Commit Graph

426 Commits

Author SHA1 Message Date
Juergen Hoeller
ace25d4987 Polishing
(cherry picked from commit ed19dc7)
2016-09-16 14:30:41 +02:00
Juergen Hoeller
7ddaf49eb2 StringUtils.parseLocaleString accepts Java 7 variants
Issue: SPR-14718
(cherry picked from commit f24ce76)
2016-09-16 10:58:51 +02:00
Juergen Hoeller
086e764845 ResolvableType.java.forRawClass(Class) supports isAssignableFrom(ResolvableType) as well
Issue: SPR-14648
(cherry picked from commit 1a30252)
2016-08-31 14:57:27 +02:00
Juergen Hoeller
696f687419 Moved encodeHttpHeaderFieldParam method to HttpHeaders itself (including tests)
This commit also sets the test source encoding to UTF-8.

Issue: SPR-14547
(cherry picked from commit a8f7f75)
2016-08-26 12:55:34 +02:00
Brian Clozel
9b91b9db8c Add RFC5987 support for HTTP header field params
This commit adds support for HTTP header field parameters encoding, as
described in RFC5987.
Note that the default implementation still relies on US-ASCII encoding,
as the latest rfc7230 Section 3.2.4 says that:

> Newly defined header fields SHOULD limit their field values to
  US-ASCII octets

Issue: SPR-14547
Cherry-picked from: f2faf84f31
2016-08-25 14:39:07 +02:00
Juergen Hoeller
1932a9d729 Polishing
(cherry picked from commit de91b1a)
2016-08-17 21:30:30 +02:00
Juergen Hoeller
cc93c2a559 Polishing 2016-07-26 00:03:21 +02:00
Juergen Hoeller
95c2e0a996 Upgrade JiBX (and BCEL) towards support of 1.8 bytecode level
This commit also drops the now-unused CUSTOM_COMPILATION test group and the outdated JavaVersion class.

Issue: SPR-10423
(cherry picked from commit 01110f6)
2016-07-25 14:58:43 +02:00
Juergen Hoeller
64595597f3 LinkedCaseInsensitiveMap properly overrides HashMap.clone()
Issue: SPR-14509
(cherry picked from commit dd65f7a)
2016-07-25 12:14:51 +02:00
Juergen Hoeller
36e1c82ef5 Backported refinements and polishing 2016-07-20 21:46:25 +02:00
Juergen Hoeller
1ca4b81856 Reinstated tests for implicit aliases
Issue: SPR-14437
(cherry picked from commit 5ea8c26)
2016-07-20 21:40:33 +02:00
Juergen Hoeller
4d6d5e0ddd Consistent alias processing behind AnnotatedTypeMetadata abstraction (also for ASM)
Issue: SPR-14427
(cherry picked from commit 3d3407c)
2016-07-15 15:40:01 +02:00
Juergen Hoeller
a1b58ee601 DigestUtils processes InputStream with buffered read instead of full copy
Issue: SPR-14427
(cherry picked from commit 61db8e9)
2016-07-15 15:34:36 +02:00
Juergen Hoeller
e30429051d Polishing
(cherry picked from commit 6aa5931)
2016-07-15 15:34:27 +02:00
Juergen Hoeller
92d78c10a2 Polishing (backported from master) 2016-07-06 17:13:18 +02:00
Sam Brannen
8389e3fc35 Suppress warnings in Gradle build 2016-07-01 13:08:39 +02:00
Juergen Hoeller
1d42009c0a Deprecated getPropertyAsClass and refined PropertySourcesPropertyResolver's logging
Issue: SPR-14370
2016-06-29 10:51:06 +02:00
Juergen Hoeller
a979885d17 SettableListenableFuture properly rethrows Error
Issue: SPR-14298
2016-05-26 19:39:22 +02:00
Juergen Hoeller
42d32ba396 ResourceRegion fits better in core.io.support (next to EncodedResource)
Issue: SPR-14221
2016-05-06 12:02:51 +02:00
Brian Clozel
3c92ddc94b Fix path matching for paths containing spaces
Prior to this commit, the latest optimizations introduced in SPR-13913
would prevent matching when patterns contained spaces. Indeed, the
optimized path would not fully tokenize the paths nor trim the tokens,
as the "longer" code path does.

This commit disables this optimized path when the `trimTokens` option is
set to `true`.

Also, the `trimTokens` setting is now set to `false` by default.

