[ 
https://issues.apache.org/jira/browse/NIFI-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Koji Kawamura resolved NIFI-6271.
---------------------------------
       Resolution: Fixed
    Fix Version/s: 1.10.0

Thanks for your contribution [~hondawei]! I've added 'Contributor' role to your 
Jira account, so that you can assign yourself to NiFi Jira tickets.

> ExecuteSQL incoming flowfile attributes not copied into output flowfiles when 
> Output Batch Size is set
> ------------------------------------------------------------------------------------------------------
>
>                 Key: NIFI-6271
>                 URL: https://issues.apache.org/jira/browse/NIFI-6271
>             Project: Apache NiFi
>          Issue Type: Bug
>          Components: Core Framework
>    Affects Versions: 1.9.2
>            Reporter: Arnaud Rivero
>            Assignee: HondaWei
>            Priority: Major
>              Labels: easyfix, features, usability
>             Fix For: 1.10.0
>
>   Original Estimate: 0.5h
>          Time Spent: 2h
>  Remaining Estimate: 0h
>
> When using the executeSQL and executeSQLRecord processors, we can use input 
> flowfiles with a certain number of attributes. If we don't set the Output 
> Batch Size, all these attributes are copied to the output flowfile. However, 
> if we set it, only the flowfiles from the first batch will have the 
> attributes copied to. The flowfiles in the following batches will only have 
> the default attributes. 
> h2. Root cause
> In the source code of the method _onTrigger_ in the class 
> _AbstractExecuteSQL,_ we have the following piece of code that is supposed to 
> create an output flowfile and copy the original attributes into it: 
> {code:java}
> FlowFile resultSetFF;
> if (fileToProcess == null) {
>   resultSetFF = session.create();
> } else {
>   resultSetFF = session.create(fileToProcess);
>   resultSetFF = session.putAllAttributes(resultSetFF, 
> fileToProcess.getAttributes());
> }
> {code}
> However the fix for the issue NIFI-6040 introduced this snippet way below in 
> the same method: 
>  
> {code:java}
> // If we've reached the batch size, send out the flow files
> if (outputBatchSize > 0 && resultSetFlowFiles.size() >= outputBatchSize) {
>   session.transfer(resultSetFlowFiles, REL_SUCCESS);
>   // Need to remove the original input file if it exists
>   if (fileToProcess != null) {
>     session.remove(fileToProcess);
>     fileToProcess = null;
>   }
>   session.commit();
>   resultSetFlowFiles.clear();
> }
> {code}
> As you can see, it sets the variable fileToProcess to null, preventing the 
> flowfiles in the next batch to copy its attributes
>  
> h2.  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to