[
https://issues.apache.org/jira/browse/SLING-5373?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15192280#comment-15192280
]
Konrad Windszus commented on SLING-5373:
----------------------------------------
Yes, good idea. Still to make life easier for Validator developers I am not too
sure that it is a good idea, that they must transfer all relevant information
from {{ValidationContext}} to {{ValidationResult}}. Currently it is location
and severity.
With having an additional constructor lacking the severity, it is very easy to
forget to transfer the {{severity}} into the failure.
One optimization possibility would be two different {{ValidationResult}} types.
One is for the consumer, containing all information, the other one contains
only the validator-specific information. So the Validator developers would only
need to fill the latter and the {{ValidationServiceImpl}} would transform that
into the former. But then it is no longer possible for a {{Validator}} to
overwrite e.g. the severity.
Another possibility is to provide a constructor for {{DefaultValidationResult}}
taking the whole {{ValidationContext}}.
Any opinions?
> Allow different severities per validator
> ----------------------------------------
>
> Key: SLING-5373
> URL: https://issues.apache.org/jira/browse/SLING-5373
> Project: Sling
> Issue Type: Bug
> Components: Validation
> Reporter: Konrad Windszus
> Fix For: Validation 1.0.0
>
>
> It would be good to be able to define a severity either on the validator
> itself or within the validation model. That would allow to later on
> differentiate the handling on how to treat those violations.
> Depending on the violation severity one could decide to either throw an
> exception or just expose the violation failure message for informational
> purposes.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
