[
https://issues.apache.org/jira/browse/NIFI-13591?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Anders updated NIFI-13591:
--------------------------
Description:
JsonRecordSetWriter has an option for Output Grouping. Either array or "One
Line Per Object" (jsonlines), where the jsonlines output is a fairly common
usecase.
The JSLTTransformJSON processor has a setting for Transformation Strategy:
"Each JSON Object"
This works on each json object in an array, and works as expected.
"Entire Flowfile"
This sort-of works with the jsonlines/NDJSON flowfiles, and each json object
gets transformed. However, it strips the newline, so the flowfile just becomes
one long line with all the objects one after the other.
{code:java}
{..}
{..}
{..}
{..}
{code}
becomes
{code:java}
{..} {..} {..} {..}
{code}
It would be useful if the newline was presereved. Either by changing the
"Entire FlowFile" behaviour, or adding some new setting.
Attached a simple template.
was:
JsonRecordSetWriter has an option for Output Grouping. Either array or "One
Line Per Object" (jsonlines), where the jsonlines output is a fairly common
usecase.
The JSLTTransformJSON processor has a setting for Transformation Strategy:
"Each JSON Object"
This works on each json object in an array, and works as expected.
"Entire Flowfile"
This sort-of works with the jsonlines/NDJSON flowfiles, and each json object
gets transformed. However, it strips the newline, so the flowfile just becomes
one long line with all the objects one after the other.
{code:java}
{..}
{..}
{..}
{..}
{code}
becomes
{code:java}
{..} {..} {..} {..}
{code}
It would be useful if the newline was presereved. Either by changing the
"Entire FlowFile" behaviour, or adding some new setting.
> Honor newline in JSLTTransformJSON
> ----------------------------------
>
> Key: NIFI-13591
> URL: https://issues.apache.org/jira/browse/NIFI-13591
> Project: Apache NiFi
> Issue Type: Improvement
> Affects Versions: 1.27.0
> Reporter: Anders
> Priority: Minor
> Attachments: JSLT.json
>
>
> JsonRecordSetWriter has an option for Output Grouping. Either array or "One
> Line Per Object" (jsonlines), where the jsonlines output is a fairly common
> usecase.
> The JSLTTransformJSON processor has a setting for Transformation Strategy:
> "Each JSON Object"
> This works on each json object in an array, and works as expected.
> "Entire Flowfile"
> This sort-of works with the jsonlines/NDJSON flowfiles, and each json object
> gets transformed. However, it strips the newline, so the flowfile just
> becomes one long line with all the objects one after the other.
> {code:java}
> {..}
> {..}
> {..}
> {..}
> {code}
> becomes
> {code:java}
> {..} {..} {..} {..}
> {code}
> It would be useful if the newline was presereved. Either by changing the
> "Entire FlowFile" behaviour, or adding some new setting.
> Attached a simple template.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
