This commit fixes a regression introduced by gh-21139
via the usage of Kotlin reflection to invoke HTTP
handler methods. It ensures that kotlin.Unit is treated
as void by returning null.
It also polishes CoroutinesUtils to have a consistent
handling compared to the regular case, and adds related
tests to prevent future regressions.
Closes gh-31648
In order to prevent leaks of large amounts of non-heap
memory (and potential other efficiency and performance side
effects), this commit updates ResourceUtils#useCachesIfNecessary
to leave the caching flag to its JVM default value for instances
of JarURLConnection.
The previous behavior was originally introduced via gh-9316 and
gh-13755 to avoid I/O failure during webapp hot reloading in
Servlet containers. This is not a popular deployment mode anymore
and we have not been able to reproduce the original issue with
a Java 17 JVM and Tomcat 10.
Closes gh-30955
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-31628
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
Allows Kotlin caller to get a non-nullable property,
using a default value when the property is not set.
A method already exists on PropertyResolver which does this,
but takes a Class parameter.
The extension method can eliminate the parameter via reified type,
similar to the other extension methods which exist already.
Closes gh-31523
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
Sébastien Deleuze
“7fantasy7”
Stéphane Nicoll
Sam Brannen
Juergen Hoeller
PiotrFLEURY
rstoyanchev
Arjen Poutsma
Jason
lorenzsimon
wakingrufus
Johnny Lim
Brian Clozel