[
https://issues.apache.org/jira/browse/ATLAS-4623?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Allen Liang updated ATLAS-4623:
-------------------------------
Summary: Different business metadata behavior of applicableEntityTypes
compared to classification (was: Different business metadata behavior of
applicableEntityTypes compared to classfication)
> Different business metadata behavior of applicableEntityTypes compared to
> classification
> ----------------------------------------------------------------------------------------
>
> Key: ATLAS-4623
> URL: https://issues.apache.org/jira/browse/ATLAS-4623
> Project: Atlas
> Issue Type: Improvement
> Components: atlas-core
> Affects Versions: 2.2.0
> Reporter: Allen Liang
> Priority: Major
>
> Hi there,
> When compared with the implementation of classification in terms of
> applicable entity type, business metadata has a different implementation that
> 1) When applicable entity type is set to empty, none of entity types will be
> applicable (while all entity types will be applicable in classification)
> 2) When applicable entity type is set to a specific entity type, that
> specific entity type and each of its derived entity types will be added the
> business attributes in their entity definition response (while classification
> doesn't have this behavior)
>
> I have the following questions
> For 1), would it make more sense to align with classification's behavior
> (i.e., empty/null means no restriction to all entity types)?
> For 2), Due to 1) if one wants to make business attributes apply to all
> entity types today, the closest thing he could do would be setting applicable
> type to "Referenceable", and that would cause all entity type definition
> expansion due to 2). What's the best practice here?
>
> And in general, I am trying to understand what's the rationale behind this
> different applicableEntityType implementation in business metadata compared
> to classification? And do we consider changing to the similar implementation
> to align with classification's behavior (which seems more reasonable) as an
> improvement in future release?
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