Issue: SPR-14247
2016-05-04 22:28:06 +02:00
Brian Clozel
5ac31fb39d Refactor HTTP Range support with ResourceRegion
Prior to this commit, the `ResourceHttpMessageConverter` would support
all HTTP Range requests and `MethodProcessors` would "wrap" controller
handler return values with a `HttpRangeResource` to support that use
case in Controllers.

This commit refactors that support in several ways:
* a new ResourceRegion class has been introduced
* a new, separate, ResourceRegionHttpMessageConverter handles the HTTP
range use cases when serving static resources with the
ResourceHttpRequestHandler
* the support of HTTP range requests on Controller handlers has been
removed until a better solution is found

Issue: SPR-14221, SPR-13834
2016-05-02 19:00:52 +02:00
Juergen Hoeller
bf7b475267 SocketUtils considers port range including maxPort
Issue: SPR-14211
2016-04-26 17:09:57 +02:00
Juergen Hoeller
9bf5a5cbcb FastByteArrayInputStream returns correct count from read(byte[])
Issue: SPR-14209
2016-04-26 17:09:57 +02:00
Rossen Stoyanchev
50c11028d5 Improve illegal MimeType checks
Issue: SPR-14124
2016-04-18 10:01:18 -04:00
Brian Clozel
368f29d5bc Fix AntPathMatcher multiple segments matching
Prior to this commit, the new match algorithm wouldn't work for multiple
consecutive path separators.
This commit separately matches path segments and path separators and
allows for multiple, consecutive path separators.

Issue: SPR-14141
2016-04-15 23:10:29 +02:00
Juergen Hoeller
f1cb793ccb Rename MimeType's getCharSet() to getCharset()
Issue: SPR-14172
2016-04-14 16:38:59 +02:00
Juergen Hoeller
74608e6b49 Polishing 2016-04-12 16:03:57 +02:00
Sam Brannen
6826bdd111 Document disabled test for hybrid annotation attribute overrides
This commit documents the @Ignore'd test to explain that Spring does not
support a hybrid approach for annotation attribute overrides with
transitive implicit aliases.

Issue: SPR-13554
2016-03-31 14:38:15 +02:00
Sam Brannen
fee056a1b1 Simplify search algorithms in AnnotatedElementUtils
This commit introduces a boolean alwaysProcesses() method in the
Processor API which allows for simplification across all search
algorithms within AnnotatedElementUtils.

Specifically, it is no longer necessary for process() methods to verify
that the supplied annotation is actually the sought target annotation.

In addition, duplicated code has been extracted into common methods
(e.g., hasMetaAnnotationTypes()) and common Processor implementations
(e.g., AlwaysTrueBooleanAnnotationProcessor).
2016-03-25 15:59:54 +01:00
Sam Brannen
46e0484bf8 Support searches for merged repeatable annotations with "get" semantics
This commit picks up where 2535469099 left off with added support for
"get" search semantics for merged repeatable annotations.

Specifically, this commit introduces a new
getMergedRepeatableAnnotations() method in AnnotatedElementUtils.

Issue: SPR-13973
2016-03-24 19:27:32 +01:00
Sam Brannen
8d0083ca96 Polishing 2016-03-24 19:27:32 +01:00
Juergen Hoeller
517ebd1d3e Consistent formatting 2016-03-24 19:22:50 +01:00
Sam Brannen
1ec35e9c62 Support searches for multiple merged annotations with "get" semantics
This commit picks up where a5139f3c66 left off with added support for
"get" search semantics for multiple merged annotations.

Specifically, this commit introduces a new getAllMergedAnnotations()
method in AnnotatedElementUtils.

Issue: SPR-13486
2016-03-24 14:46:56 +01:00
Sam Brannen
2535469099 Support repeatable annotations as composed annotations
Prior to this commit, AnnotationUtils supported searching for
repeatable annotations even if the repeatable annotation was declared
on a custom stereotype annotation. However, there was no support for
merging of attributes in composed repeatable annotations. In other
words, it was not possible for a custom annotation to override
attributes in a repeatable annotation.

This commit addresses this by introducing
findMergedRepeatableAnnotations() methods in AnnotatedElementUtils.
These new methods provide full support for explicit annotation
attribute overrides configured via @AliasFor (as well as
convention-based overrides) with "find semantics".

Issue: SPR-13973
2016-03-23 21:58:34 +01:00
Sam Brannen
a5139f3c66 Support searches for multiple merged composed annotations
Prior to this commit it was possible to search for the 1st merged
annotation above an annotated element. It was also possible to search
for annotation attributes aggregated from all annotations above an
annotated element; however, it was impossible to search for all
composed annotations above an annotated element and have the results as
synthesized merged annotations instead of a multi-map of attributes.

