From 184085284185011d7cc6d054b54d2d38eaf1dd77 Mon Sep 17 00:00:00 2001 From: Sean Owen Date: Sun, 13 Mar 2016 21:03:49 -0700 Subject: [SPARK-13823][CORE][STREAMING][SQL] Always specify Charset in String <-> byte[] conversions (and remaining Coverity items) ## What changes were proposed in this pull request? - Fixes calls to `new String(byte[])` or `String.getBytes()` that rely on platform default encoding, to use UTF-8 - Same for `InputStreamReader` and `OutputStreamWriter` constructors - Standardizes on UTF-8 everywhere - Standardizes specifying the encoding with `StandardCharsets.UTF-8`, not the Guava constant or "UTF-8" (which means handling `UnuspportedEncodingException`) - (also addresses the other remaining Coverity scan issues, which are pretty trivial; these are separated into commit https://github.com/srowen/spark/commit/1deecd8d9ca986d8adb1a42d315890ce5349d29c ) ## How was this patch tested? Jenkins tests Author: Sean Owen Closes #11657 from srowen/SPARK-13823. --- .../apache/spark/sql/catalyst/parser/ParseUtils.java | 4 +++- .../spark/sql/catalyst/expressions/literals.scala | 3 ++- .../org/apache/spark/sql/catalyst/util/package.scala | 3 ++- .../catalyst/expressions/LiteralExpressionSuite.scala | 4 +++- .../sql/catalyst/expressions/MathFunctionsSuite.scala | 6 ++++-- .../sql/catalyst/expressions/MiscFunctionsSuite.scala | 19 +++++++++++++------ .../expressions/UnsafeRowConverterSuite.scala | 14 ++++++++------ .../codegen/GeneratedProjectionSuite.scala | 5 ++++- 8 files changed, 39 insertions(+), 19 deletions(-) (limited to 'sql/catalyst/src') diff --git a/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/parser/ParseUtils.java b/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/parser/ParseUtils.java index 2520c7bb8d..01f89112a7 100644 --- a/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/parser/ParseUtils.java +++ b/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/parser/ParseUtils.java @@ -18,6 +18,8 @@ package org.apache.spark.sql.catalyst.parser; +import java.nio.charset.StandardCharsets; + /** * A couple of utility methods that help with parsing ASTs. * @@ -76,7 +78,7 @@ public final class ParseUtils { byte bVal = (byte) ((i3 - '0') + ((i2 - '0') * 8) + ((i1 - '0') * 8 * 8)); byte[] bValArr = new byte[1]; bValArr[0] = bVal; - String tmp = new String(bValArr); + String tmp = new String(bValArr, StandardCharsets.UTF_8); sb.append(tmp); i += 3; continue; diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/literals.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/literals.scala index a76517a89c..e6804d096c 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/literals.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/literals.scala @@ -17,6 +17,7 @@ package org.apache.spark.sql.catalyst.expressions +import java.nio.charset.StandardCharsets import java.sql.{Date, Timestamp} import org.json4s.JsonAST._ @@ -109,7 +110,7 @@ object Literal { case DateType => create(0, DateType) case TimestampType => create(0L, TimestampType) case StringType => Literal("") - case BinaryType => Literal("".getBytes) + case BinaryType => Literal("".getBytes(StandardCharsets.UTF_8)) case CalendarIntervalType => Literal(new CalendarInterval(0, 0)) case arr: ArrayType => create(Array(), arr) case map: MapType => create(Map(), map) diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/package.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/package.scala index d9a9b6151a..b11365b297 100644 --- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/package.scala +++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/package.scala @@ -18,6 +18,7 @@ package org.apache.spark.sql.catalyst import java.io._ +import java.nio.charset.StandardCharsets import org.apache.spark.sql.catalyst.expressions._ import org.apache.spark.sql.types.