[
https://issues.apache.org/jira/browse/BEAM-11838?focusedWorklogId=601943&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-601943
]
ASF GitHub Bot logged work on BEAM-11838:
-----------------------------------------
Author: ASF GitHub Bot
Created on: 25/May/21 20:47
Start Date: 25/May/21 20:47
Worklog Time Spent: 10m
Work Description: chamikaramj commented on pull request #14829:
URL: https://github.com/apache/beam/pull/14829#issuecomment-848249243
Yes, building and releasing a new expansion service jar has been the usual
pattern so far. We usually try to group related multi-language transforms to
use a single expansion service. For example:
https://github.com/apache/beam/tree/master/sdks/java/extensions/sql/expansion-service
But @ihji has been doing a lot of work related to dependencies recently and
the hope is that the transform will be able to specify multiple dependencies
that have to be staged instead of defining and specifying a single shaded jar.
Heejong, do we have API and implementation in place to specify individual
dependencies to be staged instead of a single shaded expansion service jar ?
--
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: 601943)
Time Spent: 8h (was: 7h 50m)
> An x-lang wrapper for DebeziumIO
> --------------------------------
>
> Key: BEAM-11838
> URL: https://issues.apache.org/jira/browse/BEAM-11838
> Project: Beam
> Issue Type: Bug
> Components: cross-language, io-java-debezium
> Reporter: Pablo Estrada
> Assignee: Benjamin Gonzalez
> Priority: P2
> Time Spent: 8h
> Remaining Estimate: 0h
>
> DebeziumIO outputs data in JSON format by default, so it's a perfect
> candidate to be made available as a xlang transform.
> The DebeziumIO module provides a base to implement this, but for an
> effective, easy-to-use connector in Python, we will need the following things:
> * A new Java module that includes the per-database dependencies (Debezium's
> MySQL, Postgres, DB2, Oracle connector dependencies). This is necessary so
> that users can simply specify a database on the Python side, and the Java
> side will automatically provide the dependency.
> * A thorough review of the current default JSON output from DebeziumIO
> * A set of parameters to be received by the Python wrapper. Maybe we'll use
> the same format as the existing DebeziumIO (i.e. mandatory parameters: host,
> user, password; and keyword parameters)
>
> A short guide:
> [https://beam.apache.org/documentation/programming-guide/#mulit-language-pipelines]
> An example of this kind of change: [https://github.com/apache/beam/pull/12145]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
