[jira] [Created] (FLINK-3668) Potential null deference in HadoopInputFormatBase#createInputSplits()

[Shang Yuanchun (Jira)]

Ted Yu created FLINK-3668:
-----------------------------

             Summary: Potential null deference in HadoopInputFormatBase#createInputSplits()
                 Key: FLINK-3668
                 URL: https://issues.apache.org/jira/browse/FLINK-3668
             Project: Flink
          Issue Type: Bug
            Reporter: Ted Yu
            Priority: Minor


Here is the code:
{code}
    List<org.apache.hadoop.mapreduce.InputSplit> splits;
    try {
      splits = this.mapreduceInputFormat.getSplits(jobContext);
    } catch (InterruptedException e) {
      throw new IOException("Could not get Splits.", e);
    }
    HadoopInputSplit[] hadoopInputSplits = new HadoopInputSplit[splits.size()];
{code}
If InterruptedException is caught, splits would be null.
Yet, the next line accesses splits.size() without null check.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)