[ https://issues.apache.org/jira/browse/MAPREDUCE-6643?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15326554#comment-15326554 ]
Akira AJISAKA commented on MAPREDUCE-6643: ------------------------------------------ Thanks [~lewuathe] for updating the patch! Mostly looks good to me. Some comments: {code} if (start == 0) { adjStart = Math.max(0L, start - FIRST_BZIP2_BLOCK_MARKER_POSITION); {code} 1. It seems that adjStart is always zero. If that is true, we should set zero instead of using max function. {code} // Other than the fist of file, the marker size is 6 bytes. {code} 2. fist -> first 3. Would you fix the checkstyle warning? {code} // corner case when we have byte alignment and position of stream are same int[] lengths = {471507, 473608}; int[] numSplits = {218, 110}; for (int i = 0; i < 2; i++) { verifyPartitions(lengths[i], numSplits[i], file, codec, conf); } {code} 4. (nit) The following code is simpler for me: {code} // corner case when we have byte alignment and position of stream are same verifyPartition(471507, 218, file, codec, conf); verifyPartition(473608, 110, file, codec, conf); {code} > org.apache.hadoop.mapred.TestTextInputFormat.testSplitableCodecs failed > ----------------------------------------------------------------------- > > Key: MAPREDUCE-6643 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-6643 > Project: Hadoop Map/Reduce > Issue Type: Bug > Reporter: Haibo Chen > Assignee: Kai Sasaki > Attachments: MAPREDUCE-6643.01.patch, MAPREDUCE-6643.02.patch, > MAPREDUCE-6643.03.patch > > > Unit test TestTextInputFormat.testSplitableCodecs() failed when the seed is > 1313094493. > Stacktrace > java.lang.AssertionError: Key in multiple partitions. > at org.junit.Assert.fail(Assert.java:88) > at org.junit.Assert.assertTrue(Assert.java:41) > at org.junit.Assert.assertFalse(Assert.java:64) > at > org.apache.hadoop.mapred.TestTextInputFormat.testSplitableCodecs(TestTextInputFormat.java:223) -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: mapreduce-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: mapreduce-issues-h...@hadoop.apache.org