[
https://issues.apache.org/jira/browse/TINKERPOP-3047?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17942035#comment-17942035
]
ASF GitHub Bot commented on TINKERPOP-3047:
-------------------------------------------
Cole-Greer commented on PR #3091:
URL: https://github.com/apache/tinkerpop/pull/3091#issuecomment-2787916982
> > It might be nice to have some (feature?) tests which T and Direction
remain valid map keys and have precedence over string keys.
>
> not sure feature tests will work for this. bytecode and the translator
will convert/normalize them to the long form so it won't actually get tested.
might need a new class of pure "script" tests? either that or make a
configuration of the translators that will use the short form? JIRA?
I didn't consider that, it can be left out of scope here. I'm not sure that
such a test case is enough to motivate a new class of tests or translator
config.
> Grammar does not parse keywords into Map keys
> ---------------------------------------------
>
> Key: TINKERPOP-3047
> URL: https://issues.apache.org/jira/browse/TINKERPOP-3047
> Project: TinkerPop
> Issue Type: Bug
> Components: language
> Affects Versions: 3.7.1
> Reporter: Stephen Mallette
> Priority: Critical
>
> {{[keys: ["a","b"]}} won't work because "keys" ends up being parsed to
> {{Column.keys}}. another issue at play is the use of parens to wrap certain
> key definitions but not others. it doesn't feel consistent. like, it will
> work for {{T}} values but not for something like "edges" which is just a
> keyword token. Not sure it's wrong but it requires some examination.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
