[
https://issues.apache.org/jira/browse/NIFI-911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14723716#comment-14723716
]
Matt Gilman commented on NIFI-911:
----------------------------------
[~markap14] - Wouldn't this have been the case when the default value was still
(.*) as well?
I ran through a multi line case before making any changes and it resulted in
"Hello WorldHello World" on each line (even with ENTIRE_TEXT as the Evaluation
Mode - due to the line terminators not matching). The goal of this ticket was
to prevent the double replacement by default. We can certainly make the change
your suggesting, in fact I think it makes sense to. Just was outside the scope
of the ticket as stated. Should I update the patch?
> ReplaceText, with default configuration, will cause the replacement to be
> added twice
> -------------------------------------------------------------------------------------
>
> Key: NIFI-911
> URL: https://issues.apache.org/jira/browse/NIFI-911
> Project: Apache NiFi
> Issue Type: Improvement
> Components: Extensions
> Reporter: Mark Payne
> Assignee: Matt Gilman
> Labels: beginner, newbie
> Fix For: 0.3.0
>
> Attachments: 0001-NIFI-911.patch
>
>
> Currently, the default value for the Regular Expression property is (.\*)
> which is intended to match any text. However, due to the way that the regular
> expression is interpreted, it first matches the entire content of the
> FlowFile and then again matches 0 characters at the end of the content
> (because .* will match 0 or more characters of any character).
> This results in the replacement value being added to the FlowFile twice. For
> example, if the replacement value is "Hello World", the output FlowFile will
> contain the text "Hello WorldHello World".
> We should instead change the regex pattern to (^.*$) to in order to ensure
> that we do not duplicate the output
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
