You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@beam.apache.org by "Innocent (JIRA)" <ji...@apache.org> on 2017/06/14 02:44:00 UTC

[jira] [Commented] (BEAM-765) TriggerStateMachine does not need to be separate from OnceTriggerStateMachine

    [ https://issues.apache.org/jira/browse/BEAM-765?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16048632#comment-16048632 ] 

Innocent commented on BEAM-765:
-------------------------------

Hi Kenneth, would you please add more details? I noticed that some classes extends TriggerStateMachine and other extends OnceTriggerStateMachine. In both cases everything is consistent with the associate semantic.(OnceTrigger for Triggers which fire at most once). Is the intention to make all implementation classes derive from TriggerStateMachine and use another mechanism to enforce the OnceTrigger Semantic?

> TriggerStateMachine does not need to be separate from OnceTriggerStateMachine
> -----------------------------------------------------------------------------
>
>                 Key: BEAM-765
>                 URL: https://issues.apache.org/jira/browse/BEAM-765
>             Project: Beam
>          Issue Type: Improvement
>          Components: runner-core
>            Reporter: Kenneth Knowles
>            Assignee: Innocent
>            Priority: Trivial
>
> In runners-core, the state machine implementation of triggers does not need to have the fine-grained type-like enforcement of whether a trigger is a {{OnceTrigger}} or not. It may simplify the code.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)