[
https://issues.apache.org/jira/browse/BEAM-4135?focusedWorklogId=92747&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-92747
]
ASF GitHub Bot logged work on BEAM-4135:
----------------------------------------
Author: ASF GitHub Bot
Created on: 19/Apr/18 17:47
Start Date: 19/Apr/18 17:47
Worklog Time Spent: 10m
Work Description: huygaa11 commented on a change in pull request #5177:
[BEAM-4135] Stop taking the whole result in WatermarkManager
URL: https://github.com/apache/beam/pull/5177#discussion_r182829877
##########
File path:
runners/direct-java/src/main/java/org/apache/beam/runners/direct/WatermarkManager.java
##########
@@ -965,15 +967,19 @@ private void applyNUpdates(int numUpdates) {
}
private void applyPendingUpdate(PendingWatermarkUpdate<ExecutableT> pending)
{
- CommittedResult<ExecutableT> result = pending.getResult();
- ExecutableT transform = result.getExecutable();
+ ExecutableT transform = pending.getExecutable();
Review comment:
Are transform and executable are interchangeable term? Or, is transform a
subset of executable?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 92747)
Time Spent: 50m (was: 40m)
> Remove Use of Java SDK Types in the DirectRunner "engine"
> ---------------------------------------------------------
>
> Key: BEAM-4135
> URL: https://issues.apache.org/jira/browse/BEAM-4135
> Project: Beam
> Issue Type: New Feature
> Components: runner-direct
> Reporter: Thomas Groh
> Assignee: Thomas Groh
> Priority: Major
> Labels: portability
> Time Spent: 50m
> Remaining Estimate: 0h
>
> The "engine" consists of the components which determine where to schedule
> work and route it to the appropriate processors, such as WatermarkManager,
> DirectBundleProcessor, and associated.
>
> These engine components never inspect the actual characteristics of the
> packaged work (e.g. the PCollection is a token, rather than a rich object),
> so they should not require use of a PCollection directly - instead, they can
> be generic.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
