You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Timo Walther (Jira)" <ji...@apache.org> on 2020/03/10 12:32:00 UTC
[jira] [Commented] (FLINK-15958) Fully support RAW types in the API
[ https://issues.apache.org/jira/browse/FLINK-15958?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055894#comment-17055894 ]
Timo Walther commented on FLINK-15958:
--------------------------------------
As can be seen in the pull request, the direct API changes are minimal. I updated the JIRA issue description accordingly.
> Fully support RAW types in the API
> ----------------------------------
>
> Key: FLINK-15958
> URL: https://issues.apache.org/jira/browse/FLINK-15958
> Project: Flink
> Issue Type: Sub-task
> Components: Table SQL / API, Table SQL / Planner
> Reporter: Timo Walther
> Assignee: Timo Walther
> Priority: Major
> Labels: pull-request-available
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Currently, the Table API does not expose a way of creating RAW types without a {{DataTypeFactory}}. The reason for that is that RAW types need to be resolved at a later stage. This is similar to user-defined types that need to be resolved from the catalog.
> API changes:
> - Introduction of {{AbstractDataType}} as the highest type of resolved and unresolved data types
> - Introduction of {{DataTypes.of()}} for classes and names and {{DataTypes.RAW(Class)}} for unresolved types
> - Helper classes and methods for constructing unresolved types type-safe
> We also need better support in the Blink planner for that.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)