Previously, KotlinDelegate.getReturnType() and getGenericReturnType()
resolved the KFunction of every Kotlin method, which materializes the
members of the declaring KClass and is costly at startup. Kotlin
reflection is only needed to resolve the return type of suspending
functions, so both methods now only use it when
KotlinDetector.isSuspendingFunction() returns true.
Closes gh-37300
Signed-off-by: Sébastien Deleuze <sdeleuze@users.noreply.github.com>
- Remove redundant super() calls from constructors
- Add missing @Override annotations
- Use switch rules in JdkClientHttpRequest and RfcUriParser
- Use instanceof pattern matching
- Use method references instead of trivial lambda expressions
- Use lambda expressions instead of anonymous inner classes
- Remove unused code and redundant semicolons
- Use braces with if-blocks
This commit adds a test to MergedAnnotationsTests that verifies a
single primitive attribute in a composed annotation can be aliased via
@AliasFor to a primitive array attribute in a meta-annotation.
See gh-37349
Prior to this commit, `adaptForAttribute(Method, Object)` created
the wrapping array from `value.getClass()` when a single non-array
value was provided for an array attribute. This worked for object
array types but failed for primitive array types: wrapping a boxed
value produced a boxed array, which then failed the compatibility
check and threw an `IllegalStateException`.
This commit derives the component type from the declared attribute
type when it is assignable from the value type, falling back to
`value.getClass()` otherwise. The existing adaptation path for
object array types is therefore preserved, and all primitive array
types now accept a single value.
The accompanying test covers single-value wrapping for every array
type declared by ArrayTypes.
Closes gh-37349
Signed-off-by: Chengang Guan <guanchengang@qq.com>
computeIfAbsent registered the case-insensitive key before invoking the
mapping function. If the function returned null or threw an exception,
no mapping was recorded in the target map, but the key registration was
left behind. The map then reported containsKey(key) as true while size()
was 0, keySet() was empty, and get(key) returned null. A later insertion
with a different casing also reused the stale casing of the failed call,
since the existing-key branch resolved to the registered key.
The case-insensitive key is now only looked up up front. For a new key,
it is registered from within the mapping function once a non-null value
has been computed, which is still before the entry is inserted. A null
result or an exception therefore leaves both maps untouched, while a
removeEldestEntry override that evicts the new entry right away still
removes the registration, as it does for put. The existing-key path
keeps computing under the stored casing.
Closes gh-37351
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
`RetryTemplateTests` can, under load, break the build because of a flaky
test: "retryableWithTimeoutExceededAfterSecondRetry".
This usually happens when many cores are busy and the timeout check
runs before each retry attempt.
This commit extends the timeout to avoid such cases and failures.
Prior to this commit, clear() polled the eviction queue to remove
entries and only afterward drained the pending write operations queue.
Consequently, a put() whose AddTask had not yet been linked into the
eviction queue -- for example, because it lost the race to self-drain
while clear() held the eviction lock -- would only be applied by that
trailing drain, linking the entry into the eviction queue right after
clear() had already finished removing everything it could see.
The practical effect was that an entry already fully added to the cache
could still be present immediately after clear() returned, with no
further concurrent activity required at that point.
To address that, this commit revises clear() so that it drains the
pending write operations queue before polling the eviction queue, so
any write that was already queued gets cleaned up along with everything
else. However, a put() that is genuinely concurrent with an in-progress
clear() call can still survive, which is consistent with the cache's
weak-consistency design.
Thanks to @guanchengang for raising gh-37286, which prompted this fix.
Closes gh-37287
markAsRemoved() transitions a node to the removed state and decrements
the current size, but it did not check whether the node had already
been removed. The eviction path and an explicit removal can process
the same node in sequence: when a write drain runs a queued AddTask
whose eviction polls a node that a concurrent remove(K) has already
taken out of the cache, the eviction decrements the size, and the
queued RemovalTask for the same node decrements it again. The sibling
transition markForRemoval() guards against invalid transitions; this one
did not.