{NumericType, StringType} @@ -118,7 +119,7 @@ package object util { val writer = new PrintWriter(out) t.printStackTrace(writer) writer.flush() - new String(out.toByteArray) + new String(out.toByteArray, StandardCharsets.UTF_8) } def stringOrNull(a: AnyRef): String = if (a == null) null else a.toString diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/LiteralExpressionSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/LiteralExpressionSuite.scala index 124172bd66..450222d8cb 100644 --- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/LiteralExpressionSuite.scala +++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/LiteralExpressionSuite.scala @@ -17,6 +17,8 @@ package org.apache.spark.sql.catalyst.expressions +import java.nio.charset.StandardCharsets + import org.apache.spark.SparkFunSuite import org.apache.spark.sql.Row import org.apache.spark.sql.catalyst.util.DateTimeUtils @@ -54,7 +56,7 @@ class LiteralExpressionSuite extends SparkFunSuite with ExpressionEvalHelper { checkEvaluation(Literal.default(FloatType), 0.0f) checkEvaluation(Literal.default(DoubleType), 0.0) checkEvaluation(Literal.default(StringType), "") - checkEvaluation(Literal.default(BinaryType), "".getBytes) + checkEvaluation(Literal.default(BinaryType), "".getBytes(StandardCharsets.UTF_8)) checkEvaluation(Literal.default(DecimalType.USER_DEFAULT), Decimal(0)) checkEvaluation(Literal.default(DecimalType.SYSTEM_DEFAULT), Decimal(0)) checkEvaluation(Literal.default(DateType), DateTimeUtils.toJavaDate(0)) diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/MathFunctionsSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/MathFunctionsSuite.scala index 4ad65db097..fba5f53715 100644 --- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/MathFunctionsSuite.scala +++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/MathFunctionsSuite.scala @@ -17,6 +17,8 @@ package org.apache.spark.sql.catalyst.expressions +import java.nio.charset.StandardCharsets + import com.google.common.math.LongMath import org.apache.spark.SparkFunSuite @@ -440,7 +442,7 @@ class MathFunctionsSuite extends SparkFunSuite with ExpressionEvalHelper { checkEvaluation(Hex(Literal(100800200404L)), "177828FED4") checkEvaluation(Hex(Literal(-100800200404L)), "FFFFFFE887D7012C") checkEvaluation(Hex(Literal.create(null, BinaryType)), null) - checkEvaluation(Hex(Literal("helloHex".getBytes())), "68656C6C6F486578") + checkEvaluation(Hex(Literal("helloHex".getBytes(StandardCharsets.UTF_8))), "68656C6C6F486578") // scalastyle:off // Turn off scala style for non-ascii chars checkEvaluation(Hex(Literal("δΈ‰ι‡ηš„".getBytes("UTF8"))), "E4B889E9878DE79A84") @@ -452,7 +454,7 @@ class MathFunctionsSuite extends SparkFunSuite with ExpressionEvalHelper { test("unhex") { checkEvaluation(Unhex(Literal.create(null, StringType)), null) - checkEvaluation(Unhex(Literal("737472696E67")), "string".getBytes) + checkEvaluation(Unhex(Literal("737472696E67")), "string".getBytes(StandardCharsets.UTF_8)) checkEvaluation(Unhex(Literal("")), new Array[Byte](0)) checkEvaluation(Unhex(Literal("F")), Array[Byte](15)) checkEvaluation(Unhex(Literal("ff")), Array[Byte](-1)) diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/MiscFunctionsSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/MiscFunctionsSuite.scala index 75131a6170..60d50baf51 100644 --- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/MiscFunctionsSuite.scala +++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/MiscFunctionsSuite.scala @@ -17,6 +17,8 @@ package org.apache.spark.sql.catalyst.expressions +import java.nio.charset.StandardCharsets + import org.apache.commons.codec.digest.DigestUtils import org.apache.spark.SparkFunSuite @@ -27,7 +29,8 @@ import org.apache.spark.sql.types._ class MiscFunctionsSuite extends SparkFunSuite with ExpressionEvalHelper { test("md5") { - checkEvaluation(Md5(Literal("ABC".getBytes)), "902fbdd2b1df0c4f70b4a5d23525e932") + checkEvaluation(Md5(Literal("ABC".getBytes(StandardCharsets.UTF_8))), + "902fbdd2b1df0c4f70b4a5d23525e932") checkEvaluation(Md5(Literal.create(Array[Byte](1, 2, 3, 4, 5, 6), BinaryType)), "6ac1e56bc78f031059be7be854522c4c") checkEvaluation(Md5(Literal.create(null, BinaryType)), null) @@ -35,27 +38,31 @@ class MiscFunctionsSuite extends SparkFunSuite with ExpressionEvalHelper { } test("sha1") { - checkEvaluation(Sha1(Literal("ABC".getBytes)), "3c01bdbb26f358bab27f267924aa2c9a03fcfdb8") + checkEvaluation(Sha1(Literal("ABC".getBytes(StandardCharsets.UTF_8))), + "3c01bdbb26f358bab27f267924aa2c9a03fcfdb8") checkEvaluation(Sha1(Literal.create(Array[Byte](1, 2, 3, 4, 5, 6), BinaryType)), "5d211bad8f4ee70e16c7d343a838fc344a1ed961") checkEvaluation(Sha1(Literal.create(null, BinaryType)), null) - checkEvaluation(Sha1(Literal("".getBytes)), "da39a3ee5e6b4b0d3255bfef95601890afd80709") + checkEvaluation(Sha1(Literal("".getBytes(StandardCharsets.UTF_8))), + "da39a3ee5e6b4b0d3255bfef95601890afd80709") checkConsistencyBetweenInterpretedAndCodegen(Sha1, BinaryType) } test("sha2") { - checkEvaluation(Sha2(Literal("ABC".getBytes), Literal(256)), DigestUtils.sha256Hex("ABC")) + checkEvaluation(Sha2(Literal("ABC".getBytes(StandardCharsets.UTF_8)), Literal(256)), + DigestUtils.sha256Hex("ABC")) checkEvaluation(Sha2(Literal.create(Array[Byte](1, 