This commit removes the dependency that the spring-web module has on
Jakarta Mail.
In FormHttpMessageConverter, a dependency on
jakarta.mail.internet.MimeUtility was replaced by existing encoding
logic in ContentDisposition.
In StandardMultipartHttpServletRequest, a dependency on the same
MimeUtility was replaced by new quoted-printable decoding logic in
ContentDisposition.
Closes gh-28392
This commit introduces the PartEvent API. PartEvents are either
- FormPartEvents, representing a form field, or
- FilePartEvents, representing a file upload.
The PartEventHttpMessageReader is a HttpMessageReader that splits
multipart data into a stream of PartEvents. Form fields generate one
FormPartEvent; file uploads produce at least one FilePartEvent. The last
element that makes up a particular part will have isLast set to true.
The PartEventHttpMessageWriter is a HttpMessageWriter that writes a
Publisher<PartEvent> to a outgoing HTTP message. This writer is
particularly useful for relaying a multipart request on the server.
Closes gh-28006
This commit makes sure that TomcatServerHttpRequest::readFromInputStream
follows the same contract as the method it overrides, and returns
AbstractListenerReadPublisher.EMPTY_BUFFER when 0 bytes are read.
See gh-28241
This commit fixes a regression that added the Accept-Ranges header on
both client and server. Accept-Ranges is response header, so we now make
sure it only appears on the server side.
See gh-28291
This commit introduces explicit HEAD support in Spring
MVC's ResourceHttpRequestHandler and WebFlux's ResourceWebHandler,
adding just headers but no body.
Closes gh-28291
This commit ensures that any resources created for multipart handling,
obtained via ServerWebExchange.getMultipartData(), are automatically
deleted after handling the completing the response.
Resource for parts obtained via BodyExtractors::toMultipartData and
BodyExtractors::toParts are not cleaned automatically, and
should be cleaned via Part::delete.
Closes gh-27633
If we read 0 bytes, e.g. chunked encoding markup read but not the
actual data within it, don't stop reading since the server may or
may not consider it necessary to call onDataAvailable again.
Instead, we keep on reading, and although isReady likely returns
false on the next iteration, it eliminates ambiguity and ensures
the server will call onDataAvailable when more data arrives.
Closes gh-28241
This commit contains changes made because of the introduction of
HttpStatusCode. In general, methods that used to return a HttpStatus
now return HttpStatusCode instead, and methods that returned raw status
codes are now deprecated.
See gh-28214
This commit introduces HttpStatusCode, an interface implemented by
HttpStatus. Instances of HttpStatusCode are obtained via static
valueOf(int) factory method, returning a HttpStatus enum entry if
available, and a default implementation otherwise.
The underlying reason behind this change is HTTP status codes are not
enumerable, but instead range from 100-999.
Closes gh-28214
With a Java 8 baseline in place for quite some time now, it no longer
makes sense to refer to features such as annotations as "Java 5
annotations".
This commit also removes old `Tiger*Tests` classes, thereby avoiding
duplicate execution of various tests.
Prior to this commit, `AsyncRestTemplate` would log errors (including
simple 404s) with WARN level. Such errors are quite common and should
not clutter logs.
This commit aligns the logging strategy with RestTemplate, using the
DEBUG level for such cases.
Fixes gh-28049
All Spring MVC exceptions from spring-web, now implement ErrorResponse
and expose HTTP error response information, including an RFC 7807 body.
See gh-27052
ErrorResponse represents a complete error response with status, headers,
and an RFC 7807 ProblemDetail body.
ErrorResponseException implements ErrorResponse and is usable on its
own or as a base class. ResponseStatusException extends
ErrorResponseException and now also supports RFC 7807 and so does its
sub-hierarchy.
ErrorResponse can be returned from `@ExceptionHandler` methods and is
mapped to ResponseEntity.
See gh-27052
ProblemDetail is a representation of an RFC 7807 "problem", and this
commits adds support for it in Spring MVC and WebFlux as a return value
from `@ExceptionHandler` methods, optionally wrapped with
ResponseEntity for headers.
