This fixes a bug in the sample, where the AuthenticationManager it builds
is a local one for the filter chain containing "/login", whereas it was
expecting to override the Boot default, which is "global". The fix is
to extract the authentication configuration out into a
GlobalAuthenticationConfigurerAdapter.
Fixes gh-699
User can now set up default properties in the Environment using
@IntegrationTest("foo:bar", "x:y") etc. Using "server.port:0" you
can get Tomcat or Jetty to spin up on a random port. We also add
a test listener that populates "local.server.port" with the actual
port the server started on so you can @Value("${local.server.port}")
inject it into the test case.
See gh-607 (this should make the extension of that PR to samples
much easier)
Apparently springloaded 1.1.5 was released to Maven Central with a
different groupId than to repo.spring.io. It would have worked in
milestones because the repository was declared.
See gh-648
Spring Security doesn't know that Spring MVC maps /foo, /foo.json
and /foo/ all to the same handler. This change explicitly adds
suffixes to the actuator endpoint matchers so they are properly
protected.
Rename the RestTemplates to TestRestTemplate to help indicate that it's
primarily intended for testing. Also now extend RestTemplate to allow
direct use, rather than via factory methods.
Fixes gh-599
If the user explicitly disables the basic security features and forgets to
@EnableWebSecurity, and yet still wants a bean of type
WebSecurityConfigurerAdapter, he is trying to use a custom
security setup and the app would fail in a confusing way without
this change.
Fixes gh-568
If the user sets security.basic.enabled=false he has to remember
to @EnableWebSecurity. Possibly we could be more helpful about the
exception, but I think this might be pilot error.
Fixes gh-568
Remove README files that have been since been migrated to the reference
documentation. Also updated remaining markdown files to asciidoctor to
save having a mix of different formats.
Fixed gh-503
Makes them a lot more readable IMO, and also enables @Autowiring
from the context into the test case (sweeet). I added @DirtiesContext
to all of them as well to be on the safe side, but possbly that can be
optimized in some way as well.
The management.contextPath property should now be respected in a
secure application, whether or not the management.port is different.
Added some test cases in the sample to verify.
Fixes gh-469