Move the nested InternalParser class up, merging it with the top-level
MultipartParser, make the constructor private, and expose a static
parse method.
Prior to this commit, comments sent with Server Sent Events could break
the wire format when sent over the network when comments contained line
breaks.
While comments are mainly used for sending keepalive messages, they can
also be used for sending debug data. This commit ensures that line
breaks are properly handled in comments.
Fixes gh-36866
When parsing a relative URI such as `path2#foo`, RfcUriParser's
SCHEME_OR_PATH state advanced to FRAGMENT without moving the component
index past the `#` character, so the captured fragment included the
entire input (`path2#foo`) instead of just `foo`. As a result,
`toUriString()` produced `path2#path2#foo`.
Update the SCHEME_OR_PATH `#` transition to advance the component
index to `i + 1`, matching the sibling `?` -> QUERY transition in the
same state and every other `-> FRAGMENT` transition in the parser.
URIs with a `/` in the path are unaffected because they leave
SCHEME_OR_PATH for PATH on the first slash; the WhatWG parser already
handled this case correctly.
Closes gh-36762
Signed-off-by: daguimu <daguimu.geek@gmail.com>
Prior to this commit, gh-36328 avoided using RFC 2047 encoding for the
"filename" parameter and use ISO-8859-1 only. This change unfortunately
caused issues because some implementations might try and detect the
encoding automatically.
This commit restricts the filename parameter to ASCII encoding only by:
* transliterating characters to the closes ASCII character
("é"->"e", "ä"->"ae"...)
* falling back to "_" for other chacacters with non latin alphabet or
emojis
Fixes gh-36805
Prior to this commit, the `PartGenerator` would allow requesting
additional part tokens while in the `CreateFileState`. This is invalid
as any new token emitted would be rejected and would fail the entire
process.
This would only happen if the tmp file creation is slow enough for a new
token to be parsed and emitted.
This commit ensures that no new part token is requested while creating
the temporary file.
This change also fixes lifecycle issues and ensures that buffer
resources are cleaned in case of errors.
Fixes gh-36694
This commit reverts the changes made to WebDataBinder's doBind()
implementation in e4d03f6625 and instead implements the skipping logic
directly in checkFieldDefaults(), checkFieldMarkers(), and
adaptEmptyArrayIndices() by preemptively checking if the corresponding
field is allowed.
This commit also improves the Javadoc and adds missing tests.
Fixes gh-36625
Prior to this commit, the implementation of HttpMethod.valueOf()
aligned with the semantics of Enum#valueOf() which requires an exact
match for the enum constant name.
However, since HttpMethod is no longer an enum, that restriction is no
longer necessary. Consequently, this commit revises the implementation
of valueOf() to perform a case-insensitive lookup for predefined
constants.
In other words, HttpMethod.valueOf("GET") and HttpMethod.valueOf("get")
now both resolve to HttpMethod.GET.
Closes gh-36518
Prior to this commit, fields that are not allowed for binding were
always skipped and would not be bound. But the field and default marker
support (with the "_" and "!" prefixes) would be still considered and
could trigger collection instantiation/autogrow.
While this does not cause unwanted binding, this allocates memory for no
reason. This commit revisits the binding algorithm to only consider
default and field marker support if the regular field is allowed.
Fixes gh-36625
Prior to this commit, all `with**Converter` methods in
`HttpMessageConverters` builders would only apply if the
`registerDefaults()` option is on. While the original intent is to allow
overrides for auto-detected defaults, this creates an unnecessary
discrepancy with `addCustomConverter` which is always effective.
Allowing sich registrations when defaults are off should not introduce
invalid states and only reflects the intent of the developer. This
commit now allows such cases.
Fixes gh-36579
As announced in "the state of HTTP clients in Spring" blog post
(https://spring.io/blog/2025/09/30/the-state-of-http-clients-in-spring),
the deprecation timeline for `RestTemplate` was announced last November
and docs were updated accordingly.
This commit `@Deprecate` `RestTemplate` and related types for removal to
send a stronger signal to our community.
The actual removal is scheduled for Spring Framework 8.0 (not yet
scheduled).
Closes gh-36574
Now that `RestClient` offers a better alternative, this commit revisits
our integration tests to use `RestClient` instead of `RestTemplate`.
Closes gh-36573
This commit adds integration tests and reference documentation
for multipart support in `RestClient` and `RestTestClient`.
Closes gh-35569
Closes gh-33263
Prior to this commit, the `FormHttpMessageConverter` would only support
`MultiValueMap` payloads for reading and writing. While this can be
useful when web forms contain multiple values under the same key, this
prevent developers from using a common `Map` type and factory methods
like `Map.of(...)`.
This commit relaxes constraints in `FormHttpMessageConverter` and
ensures that `Map` types are supported for reading and writing URL
encoded forms.
Note that when reading forms to a `Map`, only the first value for each
key will be considered and other values will be dropped if they exist.
Closes gh-36408
Prior to this commit, gh-36255 introduced the new
`MultipartHttpMessageConverter`, focusing on multipart message
conversion in a separate converter. The `FormHttpMessageConverter` did
conflate URL encoded forms and multipart messages in the same converter.
With the introduction of the new converter and related types in the same
package (with `Part`, `FormFieldPart` and `FilePart`), we can now
revisit this arrangement.
This commit restricts the `FormHttpMessageConverter` to URL encoded
forms only and as a result, changes its implementation to only consider
`MultiValueMap<String, String>` types for reading and writing HTTP
messages. Because type erasure, this converter is now a
`SmartHttpMessageConverter` to get better type information with
`ResolvableType`.
As a result, the `AllEncompassingFormHttpMessageConverter` is formally
deprecated and replaced by the `MultipartHttpMessageConverter`, by
setting part converters explicitly in its constructor.
Closes gh-36256
Prior to this commit, the `FormHttpMessageConverter` would write, but
not read, multipart HTTP messages. Reading multipart messages is
typicaly performed by Servlet containers with Spring's
`MultipartResolver` infrastructure.
This commit introduces a new `MultipartHttpMessageConverter` that copies
the existing feature for writing multipart messages, borrowed from
`FormHttpMessageConverter`. This also introduces a new `MultipartParser`
class that is an imperative port of the reactive variant, but keeping it
based on the `DataBuffer` abstraction. This will allow us to maintain
both side by side more easily.
This change also adds new `Part`, `FilePart` and `FormFieldPart` types
that will be used when converting multipart messages to
`MultiValueMap<String, Part>` maps.
Closes gh-36255
Prior to this commit, the `DefaultRestClient` implementation would
look at the type of the returned value to decide whether the HTTP
response stream needs to be closed before returning.
As of gh-36380, this checks for `InputStream` and `InputStreamResource`
return types, effectively not considering `ResponseEntity<InputStream>`
and variants.
This commit ensures that `ResponseEntity` types are unwrapped before
checking for streaming types.
Fixes gh-36492
rstoyanchev
Sam Brannen
Sébastien Deleuze
Brian Clozel
daguimu
Vinod Kumar
Juergen Hoeller
Vishnutheep B