Use the checkpoint operator at various places in WebFlux to insert
information that Reactor then uses to enrich exceptions, via suppressed
exceptions, when error signals flow through the operator.
Closes gh-22105
In line with the general trend toward favoring core JDK APIs for common
tasks in Spring Framework 5.2, this commit replaces handcrafted
statements with Math.min() and Math.max() were applicable.
Prior to this commit, calls to `MimeType` and `MediaType` would create a
significant amount of garbage:
* during startup time, in the static sections of `MimeType` and
`MediaType` when creating well-known types
* at runtime, when parsing media types for content negotiation or
writing known media types as strings in HTTP response headers
This commit does the following:
* Avoid parsing the well-known types and use regular constructors
instead
* Cache types in a simple LRU cache once they've been parsed, since an
application is likely to deal with a limited set of types
* Avoid using `java.util.stream.Stream` in hot code paths
Benchmarks show that a complete revision of the `MimeTypeUtils` parser
is not required, since the LRU cache is enough there.
Closes gh-22340
This commit improves the message for HttpStatusCodeException so that it
defaults to the HttpStatus reason phrase if a status text is not
provided.
This commit also fixes SimpleClientHttpResponse so that it does not
return null for getStatusText().
Fixed#22162
Commit #93b7a4 added support for pre-configuring URI variables at the
UriComponentsBuilder level, and also changed toUriString to encode
template and URI variables separately. However this went a bit too far
causing side effects for URLs with curly braces that don't represent
URI variables.
This commit restores the original toUriString behavior which is to
encode template and URI variables sepraately only if URI variables have
been pre-configured.
Issue: SPR-17630
Aalto's InputFactoryImpl already disables loading of external entities
by default (property "javax.xml.stream.isSupportingExternalEntities").
This commit goes further by applying the same defensive measures as we
do elsewhere for XMLInputFactory, which disables DTD completely.
Arguably there is no good reason to enable that by default in WebFlux.
Typically a straight up equals as well as Collections#contains
checks for MediaType.ALL is susceptible to the presence of
media type parameters.
This commits adds equalsTypeAndSubtype as well as an
isPresentIn(Collection<MimeType>) methods to MimeType to faciliate
with checks for MediaType.ALL.
Issue: SPR-17550
Prior to this commit, one could write a `CharSequence` to an existing
`DataBuffer` instance by turning it into a byte array or `ByteBuffer`
first. This had the following disadvantages:
1. Memory allocation was not efficient (not leveraging pooled memory
when available)
2. Dealing with `CharsetEncoder` is not always easy
3. `DataBuffer` implementations, like `NettyDataBuffer` can use
optimized implementations in some cases
This commit adds a new `DataBuffer#write(CharSequence, Charset)` method
for those cases and also an `ensureCapacity` method useful for checking
that the current buffer has enough capacity to write to it..
Issue: SPR-17558
This commit makes TomcatServerHttpRequest aware of
HttpServletRequestWrappers, and TomcatServerHttpResponse aware of
HttpServletResponseWrappers.
Issue: SPR-17611
Commit #c187cb2 introduced proactive rejection of multiple subscribers
in ReactorClientHttpResponse, instead of hanging indefinitely as per
https://github.com/reactor/reactor-netty/issues/503.
However FluxReceive also rejects subsequent subscribers if the response
is consumed fully, as opposed to being canceled, e.g. as with
bodyToMono(Void.class). In that case, a subsequent subscriber causes
two competing error signals to be sent, and one gets dropped and
logged by reactor-core.
This fix ensures that a rejection is raised in
ReactorClientHttpResponse only after a cancel() was detected.
Issue: SPR-17564
Response status 415 (unsupported media type) reported as of 416 (which is Range Not Satisfiable), mismatching with superclass constructor parameter HttpStatus.UNSUPPORTED_MEDIA_TYPE
This commit introduces a new readMessageSize(DataBuffer input) private
method, inspired from CodedInputStream#readRawVarint32(int, InputStream)
and adapted for DataBuffer using MessageDecoderFunction fields in
order to support use cases where the message size is split between
distinct chunks.
It also fixes handling of end of streams by using
DataBuffer#readableByteCount instead of -1 which is only relevant with
InputStream.
Issue: SPR-17429
Prior to this commit, when errors happened before the response was
committed, the `Content-Length` response header would be left as is.
This can be problematic since the error can be handled later in the
chain and the response body changed accordingly. For example, Spring
Boot renders error pages in those cases. If the `Content-Length` is set,
HTTP clients can get confused and only consider part of the error
response body.
