[
https://issues.apache.org/jira/browse/NIFI-1073?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15005397#comment-15005397
]
Tony Kurc commented on NIFI-1073:
---------------------------------
[~bbende] funny that you should mention that. I was meandering around the code
and somewhere, I don't remember where, but I think it was [~markap14]'s code
there was a big comment block with what I think was a bufferedwriter that had
its underlying outputstream closed before the bufferedwriter closed to prevent
the flush in the close from taking place. in any case, it prompted me to look
at the javadoc for bufferedwriter and I shook my fist as the javadoc said
"Closes the stream, flushing it first", and felt that "the stream" was not
strongly enough defined in the javadoc. Anyhow, opened the code for
BufferedWriter in the openjdk and it is clearly shutting the underlying stream
passed into the constructor.
Long story short, if someone handed me the contract for BufferedReader and
InputStreamReader as specified in the javadocs, I'd ask them to clarify what
"the stream" was before I signed it. That being said, the code checks out.
> Fix issues found by coverity static analysis
> ---------------------------------------------
>
> Key: NIFI-1073
> URL: https://issues.apache.org/jira/browse/NIFI-1073
> Project: Apache NiFi
> Issue Type: Bug
> Components: Core Framework, Extensions
> Affects Versions: 0.3.0
> Reporter: Tony Kurc
> Assignee: Tony Kurc
> Fix For: 0.4.0
>
> Attachments: NIFI-1073.000.patch
>
>
> Ran static analysis on nifi using Coverity Scan. Fix high priority issues.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