Each extra decrement makes currentSize permanently smaller than the
number of cached entries, so eviction stops triggering and the cache
exceeds its capacity for good, silently. A bounded two-thread stress
run accumulates the drift reliably: before the change the cache
stabilized far above its capacity in 20 out of 20 runs.
markAsRemoved() now returns without decrementing when the entry is
already in the removed state, mirroring the guard in markForRemoval().
The removed state is terminal, so each node is counted down exactly
once. The new test races explicit removals against eviction and then
verifies that the cache converges back to its capacity; it also
asserts that the racing thread ran and terminated cleanly.
Closes gh-37268
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
Prior to this commit, LogAccessor's CharSequence-based logging methods
delegated directly to the corresponding method on the underlying
commons-logging Log instance without first checking whether the target
level was enabled. This differed from the Supplier-based overloads,
which have checked isXxxEnabled() before delegating since Spring
Framework 5.2.9.
That asymmetry was harmless as long as spring-jcl supplied the
underlying Log implementation, since its SLF4J adapter itself checked
the level before rendering the message. However, since Spring Framework
7 replaced spring-jcl with Apache commons-logging, whose SLF4J adapters
call String.valueOf(message) unconditionally, any CharSequence argument
-- most notably a LogMessage supplied via LogMessage.format(...) or
LogMessage.of(...) -- is now rendered eagerly, even when the
corresponding level is disabled. Since LogMessage exists specifically
to defer that work, and the idiom is used extensively throughout the
framework and its portfolio projects, this leads to unnecessary
computation and allocation whenever logging is disabled.
To address this, this commit adds the same isXxxEnabled() guard to all
twelve CharSequence-based methods in LogAccessor, matching the
existing Supplier-based overloads and making LogAccessor's laziness
guarantee independent of the underlying Log implementation.
This commit also introduces LogAccessorTests, which verifies that a
lazily rendering LogMessage passed to one of the CharSequence-based
methods is only rendered when the corresponding level is enabled.
See gh-25741
Closes gh-37266
The Mutiny Uni adapter registers its empty-value supplier as
Uni.createFrom().nothing(), which returns a Uni that never signals an
item, a failure, or completion. Every sibling registration supplies an
empty value that completes immediately: Mono.empty(), Maybe.empty(),
Completable.complete(), and CompletableDeferred(null); the Multi
registration uses Multi.createFrom().empty() as well.
ReactiveAdapter.toPublisher(null) substitutes that empty value whenever
a null source needs to be adapted, for example when a WebFlux handler
method with a Uni return type returns null. With a never-completing
empty value the resulting Publisher emits no signal at all, so the
response is never written and the request hangs until a timeout,
whereas the same handler declared with Mono completes empty. The
adapter also becomes asymmetric with its own fromPublisher function,
which adapts an empty Publisher to a Uni that completes with a null
item.
The supplier now uses Uni.createFrom().nullItem(), whose conversion to
a Publisher completes without emitting an item, matching the sibling
adapters and the round-trip through fromPublisher. The descriptor is
shared by the Mutiny 1 and Mutiny 2 registrations, so both paths are
covered.
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
Prior to this commit, ConversionService#canConvert(Class, Class)
threw an IllegalArgumentException when invoked with Enum.class as the
target type (i.e., `canConvert(String.class, Enum.class)`), because
ConverterFactory#getConverter() in StringToEnumConverterFactory and
IntegerToEnumConverterFactory eagerly resolved the concrete enum type.
To address that, StringToEnumConverterFactory and
IntegerToEnumConverterFactory now implement ConditionalConverter so
that matches() can reject non-concrete-enum targets before
getConverter() is ever invoked.
Closes gh-34532
Signed-off-by: MoonFruit <dkmoonfruit@gmail.com>
Co-authored-by: Sam Brannen <104798+sbrannen@users.noreply.github.com>
PrimitiveDelegate generated code via "$LF" for Float and "(double) $L"
for Double, which emit the value's toString() verbatim. For NaN and
infinities this produced non-compilable source such as "NaNF" or
"(double) Infinity", causing the generated AOT sources to fail to
compile.