This commit ensures that any `Content-Length` response header is removed
in case of errors, if the response is not already committed.
This is done at the `AbstractServerHttpResponse` level, since errors can
be handled in multiple places and the response itself is the safest
place to handle this case.
As a consequence, this commit also removes `Content-Length` checks in
`EncoderHttpMessageWriter` since we now consider that we should rely on
the response body we're about to write rather than any previously set
value.
Issue: SPR-17502
Update the ServerHttpRespnose contract to indicate that server specific
sub-classes should fall back on the default status, if a status code
has not been set explicitly.
Issue: SPR-17368
This commit makes the 3 existing InvocableHandlerMethod types more
consistent and comparable with each other.
1. Use of consistent method names and method order.
2. Consistent error formatting.
3. Explicit for loops for resolving argument values in webflux variant
because that makes it more readable, creates less garabage, and it's
the only way to bring consistency since the other two variants cannot
throw exceptions inside Optional lambdas (vs webflux variant which can
wrap it in a Mono).
4. Use package private HandlerMethodArgumentComposite in webflux
variant in order to pick up the resolver argument caching that the
other two variants have.
5. Polish tests.
6. Add missing tests for messaging variant.
This commit fixes a memory leak in ServerSentEventHttpMessageWriter
that occurs when the input stream contains an error. Test added as well.
Issue: SPR-17419
The fix for SPR-17178 switched from debug to warn level warning for
all sub-classes of AbstractHandlerExceptionResolver where the request
concerned the DefaultHandlerExceptionResolver only.
This commit restores the original DEBUG level logging that was in
AbstractHandlerExceptionResolver from before SPR-17178. In addition
DefaultHandlerExceptionResolver registers a warnLogCategory by default
which enables warn logging and hence fulfilling the original goal
for SPR-17178.
Issue: SPR-17383
Review and update Servlet and Undertow adapters to release any data
buffers they be holding on to at the time of error or cancellation.
Also remove onDiscard hooks from Reactor and Undertow request body.
For Reactor we expect it to be handled. For Undertow there isn't
any Reactor Core upstream for the callback to be useful.
Issue: SPR-17410
This commit harmonizes the `HeadersAdapter` implementations across all
supported servers with regards to the `get(Object key)` contract; some
server implementations are not sticking to a `Map`-like contract and
return empty `List` instead of `null` when a header is not present.
This also fixes the `size()` implementations to reflect the number of
header keys, as some implementations consider multiple values for the
same header as different entries.
Issue: SPR-17396
This commit adds special processing of some HTTP response headers in
Jetty and Tomcat; they both consider some headers like "Content-Length"
as specific and require explicit calls on the `HttpServletResponse`
itself on top of setting the HTTP response header.
Issue: SPR-17250
This commit avoids copying HTTP headers when mutating an HTTP request.
Instead, we're now unwrapping the `ReadOnlyHttpHeaders` (which is most
likely backed by the native request headers).
Issue: SPR-17250
Several benchmarks underlined a few hotspots for CPU and GC pressure in
the Spring Framework codebase:
1. `org.springframework.util.MimeType.<init>(String, String, Map)`
2. `org.springframework.util.LinkedCaseInsensitiveMap.convertKey(String)`
Both are linked with HTTP request headers parsing and response headers
writin during the exchange processing phase.
1) is linked to repeated calls to `HttpHeaders.getContentType`
within a single request handling. The media type parsing operation
is expensive and the result doesn't change between calls, since
the request headers are immutable at that point.
This commit improves this by caching the parsed `MediaType` for the
`"Content-Type"` request header in the `ReadOnlyHttpHeaders` class.
This change is available for both Spring MVC and Spring WebFlux.
2) is linked to insertions/lookups in the `LinkedCaseInsensitiveMap`,
which is the data structure behind `HttpHeaders`.
Those operations are creating a lot of garbage (including a lot of
`String` created by `toLowerCase`). We could choose a more efficient
data structure for storing HTTP headers data.
As a first step, this commit is focusing on Spring WebFlux and
introduces `MultiValueMap` implementations mapped by native HTTP headers
for the following servers: Tomcat, Jetty, Netty and Undertow.
Such implementations avoid unnecessary copying of the headers
and leverages as much as possible optimized operations provided by the
native implementations.
