclint:syntax` instead of `Xdoclint:none` (warnings were still
visible in some cases, e.g. when individually building the javadoc for
a specific module).
Global javadoc-building task `api` now combines `syntax` and `reference`
`Xdoclint` groups, allowing to raise truly invalid references even when
all the modules have been aggregated.
This commit also fixes the 20+ errors which appeared following the later
change in doclet configuration.
Closes gh-30428
In order to restore native support with Hibernate 6.2, this
change updates the PersistenceUnitInfoDescriptor instance
created by SpringHibernateJpaPersistenceProvider in order to
skip the class transformer for native images.
Closes gh-30492
Bean post processors that use InjectionMetadata checks if a property
value for the element it is about to inject is set and skip it, so
that the property value is used. Previously, the AOT contribution for
the same behavior did not check if a matching property value is set
and therefore override the user-defined value.
This commit introduces an additional method that filters the injected
element list so that only the elements that should be processed are
defined.
Closes gh-30476
Prior to this commit, test AOT processing failed when using the GraalVM
tracing agent and GraalVM Native Build Tools (NBT) plugins for Maven
and Gradle.
The reason is that the AOT support in the TestContext framework (TCF)
relied on AotDetector.useGeneratedArtifacts() which delegates
internally to NativeDetector.inNativeImage() which does not
differentiate between values stored in the
"org.graalvm.nativeimage.imagecode" JVM system property.
This commit addresses this issue by introducing a TestAotDetector
utility that is specific to the TCF. This detector considers the
current runtime to be in "AOT runtime mode" if the "spring.aot.enabled"
Spring property is set to "true" or the GraalVM
"org.graalvm.nativeimage.imagecode" JVM system property is set to any
non-empty value other than "agent".
Closes gh-30281
Restores proper event type propagation to parent context.
Selectively applies payload type to given payload object.
Also reuses cached type for regular ApplicationEvent now.
Closes gh-30360
This commit fixes the check by avoiding a fallback to eventType's
hasUnresolvableGenerics(). This could previously lead to checking a
generic event type `A<T>` against a listener which accepts unrelated
`B` and return `true` despite the inconsistency.
Note that this wouldn't necessarily surface to the user because there is
a `catch (ClassCastException e)` down the line, which was primarily put
in place to deal with lambda-based listeners but happens to catch an
exception thrown due to the bad result of `supportsEventType`.
The `supportsEventType` now matches generic `PayloadApplicationEvent`
types with a raw counterpart, using the above fallback only in that case
and otherwise ultimately returning `false`.
Closes gh-30399
This commit adapts the generated code so that each injection point has
a dedicated namespace in the form of a private method. That prevents
the same variable to be reused twice which lead to a compilation failure
previously.
Closes gh-30437
When ReactorNetty2StompBrokerRelayIntegrationTests fail, typically there
are multiple exceptions "Connection refused: /127.0.0.1:61613" that
appear after we've conneted, sent CONNECT, and expecting CONNECTED, but
that does not come within the 10 second timeout.
61613 is the default port for STOMP. However, in all integration tests
we start ActiveMQ with port 0 which results in a random port. Moreover,
the stacktrace is for Netty 4 (not 5), and the eventloop thread id's
are different than the one where the connection to the correct, random
port was established.
The suspicion is that these are log messages from
MessageBrokerConfigurationTests which focuses on testing configuration
but nevertheless as a bean starts and attempts to connect to the default
port and fails. Perhaps those attempts to connect on the default port
somehow affect the ActiveMQ server, and it stops responding.
This change adds a no-op TcpClient in MessageBrokerConfigurationTests
to avoid unnecessary attempts to connect that are not needed.
See gh-29287
The current test were not catching the issue because they request 1
via StepVerifier, wait for it, and then cancel. In the case of
StringDecoder it means all chunks are used up to produce that first
String and so the cancel doesn't catch any cached chunks.
Closes gh-30299
This commit augment the AOT section with best practices we have
experienced while working on the AOT engine. In particular, it describes
how a FactoryBean with an unresolved generic should be handled.
Closes gh-30434
This commit reviews BeanInstanceSupplier to reuse more code from
ConstructorResolver. Previously, the autowired argument resolution was
partially duplicated and this commit introduces a new common path via
RegisteredBean#resolveAutowiredArgument.
Closes gh-30401
This commit adds a workaround for oracle/graal#6529
triggered by b374824319.
When the GraalVM fix will have reached a wide enough
audience, it should be removed via gh-30394.
Closes gh-30407
This commit handles AutowiredCandidateQualifier instances, rather than
relying on qualifiers being statically defined and meta-annotated with
`@Qualifier`.
Closes gh-30410
ServletContext has sets of major/minor version properties that we have
not updated in MockServletContext in several years.
Since we upgraded the baseline to Servlet 6.0 in Spring Framework 6.0,
now seems like a good time to update those version properties.
Closes gh-30395
After b374824319 related
to gh-29246, `"queryAllDeclaredMethods": true` is now added
on all registered beans.
This legit change triggers oracle/graal#6510. This
commit workarounds this GraalVM bug, and should be
removed once the GraalVM fix has reached a wide enough
audience.
Closes gh-30383
This commit makes AOT class names for bean definitions shorter.
Previously, the "__BeanDefinitions" suffix was applied for all
classes, but it was unnecessary for inner classes as the container
class already has the qualifier.
Closes gh-29846
Environment.acceptsProfiles(String...) was deprecated in 5.1 in
conjunction with gh-17063 which introduced a new
acceptsProfiles(Profiles) method to replace it. The deprecated method
only supports OR semantics; whereas, the new method supports profile
expressions. Thus, the goal was to encourage people to use the more
powerful profile expressions instead of the limited OR support with
profile names.
However, there are use cases where it is difficult (if not impossible)
to provide a Profiles instance, and there are use cases where it is
simply preferable to provide profile expressions directly as strings.
To address these issues, this commit introduces a new matchesProfiles()
method in Environment that accepts a var-args list of profile
expressions.
Closes gh-30206
Prior to gh-30325, supplying a null reference for a SpEL expression was
effectively equivalent to supplying the String "null" as the
expression. Consequently, evaluation of a null reference expression
always evaluated to a null reference. However, that was accidental
rather than by design.
Due to the introduction of the checkExpressionLength(String) method in
InternalSpelExpressionParser (in conjunction with gh-30325), an attempt
to evaluate a null reference as a SpEL expression now results in a
NullPointerException.
To address both of these issues,
TemplateAwareExpressionParser.parseExpression() and
SpelExpressionParser.parseRaw() now reject null and empty SpEL
expressions.
Closes gh-30371
* use forEach and putIfAbsent to copy headers in DefaultClientRequestBuilder
* use forEach in ReactorClientHttpRequest and ReactorNetty2ClientHttpRequest
* circumvent ReadOnlyHttpHeaders.entrySet()
* ensure the fast path to LinkedCaseInsensitiveMap for forEach and putIfAbsent exists
Closes gh-29972
This commit polishes an external contribution, ensuring that not just
spaces are encoded as underscores, and that underscores are encoded
as non-printable.
See gh-30252
This commit introduces infrastructure to differentiate between
programmatic setting of a variable in an EvaluationContext versus the
assignment of a variable within a SpEL expression using the assignment
operator (=). In addition, this commit disables variable assignment
within expressions when using the SimpleEvaluationContext.
Closes gh-30326
This commit introduces support for limiting the maximum length of a
string resulting from the concatenation operator (+) in SpEL
expressions.
Closes gh-30324
This commit changes the max regex length in SpEL expressions from 1024
to 1000 in order to consistently use "round" numbers for recently
introduced limits.
See gh-30265
This commit adds a note to an exception in `ConstructorResolver`'s
`autowireConstructor` method hinting that attention should be paid to
cases that mix indexed arguments and named arguments. This is especially
when inheriting bean definitions in xml.
