Repository: spark Updated Branches: refs/heads/branch-1.3 cbf232daa -> edcd3643a
[SPARK-7330] [SQL] avoid NPE at jdbc rdd Thank nadavoosh point this out in #5590 Author: Daoyuan Wang <daoyuan.w...@intel.com> Closes #5877 from adrian-wang/jdbcrdd and squashes the following commits: cc11900 [Daoyuan Wang] avoid NPE in jdbcrdd (cherry picked from commit ed9be06a4797bbb678355b361054c8872ac20b75) Signed-off-by: Yin Huai <yh...@databricks.com> Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/edcd3643 Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/edcd3643 Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/edcd3643 Branch: refs/heads/branch-1.3 Commit: edcd3643aa2bc39571dce17b7517eb40ced2a7f5 Parents: cbf232d Author: Daoyuan Wang <daoyuan.w...@intel.com> Authored: Thu May 7 10:05:01 2015 -0700 Committer: Yin Huai <yh...@databricks.com> Committed: Thu May 7 10:17:58 2015 -0700 ---------------------------------------------------------------------- .../org/apache/spark/sql/jdbc/JDBCRDD.scala | 8 ++++++- .../org/apache/spark/sql/jdbc/JDBCSuite.scala | 25 ++++++++++++++++++++ 2 files changed, 32 insertions(+), 1 deletion(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/spark/blob/edcd3643/sql/core/src/main/scala/org/apache/spark/sql/jdbc/JDBCRDD.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/main/scala/org/apache/spark/sql/jdbc/JDBCRDD.scala b/sql/core/src/main/scala/org/apache/spark/sql/jdbc/JDBCRDD.scala index d58b686..2881d9a 100644 --- a/sql/core/src/main/scala/org/apache/spark/sql/jdbc/JDBCRDD.scala +++ b/sql/core/src/main/scala/org/apache/spark/sql/jdbc/JDBCRDD.scala @@ -357,7 +357,13 @@ private[sql] class JDBCRDD( conversions(i) match { case BooleanConversion => mutableRow.setBoolean(i, rs.getBoolean(pos)) case DateConversion => - mutableRow.update(i, DateUtils.fromJavaDate(rs.getDate(pos))) + // DateUtils.fromJavaDate does not handle null value, so we need to check it. + val dateVal = rs.getDate(pos) + if (dateVal != null) { + mutableRow.update(i, DateUtils.fromJavaDate(dateVal)) + } else { + mutableRow.update(i, null) + } case DecimalConversion => val decimalVal = rs.getBigDecimal(pos) if (decimalVal == null) { http://git-wip-us.apache.org/repos/asf/spark/blob/edcd3643/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala ---------------------------------------------------------------------- diff --git a/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala index e8f7932..7719a76 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala @@ -102,6 +102,8 @@ class JDBCSuite extends FunSuite with BeforeAndAfter { ).executeUpdate() conn.prepareStatement("insert into test.timetypes values ('12:34:56', " + "'1996-01-01', '2002-02-20 11:22:33.543543543')").executeUpdate() + conn.prepareStatement("insert into test.timetypes values ('12:34:56', " + + "null, '2002-02-20 11:22:33.543543543')").executeUpdate() conn.commit() sql( s""" @@ -125,6 +127,23 @@ class JDBCSuite extends FunSuite with BeforeAndAfter { |OPTIONS (url '$url', dbtable 'TEST.FLTTYPES', user 'testUser', password 'testPass') """.stripMargin.replaceAll("\n", " ")) + conn.prepareStatement( + s""" + |create table test.nulltypes (a INT, b BOOLEAN, c TINYINT, d BINARY(20), e VARCHAR(20), + |f VARCHAR_IGNORECASE(20), g CHAR(20), h BLOB, i CLOB, j TIME, k DATE, l TIMESTAMP, + |m DOUBLE, n REAL, o DECIMAL(40, 20)) + """.stripMargin.replaceAll("\n", " ")).executeUpdate() + conn.prepareStatement("insert into test.nulltypes values (" + + "null, null, null, null, null, null, null, null, null, " + + "null, null, null, null, null, null)").executeUpdate() + conn.commit() + sql( + s""" + |CREATE TEMPORARY TABLE nulltypes + |USING org.apache.spark.sql.jdbc + |OPTIONS (url '$url', dbtable 'TEST.NULLTYPES', user 'testUser', password 'testPass') + """.stripMargin.replaceAll("\n", " ")) + // Untested: IDENTITY, OTHER, UUID, ARRAY, and GEOMETRY types. } @@ -252,6 +271,7 @@ class JDBCSuite extends FunSuite with BeforeAndAfter { val rows = TestSQLContext.jdbc(urlWithUserAndPass, "TEST.TIMETYPES").collect() val cachedRows = TestSQLContext.jdbc(urlWithUserAndPass, "TEST.TIMETYPES").cache().collect() assert(rows(0).getAs[java.sql.Date](1) === java.sql.Date.valueOf("1996-01-01")) + assert(rows(1).getAs[java.sql.Date](1) === null) assert(cachedRows(0).getAs[java.sql.Date](1) === java.sql.Date.valueOf("1996-01-01")) } @@ -264,6 +284,11 @@ class JDBCSuite extends FunSuite with BeforeAndAfter { assert(cachedRows(0).getAs[java.sql.Date](1) === java.sql.Date.valueOf("1996-01-01")) } + test("test types for null value") { + val rows = TestSQLContext.jdbc(urlWithUserAndPass, "TEST.NULLTYPES").collect() + assert((0 to 14).forall(i => rows(0).isNullAt(i))) + } + test("H2 floating-point types") { val rows = sql("SELECT * FROM flttypes").collect() assert(rows(0).getDouble(0) == 1.00000000000000022) // Yes, I meant ==. --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org For additional commands, e-mail: commits-h...@spark.apache.org