[
https://issues.apache.org/jira/browse/BEAM-9363?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17135474#comment-17135474
]
duli commented on BEAM-9363:
----------------------------
[~amaliujia], thanks for your reply.
If we use TVF for stream-stream join, it means two bounded dataset transfromed
by streaming with window and then join each other, or two stream join each
other and then operate the result on window? Currently, stream-stream join
with window is the latter in Flink, but we want the former from syntax
perspective.
How do you think?
> BeamSQL windowing as TVF
> ------------------------
>
> Key: BEAM-9363
> URL: https://issues.apache.org/jira/browse/BEAM-9363
> Project: Beam
> Issue Type: New Feature
> Components: dsl-sql, dsl-sql-zetasql
> Reporter: Rui Wang
> Assignee: Rui Wang
> Priority: P2
> Time Spent: 10h 10m
> Remaining Estimate: 0h
>
> This Jira tracks the implementation for
> https://s.apache.org/streaming-beam-sql
> TVF is table-valued function, which is a SQL feature that produce a table as
> function's output.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
