[jira] [Commented] (GROOVY-7734) ClassNode tryFindPossibleMethod only defined for TupleExpression

["Paul King (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/GROOVY-7734?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15173323#comment-15173323 ] 

Paul King commented on GROOVY-7734:
-----------------------------------

Well, it is a public method (in the java visibility sense) so we'd not normally delete such a method outright. There should be some analysis done. Even though we no longer use the method, do we expect others to use it? Is the existing behavior incorrect or incomplete? But are others potentially relying on that behavior anyway? It is in the org.codehaus.groovy package (some what internal informally speaking) so we have a tiny bit of wiggle room. But to cut a long story short - if we expect others to use it, it should have a unit test (and perhaps some javadoc) and if not, it should be deprecated. I suspect at the time we might not have realized that the last usage was gone and weren't vigilant checking checkstyle reports..

> ClassNode tryFindPossibleMethod only defined for TupleExpression
> ----------------------------------------------------------------
>
>                 Key: GROOVY-7734
>                 URL: https://issues.apache.org/jira/browse/GROOVY-7734
>             Project: Groovy
>          Issue Type: Bug
>    Affects Versions: 2.4.5
>            Reporter: Anton
>
> As seen here https://github.com/apache/groovy/blob/master/src/main/org/codehaus/groovy/ast/ClassNode.java#L1257-L1258 if {{arguments}} is not a {{TupleExpression}} the method just returns.
> Shouldn't this be defined for other expression types as well?



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)