This change has a few consequences:
* `HttpHeaders` can now wrap a `MultiValueMap` directly
* The default constructor of `HttpHeaders` is still backed by a
`LinkedCaseInsensitiveMap`
* The HTTP request headers for the websocket HTTP handshake now need to
be cloned, because native headers are likely to be pooled/recycled by
the server implementation, hence gone when the initial HTTP exchange is
done
Issue: SPR-17250
In order to be able to leverage WebFlux configuration in a functional
way, WebHttpHandlerBuilder and RouterFunctionMapping should leverage
new ObjectProvider capabilities to get a sorted list of beans by type
instead of using autowired containers.
Issue: SPR-17327
1. Helper method to eliminate duplication in formatting (de-)serialized
values for logging introduced with prior commit #e62298.
2. Helper method for TRACE vs DEBUG logging with different details.
Issue: SPR-17254
At DEBUG show up to 100 chars, at TRACE show full formatted value.
Note that the formatValue helper method is duplicated a number of times
in this commit. A utility method will likely be added in spring-core
through an extra commit.
Issue: SPR-17254
1. Rename globalResources to useGlobalResources.
2. Use of global resources is mutually exlusive with explicit config.
3. Allow Consumer<HttpResources> to configure global resources.
4. Allow ConnectionProvider + LoopResources Supplier to customize
creation and initialization.
5. Do not manage externally provided ConnectionProvider + LoopResources
instances.
Issue: SPR-17243
This commit adds decoder/message-reader tests for errors in
the source data buffer publisher. Because the tests extend
AbstractDataBufferAllocatingTestCase, they also check whether
the buffers that precede the error in the stream are properly
released.
Issue: SPR-17025
This commit optimizes Flux <-> Mono conversions in our codebase by
avoiding to hide that conversion from Reactor.
This tries to keep conversions sequentially so that they can be detected
by Reactor and optimized. In Spring WebFlux, this means keeping the
conversions at the edges of a method implementation (right when getting
an input parameter, and before returning it as a result). If those
conversions are made between other operators, Reactor might not be able
to detect those conversions and optimize them.
Issue: SPR-17203
Prior to this commit, an bug introduced in SPR-16949 prevented
`Mono.empty` bodies from being written to the response.
This commit ensures that empty bodies still trigger the writing to the
response and does not hang the processing of the exchange.
Issue: SPR-17220
This commit represents a best effort attempt at fixing remaining
"a" vs. "an" grammatical errors related links specified via a fully
qualified class name.
Issue: SPR-17208
When used as global Netty resources, ReactorResourceFactory creates and
sets those resources on Reactor's HttpResources directly.
When that ReactorResourceFactory bean is destroyed, those resources are
disposed but HttpResources still holds a reference to those and may try
to use them again.
This commit uses HttpResources to clear those resources and its
references to it, when the ReactorResourceFactory is treating those as
global.
Issue: SPR-17199
JettyResourceFactory, similar to ReactorResourceFactory, allows
to share resources (Executor, ByteBufferPool, Scheduler) between
Jetty clients and servers.
Issue: SPR-17179
Prior to this commit, when using the `SimpleClientHttpRequestFactory`
as a driver for `RestTemplate`, the HTTP response body would only be
drained if there was an attempt to read it in the first place.
This commit ensures that, even if there's no attempt at reading the
response body, it is properly drained when the response is closed to
make sure that the connection is released in a proper state and can be
put back in the connection pool for reuse.
Issue: SPR-17181
When dealing with `Optional` values in a Controller handler (for
example, values coming from a Spring Data repository), developers might
reuse this code snippet quite often:
```
@GetMapping("/user")
public ResponseEntity<Optional<User>> fetchUser() {
Optional<User> user = //...
return user.map(ResponseEntity::ok).orElse(notFound().build());
}
```
This commit adds a new static method on `ResponseEntity` for that,
simplifying the previous snippet with `return ResponseEntity.of(user);`
Note that in case more specific HTTP response headers are required by
the application, developers should use other static methods to
explicitly tell which headers should be used in each case.
Issue: SPR-17187
Rename "Builder" sub-section to "Configuration" and move it in the
beginning before all others since it explains how to create a client
in the first place.
Update content on Reactor Netty connector based on the API in 0.8 and
specifically address Reactor Netty resources and lifecycle.
Issue: SPR-16963
Arjen Poutsma
Rossen Stoyanchev
Sam Brannen
Juergen Hoeller
Lars Grefer
stsypanov
www
Johnny Lim
Sebastien Deleuze
Brian Clozel
Marten Deinum
王诗峣
Violeta Georgieva
igor-suhorukov
wonwoo
Francesco Komauli
Stephane Nicoll
Hanope
Dimitrios Liapis
陈灵敏
Sam Brannen