mihailom-db commented on code in PR #46801: URL: https://github.com/apache/spark/pull/46801#discussion_r1631034190
########## sql/core/src/test/scala/org/apache/spark/sql/CollationExpressionWalkerSuite.scala: ########## @@ -0,0 +1,294 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql + +import org.apache.spark.SparkFunSuite +import org.apache.spark.sql.catalyst.expressions.{EmptyRow, EvalMode, ExpectsInputTypes, Expression, GenericInternalRow, Literal} +import org.apache.spark.sql.internal.types.{AbstractArrayType, StringTypeAnyCollation, StringTypeBinaryLcase} +import org.apache.spark.sql.test.SharedSparkSession +import org.apache.spark.sql.types.{AbstractDataType, AnyDataType, AnyTimestampType, ArrayType, BinaryType, BooleanType, DataType, DatetimeType, Decimal, DecimalType, IntegerType, LongType, MapType, NumericType, StringType, StructType, TypeCollection} +import org.apache.spark.unsafe.types.UTF8String +import org.apache.spark.util.Utils + +/** + * This suite is introduced in order to test a bulk of expressions and functionalities related to + * collations + */ +class CollationExpressionWalkerSuite extends SparkFunSuite with SharedSparkSession { + + // Trait to distinguish different cases for generation + sealed trait CollationType + + case object Utf8Binary extends CollationType + + case object Utf8BinaryLcase extends CollationType + + /** + * Helper function to generate all necesary parameters + * + * @param inputEntry - List of all input entries that need to be generated + * @param collationType - Flag defining collation type to use + * @return + */ + def generateData( + inputEntry: Seq[Any], + collationType: CollationType): Seq[Any] = { + inputEntry.map(generateSingleEntry(_, collationType)) + } + + /** + * Helper function to generate single entry of data. + * @param inputEntry - Single input entry that requires generation + * @param collationType - Flag defining collation type to use + * @return + */ + def generateSingleEntry( + inputEntry: Any, + collationType: CollationType): Any = + inputEntry match { + case e: Class[_] if e.isAssignableFrom(classOf[Expression]) => + generateLiterals(StringTypeAnyCollation, collationType) + case se: Class[_] if se.isAssignableFrom(classOf[Seq[Expression]]) => + Seq(generateLiterals(StringTypeAnyCollation, collationType), + generateLiterals(StringTypeAnyCollation, collationType)) + case oe: Class[_] if oe.isAssignableFrom(classOf[Option[Any]]) => None + case b: Class[_] if b.isAssignableFrom(classOf[Boolean]) => false + case dt: Class[_] if dt.isAssignableFrom(classOf[DataType]) => StringType + case st: Class[_] if st.isAssignableFrom(classOf[StructType]) => StructType + case em: Class[_] if em.isAssignableFrom(classOf[EvalMode.Value]) => EvalMode.LEGACY + case m: Class[_] if m.isAssignableFrom(classOf[Map[_, _]]) => Map.empty + case c: Class[_] if c.isAssignableFrom(classOf[Char]) => '\\' + case i: Class[_] if i.isAssignableFrom(classOf[Int]) => 0 + case l: Class[_] if l.isAssignableFrom(classOf[Long]) => 0 + case adt: AbstractDataType => generateLiterals(adt, collationType) + case Nil => Seq() + case (head: AbstractDataType) :: rest => generateData(head :: rest, collationType) + } + + /** + * Helper function to generate single literal from the given type. + * + * @param inputType - Single input literal type that requires generation + * @param collationType - Flag defining collation type to use + * @return + */ + def generateLiterals( + inputType: AbstractDataType, + collationType: CollationType): Expression = + inputType match { + // TODO: Try to make this a bit more random. + case AnyTimestampType => Literal("2009-07-30 12:58:59") + case BinaryType => Literal(new Array[Byte](5)) + case BooleanType => Literal(true) + case _: DatetimeType => Literal(1L) + case _: DecimalType => Literal(new Decimal) + case IntegerType | NumericType => Literal(1) + case LongType => Literal(1L) + case _: StringType | StringTypeAnyCollation | StringTypeBinaryLcase | AnyDataType => + collationType match { + case Utf8Binary => + Literal.create("dummy string", StringType("UTF8_BINARY")) + case Utf8BinaryLcase => + Literal.create("DuMmY sTrInG", StringType("UTF8_BINARY_LCASE")) + } + case TypeCollection(typeCollection) => + val strTypes = typeCollection.filter(hasStringType) + if (strTypes.isEmpty) { + // Take first type + generateLiterals(typeCollection.head, collationType) + } else { + // Take first string type + generateLiterals(strTypes.head, collationType) + } + case AbstractArrayType(elementType) => + generateLiterals(elementType, collationType).map( + lit => Literal.create(Seq(lit.asInstanceOf[Literal].value), ArrayType(lit.dataType)) + ).head + case