This commit counterbalances the retain in the MessagingRSocket
handleConnectionSetupPayload method with a conditional release on SETUP
frame type in handleNoMatch, preventing Netty buffer leaks.
Closes gh-32424
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-32669
Co-authored-by: Juergen Hoeller <juergen.hoeller@broadcom.com>
Previously, a UriComponents build based on a method would require the
given method to be the annotated one as method parameter resolution only
applied locally. This was a problem when a controller was specified on
a method whose mapping is defined in a parent.
This commit harmonies the lookup using AnnotatedMethod who provides
support for a synthesized method parameter that takes annotations from
parent parameter into account.
Closes gh-32553
Previously, a BeanDefinition that has an instance supplier would be
used irrespective of custom arguments being provided. This commit only
applies the instance supplier if no arguments are provided. With
custom arguments, the regular lookup takes place based on the
information provided in the bean factory.
Closes gh-32657
Previously, if a factory method is defined on a parent, the generated
code would blindly use the method's declaring class for both the target
of the generated code, and the signature of the method.
This commit improves the resolution by considering the factory metadata
in the BeanDefinition.
Closes gh-32609
Prior to this commit, `RestClientException` thrown by status handlers
would not be registered as observation errors. This commit ensures that
such exceptions are first caught, registered in the observation and
rethrown as expected.
Closes gh-32575
Prior to this commit, the `RestClient` observations would be stopped as
soon as the exchange function was called. This means that all errors
related to response decoding or mapping would not be recorded by the
obsevations.
This commit extends the observation recording to the `ResponseSpec` DSL
calls as well as custom exchange functions.
Fixes gh-32575
Prior to this commit, the `BufferingClientHttpRequestFactory`, through
the `AbstractBufferingClientHttpRequest`, would set a "Content-Length"
header value, even if the buffered body was empty.
This behavior is invalid since no request body would be set by the
client code in the first place.
This commit ensures that this header is only set if a request body has
been buffered and is about to be written to the request.
Fixes gh-32650
This commit rephrases the `DefaultTransactionDefinition#setTimeout`
exception message to better reflect the fact that 0 is a valid input
value.
Even though this often leads to a transaction immediately timing out
after opening, there is one notable case where this has another
meaningful effect: in Jakarta transactions (`UserTransaction`), when
passing 0 "the transaction service restores the default value".
Closes gh-32635
This change improves the message of several parsing-related exceptions
that would previously entirely swallow the original exception's message
and sometimes have a slightly misleading message as a result.
This is done by appending the cause's `toString` representation to the
IllegalArgumentException messages instead of an hardcoded "cause".
Closes gh-32636
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.
Closes gh-31217
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-32616
Prior to this commit, `BufferingClientHttpRequestWrapper` would always
write to the actual client request body, even if the buffered content
was empty (empty byte array).
This would cause issues with specific client request factories,
especially the OkHttp variant, that would consider empty byte arrays as
non-empty body and would reject such cases for GET requests with an
"IllegalArgumentException: method GET must not have a request body".
This commit only writes to the request if the buffered content is not
empty.
Fixes gh-32612
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
Prior to this commit, the `JmsTemplate#sendAndReceive` method would not
instrument the JMS session. This means that no metric would be recorded
when sending the message and no trace would be propagated downstream.
This commit ensures that the JMS session is instrumented in this case as
well. Note, the reception of the response message does not create a
`"jms.message.process"` observation as the session is only receiving the
message, no listener has been configured on the message consumer.
Fixes gh-32606
Simon Baslé
Sam Brannen
Juergen Hoeller
Stéphane Nicoll
Johnny Lim
Brian Clozel
rstoyanchev
yhao3
xumengqi
Arjen Poutsma
Spring Builds
Sébastien Deleuze