2, 3, 4, 5, 6), BinaryType), Literal(384)), DigestUtils.sha384Hex(Array[Byte](1, 2, 3, 4, 5, 6))) // unsupported bit length checkEvaluation(Sha2(Literal.create(null, BinaryType), Literal(1024)), null) checkEvaluation(Sha2(Literal.create(null, BinaryType), Literal(512)), null) - checkEvaluation(Sha2(Literal("ABC".getBytes), Literal.create(null, IntegerType)), null) + checkEvaluation(Sha2(Literal("ABC".getBytes(StandardCharsets.UTF_8)), + Literal.create(null, IntegerType)), null) checkEvaluation(Sha2(Literal.create(null, BinaryType), Literal.create(null, IntegerType)), null) } test("crc32") { - checkEvaluation(Crc32(Literal("ABC".getBytes)), 2743272264L) + checkEvaluation(Crc32(Literal("ABC".getBytes(StandardCharsets.UTF_8))), 2743272264L) checkEvaluation(Crc32(Literal.create(Array[Byte](1, 2, 3, 4, 5, 6), BinaryType)), 2180413220L) checkEvaluation(Crc32(Literal.create(null, BinaryType)), null) diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/UnsafeRowConverterSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/UnsafeRowConverterSuite.scala index 68545f33e5..1265908182 100644 --- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/UnsafeRowConverterSuite.scala +++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/UnsafeRowConverterSuite.scala @@ -17,6 +17,7 @@ package org.apache.spark.sql.catalyst.expressions +import java.nio.charset.StandardCharsets import java.sql.{Date, Timestamp} import org.scalatest.Matchers @@ -77,16 +78,16 @@ class UnsafeRowConverterSuite extends SparkFunSuite with Matchers { val row = new SpecificMutableRow(fieldTypes) row.setLong(0, 0) row.update(1, UTF8String.fromString("Hello")) - row.update(2, "World".getBytes) + row.update(2, "World".getBytes(StandardCharsets.UTF_8)) val unsafeRow: UnsafeRow = converter.apply(row) assert(unsafeRow.getSizeInBytes === 8 + (8 * 3) + - roundedSize("Hello".getBytes.length) + - roundedSize("World".getBytes.length)) + roundedSize("Hello".getBytes(StandardCharsets.UTF_8).length) + + roundedSize("World".getBytes(StandardCharsets.UTF_8).length)) assert(unsafeRow.getLong(0) === 0) assert(unsafeRow.getString(1) === "Hello") - assert(unsafeRow.getBinary(2) === "World".getBytes) + assert(unsafeRow.getBinary(2) === "World".getBytes(StandardCharsets.UTF_8)) } test("basic conversion with primitive, string, date and timestamp types") { @@ -100,7 +101,8 @@ class UnsafeRowConverterSuite extends SparkFunSuite with Matchers { row.update(3, DateTimeUtils.fromJavaTimestamp(Timestamp.valueOf("2015-05-08 08:10:25"))) val unsafeRow: UnsafeRow = converter.apply(row) - assert(unsafeRow.getSizeInBytes === 8 + (8 * 4) + roundedSize("Hello".getBytes.length)) + assert(unsafeRow.getSizeInBytes === + 8 + (8 * 4) + roundedSize("Hello".getBytes(StandardCharsets.UTF_8).length)) assert(unsafeRow.getLong(0) === 0) assert(unsafeRow.getString(1) === "Hello") @@ -175,7 +177,7 @@ class UnsafeRowConverterSuite extends SparkFunSuite with Matchers { r.setFloat(6, 600) r.setDouble(7, 700) r.update(8, UTF8String.fromString("hello")) - r.update(9, "world".getBytes) + r.update(9, "world".getBytes(StandardCharsets.UTF_8)) r.setDecimal(10, Decimal(10), 10) r.setDecimal(11, Decimal(10.00, 38, 18), 38) // r.update(11, Array(11)) diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/codegen/GeneratedProjectionSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/codegen/GeneratedProjectionSuite.scala index 1522ee34e4..e2a8eb8ee1 100644 --- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/codegen/GeneratedProjectionSuite.scala +++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/codegen/GeneratedProjectionSuite.scala @@ -17,6 +17,8 @@ package org.apache.spark.sql.catalyst.expressions.codegen +import java.nio.charset.StandardCharsets + import org.apache.spark.SparkFunSuite import org.apache.spark.sql.catalyst.InternalRow import org.apache.spark.sql.catalyst.expressions._ @@ -107,7 +109,8 @@ class GeneratedProjectionSuite extends SparkFunSuite { val fields = Array[DataType](StringType, struct) val unsafeProj = UnsafeProjection.create(fields) - val innerRow = InternalRow(false, 1.toByte, 2.toShort, 3, 4.0f, "".getBytes, + val innerRow = InternalRow(false, 1.toByte, 2.toShort, 3, 4.0f, + "".getBytes(StandardCharsets.UTF_8), UTF8String.fromString("")) val row1 = InternalRow(UTF8String.fromString(""), innerRow) val unsafe1 = unsafeProj(row1).copy() -- cgit v1.2.3