Detect NaN (via isNaN, since NaN is never equal to itself) and the
positive/negative infinities, emitting the corresponding constant
field references (Float.NaN, Double.POSITIVE_INFINITY, etc.) through
the "$T" placeholder. Finite values keep their existing handling.
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
GeneratedClass.reserveMethodNames(String...) passed the entire varargs
array to MethodName.of() inside the per-name loop instead of the current
element. Since MethodName.of(String...) joins all parts into a single
camel-case name, reserving two or more names (for example "apply" and
"test") produced "applyTest", and the per-element check
Assert.state(generatedName.equals(reservedMethodName)) failed with an
IllegalStateException. Single-name calls worked only by accident.
Reserve each supplied name individually by passing the loop variable.
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
Prior to this commit, AbstractXMLStreamReader.getTextCharacters(int
sourceStart, char[], int, int) capped the copy length with
Math.min(length, source.length), ignoring sourceStart. When sourceStart
> 0 and sourceStart + length exceeds the text length, System.arraycopy
read past the end of the source array and threw
ArrayIndexOutOfBoundsException, contrary to the
XMLStreamReader#getTextCharacters contract (copy up to length
characters starting at sourceStart and return the number copied).
To address that, this commit caps the length by the number of
characters remaining from sourceStart.
Closes gh-36914
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
ClassFileMethodMetadata's toString() formatted method parameter types
as packageName() + "." + displayName(). Since ClassDesc.packageName()
is empty for primitive, array and default-package types, these rendered
with a leading dot (for example ".int" and ".String[]") and reference
arrays lost their package. The return type already uses
ClassFileAnnotationMetadata.resolveTypeName(); apply it to the
parameters as well.
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
Prior to this commit, only the very first AsynchronousFileChannel#write
call in DataBufferUtils$WriteCompletionHandler#hookOnNext(DataBuffer)
was guarded against exceptions escaping synchronously, and even then
only via `catch (RuntimeException ex)`, per the original fix for
gh-36184.
While widening that guard to match the read side's `catch (Throwable
ex)` combined with `Exceptions.throwIfFatal(ex)` (see gh-37143), we
discovered that completed(Integer, Attachment) contains two more direct
`this.channel.write(...)` calls -- for continuing a partial write and
for advancing to the next ByteBuffer within the same DataBuffer's
iterator -- neither of which was guarded at all. Since completed() is
invoked by the channel's own completion callback, typically on a
different thread than the one that issued the original write, a
synchronous exception escaping either of those calls has no path back
to the FluxSink, and the resulting Flux hangs indefinitely, exactly as
described in gh-37143, for any write that receives a partial OS write
or spans multiple ByteBuffers.
To address that, this commit extracts a private write(ByteBuffer, long,
Attachment) helper that wraps the channel.write(...) call with a
try/catch block, routing any non-fatal Throwable -- via
Exceptions.throwIfFatal() -- to the existing failed(Throwable,
Attachment) handler. All three call sites (hookOnNext() and both
branches in completed()) now go through this helper, ensuring the
Flux always terminates with a proper error signal instead of hanging
silently, regardless of which write attempt fails or which thread it
fails on.
See gh-36184
See gh-37143
Closes gh-37145
Prior to this commit, DataBufferUtils$ReadCompletionHandler#read()
invoked AsynchronousFileChannel#read(ByteBuffer, long, Attachment,
CompletionHandler) without guarding against exceptions thrown directly
by that call. Although that method is documented to report failures
asynchronously via the supplied CompletionHandler, some platform-
specific implementations can instead throw synchronously – for
example, on Windows with JDK 25, when the JDK rejects a ByteBuffer
backed by a closeable shared memory Arena, as produced by Netty 4.2's
off-heap buffer allocation.
