Commit Graph

50613 Commits

Author SHA1 Message Date
Andy Wilkinson
8810450e64 Upgrade to Spring Data Hopper SR11
Closes gh-9424
2017-06-07 13:23:17 +01:00
Andy Wilkinson
dc7498244d Merge branch '1.5.x' 2017-06-07 10:41:05 +01:00
Andy Wilkinson
197c0cefbb Merge pull request #9259 from Ethan Rubinson
* gh-9259:
  Polish "Copy conversion service when performing environment conversion"
  Copy conversion service when performing environment conversion
2017-06-07 10:27:14 +01:00
Andy Wilkinson
5b30269ac9 Polish "Copy conversion service when performing environment conversion"
Closes gh-9246
2017-06-07 10:24:15 +01:00
Rubinson,Ethan(erubinson)
a424081803 Copy conversion service when performing environment conversion
Previously, when a web environment was converted to a
StandardEnvironment, any customizations of the source environment's
ConversionService were lost.

This commit updates the logic that performs the conversion to copy
the source's ConversionService to the converted environment, thereby
ensuring that any customizations are retained.

Closes gh-9259
See gh-9246
2017-06-07 10:21:56 +01:00
Andy Wilkinson
563c9dc58c Merge branch '1.5.x' 2017-06-06 22:24:31 +01:00
Andy Wilkinson
84721666c5 Stop DuplicateJsonObjectContextCustomizer breaking context caching
See gh-9248
2017-06-06 22:23:43 +01:00
Andy Wilkinson
61a1798296 Merge branch '1.5.x' 2017-06-06 21:39:09 +01:00
Andy Wilkinson
21a5ab7875 Log a warning during tests when multiple JSONOjects on the class path
Closes gh-9248
2017-06-06 21:35:33 +01:00
Stephane Nicoll
05fbd5dc46 Merge branch '1.5.x' 2017-06-06 16:04:29 +02:00
Stephane Nicoll
cbdab9edb3 Add support for custom ProtocolResolver with Devtools
Prior to this commit, custom `ProtocolResolvers` set on the
`ApplicationContext` were lost when Devtools is used as the customized
`ResourceLoader` did not copy any customization made to the default
resource loader.

This commit makes sure to copy any `ProtocolResolver` set on the context.

Closes gh-9331
2017-06-06 16:04:12 +02:00
Andy Wilkinson
d844a0cf66 Merge branch '1.5.x' 2017-06-06 14:57:40 +01:00
Andy Wilkinson
aa6dbdbae2 Ensure that listeners are called when application fails to run
Closes gh-9054
2017-06-06 14:53:05 +01:00
Andy Wilkinson
d0c8f80117 Merge branch '1.5.x' 2017-06-06 11:46:32 +01:00
Andy Wilkinson
ad629055fa Ensure that custom static resource locations end with /
Closes gh-9360
2017-06-06 11:39:04 +01:00
Andy Wilkinson
cdebfcdeef Merge branch '1.5.x' 2017-06-06 11:22:31 +01:00
Andy Wilkinson
5be5b13775 Clarify need for Apache HTTP Client to disable redirects in TestRestTemplate
Closes gh-9410
2017-06-06 11:14:15 +01:00
Andy Wilkinson
9464eea7ae Merge branch '1.5.x' 2017-06-06 10:43:11 +01:00
Andy Wilkinson
a666919acf Increase prominence of warning about fully-executable jar compatibility
Closes gh-8927
2017-06-06 10:42:30 +01:00
Stephane Nicoll
f61ad19299 Merge pull request #9411 from vpavic:improve-quartz-support
* pr/9411:
  Polish contribution
  Ensure `QuartzDatabaseInitializer` is initialized before `Scheduler`
2017-06-06 10:59:52 +02:00
Stephane Nicoll
7f420d1268 Polish contribution
Closes gh-9411
2017-06-06 10:52:31 +02:00
Vedran Pavic
db060c847d Ensure QuartzDatabaseInitializer is initialized before Scheduler
If the auto-configured `Scheduler` instance backed by JDBC job store is
used as a dependency in an application component, the initialization of
`Scheduler` will be triggered before `QuartzDatabaseInitializer`. This
will result in failure due to schema not being prepared in time for
`Scheduler` to populate job details.

This commit ensures `QuartzDatabaseInitializer` is initialized before the
auto-configured `Scheduler` by introducing a dependency between the two.

