[
https://issues.apache.org/jira/browse/NIFI-4356?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16260245#comment-16260245
]
Henry Sowell commented on NIFI-4356:
------------------------------------
+1 on Andy’s proposal. Under the current logic the processor is significantly
handicapped in scenarios where you want to continue working with the original
HTML content. An example would be using HTML elements in attributes for things
like routing the original content based attributes extracted with this
processor.
> GetHTMLElement removes content from resulting flowfiles even when destination
> is attribute
> ------------------------------------------------------------------------------------------
>
> Key: NIFI-4356
> URL: https://issues.apache.org/jira/browse/NIFI-4356
> Project: Apache NiFi
> Issue Type: Bug
> Components: Extensions
> Affects Versions: 1.3.0
> Reporter: Andy LoPresto
> Labels: html
>
> As noted in
> [StackOverflow|https://stackoverflow.com/questions/46070288/multiple-gethtml-attributes-in-apache-nifi],
> {{GetHTMLElement}} processors cannot be chained because the success
> relationship clears the flowfile content even if the destination is an
> attribute. I propose the logic be changed to the following:
> * Destination = content
> * Replace the flowfile content for the *success* relationship
> * Maintain the content for *original*
> * Destination = attribute
> * Maintain the content for *success* and *original*
> * Add a property descriptor for the attribute name
> * Name attributes {{Attribute.1}} as in {{Extract*}} in the event that
> multiple elements are matched
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