Closes gh-29976
Close gh-PR
In some application setups, the WebSocket server does not transmit
the disconnect message to the client, so that the client has no idea
that the established connection has been terminated.
This issue arises when the application uses SimpleBrokerMessageHandler
and the error handler is set to the instance of
StompSubProtocolErrorHandler or an extended class that does not
override the handleErrorMessageToClient method.
The commit fixes disconnect message population so that
`java.lang.IllegalArgumentException: No StompHeaderAccessor` exception
is not thrown in the handleErrorMessageToClient method in
StompSubProtocolErrorHandler class.
See gh-30120
Prior to this commit, `WebClient` observations would be recorded as
aborted (with tags "outcome":"UNKNOWN", "status":"CLIENT_ERROR")
for use cases like this:
```
Flux<String> result = client.get()
.uri("/path")
.retrieve()
.bodyToFlux(String.class)
.take(1);
```
This is due to operators like `take` or `next` that consume *some*
`onNext` signals and then cancels the subscription before completion.
This means the subscriber is only partially interested in the response
and we should not count this as a client error.
This commit ensures that observations are only recorded as aborted if
the response was not published at the time the CANCEL signal was
received.
The code snippet above will now publish observations with
"outcome":"SUCCESS" and "status":"200" tags, for example.
Closes gh-30070
Prior to this commit, DisposableBeanAdapter attempted to invoke a
configured default-destroy-method on every bean, including beans that
do not declare the named destroy method, resulting in a
NullPointerException being thrown and logged at WARN level.
This commit addresses this by effectively ignoring any nonexistent
destroy method.
Closes gh-30301
This commit ensures that HTTP headers like "text/event-stream"
are correctly forwarded to the converter used in
SseServerResponse for proper pretty print handling.
Close gh-30277
This commit increases the max regex length in SpEL expressions from 256
to 1024 in order to support use cases where a regex may be rather long
without necessarily increasing the complexity of the regex.
Closes gh-30265
This commit adds assertions to MockMvc's CookieresultMatchers:
- `attribute` for arbitrary attributes
- `sameSite` for the SameSite well-known attribute
Note that the `sameSite` methods delegate to their `attribute`
counterparts. Note also that Jakarta's `Cookie#getAttribute` method is
case-insensitive, which is reflected in the documentation of the
`attribute` assertion method and the tests.
Closes gh-30285
ObjectUtils.nullSafeToString(Object) exists for generating a string
representation of various objects in a "null-safe" manner, including
support for object graphs, collections, etc.
However, there are times when we would like to generate a "concise",
null-safe string representation that does not include an entire object
graph (or potentially a collection of object graphs).
This commit introduces ObjectUtils.nullSafeConciseToString(Object) to
address this need and makes use of the new feature in FieldError and
ConversionFailedException.
Closes gh-30286
StringUtils.truncate() serves as central, consistent way for truncating
strings used in log messages and exception failure messages, for
immediate use in LogFormatUtils and ObjectUtils.
See gh-30286
Closes gh-30290
This commit refactors some AssertJ assertions into more idiomatic and
readable ones. Using the dedicated assertion instead of a generic one
will produce more meaningful error messages.
For instance, consider collection size:
```
// expected: 5 but was: 2
assertThat(collection.size()).equals(5);
// Expected size: 5 but was: 2 in: [1, 2]
assertThat(collection).hasSize(5);
```
Closes gh-30104
This commit adds mapping for two types from the `java.time` package,
complementing the types that are already translatable to Sql types
TIME, DATE and TIMESTAMP:
- `OffsetTime` maps to a `TIME_WITH_TIMEZONE`
- `OffsetDateTime` maps to a `TIMESTAMP_WITH_TIMEZONE`
This is in accordance with the B.4 table provided in the JDBC 4.2
specification.
When preparing statements, these `java.time` types use the `setObject`
method. Tests covering the 5 `java.time` classes have also been added.
See gh-28778
See gh-28527
Closes gh-30123
This commit changes the name of two recently introduced methods in the
`MockRestRequestMatchers` class for header and queryParam. These have
been found to cause false negatives in user tests, due to the new
overload taking precedence in some cases.
Namely, using a `Matcher` factory method which can apply to both `List`
and `String` will cause the compiler to select the newest list overload,
by instantiating a `Matcher<Object>`.
This can cause false negatives in user tests, failing tests that used
to pass because the Matcher previously applied to the first String in
the header or queryParam value list. For instance, `equalsTo("a")`.
The new overloads are recent enough and this has enough potential to
cause an arbitrary number of user tests to fail that we break the API
to eliminate the ambiguity, by renaming the methods with a `*List`
suffix.
Closes gh-30220
Closes gh-30238
See gh-29953
See gh-28660
This commit documents the fact that default status handlers configured
on the `WebClient` are not applied to `exchangeTo*` methods as those
variants give full access to the client response.
Applying them here would restrict the ability to adapt the behavior
depending on the HTTP response status.
Closes gh-30059
Prior to this commit, an error thrown by a `ExchangeFilterFunction`
configured on a `WebClient` instance would be recorded as such by the
client observation, but the response details would be missing from the
observation.
All filter functions and the exchange function (performing the HTTP
call) would be merged into a single `ExchangeFunction`; this instance
was instrumented and osberved. As a result, the instrumentation would
only get the error signal returned by the filter function and would not
see the HTTP response even if it was received. This means that the
recorded observation would not have the relevant information for the
HTTP status.
This commit ensures that between the configured `ExchangeFilterFunction`
and the `ExchangeFunction`, an instrumentation `ExchangeFilterFunction`
is inserted. This allows to set the client response to the observation
context, even if a later error signal is thrown by a filter function.
Note that with this change, an error signal sent by a filter function
will be still recorded in the observation.
See gh-30059
This commit updates AbstractMessageWriterResultHandler#writeBody in
order to use the declared bodyParameter instead of
ResolvableType.forInstance(body) when the former has unresolvable
generics.
Closes gh-30214
After this commit, DisposableBeanAdapter can find destruction related methods
even when hints are just specified at interface level, which is typically the
case when a bean is exposed via one of its interfaces.
Closes gh-29545
This commit ensures that init/destroy methods that are provided as
default methods from interfaces are properly covered by runtime hints at
runtime.
Closes gh-29246
Prior to this commit, reflection hints registered for beans was
selectively applied to only consider the methods that we'll actually
need reflection on at runtime. This would rely on an undocumented
behavior of GraalVM Native where calling `getDeclaredMethods` on a type
would only return known metadata at runtime, ignoring the ones that were
not registered during native compilation.
As of oracle/graal#5171, this behavior is now fixed in GraalVM and
aligns with the JVM behavior: all methods will be returned. This means
that if during native compilation, introspection was not registered for
the type a new `MissingReflectionMetadataException` will be raised.
As a follow up of #29205, this commit contributes the "introspection on
declared method" reflection hint for all registered beans.
Closes gh-29246
Updated the Spring Framework documentation to include an example of using Vavr's Try monad to trigger transaction rollbacks when a @Transactional-annotated method returns a Failure. The modified documentation demonstrates how to use Try in a transactional method and how to check if an exception has been wrapped inside a Try.Failure instance. Additionally, a link to the official Vavr documentation was added to provide more information on the Try method.
This commit adds a reflection hint for HttpMethod#valueOf in order
to be able to support conversion required for Spring Boot
configuration properties binding as described in
https://github.com/spring-projects/spring-boot/issues/34483.
Closes gh-30201
The `echoWebSocketHandler()` method is not defined in the
`WebSocketConfigurer` interface and should therefore be annotated with
`@Bean` instead of `@Override`.
Closes gh-30183
After setting up a JDK 20 CI build pipeline, numerous tests involving
date/time parsing and formatting began to fail [0]. (The failing JPA
tests are not specific to JDK 20.)
For example, we encounter visually confusing assertion failures such as
the following.
org.opentest4j.AssertionFailedError:
expected: "12:00 PM"
but was: "12:00 PM"
The expected string contains a normal space (which has always been the
case prior to JDK 20); whereas, the actual string now contains a narrow
non-breaking space.
