Prior to this change, location paths used for resource handling would
not allow "non-cleaned, relative paths" such as
`file://home/user/static/../static/`. When checking if the resolved
resource's path starts with the location path, a mismatch would happen
when comparing for example:
* the location `file://home/user/static/../static/`
* and the resource `file://home/user/static/resource.txt`
This commit cleans the location path before comparing it to the resource
path.
Issue: SPR-12624
This commit allows the use of "protcol relative URLs" (i.e. URLs without
scheme, starting with `//`), often used to serve resources automatically
from https or http with third party domains.
This syntax is allowed by RFC 3986.
Issue: SPR-12632
This commit fixes the default value for the contextRelative attribute of
a RedirectView, when this view is registered via a
RedirectViewController in XML. The value is set to true.
Note that the default value for this is correctly documented in
spring-mvc-4.1.xsd. Also, the documentation and implementation for its
javadoc counterpart also enforces true as a default value.
Issue: SPR-12607
A logical follow-up on commit 43d937, this change also removes (or
rather deprecates for now) writePrelude that is only of concern to
streaming SockJS session implementations.
Issue: SPR-12427
This change removes the need for the isStreaming field from the base
class AbstractHttpSockJsSession. This field was used to account for
differences between polling vs streaming SockJS sessions without having
to expose to sub-classes private fields that are otherwise protected
from concurrent access by the base class. The change manages to delegate
to sub-classes without providing direct access to protected fields.
Issue: SPR-12427
This commit overhauls several of the tests that interact with an
MBeanServer with the goal of increasing the reliability of these tests.
- MBeanClientInterceptorTests now uses JUnit "assumptions" instead of
preemptively returning from test methods, thus allowing such methods
to be properly marked as "ignored" instead of "passed".
- MBeanClientInterceptorTests now uses JUnit's support for expected
exceptions where appropriate.
- MBeanClientInterceptorTests and RemoteMBeanClientInterceptorTests now
use Spring's SocketUtils to find an available TCP port when starting
an MBeanServer instead of aborting the tests when the default JMX
port is not available.
Issue: SPR-12601
Commit 65d163e changed the textual message of an exception thrown by
ScheduledAnnotationBeanPostProcessor.afterSingletonsInstantiated(), and
this in turn caused the withAmbiguousTaskSchedulers_andSingleTask()
method in EnableSchedulingTests to start failing (albeit only during
'Performance' builds).
This commit updates the assertion to match the current implementation of
ScheduledAnnotationBeanPostProcessor.
In order to allow DefaultActiveProfilesResolver to be reused (e.g., via
extension or delegation), the check which asserts that the 'resolver'
attribute of @ActiveProfiles is not set to a customer resolver class
has been removed.
Issue: SPR-12611
JUnit 4.9 introduced a regression in BlockJUnit4ClassRunner.runChild()
such that exceptions thrown from methodBlock() cause the current test
execution to abort immediately. As a result, the failing test method is
unrooted, and subsequent test methods are never invoked. Furthermore,
RunListeners registered with JUnit are not properly notified.
In conjunction with SPR-11908, SpringJUnit4ClassRunner was updated to
use the aforementioned changes to BlockJUnit4ClassRunner.runChild().
Consequently, SpringJUnit4ClassRunner now suffers from the same
regression.
This commit addresses this issue by ensuring that any exceptions thrown
during the invocation of methodBlock() are properly wrapped in a JUnit
Fail Statement.
Issue: SPR-12613
- Copyright end date is now dynamic, based on the current year.
- Added missing spaces in front of web link to ASL 2.0 license file.
- Changed @since tag to 4.2.
Prior to this commit, HTTP responses without body (response status 204
or 304, Content-Length: 0) were handled properly by RestTemplates. But
some other cases were not properly managed, throwing exceptions for
valid HTTP responses.
This commit better handles HTTP responses, using a response wrapper that
can tell if a response:
* has no message body (HTTP status 1XX, 204, 304 or Content-Length:0)
* has an empty message body
This covers rfc7230 Section 3.3.3.
Issue: SPR-8016
Prior to this change, the ShallowEtagHeaderFilter would use a
ResizableByteArrayOutputStream to internally write data and calculate
the ETag. While that implementation is faster than the regular
ByteArrayOutputStream (since it has a better strategy for growing the
internal buffer), a lot of buffer copying/writing still happens.
This change adds a new FastByteArrayOutputStream implementation that
internally uses a LinkedList<Byte[]> to store the content. So when
writing bytes to that OutputStream implementation, new byte[] are
added to the list when the previous ones are full. This saves most
of the instantiating/copying operations.
Note that new methods were added in DigestUtils to allow usage of
Streams instead of byte[], which is more efficient in our case.
Fixes#653
Issue: SPR-12081
The test was verifying that when a DISCONNECT frame is sent to the
broker, there will be no further messages on the clientOutboundChannel.
This is generally true, however in some cases when the broker receives
a DISCONNECT it may close its connection fast enough (before we do) in
which case we send an ERROR message downstream to ensure the WebSocket
side is cleaned up. Either way the downstream should be idempotent
with regards to cleaning up sessions.
This change designates Jetty SockJS integration tests to run as part of
the "performance", but not the main "publication", CI build due to
recurring low-level failures suspected to be Jetty issues, e.g.
"java.io.IOException: Cannot append to finished buffer" or
"java.io.IOException: Out of order Continuation frame encountered".
The tests will still run at once a day with the performance build but
should not fail the main build with false negatives. Also note that
an Undertow variant of the exact same tests, which hasn't been failing,
will continue to run as part of the main build.
Update HttpComponents wrapper to merge local customizations with the
default of the current HttpClient instead of overriding everything.
This is available as from HttpComponents 4.4. that exposes the default
request config from the client via the Configurable interface. If the
client does not implement such interface, the previous behaviour is
applied
Issue: SPR-12583
Fix a regression introduced by 961574bd17 that prevents a proper lookup
of the default transaction manager by name as the absence of a qualifier
is represented by an empty string (passing the faulty null check).
Issue: SPR-12577
Prior to this commit, some requests would be wrongly mapped to "/" when:
* requests ends with "//" such as "/foo/bar//"
* the DispatcherServlet is mapped as the default ("/")
* the app container sanitizes the servletPath
In those cases, the path extraction algorithm was wrongly guessing the
path of the current request.
This commit detects if the app container sanitized the servletPath for
the current request. If so, it uses a sanitized version of the
requestUri to extract the path information.
Issue: SPR-12372
Juergen Hoeller
Rossen Stoyanchev
Brian Clozel
Sebastien Deleuze
Stephane Nicoll
Maciej Ziarko
Sam Brannen
Craig Andrews
John
Xiaolong Zuo