See gh-27052
SocketUtils was introduced in Spring Framework 4.0, primarily to assist
in writing integration tests which start an external server on an
available random port. However, these utilities make no guarantee about
the subsequent availability of a given port and are therefore
unreliable. Instead of using SocketUtils to find an available local
port for a server, it is recommended that users rely on a server's
ability to start on a random port that it selects or is assigned by the
operating system. To interact with that server, the user should query
the server for the port it is currently using.
SocketUtils is now deprecated in 5.3.16 and will be removed in 6.0.
Closes gh-28052
In a small minority of cases, the multipart boundary can spread across
three incoming buffers.
Prior to this commit, MultipartParser.BodyState only supported two
buffers. If the boundary is spread across three buffers, the first
buffer of the three is sent as a whole, even though it contains the
first bytes of the boundary.
This commit fixes this bug, by enqueuing all prior buffers in a queue,
and emitting the ones that cannot contain boundary bytes.
Closes gh-27939
The failing test is for Apache HttpComponents where we cannot apply the
Content-Length together with other headers, and must pass it instead
explicitly to ReactiveEntityProducer when writing at which point it gets
set in the native headers.
Closes gh-27768
NettyHeadersAdapter::putAll uses HttpHeaders::add, which does not
overwrite existing headers. putAll should use HttpHeaders::set instead,
just like NettyHeadersAdapter::put does.
Closes gh-27887
Where unfeasible, this commit adds inline comments to explain why
try-with-resources must not be used in certain scenarios. The purpose
of the comments is to avoid accidental conversion to try-with-resources
at a later date.
Closes gh-27823
With JDK 9, java.util.regex.Matcher received overloaded variants of
appendReplacement() and appendTail() that accept a StringBuilder,
allowing us to replace the use of StringBuffer with StringBuilder.
Closes gh-27822
This commit replaces the use of Collections.unmodifiableList/Set/Map
with the corresponding 'of(...)' factory methods introduced in Java 9.
Closes gh-27824
This commit refactors HttpMethod from a Java enum into a class. The
underlying reason being that HTTP methods are not enumerable, but
instead an open range and not limited to the predefined values in the
specifications.
Closes gh-27697
The current implementation suggests that the request's headers are not
expected to change. Hence, it's not necessary to copy them.
Furthermore, it might be costly to do so if there are many headers.
Instead, cache only the request's header names for method getHeaderNames.
Methods getHeader and getHeaders delegate to the respective methods of
request if the header name is not in FORWARDED_HEADER_NAMES. Otherwise,
they return null or an empty Enumeration respectively.
See gh-27466
This commit makes several changes to MimeType and MediaType
related to the topic of specificity.
This commit deprecates the MimeType and MediaType Comparators.
Comparators require a transitive relationship, and the desired order for
these types is not transitive (see #27488).
Instead, this commit introduces two new MimeType methods: isMoreSpecific
and isLessSpecific, both of which return booleans. MediaType overrides
these methods to include the quality factor (q) in the comparison.
All MediaType sorting methods have been deprecated in favor of
MimeTypeUtils::sortBySpecificity. This sorting method now uses
MimeType::isLessSpecific in combination a bubble sort algorithm (which
does not require a transitive compare function).
Closes gh-27580
This commit ensures that only HTML references of length <
MAX_REFERENCE_SIZE are considered as potential references. This check is
possible because reference longer than 10 digits are out of bounds for
Integers.
Closes gh-1249
Alex Lei
Arjen Poutsma
rstoyanchev
binchoo
Olga Maciaszek-Sharma
Stephane Nicoll
izeye
Sam Brannen
Juergen Hoeller
Brian Clozel
a.yazychyan
Phillip Webb
heowc
shirohoo
Max Demydenko
sokomishalov
Marten Deinum
liuzhifei
ydh6226
Frederick Zhang
Daniel Le
김보배(Bobae Kim)/Platform Engineering팀/11ST
Arjen Poutsma
Julien Eyraud
d4ksn
Christoph Dreis
SungMin
happyWilliam0