The cause of this is mentioned in the JDK 20 Release Notes [1] as "NBSP
prefixed to a, instead of a normal space". Note, however, that the
links for the first two bullet points in that section are mixed up.
"NBSP prefixed to a, instead of a normal space" should point to [2].
Furthermore, the new whitespace character is not a non-breaking space
(NBSP) but rather a narrow non-breaking space (NNBSP). In addition, the
second bullet point should technically read "NNBSP prefixed to `a`,
instead of a normal space" -- even though `a` provides limited value to
most readers.
The downside for the Java community is that this constitutes a breaking
change for parsing and formatting date/time values that include "AM"
and "PM" units (any may potentially apply to other date/time
parsing/formatting scenarios). In Spring Framework's test suite we have
witnessed this in conjunction with Spring's @DateTimeFormat and
DateTimeFormatterFactory infrastructure as well as with Google's
Gson-to-JSON support.
A colleague who works at Oracle graciously informed me that one can use
"legacy locale data" by supplying `-Djava.locale.providers=COMPAT` as a
JVM argument, noting however that this option limits some newer
functionalities (but without enumerating which new functionalities one
might be missing when using this option).
In any case, this commit adds that JVM argument to our Gradle toolchain
builds so that our test suite passes on JDK 20, and we will continue to
investigate further options for our builds and for our users.
Note, however, that one must manually configure the
`-Djava.locale.providers=COMPAT` JVM argument when running affected
tests within an IDE.
See gh-30185
[0] https://ge.spring.io/s/kmiq2bz2afafs/tests/overview?outcome=failed
[1] https://jdk.java.net/20/release-notes#JDK-8284840
[2] https://unicode-org.atlassian.net/browse/CLDR-14032
Prior to this commit, the EntityManager was not closed in
SharedEntityManagerCreator.DeferredQueryInvocationHandler's
invoke(Object, Method, Object[]) method if an invocation of
getOutputParameterValue(*) threw an exception other than
IllegalArgumentException, which could lead to a connection leak.
This commit addresses this by catching RuntimeException instead of
IllegalArgumentException.
Closes gh-30161
This commit turns some stream-based iterations back into simpler
enhanced for loops.
For simple use cases like these, where the stream API is merely used to
map/filter + collect to a List, a for loop is more efficient.
This is especially true for small collections like the ones we deal
with in BodyInserters/BodyExtractors here (in the order of 50ns/op vs
5ns/op). These cases are also simple enough that they don't lose in
readability after the conversion.
Closes gh-30136
This commit ensures that the HttpMethod, exposed through
ServerHttpRequest::getMethod, is cached in AbstractServerHttpRequest so
that potentially expensive HTTP method lookups are only done once.
Closes gh-30139
Supplying a large regular expression to the `matches` operator in a
SpEL expression can result in errors that are not very helpful to the
user.
This commit improves the diagnostics in SpEL for the `matches` operator
by throwing a SpelEvaluationException with a meaningful error message
to better assist the user.
Closes gh-30144
Attempting to create repeated text in a SpEL expression using the
repeat operator can result in errors that are not very helpful to the
user.
This commit improves the diagnostics in SpEL for the repeat operator by
throwing a SpelEvaluationException with a meaningful error message in
order to better assist the user.
Closes gh-30142
Prior to this commit, the pattern cache for the SpEL `matches` operator
only applied to expressions such as the following where the same
`matches` operator is invoked multiple times with different input:
"map.keySet().?[#this matches '.+xyz']"
The pattern cache did not apply to expressions such as the following
where the same pattern ('.+xyz') is used in multiple `matches`
operations:
"foo matches '.+xyz' AND bar matches '.+xyz'"
This commit addresses this by moving the instance of the pattern cache
map from OperatorMatches to InternalSpelExpressionParser so that the
cache can be reused for all `matches` operations for the given parser.
Closes gh-30140
This commit adds a link to the now published information on spring.io
about the GPG key used to sign the Spring artifacts published on Maven
Central.
Closes gh-23434
This commit adds an `anchor-rewrite.properties` file in the reference
documentation sources. This allows to redirect from one anchor to
another when they are renamed.
This also populates the file with anchors that were renamed in the AOT
section of the documentation.
Closes gh-30132
Prior to this commit, the `RequestedContentTypeResolverBuilder` would
create a `RequestedContentTypeResolver` that internally delegates to a
list of resolvers. Each resolver would either return the list of
requested media types, or a singleton list with the "*/*" media type; in
this case this signals that the resolver cannot find a specific media
type requested and that we should continue with the next resolver in the
list.
Media Types returned by resolvers can contain parameters, such as the
quality factor. If the HTTP client requests "*/*;q=0.8", the
`HeaderContentTypeResolver` will return this as a singleton list. While
this has been resolved from the request, such a media type should not be
selected over other media types that could be returned by other
resolvers.
This commit changes the `RequestedContentTypeResolverBuilder` so that it
does not select "*/*;q=0.8" as the requested media type, but instead
continues delegating to other resolvers in the list. This means we need
to remove the quality factor before comparing it to the "*/*" for
equality check.
Fixes gh-29915
This commit ensures that methods declared in java.lang.Object (such as
toString() can be invoked on a JDK proxy instance in a SpEL expression.
Closes gh-25316
This commit picks up where the two previous commits left off.
Specifically, this commit:
- Removes the "severity=warning" configuration to ensure that violations
actually fail the build.
- Fixes regular expressions for suppressions by matching forward
slashes using `[\\/]` instead of `\/`.
- Moves the configuration for newly introduced checks to locations in
checkstyle.xml that align with the existing organization of that file.
- Renames the IDs for RegexpSinglelineJava checks from
javaDocPackageNonNullApiAnnotation/javaDocPackageNonNullFieldsAnnotation
to packageLevelNonNullApiAnnotation/packageLevelNonNullFieldsAnnotation,
respectively, since these checks are not related to Javadoc.
- Simplifies the null-safety annotation checks to match against
imported annotation types, which enforces consistency across
package-info.java files for the annotation declarations.
- Simplifies the RegEx for JavadocPackage suppressions to only exclude
packages not under src/main/java (vs src/main) and those in the
framework-docs module.
- Consistently suppresses all checks for the `asm`, `cglib`, `objenesis`,
and `javapoet` packages in spring-core.
- Adds explicit suppressions for null-safety annotations for the `lang`
package in spring-core.
- Adds explicit suppressions for null-safety annotations for the
`org.aopalliance` package in spring-aop.
- Revises the RegEx for null-safety annotation suppressions to only
exclude package-info.java files not under src/main/java and
additionally to exclude package-info.java files in the framework-docs
module as well as those in the spring-context-indexer,
spring-instrument, and spring-jcl modules.
- Adds all missing package-info.java files.
- Adds null-safety annotations to package-info.java files where
appropriate.
Closes gh-30069
This commit updates the project's checkstyle configuration to check
that package-info.java files contain the @NonNullApi and @NonNullFields
null-safety annotations.
See gh-30069
This commit updates the project's checkstyle configuration to require
the existence of a package-info.java file for all packages within the
src/main directory while excluding the framework-docs module and
certain packages inside the spring-core module. A missing
package-info.java file will result in emitting a warning.
See gh-30069
This change fixes a situation where error handling was skipped during
`processCommit()` in case the `doCommit()` failed. The error handling
was set up via an `onErrorResume` operator that was nested inside a
`then(...)`, applied to an inner `Mono.empty()`. As a consequence,
it would never receive an error signal (effectively decoupling the
onErrorResume from the main chain).
This change simply moves the error handling back one level up. It also
simplifies the `doCommit` code a bit by getting rid of the steps that
artificially introduce a `Mono<Object>` return type, which is not really
needed.