When such an exception is thrown from a recursive read() invocation
triggered from completed() – which happens once a resource requires
more than a single chunk – the exception has no path back to the
FluxSink: it escapes on whatever thread invoked the CompletionHandler,
and the resulting Flux never signals onError or onComplete. In
practice, this surfaced as an indefinite hang when serving a Resource
whose HTTP response is not a ZeroCopyHttpOutputMessage, since
ResourceHttpMessageWriter falls back to ResourceEncoder, which reads
the resource via DataBufferUtils.
To address that, this commit wraps the channel.read(...) call in a
try/catch block and routes any non-fatal Throwable to the existing
failed(Throwable, Attachment) handler, via Exceptions.throwIfFatal(),
mirroring the equivalent fix already applied to the write side for
gh-36184. This ensures the allocated DataBuffer is released and the
Flux always terminates with a proper error signal instead of leaking a
buffer or hanging silently.
See gh-36184
Closes gh-37143
DataBufferUtils.TwoByteMatcher inherited AbstractNestedMatcher.match(byte)
without providing the mismatch fallback that its siblings implement
(KnuthMorrisPrattMatcher backtracks via its suffix-prefix table, and
SingleByteMatcher is stateless). As a result, once the first delimiter
byte had matched, the match counter stayed at 1 across any number of
intervening non-matching bytes, so a later occurrence of the second
delimiter byte falsely completed the match.
For a two-byte delimiter such as \r\n this made the matcher report a
match across non-contiguous bytes. CompositeMatcher prefers the longest
delimiter that matches at a position, so the false \r\n match was chosen
over a real single \n, causing StringDecoder to strip two bytes and drop
the character preceding a lone \n whenever a line contained a stray \r.
TwoByteMatcher now overrides match(byte) to reset the counter to 0 when
the incoming byte is not the expected next delimiter byte before
delegating to super.match(), mirroring KnuthMorrisPrattMatcher. A
genuine contiguous delimiter is unaffected.
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
Prior to this commit, `ClassFileAnnotationDelegate#parseArrayValue`
would only consider `int[]`, `double[]` and `long[]` array
annotation attributes; other primitive array types like `byte[]`
would use a generic path that would use boxed types.
This commit ensures that a comprehensive pass is made for all
primitive typed arrays. Because the `AnnotationValue` hierarchy
is sealed, we can now maje sure that the implementation is
exhaustive.
Closes gh-37083
The taskTerminationTimeoutWithImmediateCancel test submitted a task and
immediately closed the executor, then asserted that the future was
cancelled. The cancellation flag is set by close() on the calling thread,
while it is checked at the start of the task on a separate worker thread.
With no ordering guarantee between the two, a quickly scheduled worker
could pass the cancellation check before close() set the flag, complete
the trivial task normally, and leave the future uncancelled, making the
test fail intermittently under load.
Override doExecute to capture the task-tracking wrapper instead of
running it on a background thread, then run it on the test thread after
close() has set the cancellation flag. This exercises the same
cancellation path deterministically, with no reliance on thread
scheduling.
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
Prior to this commit, FileNativeConfigurationWriter did not write reachability-metadata.json
when a RuntimeHints instance contained only lambda hints, because
NativeConfigurationWriter.hasAnyHint() omitted
ReflectionHints.lambdaHints() from its checks. As a result, lambda
metadata emitted by RuntimeHintsWriter was silently dropped.
This commit addresses that by including lambda hints in hasAnyHint() so
the configuration file is written whenever lambda hints are present.
Closes gh-36989
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
This commit addresses warnings across the code base related to:
- internal and public deprecations in Spring Framework
- deprecated Locale constructors
- deprecated URL constructors
- deprecated Thread#getId method
Prior to this commit, FileNativeConfigurationWriter wrote native-image
configuration files using a plain FileWriter, which encodes with the
JVM platform default charset. On a non-UTF-8 platform (for example a
Windows JVM, where the default charset is not UTF-8 prior to JDK 18)
non-ASCII characters in resource patterns or bundle names were written
with the wrong encoding, while GraalVM expects the configuration files
to be UTF-8.
