[
https://issues.apache.org/jira/browse/NIFI-4892?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16369850#comment-16369850
]
Pierre Villard commented on NIFI-4892:
--------------------------------------
Hi [~Kolpakov],
Thanks for reporting this. I just had a look and added a unit test to confirm
the issue.
It seems to be on purpose while reading the columns:
[https://github.com/super-csv/super-csv/blob/master/super-csv/src/main/java/org/supercsv/io/Tokenizer.java#L211]
But I just submitted a PR that slightly changes the way we are dealing with
individual invalid lines. This way it solves the issue you reported.
> ValidateCSV Processor - no doublequote escaping in "invalid" output
> -------------------------------------------------------------------
>
> Key: NIFI-4892
> URL: https://issues.apache.org/jira/browse/NIFI-4892
> Project: Apache NiFi
> Issue Type: Bug
> Components: Extensions
> Affects Versions: 1.4.0, 1.5.0
> Reporter: Eugeny Kolpakov
> Assignee: Pierre Villard
> Priority: Major
> Attachments: ValidateCSV_LineByLine_EscapingBug.xml
>
>
> ValidateCSV processor fails to preserve field escaping for flow files routed
> to "invalid" relation when "Line by Line" validation strategy is used.
>
> Minimal sample flow attached.
>
> Observed in 1.4.0, but 1.5.0 might be affected as well - I'll update this
> ticket after upgrading to 1.5.0 (unless someone could do that earlier)
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