A pre-existing test was missing the fact that the rollback didn't occur,
which is now fixed. Another dedicated test is introduced building upon
the `ReactiveTestTransactionManager` class.
Closes gh-30096
Prior to this commit, using a dynamic `CaffeineCacheManager` would rely
on `ConcurrentHashMap#computeIfAbsent` for retrieving and creating cache
instances as needed. It turns out that using this method concurrently
can cause lock contention even when all known cache instances are
instantiated.
This commit avoids using this method if the cache instance already
exists and avoid storing `null` entries in the map. This change reduces
lock contention and the overall HashMap size in the non-dynamic case.
Fixes gh-30066
This commit reverts the changes done via 50a0094a47
in order to ensure consistency across modules (those changes were specific to
spring-core).
Reproducible builds are likely desirable, but it is a complex topic
(see gradle/gradle#14819) that needs to be addressed portfolio wide, with
a better control than what Gradle currently allows (for example to allow
using EPOCH instead of the current 1980 date).
Basic tests did not show an obvious impact on testing avoidance with modern
Gradle versions.
Closes gh-29633
A failure to commit a reactive transaction will complete the
transaction and clean up resources. Executing a rollback at
that point is invalid, which causes an
IllegalTransactionStateException that masks the cause of the
commit failure.
This change restructures TransactionalOperatorImpl and
ReactiveTransactionSupport to avoid executing a rollback after
a failed commit. While there, the Mono transaction handling in
TransactionalOperator is simplified by moving it to a default
method on the interface.
Closes gh-27572
If content-length is available, pass it to readNBytes in
ByteArrayHttpMessageConverter. When the content length is less than
the internal buffer size in InputStream (8192), this avoids a copy,
as readNBytes will return the buffer directly. When the content length
is greater than the buffer size used in InputStream, passing the
content-length at least avoids over-allocating the final buffer (e.g.,
if the content length were 8193 bytes, 1 byte more than the default
buffer size).
If the content length isn't present or is too large to represent as
an integer, fall back to the default behavior of readAllBytes by
passing in Integer.MAX_VALUE.
See gh-30010
Before this commit, anonymous classes could throw an
unexpected NullPointerException in
ReflectionsHint#registerType and lambdas entries could
be created in the related generated reflect-config.json.
This commit refines how anonymous classes are handled by
explicitly checking for null class and canonical name in
ReflectionTypeReference#of, while skipping such class in
ReflectionHints#registerType in order to keep a lenient
behavior.
Closes gh-29774
We no longer have any JDK 5 related limitations imposed on us as was
the case when MetadataAwareAspectInstanceFactory was introduced; however,
MetadataAwareAspectInstanceFactory will remain as a specialized
sub-interface of AspectInstanceFactory.
Rationale: changes in org.springframework.aop.framework.autoproxy.target
were previously ignored since the "target" package was ignored by the
previous "eager" pattern for "target" folders.
This commit uses a local variable for the creation of a new JMS
Connection so that a rare failure in prepareConnection(...) does not
leave the connection field in a partially initialized state.
If such a JMSException occurs, the intermediary connection is closed.
This commit further defends against close() failures at that point,
by logging the close exception at DEBUG level. As a result, the original
JMSException is always re-thrown.
Closes gh-29116
See gh-29115
HttpServiceProxyFactoryExtensions.kt has been mistakenly created
in spring-webflux module instead of spring-web, breaking JPMS for
WebFlux users.
This commit moves this file and related tests to the spring-web
module.
Closes gh-30042
An ArrayIndexOutOfBoundsException is thrown by
Validator.getConstraintsForClass when processing Kotlin beans
with extensions functions (Kotlin or Hibernate Validator bug).
This commit catches those exceptions and report them as warning
without the full stactrace, and report as well other
ones thrown as errors with the full stracktrace.
Closes gh-30037
The extensive use of shared pointcuts can be confusing to users who
read a given AspectJ example without having read the "Sharing Common
Pointcut Definitions" section of the AOP chapter.
The primary purpose of this commit is therefore to reduce the extensive
use of shared pointcuts in AspectJ examples by favoring "inline
pointcuts" over shared "named pointcuts" and by cross referencing the
declaration of shared pointcuts when they are used.
Since the affected parts of the documentation are intertwined, this
commit also refactors the examples to use a common "com.xyz" package
namespace for greater clarity.
In order to avoid the need to escape asterisks (\*) in examples, this
commit also removes the `subs="quotes"` configuration for code blocks.
Closes gh-30003
This commit refines CoroutinesUtils#invokeSuspendingFunction in order
to clarify the behavior when used on a non suspending function, and
support usages with or without the Continuation argument.
Closes gh-30005
This commit modifies PerThisAspect (which is used indirectly in
AspectJAutoProxyCreatorTests.perThisAspect()) so that it uses a shared
@Pointcut for both the @Aspect and @Around declarations, in order to
refute claims made in gh-29998 that the documentation in the reference
manual is incorrect.
This commit ensures that WebFlux's RequestMethodsRequestCondition
supports HTTP methods that are not in the RequestMethod enum.
- RequestMethod::resolve is introduced, to convert from a HttpMethod
(name) to enum values.
- RequestMethod::asHttpMethod is introduced, to convert from enum value
to HttpMethod.
- HttpMethod::valueOf replaced Map-based lookup to a switch statement
- Enabled tests that check for WebDAV methods
See gh-27697
Closes gh-29981
This commit refines ModelAttributeMethodProcessor Kotlin exception
handling in order to throw a proper MethodArgumentNotValidException
instead of a NullPointerException when Kotlin null-safety constraints
are not fulfilled, translating to an HTTP error with 400 status code
(Bad Request) instead of 500 (Internal Server Error).
Closes gh-23846
This commit refines the instance supplier check in
BeanDefinitionMethodGenerator constructor in order to allow
overriding by an AOT contribution.
Closes gh-29556
MethodArgumentTypeMismatchException and
MethodArgumentConversionNotSupportedException are TypeMismatchException
subclasses with MethodParameter information and should initialize
propertyName in TypeMismatchInformation.
Closes gh-29959
This commit makes several changes to PR #24651.
- Add byte[] getContentAsByteArray() on Resource.
- Remove getContentAsString() from Resource, as it relied on the default
charset which is not reliable.
- Add getContentAsString() to EncodedResource, as a charset is provided
through the constructor.
See gh-24651
This commit adds a `header` variant and a `queryParam` variant to the
`MockRestRequestMatchers` API which take a single `Matcher` over the
list of values.
Contrary to the vararg variants, the whole list is evaluated and the
caller can choose the desired semantics using readily-available iterable
matchers like `everyItem`, `hasItems`, `hasSize`, `contains` or
`containsInAnyOrder`...
The fact that the previous variants don't strictly check the size of the
actual list == the number of provided matchers or expected values is
now documented in their respective javadocs.
See gh-28660
Closes gh-29953
This commit ensures that CoroutineContext is properly
propagated in transactional suspending functions. Both
annotation and functional variants are supported.
Closes gh-27308
This commit introduces DataBuffer::readableByteBuffers and
DataBuffer::writableByteBuffers, allowing restricted access to the
ByteBuffer used internally by DataBuffer implementations.
Closes gh-29943
Includes small refactoring in DefaultServerWebExchange and adjustment
of initMultipartData to get involved for any "multipart/" prefixed
media type.
In addition, "multipart/related" is now in the list of media types
supported by FormHttpMessageConverter, which aligns it with
MultipartHttpMessageReader.
Closes gh-29671
Prior to this commit, the term "field" was sometimes used to refer to a
database column and sometimes used to refer to a bean property, which
lead to confusion in the Javadoc as well as within the code.
This commit addresses this by avoiding use of the term "field".
This change restricts the maximum number of forwards in MockMvcWebConnection to 100,
in case a forward is configured in a way that causes a loop. This is necessary in HtmlUnit
backed tests, unlike in classic MockMvc tests in which the forwards are not actually resolved.
