[
https://issues.apache.org/jira/browse/NIFI-8269?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18044650#comment-18044650
]
Pierre Villard commented on NIFI-8269:
--------------------------------------
[~mattyb149] - the way you have configured your XML Reader, the inferred data
would look like:
{code:java}
[ {
"sessionTask" : [ {
"taskId" : 1,
"targetTable" : "S_WDMD_WD_L_ADDITIONAL_INFO",
"column" : [ {
"name" : "email",
"dataType" : "string"
}, {
"name" : "phone",
"dataType" : "string"
} ]
}, {
"taskId" : 1,
"targetTable" : "S_WDMD_WD_L_IDENTIFIER",
"column" : [ {
"name" : "id",
"dataType" : "integer"
}, {
"name" : "fullName",
"dataType" : "string"
} ]
} ]
} ] {code}
So there is no "workflow" level anymore (it is removed by the XML reader) and
that is why the path "/workflow/sessionTask" does not work but "//sessionTask"
works.
> Improve ForkRecord to inherit schema from the array being forked
> ----------------------------------------------------------------
>
> Key: NIFI-8269
> URL: https://issues.apache.org/jira/browse/NIFI-8269
> Project: Apache NiFi
> Issue Type: Improvement
> Reporter: Pierre Villard
> Assignee: Pierre Villard
> Priority: Major
> Attachments: NIFI-8269_array_error.json
>
> Time Spent: 0.5h
> Remaining Estimate: 0h
>
> At the moment the ForkRecord processor will require the writer to be
> configured with a schema corresponding to what will be output. We could make
> things more clever to inherit the schema from the reader and change it based
> on the operation being performed.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
