Previously, if endpoints.enabled was false setting
endpoints.docs.enabled=true or endpoints.actuator.enabled=true would
have no effect as their entire configuration class was conditional
on endpoints.enabled being true.
This commit updates the conditions on the configuration class so that
it is conditional on either the actuator or docs endpoint being enabled.
Closes gh-5007
Plugin disabling logic was broken by e009d3e4. Prior to this change,
a plugin would be disabled if it or any of the implemented interfaces
in its inheritance hierarchy were configured as being disabled. The
offending commit inverted the logic so that the plugin would be
enabled if any part of it was NOT configured as being disabled.
This commit restores the logic such that the early return happens only
in the negative case.
Previously, the tests were written as though
PluginContext#getPlugin(Class) would consider the specified class
against the runtime type of the plugin (not an unreasonable
assumption); rather this method considers the broader 'plugin type'.
This commit rewrites the test to seek by plugin type and assert the
absence of the disabled plugins.
Closes gh-5032
Support for configuring an endpoint’s path separately from its id was
introduced in 97255785, but it didn’t work for a variety of reasons:
1. Some custom MVC endpoints did not have configuration properties
bound to them
2. Some generic endpoints rejected the path property as they were
configured not to ignore unknown fields
3. The property used to configure the path was dependent on the id
of the endpoint. This meant that the path property’s name would
change if the endpoint’s id was changed
This commit addresses these problems:
1. @ConfigurationProperties has been added to custom MvcEndpoints where
it was missing
2. Generic endpoints have been updated to ignore unknown fields,
allowing the path of their MVC adapter to be configured
3. Rather than using the id of a generic endpoint to determine the name
of its path property, the prefix or value of the endpoint’s
@ConfigurationProperties annotation is used instead. Any generic
endpoint that is not annotated with @ConfigurationProperties is
ignored, making its path unconfigurable.
Closes gh-5105
Rework commit b726974 to avoid exposing setters that would permit anyone
to change Spring Boot's defaults. Also, since these are configurers of a
specific instance, they should be named accordingly.
Closes gh-5138
The Jolokia auto-configuration requires ServletWrappingController from
Spring MVC to be on the classpath. This commit updates the
auto-configuration to make it conditional on the presence of this
class.
Closes gh-5153
There's a long cycle when Spring Data REST, Data JPA and Actuator
are used in an app that retrieves its DataSource from JNDI. The cycle
is:
- WebMvcAutoConfiguration
- HttpMessageConverters
- MappingJackson2HttpMessageConverter (needs an ObjectMapper)
- SpringBootRepositoryRestMvcConfiguration
- ObjectMapper
- RepositoryResourceMappings (part of a custom Jackson module)
- Repositories
- EntityManagerFactory (Triggered by application's Spring Data JPA repository)
- HibernateJpaAutoConfiguration
- JndiDataSourceAutoConfiguration
- MBeanExporter (Used to prevent export of DataSource MBean that's already in JMX)
- EndpointMBeanExportAutoConfiguration
- ObjectMapper (Used to format JSON produced by the exported endpoints)
Spring Data Rest caused the ObjectMapper to depend on JPA. JPA depends
on the DataSource. JnidDataSourceAutoConfiguration depends on the
MBeanExporter. Actuator's MBeanExporter requires an ObjectMapper to
produce JSON strings.
This commit breaks the cycle by making JndiDataSourceAutoConfiguration
access the MBeanExporter lazily. Rather than using `@Lazy`. which does
not work with `@Autowired(required=false)`, the application context
is injected and the MBeanExporter is retrieved manually when it is
needed.
Closes gh-4980
Spring Boot supports the automatic configuration of an additional
HazelcastInstance if one already exists and an explicit property has been
set to use a different configuration for caching. So three cases are
supported really: no `HazelcastInstance` exists so we need to create one
anyway or an `HazelcastInstance` already exists; in that latter case, we
should either reuse it or create a new one.
Unfortunately, the conditions that checked those three use cases were
not ordered consistently and we could easily get in a situation where
both conditions were evaluated.
This commit makes sure that we first check if an `HazelcastInstance`
exists and then (and only then) we create the missing `HazelcastInstance`
used for caching. The tests have also been improved to validate the
proper `HazelcastInstance` is used for caching.
Closes gh-5181
Update the requestsWithDisallowedMethodsAreRejected test to use PATCH
rather than HEAD. The change is to allow support for Spring Framework
4.3 which will implicitly map HEAD requests to GET.
Pre-flight requests are also only for "non-simple" HTTP methods [1]
(i.e. anything but GET, HEAD, POST) so there is really no such a thing
as a pre-flight request for HEAD.
[1] https://www.w3.org/TR/cors/#resource-preflight-requests
Rework a few parts of the diagnostics support:
- Move code from SpringApplication to FailureAnalyzers
- Allow AbstractFailureAnalyzer to take generic cause type
- Move own analyzers into a new package and make package private
See gh-4907
Andy Wilkinson
Matt Benson
Vedran Pavic
Stephane Nicoll
Phillip Webb
Gary Russell
Dave Syer