[
https://issues.apache.org/jira/browse/FLINK-9864?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jark Wu closed FLINK-9864.
--------------------------
Resolution: Fixed
I think we can close this issue now. Because we have more powerful formats now
(e.g. JSON and CSV are able to parse different timestamp strings), and powerful
function (e.g. TO_TIMESTAMP, DATE_FORMAT, FROM_UNIXTIMESTAMP).
> Make timestamp extraction more flexible in SQL Client
> -----------------------------------------------------
>
> Key: FLINK-9864
> URL: https://issues.apache.org/jira/browse/FLINK-9864
> Project: Flink
> Issue Type: New Feature
> Components: Table SQL / Client
> Reporter: Timo Walther
> Priority: Major
> Labels: auto-unassigned
>
> Currently, a timestamp must be in the top-level of a possibly nested row and
> must have a certain format. We should think about making this more flexible
> to cover most of the use cases.
> A first solution could be to allow a DOT operator syntax:
> {{myfield.nested.timestamp}}
> Other cases might be:
> - The time could also be split into several field
> - Or needs to be parsed using a [date format
> syntax|https://ci.apache.org/projects/flink/flink-docs-master/dev/table/sql.html#date-format-specifier].
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