Closes gh-29483
Closes gh-29557
Co-authored-by: Simon Baslé <sbasle@vmware.com>
When using R2dbcTransactionManager, connection will not be released if
it encounters error while doing `afterCleanup` steps. As `afterCleanup`
can use a database connection when doing `setAutoCommit(true)`, it can
fail under some conditions where the connection is not reliable.
This leads to the Connection not being released.
This commit ensures that inner steps of the `doCleanupAfterCompletion`
are protected against errors, logging the errors and continuing the
cleanup until the last step, which releases the connection.
Closes gh-29703
Co-authored-by: Simon Baslé <sbasle@vmware.com>
The `ServerHttpObservationFilter` implementations record observations
for processed HTTP exchanges. The `Observation.Context` contains various
metadata contributed by the observation convention. The instrumentation
can also mark the observation as an error by setting any `Throwable` on
the context.
Because the instrumentation is done as filters, only exceptions reaching
the filter can be considered. Any error handled at a lower level by the
Framework can, or cannot be considered as an error for an observation.
This commit documents how a web application should opt-in for
considering a handled exception as an error for the current observation.
Closes gh-29848
This commit documents how Observation instrumentation should be
activated for `RestTemplate` and `WebClient`: they both need an
`ObservationRegistry` configured to create and record actual
observations.
This is being done automatically in Spring Boot if auto-configured
builders (`RestTemplateBuilder`, `WebClient.Builder`) are used.
Closes gh-29904
Before this commit, TransactionalOperator.executeAndAwait had a rigid
null-safety handling. This commit takes advantage of Kotlin capability
to allow to deal with both non-null and nullable depending on the
nullability of the lambda.
Closes gh-29919
This commit introduces new AOT processors that look for
`@RSocketExchange` annotated methods on interfaces implemented by beans
and registers reachability metadata accordingly:
* JDK proxies for the beans themselves
* invocation reflection for annotated methods
* binding reflection for arguments and return types
This allows to compile such clients to Native Images.
Closes gh-29877
This commit ensures the ContentDisposition class prints the filename in
both in the regular filename parameter and the extended filename*
parameter (RFC 5987).
Quoted printable (RFC 2047) is used to encode any non-ASCII characters
in the regular filename parameter.
Closes gh-29861
Prior to this commit, a recent change applied in gh-29125 changed the
behavior of `MockHttpServletRequest` instances. In case of an empty
request body, the returned `InputStream` would be static and could not
be reused across requests.
This could result in `java.io.IOException: Stream closed` exceptions if
a previous request was read.
This commit ensures that a new instance of an empty stream is returned
for each request instance.
Fixes gh-29901
This commit ensures that we register a result callback when executing
the request (next to the existing response callback), which gets
notified of invalid hosts and other connection issues.
Closes gh-29156
In the checkHeaders method, if allowedHeaders contains * the result
size can be requestHeaders.size(), but if this is not the case
then the result size can only be as big as the minimum value
of requestHeaders.size() and allowedHeaders.size().
requestHeaders.size() can be potentially big (user input) and
allowedHeaders.size() could be small, so this saves memory in
that case.
Closes gh-29853
This commit leverages Hibernate Validator's own internal use
of standard Java reflection to perform validation on suspending
function, which fixes the ArrayIndexOutOfBoundsException previously
observed.
Validation of suspending function return values remains unsupported
as Hibernate Validator is not Coroutines aware.
Closes gh-23499
Add constructors to HttpMediaTypeNotSupportedException and
UnsupportedMediaTypeStatusException for a parse error that also accept
the list of supported media types to include in the response headers.
Closes gh-28062
The previous commit changed the generated default name for a JMS
subscription to <FQCN>#<method name> -- for example:
- org.example.MyListener#myListenerMethod
However, the JMS spec does not guarantee that '#' is a supported
character. This commit therefore changes '#' to '.' as the separator
between the class name and method name -- for example:
- org.example.MyListener.myListenerMethod
This commit also introduces tests and documentation for these changes.
See gh-29790
Prior to this commit, when using durable subscribers with @JmsListener
methods that do not specify a custom subscription name the generated
default subscription name was always
org.springframework.jms.listener.adapter.MessagingMessageListenerAdapter.
Consequently, multiple such @JmsListener methods were assigned the
same subscription name which violates the uniqueness requirement.
To address this, MessagingMessageListenerAdapter now implements
SubscriptionNameProvider and generates the subscription name based on
the following rules.
- if the InvocableHandlerMethod is present, the subscription name will
take the form of handlerMethod.getBeanType().getName() + "#" +
handlerMethod.getMethod().getName().
- otherwise, getClass().getName() is used, which is analogous to the
previous behavior.
Closes gh-29790
Prior to this commit, the "uri" KeyValue for low cardinality metadata
would contain the entire uri template given to the HTTP client when
creating the request. This was a breaking change for existing metrics
dashboards, as previous support was removing the protocol, host and port
parts of the URI.
Indeed, this information is available in the "client.name" and
"http.uri" KayValue.
This commit parses and removes the protocol+host+port information from
the uri template for the "uri" KeyValue.
Fixes gh-29885
Prior to this commit, client HTTP requests performed by `WebClient`
could miss the "uri" KeyValue for simple "/" requests.
This can happen when the baseUri is configured for the client with a
host and a root base path like "https://example.org/"; given the nature
of the `WebClient` API, in these cases, one can perform requests like
this:
```
WebClient client = WebClient.builder()
.observationRegistry(registry)
.baseUrl("https://example.org/")
.build();
String response = client.get().retrieve().bodyToMono(String.class).block();
```
Such a call would contribute a `"none"` value for the `"uri"` KeyValue.
While only templates should be allowed for this keyvalue, we can assume
that requests to `"/"` should be recorded anyway and won't cause
cardinality explosion.
Fixes gh-29879
Prior to this commit, we included Hibernate's Query.list() method in
SharedEntityManagerCreator's queryTerminatingMethods set but did not
include all of Hibernate's query-terminating methods.
To address this, this commit additionally includes the stream(),
uniqueResult(), and uniqueResultOptional() methods from Hibernate's
Query API in SharedEntityManagerCreator's query-terminating methods set.
Closes gh-29886
Prior to this commit, the `DefaultWebClient` would be instrumented for
client observations and would start/stop a `"http.client.requests"`
observation. This would not set this new observation as the current one
in the Reactor context under `ObservationThreadLocalAccessor.KEY`.
This means that potential child observations would not detect it as
their parent; this can happen if the Reactor Netty `HttpClient`
observation is enabled.
This commit ensures that the reactor context is properly populated for
upstream operators.
Fixes gh-29891
This commit modifies the `DefaultDatabaseClient` implementation in order
to ensure lazier usage of the `Supplier<String>` passed to the sql
method (`DatabaseClient#sql(Supplier)`).
Since technically `DatabaseClient` is an interface that could have 3rd
party implementations, the lazyness expectation is only hinted at in the
`DatabaseClient#sql` javadoc.
Possible caveat: some log statements attempt to reflect the now lazily
resolved SQL string. Similarly, some exceptions can capture the SQL that
caused the issue if known. We expect that these always occur after the
execution of the statement has been attempted (see `ResultFunction`).
At this point the SQL string will be accessible and logs and exceptions
should reflect it as before. Keep an eye out for such strings turning
into `null` after this change, which would indicate the opposite.
Closes gh-29367
Prior to this commit, the `"client.name"` key value for the
`"http.client.requests"` client HTTP observations would be considered as
high cardinality, as the URI host is technically unbounded.
In practice, the number of hosts used by a client in a given application
can be considered as low cardinality. This commit moves this keyvalue to
low cardinality so that it's present for both metrics and traces.
Closes gh-29839
This commit introduces support for Optional in the
AbstractJackson2HttpMessageConverter, similar the existing support for
collection types were supported.
Closes gh-24498
This commit introduces a BeanValidationBeanRegistrationAotProcessor
which adds reflection hints for custom ConstraintValidator discovered
on beans constructors, methods and properties.
