[
https://issues.apache.org/jira/browse/NIFI-844?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14703450#comment-14703450
]
Mark Payne commented on NIFI-844:
---------------------------------
Sean,
The GenericData.toString(Object) does have JavaDocs indicating that it will
convert the record into JSON. I realize now, after your comment, that I updated
the code in only one place to use this - in one place the patch uses
GenericData.toString(Object) and in the other place it uses
GenericRecord.toString() - I will update to replace GenericRecord.toString() to
use the GenericData.toString(Object), since it is documented.
I certainly believe that a processor to convert between two Avro schemas is
useful, and that's actually mentioned as #3 of the "First-class Avro Support"
Feature proposal
(https://cwiki.apache.org/confluence/display/NIFI/First-class+Avro+Support).
What I'm looking for here is a JSON document that has exactly the same
hierarchy as the associated Avro record. The specific use-case that I have in
mind is when using ExecuteSQL, the results of the query are output as an Avro
record(s). I want to then convert that into a JSON record that is essentially
key/value pairs.
So if I have a database that has a table named PERSONS that has two columns: ID
and NAME, and i do a "SELECT ID as id, NAME AS name FROM PERSONS" I want a JSON
record to come back in the form:
{code}
{
"id": 8,
"name": "Mark Payne"
}
{code}
I *think* what you are suggesting will result in far more complicated JSON that
does not match this schema, no?
> Create Processor to convert Avro to JSON
> ----------------------------------------
>
> Key: NIFI-844
> URL: https://issues.apache.org/jira/browse/NIFI-844
> Project: Apache NiFi
> Issue Type: Task
> Components: Extensions
> Reporter: Mark Payne
> Assignee: Mark Payne
> Fix For: 0.3.0
>
> Attachments:
> 0001-NIFI-844-Initial-import-of-ConvertAvroToJSON-process.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
