Github user pvillard31 commented on a diff in the pull request:
https://github.com/apache/nifi/pull/2037#discussion_r133908650
--- Diff:
nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/main/resources/docs/org.apache.nifi.processors.standard.ForkRecord/additionalDetails.html
---
@@ -0,0 +1,365 @@
+<!DOCTYPE html>
+<html lang="en">
+ <!--
+ Licensed to the Apache Software Foundation (ASF) under one or more
+ contributor license agreements. See the NOTICE file distributed with
+ this work for additional information regarding copyright ownership.
+ The ASF licenses this file to You under the Apache License, Version
2.0
+ (the "License"); you may not use this file except in compliance with
+ the License. You may obtain a copy of the License at
+ http://www.apache.org/licenses/LICENSE-2.0
+ Unless required by applicable law or agreed to in writing, software
+ distributed under the License is distributed on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied.
+ See the License for the specific language governing permissions and
+ limitations under the License.
+ -->
+ <head>
+ <meta charset="utf-8" />
+ <title>ForkRecord</title>
+
+ <link rel="stylesheet"
href="../../../../../css/component-usage.css" type="text/css" />
+ </head>
+
+ <body>
+ <p>
+ ForkRecord allows the user to fork a record into
multiple records. To do that, the user must specify
+ a <a
href="../../../../../html/record-path-guide.html">RecordPath</a> pointing to a
field of type
+ ARRAY containing RECORD elements. The generated flow
file will contain the records from the specified
+ array. It is also possible to add in each record all
the fields of the parent records from the root
+ level to the record element being forked. However it
supposes the fields to add are defined in the
+ schema of the Record Writer controller service.
+ </p>
+
+ <h2>Examples</h2>
+
+ <p>
+ To better understand how this Processor works, we will
lay out a few examples. For the sake of these examples, let's assume that our
input
+ data is JSON formatted and looks like this:
+ </p>
+
+<code>
+<pre>
+[{
+ "id": 1,
+ "name": "John Doe",
+ "address": "123 My Street",
+ "city": "My City",
+ "state": "MS",
+ "zipCode": "11111",
+ "country": "USA",
+ "accounts": [{
+ "id": 42,
+ "balance": 4750.89
+ }, {
+ "id": 43,
+ "balance": 48212.38
+ }]
+},
+{
+ "id": 2,
+ "name": "Jane Doe",
+ "address": "345 My Street",
+ "city": "Her City",
+ "state": "NY",
+ "zipCode": "22222",
+ "country": "USA",
+ "accounts": [{
+ "id": 45,
+ "balance": 6578.45
+ }, {
+ "id": 46,
+ "balance": 34567.21
+ }]
+}]
+</pre>
+</code>
+
+
+ <h3>Example 1 - Fork without parent fields</h3>
+
+ <p>
+ For this case, we want to create one record per
<code>account</code> and we don't care about
+ the other fields. We'll set the Record path property to
<code>/accounts</code>. The resulting
+ flow file will contain 4 records and will look like
(assuming the Record Writer schema is
+ correctly set):
+ </p>
+
+<code>
+<pre>
+[{
+ "id": 42,
+ "balance": 4750.89
+}, {
+ "id": 43,
+ "balance": 48212.38
+}, {
+ "id": 45,
+ "balance": 6578.45
+}, {
+ "id": 46,
+ "balance": 34567.21
+}]
+</pre>
+</code>
+
+
+ <h3>Example 2 - Fork with parent fields</h3>
+
+ <p>
+ Now, if we set the property "Include parent fields" to
true, this will recursively include
--- End diff --
In the use case I have that motivated this processor, being able to
"extract" each element and "merge" it with the parent fields is really useful.
I agree that this could be achieved with the UpdateRecord processor (and the
addition of [NIFI-4270](https://issues.apache.org/jira/browse/NIFI-4270)) but
it'd require one processor per type of record and can represent a lot of
configuration (in my case, I've hundreds of types of records to "extract").
What about an intermediary solution: the include parent fields property
does not really make sense since it is up to the writing schema to define what
should be in the output. I propose to rename this property in something like
"Fork strategy" with two possible values:
- Split: that would create the output you propose. It won't change the
structure but will create 1-element array.
- Extract: that would create the current output with the merging of the
parent fields (assuming the output schema does contain the parent fields).
What do you think?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