Closes gh-29823
Prior to this commit, the reactive `ResourceWebHandler` would only look
at the path within the current mapping when resolving static resources
to be served. This means that when registering a handler at
`"/resources/**"` with a `"classpath:/static/"` location, the handler
would process a `"GET /resources/file.txt"` as the `"/static/file.txt"`
classpath location.
When a developer registers a fixed pattern like `"/resources/file.txt"`
with the same location, the path within the handler mapping is empty as
there is no dynamic part in the given pattern. While the typical use
case for this feature is to register multiple resources at once with a
pattern, we should support a single registration like this.
This commit ensures that if the matching `PathPattern` for the current
request does not have a pattern syntax (i.e. no regexp, no wildcard), we
can use it to match the resource directly. Otherwise, we can use the
path within the handler mapping to resolve the resource as before.
Closes gh-29739
This commit introduces support for double-quoted HTTP header values in
HttpHeaders::getValuesAsList, as described in RFC 9110 section 5.5.
Closes gh-29785
Ensure the port used by the client in malformedResponseChunksOnBodilessEntity
and malformedResponseChunksOnEntityWithBody has correctly been set.
Closes gh-29862
This commit changes the FatalBeanException previously thrown for
an IllegalArgumentException which seems more suitable for that
use case.
Closes gh-29859
It is by design not possible to generate code that handles
bean definitions with user-provided instance suppliers because
the JVM does not allow to get a stable reference reusable at
runtime on the lambda or method reference in the code generated
AOT.
Before this commit, such instance supplier was ignored.
After this commit, an IllegalArgumentException is thrown,
allowing projects to be aware this is not supported and enforce
related refactorings.
The related issue gh-29555 describes how this limitation could
be relaxed in the future.
Closes gh-29556
When comparing empty ProducesRequestCondition, compareTo would throw an
IllegalStateException if the Accept header was invalid. This commit
fixes that behavior.
Closes gh-29794
Originally, `Base64Utils` was providing base64 encoding and decoding
utilities, bridging to commons-codecs or Java 8, if available. Since
then, only the Java 8 variant remains and Spring Framework 6 requires
now Java 17. This utility class doesn't provide additional checks or
syntactic sugar over what's in Java already.
As a result, this commit deprecates this class in favor of `Base64` and
schedules the removal of this class completely.
Closes gh-28434
Prior to this commit, `ConcurrentReferenceHashMap` subclasses extending
`Reference` would call `Reference::clear` right next to
`Reference::enqueue`; this is not needed as the latter calls it already.
Closes gh-28863
Prior to this commit, a Spring web application would not be able to
serialize a `ProblemDetail` object instance in a Native Image, as
serialization hints would be missing.
This commit adds a RuntimeHints registrar that processes the
`ProblemDetail` type accordingly during the AOT phase.
Fixes gh-29801
When logging the current connection inside R2dbcTransactionManager
doBegin, the mono object was logged instead of the connection lambda
parameter.
Other similar debug-level logs do use the actual Connection object,
so this commit does the same.
This commit takes rfc8297 into account and introduces a newer code 103
HttpStatus value which uses `Early Hints` as the more correct reason
phrase, deprecating the outdated `CHECKPOINT` enum value for 103.
Additionally:
- `HttpStatus.valueOf(103)` will return the new enum value
- `HttpStatusCode#isSameCodeAs(HttpStatusCode)` is introduced to ease
comparison of deprecated enums vs their newer counterparts (or any
instance of a more generic `HttpStatusCode`) by comparing the integer
`value()`
- `HttpStatusTests` covers the new deprecation as well as the three
previously deprecated codes, including a check with the above new
method to ensure they have comparable integer values
Supersedes and Closes gh-27960
Kotlin reflection API invocation on a specific function
may require iterating on all Java methods to find the right
Kotlin function. As a consequence, this commit adds introspection
hints on the class declared methods for all Kotlin beans since
the impact on the footprint is low.
Closes gh-29663
This commit infers the reflection hints required for converters
when they are specified with the @Convert annotation at class or
field level.
It also refines the hints generated for @Converter in order
to just generate the construct hint, not the public method one
which should be not needed.
Closes gh-29771
As the DSL internally calls `ResultActions.andExpect`, this is done with
a trick where a synthetic `ResultActions` is provided at top level which
stores each `ResultMatcher` in a mutable list.
Once the DSL usage is done, the top level DSL `andExpectAll` turns that
list into a `vararg` passed down to the actual `actions.andExpectAll`.
Closes gh-27317
Prior to this commit, extracting the path within handler mapping would
result in "" if the matching path element would be a Regex and contain
".*". This could cause issues with resource handling if the handler
mapping pattern was similar to `"/folder/file.*.extension"`.
This commit introduces a new `isLiteral()` method in the `PathElement`
abstract class that expresses whether the path element can be compared
as a String for path matching or if it requires a more elaborate
matching process.
Using this method for extracting the path within handler mapping avoids
relying on wildcard count or other properties.
Fixes gh-29712
The block title style previously used was not rendered in HTML and the
title couldn't be differentiated from the text. Though, it was in the
PDF, as italics.
Introducing delimited blocks in the open (`--`) style did introduce
styling, but the vertical alignment isn't great.
This commit turns these block titles to actual (deep) section titles.
In the final HTML, at this depth there is no numbering but bold styling
is there. The PDF rendering has also been verified to have relevant
style.
Closes gh-29694
The link was previously named "Compared to WebFlux", which is easy to
mix up with the various links to equivalent sections in the WebFlux
chapter. Here the links point to a small section comparing the Servlet
Async API to the WebFlux stack from a high perspective.
In this commit we eliminate most of these links, except at the
beginning of the Asynchronous section. We also add a small mention of
the Servlet configuration in the comparison paragraphs, since the
Configuring section is the one furthest from the comparison paragraphs
that used to have a link to it.
Closes gh-29694
This commit extracts the DTD/XSD remote lookup fallback from the
resolveEntity() method into a protected method.
A WARN-level logging statement is added to the extracted fallback in
order to make it clear that remote lookup happened.
Overriding the protected method would allow users to avoid this
fallback entirely if it isn't desirable, without the need to duplicate
the local resolution code.
Closes gh-29697
The "manipulating property sources" example in the Javadoc for
`ConfigurableEnvironment` states that `MutablePropertySources`
expect a `Map<String,String>`; whereas it expects a
`Map<String,Object>`.
Closes gh-29693
This commit ensures that HTTP methods not supported by HttpServlet, for
instance WebDAV methods, are still supported in FrameworkServlet.
Closes gh-29689
Prior to this commit, there was a bug in the implementation of
StandardRepeatableContainers.computeRepeatedAnnotationsMethod() which
has existed since Spring Framework 5.2 (when
StandardRepeatableContainers was introduced). Specifically,
StandardRepeatableContainers ignored any repeatable container
annotation if it declared attributes other than `value()`. However,
Java permits any number of attributes in a repeatable container
annotation.
In addition, the changes made in conjunction with gh-20279 made the bug
in StandardRepeatableContainers apparent when using the
getMergedRepeatableAnnotations() or findMergedRepeatableAnnotations()
method in AnnotatedElementUtils, resulting in regressions for the
behavior of those two methods.
This commit fixes the regressions and bug by altering the logic in
StandardRepeatableContainers.computeRepeatedAnnotationsMethod() so that
it explicitly looks for the `value()` method and ignores any other
methods declared in a repeatable container annotation candidate.
Closes gh-29685
This commit ensures that the same multipart codecs are registered on
both client and server. Previously, only the client enabled only sending
multipart, and the server only receiving.
Closes gh-29630
This commit introduces additional constructors in MockClientHttpRequest
and MockClientHttpResponse that were previously only present in the
internal "test fixtures" in spring-web.
This commit also aligns the mocks in spring-test with the test fixtures
in spring-web to simplify continued maintenance of the mocks and test
fixtures.