ArrayType(elementType, _) => + generateLiterals(elementType, collationType).map( + lit => Literal.create(Seq(lit.asInstanceOf[Literal].value), ArrayType(lit.dataType)) + ).head + case ArrayType => + generateLiterals(StringTypeAnyCollation, collationType).map( + lit => Literal.create(Seq(lit.asInstanceOf[Literal].value), ArrayType(lit.dataType)) + ).head + case MapType => + val key = generateLiterals(StringTypeAnyCollation, collationType) + val value = generateLiterals(StringTypeAnyCollation, collationType) + Literal.create(Map(key -> value)) + case MapType(keyType, valueType, _) => + val key = generateLiterals(keyType, collationType) + val value = generateLiterals(valueType, collationType) + Literal.create(Map(key -> value)) + case StructType => + Literal.create((generateLiterals(StringTypeAnyCollation, collationType), + generateLiterals(StringTypeAnyCollation, collationType))) + case StructType(fields) => + Literal.create(new GenericInternalRow( + fields.map(f => generateLiterals(f.dataType, collationType).asInstanceOf[Any])), + StructType(fields)) + } + + /** + * Helper function to extract types of relevance + * @param inputType + * @return + */ + def hasStringType(inputType: AbstractDataType): Boolean = { + inputType match { + case _: StringType | StringTypeAnyCollation | StringTypeBinaryLcase | AnyDataType => + true + case ArrayType => true + case MapType => true + case MapType(keyType, valueType, _) => hasStringType(keyType) || hasStringType(valueType) + case ArrayType(elementType, _) => hasStringType(elementType) + case AbstractArrayType(elementType) => hasStringType(elementType) + case TypeCollection(typeCollection) => + typeCollection.exists(hasStringType) + case StructType => true + case StructType(fields) => fields.exists(sf => hasStringType(sf.dataType)) + case _ => false + } + } + + def replaceExpressions(inputTypes: Seq[AbstractDataType], params: Seq[Class[_]]): Seq[Any] = { + (inputTypes, params) match { + case (Nil, mparams) => mparams + case (_, Nil) => Nil + case (minputTypes, mparams) if mparams.head.isAssignableFrom(classOf[Expression]) => + minputTypes.head +: replaceExpressions(inputTypes.tail, mparams.tail) + case (minputTypes, mparams) => + mparams.head +: replaceExpressions(minputTypes.tail, mparams.tail) + } + } + + test("SPARK-48280: Expression Walker for Test") { + // This test does following: + // 1) Take all expressions + // 2) Find the ones that have at least one argument of StringType + // 3) Use reflection to create an instance of the expression using first constructor + // (test other as well). + // 4) Check if the expression is of type ExpectsInputTypes (should make this a bit broader) + // 5) Run eval against literals with strings under: + // a) UTF8_BINARY, "dummy string" as input. + // b) UTF8_BINARY_LCASE, "DuMmY sTrInG" as input. + // 6) Check if both expressions throw an exception. + // 7) If no exception, check if the result is the same. + // 8) There is a list of allowed expressions that can differ (e.g. hex) + var expressionCounter = 0 + var expectsExpressionCounter = 0; + val funInfos = spark.sessionState.functionRegistry.listFunction().map { funcId => + spark.sessionState.catalog.lookupFunctionInfo(funcId) + }.filter(funInfo => { + // make sure that there is a constructor. + val cl = Utils.classForName(funInfo.getClassName) + !cl.getConstructors.isEmpty + }).filter(funInfo => { + expressionCounter = expressionCounter + 1 + val cl = Utils.classForName(funInfo.getClassName) + // dummy instance + // Take first constructor. + val headConstructor = cl.getConstructors.head + + val params = headConstructor.getParameters.map(p => p.getType) + + val args = generateData(params.toSeq, Utf8Binary) + // Find all expressions that have string as input + try { + val expr = headConstructor.newInstance(args: _*) + expr match { + case types: ExpectsInputTypes => + expectsExpressionCounter = expectsExpressionCounter + 1 + val inputTypes = types.inputTypes + inputTypes.exists(hasStringType) || inputTypes.isEmpty + } + } catch { + case _: Throwable => false + } + }).toArray + + val toSkip = List( + "parse_url", // Parse URL cannot be generalized with ExpectInputTypes + "hex" // Different inputs affect conversion Review Comment: Found it. Difference is that Hex can accept StringType, while base64 can only accept BinaryType, so base64 performs transformation to BinaryType before executing, which is using the hashing from CollationFactory, and hex is calling java's toBytes on the UTF8String. @dbatomic What behaviour is expected? I would say this is weird to have some conversions work one way and others the other way. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org