This commit introduces a new findAllMergedAnnotations() method in
AnnotatedElementUtils that finds all annotations of the specified type
within the annotation hierarchy above the supplied element. For each
such annotation found, it merges that annotation's attributes with
matching attributes from annotations in lower levels of the annotation
hierarchy and synthesizes the results back into an annotation of the
specified type. All such merged annotations are collected and returned
as a set.

Issue: SPR-13486
2016-03-20 19:18:11 +01:00
Sam Brannen
f1378aa376 Test status quo for finding multiple composed annotations
This commit introduces tests that verify the status quo for finding
multiple merged composed annotations on a single annotated element.

Issue: SPR-13486
2016-03-19 20:22:04 +01:00
Sam Brannen
d22480b0eb Ensure all aliased attributes in target annotation are overridden
Prior to this commit, it was possible that implicit aliases and
transitive implicit aliases (configured via @AliasFor) might not be
honored in certain circumstances, in particular if implicit aliases
were declared to override different attributes within an alias pair in
the target meta-annotation.

This commit addresses this issue by ensuring that all aliased
attributes in the target meta-annotation are overridden during the
merge process in AnnotatedElementUtils.

In addition, concrete default values for attributes in a
meta-annotation declaration can now be effectively shadowed by
transitive implicit aliases in composed annotations.

Issue: SPR-14069
2016-03-19 16:42:56 +01:00
Juergen Hoeller
1ca4340271 NumberUtils consistently raises overflow exception for BigInteger/BigDecimal input
Issue: SPR-14041
2016-03-18 18:51:41 +01:00
Sam Brannen
2153d88706 Add TODO in AnnotatedElementUtilsTests
Issue: SPR-14069
2016-03-18 15:29:25 +01:00
Sam Brannen
856da8e395 Introduce tests for transitive aliases with implied attribute names 2016-03-17 20:53:52 +01:00
Stephane Nicoll
f422e490b0 Polish contribution
Issue: SPR-14039 - Closes gh-998
2016-03-16 17:00:00 +01:00
Yanming Zhou
1a302e1fac Add converters between Enum and Integer 2016-03-16 17:00:00 +01:00
Sam Brannen
eb654dc177 Allow single element to override array in synthesized annotation
This commit picks up where 8ff9e818a5
left off.

Specifically, this commit introduces support that allows a single
element attribute to override an array attribute with a matching
component type when synthesizing annotations (e.g., in annotations
synthesized from attributes that have been merged from the annotation
hierarchy above a composed annotation).

Issue: SPR-13972
2016-03-07 15:22:27 +01:00
Juergen Hoeller
7a32ce317c LinkedCaseInsensitiveMap provides reliable getOrDefault implementation
Issue: SPR-13981
2016-02-25 21:42:11 +01:00
Juergen Hoeller
8ff9e818a5 Allow for a single element overriding an array attribute in a meta-annotation
Includes refinements for consistent quoting of names in exception messages.

Issue: SPR-13972
2016-02-22 23:24:30 +01:00
Phillip Webb
d8a05592e2 Allow package private classes in spring.factories
Update SpringFactoriesLoader so that package private classes can be
used.

Issue: SPR-13969
2016-02-22 10:01:35 -08:00
mnhock
e98977cbc8 Fix Format-string method arguments
Closes gh-934
2016-02-20 09:01:54 +01:00
Phillip Webb
c4117885bd Refine AntPathMatcher optimizations
Refine the optimizations made in 6f55ab69 in order to restore binary
compatibility and resolve a regression.

Tests of the form pathMatcher.match("/foo/bar/**", "/foo/bar") should
return true as this was the behavior in Spring 4.2.

Issue: SPR-13913
2016-02-18 18:49:58 -08:00
soldierkam
e77ff3c991 Improve AntPathMatcher matching performance
This commit speeds up the AntPathMatcher implementation by
pre-processing patterns and checking that candidates are likely
matches if they start with the static prefix of the pattern.

Those changes can result in a small performance penalty for positive
matches, but with a significant boost for checking candidates that don't
match. Overall, this tradeoff is acceptable since this feature is often
used to select a few matching patterns in a much bigger list.

This will lead to small but consistent performance improvements in
Spring MVC when matching a given request with the available routes.

Issue: SPR-13913
2016-02-10 11:39:38 +01:00
Juergen Hoeller
74c07d3085 Removed JavaVersion checks from spring-core tests (for JDK 9 compatibility)
Issue: SPR-13344
2016-01-15 17:45:37 +01:00