Closes gh-29670
Prior to this commit, SpEL's ConstructorReference did not provide
support for arrays when generating a string representation of the
internal AST. For example, 'new String[3]' was represented as 'new
String()' instead of 'new String[3]'.
This commit introduces support for standard array construction and array
construction with initializers in ConstructorReference's toStringAST()
implementation.
Closes gh-29665
Prior to this commit, there were two bugs in the support for quotes
within String literals in SpEL expressions.
- Two double quotes ("") or two single quotes ('') were always replaced
with one double quote or one single quote, respectively, regardless
of which quote character was used to enclose the original String
literal. This resulted in the loss of one of the double quotes when
the String literal was enclosed in single quotes, and vice versa. For
example, 'x "" y' became 'x " y'.
- A single quote which was properly escaped in a String literal
enclosed within single quotes was not escaped in the AST string
representation of the expression. For example, 'x '' y' became 'x ' y'.
This commit fixes both of these related issues in StringLiteral and
overhauls the structure of ParsingTests.
Closes gh-29604, gh-28356
This commit ensures that giving an anonymous class for reflection hints
registration does not result in a NullPointerException, since the
canonical name of anonymous classes is null.
Fixes gh-29657
Spring Framework 6.0 GA introduced a regression in the component index
support for Jakarta annotations such as @Named and @ManagedBean.
Prior to this commit, @Named and @ManagedBean components were
registered in the component index at build time; however, component
scanning failed to find those component at run time.
This commit updates ClassPathScanningCandidateComponentProvider so that
`jakarta.*` annotation types are once again supported for component
scanning via the component index at run time.
Closes gh-29641
Prior to this commit, the `ConcurrentLruCache` implementation would use
arrays of `AtomicReference` as operation buffers, and the buffer count
would be calculated with the nearest power of two for the CPU count.
This can result in significant heap memory usage as each
`AtomicReference` buffer entry adds to the memory pressure. As seen in
FasterXML/jackson-databind#3665, this can add a significant overhead for
no real added benefit for the current use case.
This commit changes the current implementation to use
`AtomicReferenceArray` as buffers and reduce the number of buffers.
JMH benchmarks results are within the error margin so we can assume that
this does not change the performance characteristics for the typical use
case in Spring Framework.
Fixes gh-29520
This commit deprecates the `"application/graphql+json"` media type in
favor of the new `"application/graphql-response+json"`, since the former
has been removed in graphql/graphql-over-http#215.
Closes gh-29617
This ensures that by default the reason is used to set the "detail"
field. It's a follow-up fix to a27f2e994b
which resolved the issue partially.
Closes gh-29608
In order to automate maintenance of copyright headers in our source
code (especially when merging PRs from external contributors), this
commit introduces an update_copyright_headers.sh script (tested on
mac OS) that will update the copyright headers of all Java, Kotlin,
and Groovy source files that have been added or modified this year
(or at least as far back as the git log history supports it).
For example, running this script currently outputs the following.
Updating copyright headers in Java, Kotlin, and Groovy source code for year 2022
warning: log for 'main' only goes back to Tue, 16 Aug 2022 16:24:55 +0200
- group code example callouts to ensure callouts are displayed for the
correct examples
- add missing callouts
- fix syntax, annotation attribute names, etc.
Prior to this commit, if AOT processing of a test's ApplicationContext
failed, the TestContext Framework (TCF) still attempted to load the
context via standard (JVM) mechanisms when running in AOT mode.
For example, if a test class uses Spring Boot's @MockBean, AOT
processing of that test's context will fail with a WARN log message,
and there will no mapping from that test class to an AOT-generated
ApplicationContextInitializer (ACI). Consequently, when the test suite
is run in AOT mode that particular test class will currently fail with
a confusing stack trace due to the fact that Spring Boot's
SpringApplication attempts to locate a "main" ACI instead of the
missing "test" ACI (missing because it was not generated during AOT
processing).
In general, the TCF should not attempt to load an ApplicationContext in
"JVM mode" while running in "AOT mode".
This commit therefore reworks the logic in
DefaultCacheAwareContextLoaderDelegate to fail fast (with a meaningful
error message) if an AOT-generated ACI cannot be found while running in
AOT mode. This avoids the aforementioned confusion when @MockBean tests
fail in AOT mode (on the JVM or within a native image), and it also
helps to diagnose build problems if AOT processing has not yet been
performed for the project's test suite.
Closes gh-29579
Reflective getHeaders calls to be revisited; see GitHub issue #8938 in Jetty project.
HttpOutput optimization commented out still in order to avoid alpha build dependency.
See gh-29575
# <img src="framework-docs/src/docs/spring-framework.png" width="80" height="80"> Spring Framework [](https://ci.spring.io/teams/spring-framework/pipelines/spring-framework-5.3.x?groups=Build") [](https://ge.spring.io/scans?search.rootProjectNames=spring)
# <img src="framework-docs/src/docs/spring-framework.png" width="80" height="80"> Spring Framework [](https://github.com/spring-projects/spring-framework/actions/workflows/build-and-deploy-snapshot.yml?query=branch%3A6.0.x) [](https://ge.spring.io/scans?search.rootProjectNames=spring)
This is the home of the Spring Framework: the foundation for all [Spring projects](https://spring.io/projects). Collectively the Spring Framework and the family of Spring projects are often referred to simply as "Spring".
// TODO Uncomment link to JUnit 5 docs once we have sorted out
// the following warning in the build.
// TODO Uncomment link to JUnit 5 docs once we execute Gradle with Java 18+.
// See https://github.com/spring-projects/spring-framework/issues/27497
//
// warning: The code being documented uses packages in the unnamed module, but the packages defined in https://junit.org/junit5/docs/5.9.1/api/ are in named modules.
The Spring Framework uses https://concourse-ci.org/[Concourse] for its CI build and other automated tasks.
The Spring team has a dedicated Concourse instance available at https://ci.spring.io with a build pipeline
for https://ci.spring.io/teams/spring-framework/pipelines/spring-framework-6.0.x[Spring Framework 6.0.x].
=== Setting up your development environment
If you're part of the Spring Framework project on GitHub, you can get access to CI management features.
First, you need to go to https://ci.spring.io and install the client CLI for your platform (see bottom right of the screen).
You can then login with the instance using:
[source]
----
$ fly -t spring login -n spring-framework -c https://ci.spring.io
----
Once logged in, you should get something like:
[source]
----
$ fly ts
name url team expiry
spring https://ci.spring.io spring-framework Wed, 25 Mar 2020 17:45:26 UTC
----
=== Pipeline configuration and structure
The build pipelines are described in `pipeline.yml` file.
This file is listing Concourse resources, i.e. build inputs and outputs such as container images, artifact repositories, source repositories, notification services, etc.
It also describes jobs (a job is a sequence of inputs, tasks and outputs); jobs are organized by groups.
The `pipeline.yml` definition contains `((parameters))` which are loaded from the `parameters.yml` file or from our https://docs.cloudfoundry.org/credhub/[credhub instance].
You'll find in this folder the following resources:
* `pipeline.yml` the build pipeline
* `parameters.yml` the build parameters used for the pipeline
* `images/` holds the container images definitions used in this pipeline
* `scripts/` holds the build scripts that ship within the CI container images
* `tasks` contains the task definitions used in the main `pipeline.yml`
=== Updating the build pipeline
Updating files on the repository is not enough to update the build pipeline, as changes need to be applied.
The pipeline can be deployed using the following command:
| Instructs Spring to ignore operating system environment variables if a Spring
@@ -43,12 +41,12 @@ for details.
| `spring.index.ignore`
| Instructs Spring to ignore the components index located in
`META-INF/spring.components`. See <<core.adoc#beans-scanning-index, Generating an Index
of Candidate Components>>.
