[
https://issues.apache.org/jira/browse/NIFI-8137?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Chris Sampson updated NIFI-8137:
--------------------------------
Description:
Record Path functions to:
* unescape JSON contained within a String field and put the resulting Record
fields into the data, e.g. UpdateRecord for {{/}} =>
{{unescapeJson(/stringified_json)}} would unescape JSON String in
{{/stringified_json}} and put the resulting Object fields into the top-level of
the Record
* collect one or more Record fields into an Object that is put into a single
stringified JSON field, e.g. UpdateRecord for {{/stringified_json}} =>
{{escapeJson( / )}} would do the reverse of the above
These would be equivalent to the existing Expression Language functions, but
mean there was no need to split records into individual flowfiles, extract the
field from the content into an attribute, unescape the JSON then replace the
text back into the content and re-merge all the records (which is fidly and
resource expensive).
was:
Record Path functions to:
* unescape JSON contained within a String field and put the resulting Record
fields into the data, e.g. UpdateRecord for {{/}} =>
{{unescapeJson(/stringified_json)}} would unescape JSON String in
{{/stringified_json}} and put the resulting Object fields into the top-level of
the Record
* collect one or more Record fields into an Object that is put into a single
stringified JSON field, e.g. UpdateRecord for {{/stringified_json}} =>
{{escapeJson(/)}} would do the reverse of the above
These would be equivalent to the existing Expression Language functions, but
mean there was no need to split records into individual flowfiles, extract the
field from the content into an attribute, unescape the JSON then replace the
text back into the content and re-merge all the records (which is fidly and
resource expensive).
> Record Path functions to un/escape JSON
> ---------------------------------------
>
> Key: NIFI-8137
> URL: https://issues.apache.org/jira/browse/NIFI-8137
> Project: Apache NiFi
> Issue Type: Improvement
> Affects Versions: 1.11.4
> Reporter: Chris Sampson
> Priority: Major
>
> Record Path functions to:
> * unescape JSON contained within a String field and put the resulting Record
> fields into the data, e.g. UpdateRecord for {{/}} =>
> {{unescapeJson(/stringified_json)}} would unescape JSON String in
> {{/stringified_json}} and put the resulting Object fields into the top-level
> of the Record
> * collect one or more Record fields into an Object that is put into a single
> stringified JSON field, e.g. UpdateRecord for {{/stringified_json}} =>
> {{escapeJson( / )}} would do the reverse of the above
> These would be equivalent to the existing Expression Language functions, but
> mean there was no need to split records into individual flowfiles, extract
> the field from the content into an attribute, unescape the JSON then replace
> the text back into the content and re-merge all the records (which is fidly
> and resource expensive).
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
