Previously, the data scripts were always run in response to the
publication of a DataSourceInitializedEvent, irrespective of
spring.datasource.initialize. While the event won't be published by
DataSourceInitializer if spring.datasource.initialize is false, it
will be published if spring.jpa.hibernate.hbm2ddl.auto has been set.

This commit updates DataSourceInitializer's handling of
DataSourceInitializedEvent to only run the data scripts if
spring.datasource.initialize is true.

Fixes #1336

The request is being made to '/' and, while the application does have
a mapping for '/', that mapping is not looked for before Spring
Security's filter rejects the request with a 401. This means that the
request is considered to be unmapped and this is reflected in the
metric's name.

See #1331 and #1333

Fixes gh-1331, Fixes gh-1333

Prior to this commit, the internal BeanWrapper used by the
RelaxedDataBinder was not using any conversion service at all. This
commit makes sure to propagate the one that has been configured
internally.

Also a dummy call to a [foo] key seems to have been left and has
been removed. This allows to support Enum as map key.

Fixes gh-1242

The ErrorPageFilter wasn't setting the response status in the case that
there was an error page mapped to the current request (i.e. for all
autoconfigured apps). N.B. this only affects non-embedded apps.

Fixes gh-1320

Fixes gh-1328

Fixes gh-1321

Previously, the ErrorPageFilter would always flush the response
buffer, irrespective of the request being asynchronous. This could
lead to a response being committed prematurely, preventing, for
example, headers being set by subsequent processing.

This commit updates ErrorPageFilter so that in the success case
(status < 400) the response buffer is only flushed if the request is
not async (determined by calling request.isAsyncStarted()). If an
exception's been thrown or the status is >= 400 the response buffer
is always flushed.

Fixes #1316

Previously, Repackager would repackage a jar file as many times as
it was asked to do so. This lead to problems if a user made a mistake
when using Maven that led to the package phase being driven twice,
for example by running "mvn clean install package".

This commit updates Repackager so that a repackage call becomes a
no-op if the source jar's manifest already contains the
Spring-Boot-Version attribute which is added by repackaging.

Fixes #1251

Closes #1307

Closes #1305

Previously, when running on Java 7+, the JDBC query driven by
DataSourceHealthIndicator resulted in a call to
ResultSet.getObject(index, Object.class). When using MySQL's JDBC
driver this failed with an SQLException with the message "Conversion
not supported for type java.lang.Object". The problem does not occur
on Java 6 as the overload of getObject that takes a type does not
exist; ResultSet.getObject(index) is called instead and MySQL happily
returns whatever type it deems to be appropriate for the column.

This commit updates DataSourceHealthIndicator so that
ResultSet.getObject(index) will always be used, irrespective of the
version of Java that Boot is running on.

Closes #1306

Closes #1304

The original fix for gh-1293 (commit 05e6af23) caused test failures due
to the fact that Spring Boot's MappingJackson2HttpMessageConverter was
added before Spring's default StringHttpMessageConverter.

This commit changes the HttpMessageConverters logic so that additional
converts are added just before any default converter of the same type.
This allows additional converters to be added whilst still retaining
the sensible ordering of the default converters.

Fixes gh-1293

It was incorrect to simply replace existing instances because you can't
tell from the instance which media types and java types it supports. This
fix just prepends the custom converters so they get higher priority.

Fixes gh-1293

Since Spring 4.0.4, it's now possible to use both classes and
locations to provide a test's configuration.

Closes #1300

Closes #1299

Previously the server was started to make the ServletContext
available, then, to prevent requests from being handled before the
application context had been started, the connectors were stopped.
Once application context startup had completed, the connectors were
then started again. In addition to being somewhat inefficient, this
caused problems on FreeBSD where stopping the connector didn't free
up the port quickly enough for the subsequent start to then be able
to bind to it.

This commit updates the Jetty startup logic to be closer to the logic
that's used for Tomcat. Before the server is started, the configured
connectors are cached and then removed. The server is then started
without any connectors. Once application context startup has
completed, the connectors are reinstated and started.

Fixes #968

Often this change will not be important because you are binding to
a bean with strongly typed properties. A bean with a Map property,
on the other hand, won't oytherwise be able to reason about the
permitted keys so it will miss any non-enumerable property sources,
including composites whose nested sources are themselves enumerable.

Fixed gh-1294

This commit changes the query in MongoHealthIndicator from serverStatus to buildInfo to avoid unprivileged access and corresponding errors.

fixes #1289

With this change I got a simple Eclipselink version of the data-jpa
sample working. I'll push that when I get time to research it a bit more
(I needed to set up a Java agent so either that might be a problem
for our integration tests if we can't work around it).

Fixes gh-1268.

Previously any custom http code mapping would remove the default mappings. With this commit the behaviour is changed so that default mappings will stay if a custom mapping is registered. Certainly a default mapping can be overridden.

fixes #1264

Added $JAVA_OPTS variable to the sample Heroku Procfile.
If this is left out it can cause memory issues when the app starts.

Fixes gh-1266

This commit fixes some inconsistent or outdated keys in the
documentation. More specifically:

* allowSessionOverride is no longer a template parameter
* templateEncoding has been renamed to charSet
* Groovy templates do not have the same configuration hierarchy, hence
  they don't share all settings
* spring.data.elasticsearch.local does not seem to exist
* flyway prefix and suffix should be sqlMigrationPrefix and suffix
* spring.rabbitmq.virtualHost had a typo
* endpoints.error.path is not a valid property
* shell.command-path-patterns had a typo
* spring.datasource.max-wait had a typo

Fixes gh-1226

Fixes gh-1259

Fixes #1247

Fixes gh-1257

Using StringUtils.commaDelimitedListToSet() does not preserve order (why?),
so we have to use commaDelimitedListToStringArray().

Fixes gh-1262