`META-INF/spring.components`. See xref:core/beans/classpath-scanning.adoc#beans-scanning-index[Generating an Index of Candidate Components]
.
| `spring.jdbc.getParameterType.ignore`
| Instructs Spring to ignore `java.sql.ParameterMetaData.getParameterType` completely.
See the note in <<data-access.adoc#jdbc-batch-list, Batch Operations with a List of Objects>>.
See the note in xref:data-access/jdbc/advanced.adoc#jdbc-batch-list[Batch Operations with a List of Objects].
| `spring.jndi.ignore`
| Instructs Spring to ignore a default JNDI environment, as an optimization for scenarios
@@ -64,17 +62,17 @@ for details.
| `spring.test.constructor.autowire.mode`
| The default _test constructor autowire mode_ to use if `@TestConstructor` is not present
on a test class. See <<testing.adoc#integration-testing-annotations-testconstructor,
Changing the default test constructor autowire mode>>.
on a test class. See xref:testing/annotations/integration-junit-jupiter.adoc#integration-testing-annotations-testconstructor[Changing the default test constructor autowire mode]
.
| `spring.test.context.cache.maxSize`
| The maximum size of the context cache in the _Spring TestContext Framework_. See
<1> References the `publicMethod` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-pointcuts-combining[Combining Pointcut Expressions].
<1> References the `publicMethod` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-pointcuts-combining[Combining Pointcut Expressions].
======
[[aop-ataspectj-advice-params-generics]]
=== Advice Parameters and Generics
Spring AOP can handle generics used in class declarations and method parameters. Suppose
value = "com.xyz.Pointcuts.publicMethod() && target(bean) && @annotation(auditable)", // <1>
argNames = "bean,auditable") // <2>
public void audit(Object bean, Auditable auditable) {
AuditCode code = auditable.value();
// ... use code and bean
}
----
<1> References the `publicMethod` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-pointcuts-combining[Combining Pointcut Expressions].
<2> Declares `bean` and `auditable` as the argument names.
value = "com.xyz.Pointcuts.publicMethod() && target(bean) && @annotation(auditable)", // <1>
argNames = "bean,auditable") // <2>
fun audit(bean: Any, auditable: Auditable) {
val code = auditable.value()
// ... use code and bean
}
----
<1> References the `publicMethod` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-pointcuts-combining[Combining Pointcut Expressions].
<2> Declares `bean` and `auditable` as the argument names.
======
If the first parameter is of type `JoinPoint`, `ProceedingJoinPoint`, or
`JoinPoint.StaticPart`, you can omit the name of the parameter from the value of the
`argNames` attribute. For example, if you modify the preceding advice to receive the join
point object, the `argNames` attribute does not need to include it:
value = "com.xyz.Pointcuts.publicMethod() && target(bean) && @annotation(auditable)", // <1>
argNames = "bean,auditable") // <2>
public void audit(JoinPoint jp, Object bean, Auditable auditable) {
AuditCode code = auditable.value();
// ... use code, bean, and jp
}
----
<1> References the `publicMethod` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-pointcuts-combining[Combining Pointcut Expressions].
<2> Declares `bean` and `auditable` as the argument names.
value = "com.xyz.Pointcuts.publicMethod() && target(bean) && @annotation(auditable)", // <1>
argNames = "bean,auditable") // <2>
fun audit(jp: JoinPoint, bean: Any, auditable: Auditable) {
val code = auditable.value()
// ... use code, bean, and jp
}
----
<1> References the `publicMethod` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-pointcuts-combining[Combining Pointcut Expressions].
<2> Declares `bean` and `auditable` as the argument names.
======
The special treatment given to the first parameter of type `JoinPoint`,
`ProceedingJoinPoint`, or `JoinPoint.StaticPart` is particularly convenient for advice
methods that do not collect any other join point context. In such situations, you may
omit the `argNames` attribute. For example, the following advice does not need to declare
<1> References the `publicMethod` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-pointcuts-combining[Combining Pointcut Expressions].
<1> References the `publicMethod` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-pointcuts-combining[Combining Pointcut Expressions].
======
[[aop-ataspectj-advice-proceeding-with-the-call]]
=== Proceeding with Arguments
We remarked earlier that we would describe how to write a `proceed` call with
arguments that works consistently across Spring AOP and AspectJ. The solution is
to ensure that the advice signature binds each of the method parameters in order.
<1> References the `inDataAccessLayer` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-common-pointcuts[Sharing Named Pointcut Definitions].
fun preProcessQueryPattern(pjp: ProceedingJoinPoint,
accountHolderNamePattern: String): Any? {
val newPattern = preProcess(accountHolderNamePattern)
return pjp.proceed(arrayOf<Any>(newPattern))
}
----
<1> References the `inDataAccessLayer` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-common-pointcuts[Sharing Named Pointcut Definitions].
======
In many cases, you do this binding anyway (as in the preceding example).
[[aop-ataspectj-advice-ordering]]
== Advice Ordering
What happens when multiple pieces of advice all want to run at the same join point?
Spring AOP follows the same precedence rules as AspectJ to determine the order of advice
execution. The highest precedence advice runs first "on the way in" (so, given two pieces
of before advice, the one with highest precedence runs first). "On the way out" from a
join point, the highest precedence advice runs last (so, given two pieces of after
advice, the one with the highest precedence will run second).
When two pieces of advice defined in different aspects both need to run at the same
join point, unless you specify otherwise, the order of execution is undefined. You can
control the order of execution by specifying precedence. This is done in the normal
Spring way by either implementing the `org.springframework.core.Ordered` interface in
the aspect class or annotating it with the `@Order` annotation. Given two aspects, the
aspect returning the lower value from `Ordered.getOrder()` (or the annotation value) has
the higher precedence.
[NOTE]
====
Each of the distinct advice types of a particular aspect is conceptually meant to apply
to the join point directly. As a consequence, an `@AfterThrowing` advice method is not
supposed to receive an exception from an accompanying `@After`/`@AfterReturning` method.
As of Spring Framework 5.2.7, advice methods defined in the same `@Aspect` class that
need to run at the same join point are assigned precedence based on their advice type in
the following order, from highest to lowest precedence: `@Around`, `@Before`, `@After`,
`@AfterReturning`, `@AfterThrowing`. Note, however, that an `@After` advice method will
effectively be invoked after any `@AfterReturning` or `@AfterThrowing` advice methods
in the same aspect, following AspectJ's "after finally advice" semantics for `@After`.
When two pieces of the same type of advice (for example, two `@After` advice methods)
defined in the same `@Aspect` class both need to run at the same join point, the ordering
is undefined (since there is no way to retrieve the source code declaration order through
reflection for javac-compiled classes). Consider collapsing such advice methods into one
advice method per join point in each `@Aspect` class or refactor the pieces of advice into
separate `@Aspect` classes that you can order at the aspect level via `Ordered` or `@Order`.
<1> References the `businessService` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-common-pointcuts[Sharing Named Pointcut Definitions].
<1> References the `businessService` named pointcut defined in xref:core/aop/ataspectj/pointcuts.adoc#aop-common-pointcuts[Sharing Named Pointcut Definitions].
======
Note that the aspect implements the `Ordered` interface so that we can set the precedence of
the aspect higher than the transaction advice (we want a fresh transaction each time we
retry). The `maxRetries` and `order` properties are both configured by Spring. The
main action happens in the `doConcurrentOperation` around advice. Notice that, for the
moment, we apply the retry logic to each `businessService`. We try to proceed,
and if we fail with a `PessimisticLockingFailureException`, we try again, unless
More information on AspectJ can be found on the https://www.eclipse.org/aspectj[AspectJ website].
_Eclipse AspectJ_ by Adrian Colyer et. al. (Addison-Wesley, 2005) provides a
comprehensive introduction and reference for the AspectJ language.
_AspectJ in Action_, Second Edition by Ramnivas Laddad (Manning, 2009) comes highly
recommended. The focus of the book is on AspectJ, but a lot of general AOP themes are
explored (in some depth).
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Stéphane Nicoll
