[
https://issues.apache.org/jira/browse/CALCITE-7448?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18067586#comment-18067586
]
Zhen Chen commented on CALCITE-7448:
------------------------------------
Indeed, as you said, this will definitely cause ambiguity. Do you have any good
ideas at the moment?
> Support colon-based field/item access syntax
> --------------------------------------------
>
> Key: CALCITE-7448
> URL: https://issues.apache.org/jira/browse/CALCITE-7448
> Project: Calcite
> Issue Type: New Feature
> Components: core
> Affects Versions: 1.41.0
> Reporter: Tamas Mate
> Priority: Major
> Labels: parser
>
> Implement support for colon field access syntax in the parser, for example:
> {code:java}
> v:field
> v:field.subfield
> v:['field name']
> arr[1]:field{code}
> Several systems (e.g. Snowflake, Databricks) support this kind of access, and
> with recent VARIANT support in Parquet it seems increasingly relevant for
> interoperability.
> One challenge is that ":" is already used in JSON constructor syntax, for
> example:
> {code:java}
> JSON_OBJECT('k': v)
> JSON_OBJECTAGG(k: v){code}
> The implementation needs to define how these cases interact, whether through
> disambiguation, dialect-specific behavior, or parser configuration.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
