https://issues.jboss.org/browse/WFCORE-508 Remove need to maintain discovery-option ordering separately from the option #501
Conversation
|
Linux Build 1215 is now running using a merge of 66b4c4c |
|
Linux Build 1215 outcome was SUCCESS using a merge of 66b4c4c |
| DiscoveryOptionsResource(Resource delegate) { | ||
| if (delegate == null) { | ||
| throw HostControllerLogger.ROOT_LOGGER.nullVar("delegate"); | ||
| } |
There was a problem hiding this comment.
No need to change this; just an FYI.
We put in quite a few of these early in AS7 days, but they can be just an assert delegate != null; If this is null it's a programming error, not a mistake in user input.
|
FYI, I found this while looking at what this branch does: https://issues.jboss.org/browse/WFCORE-565 Not something that needs to be fixed in this PR or by you; just an FYI. |
| } | ||
| discoveryOptionsOrdering.add(type, Util.getNameFromAddress(operation.get(OP_ADDR))); | ||
| list.add(type, node); |
There was a problem hiding this comment.
This produces a list of ModelType.PROPERTY instead of the specified list of ModelType.OBJECT
ModelNode element = new ModelNode();
element.get(type).set(node);
list.add(element);
will produce the expected structure.
…scovery-option ordering separately from the option
…e value type during the description generation
…nd remove handlers
aloubyansky
force-pushed
the
WFCORE-508-discovery-options
branch
from
66b4c4c
to
bb6db8b
Compare
|
Linux Build 1275 is now running using a merge of bb6db8b |
| @@ -52,6 +52,7 @@ | |||
|
|
|||
| /** | |||
| * Gets the host name of the remote domain controller. | |||
| * This method may return null. | |||
There was a problem hiding this comment.
I was not sure about this. It looks to me it's not expected to return null though.
There was a problem hiding this comment.
It's messy. It can return null until the discover() method is invoked. I commented more on WFCORE-565.
I wanted to play with this yesterday so I could see what the read-resource and read-resource-description responses looked like, including with custom-discovery set up. I found a lot of problems, but not due to this patch.
https://github.com/bstansberry/wildfly-core/commits/WFCORE-565 had some quick changes just to get things so I could do my experiments.
|
Windows Build 960 outcome was FAILURE using a merge of bb6db8b Build problems:Failed tests detected Failed tests |
|
Linux Build 1275 outcome was FAILURE using a merge of bb6db8b Build problems:Failed tests detected Failed tests |
…inition allows undefined values for its attributes
aloubyansky
force-pushed
the
WFCORE-508-discovery-options
branch
from
bb6db8b
to
bc790f6
Compare
|
Linux Build 1276 is now running using a merge of bc790f6 |
|
Linux Build 1276 outcome was FAILURE using a merge of bc790f6 Build problems:Failed tests detected Failed tests |
|
retest this please |
|
Linux Build 1277 is now running using a merge of bc790f6 |
|
Linux Build 1277 outcome was SUCCESS using a merge of bc790f6 |
bstansberry
added
the
ready-for-merge
https://issues.jboss.org/browse/WFCORE-508 Remove need to maintain discovery-option ordering separately from the option
[JBEAP-14276] upgrade picketbox to 5.0.3.Final
The first commit introduces a new attribute 'options' for discovery-options resource, which contains a LIST of various discovery options as described on the jira.
The resources discovery-option and static-discovery are still preserved in the tree, although they are PlaceholderResource's now, all the actual data is kept in the newly introduced attribute and all the resource manipulation will be translated into the corresponding manipulation of the attribute.
discovery-options attribute is also preserved for now, although it's a read-only attribute now and the read-attribute handler will pull the data from the options attribute.
The second commit fixes value type initialization for complex types with map attributes during the generation of descriptions.
Thanks,
Alexey