amaliujia commented on a change in pull request #1761: URL: https://github.com/apache/calcite/pull/1761#discussion_r412349389
########## File path: core/src/main/java/org/apache/calcite/sql/SqlSessionTableFunction.java ########## @@ -0,0 +1,71 @@ +/* + * 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.calcite.sql; + +import org.apache.calcite.rel.type.RelDataType; +import org.apache.calcite.sql.type.SqlOperandCountRanges; +import org.apache.calcite.sql.type.SqlTypeName; +import org.apache.calcite.sql.type.SqlTypeUtil; +import org.apache.calcite.sql.validate.SqlValidator; + +/** + * SqlSessionTableFunction implements an operator for per-key sessionization. It allows + * four parameters: + * 1. a table. + * 2. a descriptor to provide a watermarked column name from the input table. + * 3. a descriptor to provide a column as key, on which sessionization will be applied. + * 4. an interval parameter to specify a inactive activity gap to break sessions. + */ +public class SqlSessionTableFunction extends SqlWindowTableFunction { + public SqlSessionTableFunction() { + super(SqlKind.SESSION.name()); + } + + @Override public SqlOperandCountRange getOperandCountRange() { + return SqlOperandCountRanges.of(4); + } + + @Override public boolean checkOperandTypes(SqlCallBinding callBinding, + boolean throwOnFailure) { + final SqlNode operand0 = callBinding.operand(0); + final SqlValidator validator = callBinding.getValidator(); + final RelDataType type = validator.getValidatedNodeType(operand0); + if (type.getSqlTypeName() != SqlTypeName.ROW) { + return throwValidationSignatureErrorOrReturnFalse(callBinding, throwOnFailure); + } + final SqlNode operand1 = callBinding.operand(1); + if (operand1.getKind() != SqlKind.DESCRIPTOR) { + return throwValidationSignatureErrorOrReturnFalse(callBinding, throwOnFailure); + } + validateColumnNames(validator, type.getFieldNames(), ((SqlCall) operand1).getOperandList()); + final SqlNode operand2 = callBinding.operand(2); + if (operand2.getKind() != SqlKind.DESCRIPTOR) { + return throwValidationSignatureErrorOrReturnFalse(callBinding, throwOnFailure); + } + validateColumnNames(validator, type.getFieldNames(), ((SqlCall) operand2).getOperandList()); + final RelDataType type3 = validator.getValidatedNodeType(callBinding.operand(3)); + if (!SqlTypeUtil.isInterval(type3)) { + return throwValidationSignatureErrorOrReturnFalse(callBinding, throwOnFailure); + } + return true; + } + + @Override public String getAllowedSignatures(String opNameToUse) { + return getName() + "(TABLE table_name, DESCRIPTOR(col), " + + "DESCRIPTOR(col), datetime interval)"; Review comment: Because descriptor only provide column names so far. If there is only one descriptor in which there are two column names, it will have a problem to know which one is the timestamp column that apply windowing upon and which one is the session key(note that session could be timestamp type as well). Of course a workaround is to define, say: the first column name is for windowing, and the second column is for session key. But this is error-prone, both from user usage and documentation perspectives: users have to use the right column name on the right position and documentation needs to explain the meaning of column position. Two descriptors is better as each one has its clear semantic and require one column only. So I think it's a good API design than single descriptor. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected]