This commit specifies StandardCharsets.UTF_8 explicitly so the files
are always written as UTF-8, consistent with the UTF-8 usage already
present in the aot.generate package.
Closes gh-36972
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
Due to changes made in commit 41cd6879bd, MimeTypeUtils now raises a
StringIndexOutOfBoundsException instead of an InvalidMimeTypeException
when parsing certain invalid mime types -- for example, for a value
wrapped in double quotes which does not contain a ";" character.
To address that minor regression, this commit replaces
`mimeType.charAt(nextIndex - 1) != '\\'` with
`(nextIndex == 0 || mimeType.charAt(nextIndex - 1) != '\\')` to avoid
invoking `String#charAt` with a negative value.
See gh-36730
Closes gh-36971
XmlValidationModeDetector peeks at the start of an XML document to
choose between DTD- and XSD-based validation, skipping any DOCTYPE that
appears inside an XML comment.
Prior to this commit, consumeCommentTokens() short-circuited a line
with no start or end comment marker by returning it unchanged, even
while already inside a multi-line comment. Such a body line was then
treated as content, so a literal "DOCTYPE" word in the comment body
caused an XSD document to be misdetected as DTD-based.
This commit honors the "in comment" parse state in that early return so
a comment body line is treated as empty content, completing the fix for
gh-27915 which only covered comment markers on the same line.
Closes gh-36948
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
When an ExponentialBackOff is configured with an initialInterval of 0
and a positive jitter, the first nextBackOff() evaluated (jitter *
(interval / initialInterval)) performs integer division by zero
and throws an ArithmeticException.
Both initialInterval = 0 and jitter > 0 are individually accepted
configurations -- with jitter = 0, an initialInterval of 0 already
yields a delay of 0 -- so the combination should not throw.
This commit addresses that by guarding the division so that no jitter
scaling is applied when initialInterval is 0, leaving the behavior for
positive intervals unchanged.
Closes gh-36932
Signed-off-by: junhyeong9812 <pickjog@gmail.com>
This commit adds further fixes in the same area, since there were
similar bugs in the WriteCompletionHandler:
* databuffers were not always emitted when fully read in the onNext hook
* on completion, the iterator was closed too early, before it was fully
read
* on completion, writing the next bytebuffers from the iterator would
always reuse the first one and not update the attachment
Closes gh-36714
Prior to this commit, WritableByteChannelSubscriber.hookOnNext() called
iterator.next() exactly once. If a DataBuffer consisted of multiple NIO
ByteBuffers (e.g., NettyDataBuffer wrapping a CompositeByteBuf), only
the first buffer was written to the channel, and the remaining buffers
were silently ignored and lost.
This commit adds the missing while (iterator.hasNext()) outer loop to
ensure all fragmented buffers exposed by the iterator are completely
and safely written to the synchronous channel.
See gh-36714
Signed-off-by: KimDaehyeon <daehyeon3351@gmail.com>
Add a public accessor for the ClassLoader configured on a
DefaultDeserializer instance so that callers no longer need to read the
private field via reflection in order to forward it to a
ConfigurableObjectInputStream subclass.
See gh-36827
Closes gh-36833
Signed-off-by: seonwoo_jung <laborlawseon@kap.kr>
Prior to this commit, MIME types with parameter values that contain a
quoted pair would sometimes fail and parse an incomplete parameter
value.
This commit ensures that the quoted section of the parameter value is
correctly handled.
Fixes gh-36730
This commit fixes a regression introduced by gh-36449 for
nullable value class with an non-null value.
Closes gh-36665
Signed-off-by: Sigurd Gerke <sigurd.gerke@onedata.de>
Juergen Hoeller
Sam Brannen
Sébastien Deleuze
Chengang Guan
김준형
Brian Clozel
Hyunwoo Jung
junhyeong9812
MoonFruitandSam Brannen
Yanming Zhou
KimDaehyeon
seonwoo_jung
Dmitry Sulman
Sigurd Gerke