[CARBONDATA-2125] like% filter is giving ArrayIndexOutOfBoundException in case of table having more pages
Problem: like% filter is giving ArrayIndexOutOfBoundException in case of table having more pages Solution: In RowlevelFilter the number of rows should be filled based on the rows in a page. This closes #1909 Project: http://git-wip-us.apache.org/repos/asf/carbondata/repo Commit: http://git-wip-us.apache.org/repos/asf/carbondata/commit/50e2f2c8 Tree: http://git-wip-us.apache.org/repos/asf/carbondata/tree/50e2f2c8 Diff: http://git-wip-us.apache.org/repos/asf/carbondata/diff/50e2f2c8 Branch: refs/heads/branch-1.3 Commit: 50e2f2c8f2cc6ee4b72839b704a038666ae629ba Parents: 54b7db5 Author: dhatchayani <dhatcha.offic...@gmail.com> Authored: Fri Feb 2 10:55:19 2018 +0530 Committer: ravipesala <ravi.pes...@gmail.com> Committed: Sat Feb 3 22:55:36 2018 +0530 ---------------------------------------------------------------------- .../executer/RowLevelFilterExecuterImpl.java | 10 ++++++-- .../filter/executer/TrueFilterExecutor.java | 2 +- .../filterexpr/FilterProcessorTestCase.scala | 25 ++++++++++++++++++++ 3 files changed, 34 insertions(+), 3 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/carbondata/blob/50e2f2c8/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/RowLevelFilterExecuterImpl.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/RowLevelFilterExecuterImpl.java b/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/RowLevelFilterExecuterImpl.java index 224a69f..89489a2 100644 --- a/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/RowLevelFilterExecuterImpl.java +++ b/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/RowLevelFilterExecuterImpl.java @@ -205,7 +205,10 @@ public class RowLevelFilterExecuterImpl implements FilterExecuter { } else { // specific for restructure case where default values need to be filled pageNumbers = blockChunkHolder.getDataBlock().numberOfPages(); - numberOfRows = new int[] { blockChunkHolder.getDataBlock().nodeSize() }; + numberOfRows = new int[pageNumbers]; + for (int i = 0; i < pageNumbers; i++) { + numberOfRows[i] = blockChunkHolder.getDataBlock().getPageRowCount(i); + } } } if (msrColEvalutorInfoList.size() > 0) { @@ -217,7 +220,10 @@ public class RowLevelFilterExecuterImpl implements FilterExecuter { } else { // specific for restructure case where default values need to be filled pageNumbers = blockChunkHolder.getDataBlock().numberOfPages(); - numberOfRows = new int[] { blockChunkHolder.getDataBlock().nodeSize() }; + numberOfRows = new int[pageNumbers]; + for (int i = 0; i < pageNumbers; i++) { + numberOfRows[i] = blockChunkHolder.getDataBlock().getPageRowCount(i); + } } } BitSetGroup bitSetGroup = new BitSetGroup(pageNumbers); http://git-wip-us.apache.org/repos/asf/carbondata/blob/50e2f2c8/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/TrueFilterExecutor.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/TrueFilterExecutor.java b/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/TrueFilterExecutor.java index 92396ae..4b3738a 100644 --- a/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/TrueFilterExecutor.java +++ b/core/src/main/java/org/apache/carbondata/core/scan/filter/executer/TrueFilterExecutor.java @@ -39,7 +39,7 @@ public class TrueFilterExecutor implements FilterExecuter { BitSetGroup group = new BitSetGroup(numberOfPages); for (int i = 0; i < numberOfPages; i++) { BitSet set = new BitSet(); - set.flip(0, blockChunkHolder.getDataBlock().nodeSize()); + set.flip(0, blockChunkHolder.getDataBlock().getPageRowCount(i)); group.setBitSet(set, i); } return group; http://git-wip-us.apache.org/repos/asf/carbondata/blob/50e2f2c8/integration/spark-common-test/src/test/scala/org/apache/carbondata/spark/testsuite/filterexpr/FilterProcessorTestCase.scala ---------------------------------------------------------------------- diff --git a/integration/spark-common-test/src/test/scala/org/apache/carbondata/spark/testsuite/filterexpr/FilterProcessorTestCase.scala b/integration/spark-common-test/src/test/scala/org/apache/carbondata/spark/testsuite/filterexpr/FilterProcessorTestCase.scala index b92b379..d54906f 100644 --- a/integration/spark-common-test/src/test/scala/org/apache/carbondata/spark/testsuite/filterexpr/FilterProcessorTestCase.scala +++ b/integration/spark-common-test/src/test/scala/org/apache/carbondata/spark/testsuite/filterexpr/FilterProcessorTestCase.scala @@ -21,22 +21,30 @@ import java.sql.Timestamp import org.apache.spark.sql.Row import org.scalatest.BeforeAndAfterAll + import org.apache.carbondata.core.constants.CarbonCommonConstants import org.apache.carbondata.core.util.CarbonProperties import org.apache.spark.sql.test.util.QueryTest +import org.apache.carbondata.spark.testsuite.datacompaction.CompactionSupportGlobalSortBigFileTest + /** * Test Class for filter expression query on String datatypes * */ class FilterProcessorTestCase extends QueryTest with BeforeAndAfterAll { + val file1 = resourcesPath + "/filter/file1.csv" + override def beforeAll { sql("drop table if exists filtertestTables") sql("drop table if exists filtertestTablesWithDecimal") sql("drop table if exists filtertestTablesWithNull") sql("drop table if exists filterTimestampDataType") sql("drop table if exists noloadtable") + sql("drop table if exists like_filter") + + CompactionSupportGlobalSortBigFileTest.createFile(file1, 500000, 0) sql("CREATE TABLE filtertestTables (ID int, date Timestamp, country String, " + "name String, phonetype String, serialname String, salary int) " + @@ -279,6 +287,21 @@ class FilterProcessorTestCase extends QueryTest with BeforeAndAfterAll { sql("drop table if exists outofrange") } + test("like% test case with restructure") { + sql("drop table if exists like_filter") + sql( + """ + | CREATE TABLE like_filter(id INT, name STRING, city STRING, age INT) + | STORED BY 'org.apache.carbondata.format' + | TBLPROPERTIES('SORT_COLUMNS'='city,name', 'SORT_SCOPE'='GLOBAL_SORT') + """.stripMargin) + sql(s"LOAD DATA LOCAL INPATH '$file1' INTO TABLE like_filter OPTIONS('header'='false')") + sql( + "ALTER TABLE like_filter ADD COLUMNS(filter STRING) TBLPROPERTIES ('DEFAULT.VALUE" + + ".FILTER'='altered column')") + checkAnswer(sql("select count(*) from like_filter where filter like '%column'"), Row(500000)) + } + @@ -294,6 +317,8 @@ class FilterProcessorTestCase extends QueryTest with BeforeAndAfterAll { sql("DROP TABLE IF EXISTS big_int_basicc_Hive_1") sql("DROP TABLE IF EXISTS filtertestTablesWithNull") sql("DROP TABLE IF EXISTS filtertestTablesWithNullJoin") + sql("drop table if exists like_filter") + CompactionSupportGlobalSortBigFileTest.deleteFile(file1) CarbonProperties.getInstance() .addProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT, "dd-MM-yyyy") }