See gh-9411
2017-06-06 10:14:49 +02:00
Andy Wilkinson
b16fdafa3c Merge branch '1.5.x' 2017-06-05 11:15:43 +01:00
Andy Wilkinson
3aaf5b6463 Upgrade to Dependency Management Plugin 1.0.3.RELEASE
Closes gh-9407
Closes gh-9135
2017-06-05 11:15:25 +01:00
Stephane Nicoll
c1b08be604 Merge pull request #9404 from htynkn:feature/document-influx-db-in-reference-manual
* pr/9404:
  Polish "Document InfluxDB support"
  Document InfluxDB support
2017-06-05 11:07:10 +02:00
Stephane Nicoll
068f904f24 Polish "Document InfluxDB support"
Closes gh-9404
2017-06-05 11:06:49 +02:00
Huang YunKun
0ee5048412 Document InfluxDB support
See gh-9404
2017-06-05 11:00:27 +02:00
Stephane Nicoll
540dca7bdd Revisit InfluxDB configuration structure
This commit removes the `client` namespace for InfluxDB as the
component that is created is `InfluxDB`, not `InfluxDBClient` or
something.

This aligns with all the other url/user/password properties Spring
Boot provides already

See gh-9066
2017-06-05 11:00:14 +02:00
Andy Wilkinson
4a6b34e725 Merge branch '1.5.x' 2017-06-03 19:41:35 +01:00
Andy Wilkinson
d5184724b0 Downgrade Jersey to version that plays nicely with Docker's Java client 2017-06-03 19:40:21 +01:00
Andy Wilkinson
09a8b3c072 Merge branch '1.5.x' 2017-06-03 16:54:02 +01:00
Andy Wilkinson
ec8ae2f91e Merge branch '1.4.x' into 1.5.x 2017-06-03 16:53:48 +01:00
Andy Wilkinson
e60a261952 Do not allow failure to remove container to mask earlier failure 2017-06-03 16:53:28 +01:00
Andy Wilkinson
49fc1b831c Merge branch '1.5.x' 2017-06-03 13:13:04 +01:00
Andy Wilkinson
a215103bb5 Merge branch '1.4.x' into 1.5.x 2017-06-03 13:11:39 +01:00
Andy Wilkinson
31ff7f1846 Tolerate Successfully built being found in response other than last
Different versions of Docker produce different responses when building
and tagging an image. On CI, a response with a stream like
"Successfully built 185991ffe24a" followed by a response with a
stream like "Successfully tagged spring-boot-it/centos:6.9-a23bced6"
is received. By default, for the building of an image to be considered
successful, the Docker Java client requires the stream for the last
response item to contain "Successfully built". This means that, on the
CI server, it incorrectly believes that the building of the tagged
image has failed.

This commit uses a custom BuildImageResultCallback that doesn't
require the last response to be the one that has a stream containing
"Successfully built". Instead, it looks back through the error-free
responses (newest to oldest) looking for one with a stream containing
"Successfully built".
2017-06-03 13:00:15 +01:00
Andy Wilkinson
b8bd51fd48 Merge branch '1.5.x' 2017-06-03 09:55:47 +01:00
Andy Wilkinson
5722c149d0 Merge branch '1.4.x' into 1.5.x 2017-06-03 09:55:29 +01:00
Andy Wilkinson
0270ccafa5 Enable Docker command debug logging to improve diagnostics on CI server 2017-06-03 09:54:13 +01:00
Phillip Webb
2c7dd9f519 Polish 2017-06-02 16:00:39 -07:00
Phillip Webb
8f7004738b Merge branch '1.5.x' 2017-06-02 13:48:27 -07:00
Phillip Webb
b9fd99e268 Polish 2017-06-02 13:47:28 -07:00
Andy Wilkinson
4f747d426e Drop back to Surefire 2.19.1 as 2.19.2 was rebranded as 2.20 2017-06-02 20:15:11 +01:00
Andy Wilkinson
33924f5569 Drop back to Surefire 2.19.2 in case 2.20 has a memory leak
See gh-6299
2017-06-02 20:08:47 +01:00
Andy Wilkinson
99790f67e0 Merge branch '1.5.x' 2017-06-02 19:49:51 +01:00
Andy Wilkinson
3cf8e81a8b Merge branch '1.4.x' into 1.5.x 2017-06-02 19:49:16 +01:00
Andy Wilkinson
a69a851ae9 Drop CentOS 5 from launch script tests and upgrade to 6.9
CentOS 5 was declared EOL in March 2017 and yum on longer works
out of the box. 6.9 is the latest release of CentOS 6. Tests for
CentOS 7 have not been added as it uses systemd rather than SysVinit.

Closes gh-9395
2017-06-02 19:45:45 +01:00
Stephane Nicoll
9624bf14c4 Merge branch '1.5.x' 2017-06-02 17:58:18 +02:00
Stephane Nicoll
90498a0b2f Fix broken build 2017-06-02 17:58:14 +02:00
Stephane Nicoll
f11edd1eb3 Add missing @Configuration on EmbeddedDatabaseConfiguration 2017-06-02 17:42:18 +02:00