[
https://issues.apache.org/jira/browse/GRIFFIN-302?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16981610#comment-16981610
]
Chitral Verma edited comment on GRIFFIN-302 at 11/25/19 2:38 PM:
-----------------------------------------------------------------
I'd like to work on this.
[~wankunde], [~wankun] What do you think of this? I can start creating sub
tasks .
was (Author: chitralverma):
I'd like to work on this.
[~wankunde], [~wankun] What do you think of this?
> Standardize Data Connectors
> ---------------------------
>
> Key: GRIFFIN-302
> URL: https://issues.apache.org/jira/browse/GRIFFIN-302
> Project: Griffin
> Issue Type: Improvement
> Reporter: Chitral Verma
> Priority: Major
> Labels: features
>
> Currently, the Data Connectors have a hierarchy - Batch and Streaming.
> BatchDataConnector and StreamingDataConnector are further extended in
> individual classes for each distinct data source (example: MySql, Avro,
> ElasticSearch etc). This hierarchy has lead to a lot of duplicated code
> across individual Batch Sources and DataConnectors in general.
> I propose here a different hierarchy based on type of DataSource (example
> FileBased: see
> [GRIFFIN-297|https://issues.apache.org/jira/browse/GRIFFIN-297], JDBCBased
> etc). Additionally, we can have certain user facing abstractions, which the
> end users can extend to implement their own DataConnectors and make them
> available via `spark.jars` property.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
