[
https://issues.apache.org/jira/browse/FLINK-5852?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15888098#comment-15888098
]
ASF GitHub Bot commented on FLINK-5852:
---------------------------------------
Github user zentol commented on a diff in the pull request:
https://github.com/apache/flink/pull/3365#discussion_r103463182
--- Diff:
flink-core/src/main/java/org/apache/flink/api/common/ArchivedExecutionConfig.java
---
@@ -54,6 +54,19 @@ public ArchivedExecutionConfig(ExecutionConfig ec) {
}
}
+ public ArchivedExecutionConfig(
+ String executionMode,
+ String restartStrategyDescription,
+ int parallelism,
+ boolean objectReuseEnabled,
+ Map<String, String> globalJobParameters) {
+ this.executionMode = executionMode;
--- End diff --
That would lead to maintainability issues. The ArchivedExecutionConfigs
very purpose is to truthfully represent some state X. In order to do so it must
only impose restrictions that the original object imposed, i.e the
ExecutionConfig, otherwise we risk that an ExecutionConfig can't be archived
which doesn't make sense. This in turn would require that any change to the
conditions that the EC imposes must be propagated to the ArchviedEC, which
simply won't happen.
> Move JSON generation code into static methods
> ---------------------------------------------
>
> Key: FLINK-5852
> URL: https://issues.apache.org/jira/browse/FLINK-5852
> Project: Flink
> Issue Type: Improvement
> Components: Webfrontend
> Reporter: Chesnay Schepler
> Assignee: Chesnay Schepler
> Fix For: 1.3.0
>
>
> In order to implement the HistoryServer we need a way to generate the JSON
> responses independent of the REST API. As such i suggest to move the main
> parts of the generation code for job-specific handlers into static methods.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
