[
https://issues.apache.org/jira/browse/BEAM-8456?focusedWorklogId=333724&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-333724
]
ASF GitHub Bot logged work on BEAM-8456:
----------------------------------------
Author: ASF GitHub Bot
Created on: 24/Oct/19 21:31
Start Date: 24/Oct/19 21:31
Worklog Time Spent: 10m
Work Description: kennknowles commented on pull request #9849:
[BEAM-8456] Add pipeline option to have Data Catalog truncate sub-millisecond
precision
URL: https://github.com/apache/beam/pull/9849#discussion_r338801230
##########
File path:
sdks/java/extensions/sql/datacatalog/src/main/java/org/apache/beam/sdk/extensions/sql/meta/provider/datacatalog/DataCatalogPipelineOptions.java
##########
@@ -32,4 +32,12 @@
String getDataCatalogEndpoint();
void setDataCatalogEndpoint(String dataCatalogEndpoint);
+
+ /** Whether to truncate timestamps in tables described by Data Catalog. */
Review comment:
In this case, it only controls BigQuery. We do know that other GCP products
have high precision timestamps. So it really is dependent on the nature of the
data source. Probably the best thing would be to control it per occurrence of
the table in the query. But that is a lot harder.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 333724)
Time Spent: 2h (was: 1h 50m)
> BigQuery to Beam SQL timestamp has the wrong default: truncation makes the
> most sense
> -------------------------------------------------------------------------------------
>
> Key: BEAM-8456
> URL: https://issues.apache.org/jira/browse/BEAM-8456
> Project: Beam
> Issue Type: Improvement
> Components: dsl-sql
> Reporter: Kenneth Knowles
> Assignee: Kenneth Knowles
> Priority: Major
> Time Spent: 2h
> Remaining Estimate: 0h
>
> Most of the time, a user reading a timestamp from BigQuery with
> higher-than-millisecond precision timestamps may not even realize that the
> data source created these high precision timestamps. They are probably
> timestamps on log entries generated by a system with higher precision.
> If they are using it with Beam SQL, which only supports millisecond
> precision, it makes sense to "just work" by default.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
