Github user mattyb149 commented on a diff in the pull request:
https://github.com/apache/nifi/pull/2287#discussion_r160470497
--- Diff:
nifi-nar-bundles/nifi-elasticsearch-bundle/nifi-elasticsearch-5-processors/src/main/java/org/apache/nifi/processors/elasticsearch/PutElasticsearch5.java
---
@@ -129,6 +130,12 @@
.defaultValue("100")
.expressionLanguageSupported(true)
.build();
+
+ public static final PropertyDescriptor VERSION_ATTRIBUTE = new
PropertyDescriptor.Builder()
+ .name("Version").displayName("Version Attribute")
+ .description("The name of the attribute containing the
version for each FlowFile. Only used for index operation. Forces External
Versioning")
+
.required(false).expressionLanguageSupported(false).addValidator(StandardValidators.ATTRIBUTE_KEY_VALIDATOR)
--- End diff --
What do you think about the above suggestion, changing from an attribute
name to an Expression Language supported property with a NON_EMPTY_VALIDATOR?
Personally I'd prefer it that way, even if it doesn't behave the same as
ID_ATTRIBUTE, since they are not really used the same way anyway. A regular
EL-supporting property is more flexible and you can still use a flow file
attribute by simply wrapping the name in an expression like ${myAttribute}.
If you'd like to keep the current behavior and/or discuss further, please
let me know
---
