Fix function identify problem in converting to sql dialect (#4793) (#4807)

(cherry picked from commit 3a91e5c)

Signed-off-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

Author: opensearch-trigger-bot[bot]

core/src/main/java/org/opensearch/sql/expression/function/PPLBuiltinOperators.java

@@ -98,10 +98,10 @@
 import org.opensearch.sql.expression.function.udf.math.ConvFunction;
 import org.opensearch.sql.expression.function.udf.math.DivideFunction;
 import org.opensearch.sql.expression.function.udf.math.EulerFunction;
-import org.opensearch.sql.expression.function.udf.math.MaxFunction;
-import org.opensearch.sql.expression.function.udf.math.MinFunction;
 import org.opensearch.sql.expression.function.udf.math.ModFunction;
 import org.opensearch.sql.expression.function.udf.math.NumberToStringFunction;
+import org.opensearch.sql.expression.function.udf.math.ScalarMaxFunction;
+import org.opensearch.sql.expression.function.udf.math.ScalarMinFunction;
 
 /** Defines functions and operators that are implemented only by PPL */
 public class PPLBuiltinOperators extends ReflectiveSqlOperatorTable {
@@ -131,8 +131,8 @@ public class PPLBuiltinOperators extends ReflectiveSqlOperatorTable {
   public static final SqlOperator DIVIDE = new DivideFunction().toUDF("DIVIDE");
   public static final SqlOperator SHA2 = CryptographicFunction.sha2().toUDF("SHA2");
   public static final SqlOperator CIDRMATCH = new CidrMatchFunction().toUDF("CIDRMATCH");
-  public static final SqlOperator MAX = new MaxFunction().toUDF("MAX");
-  public static final SqlOperator MIN = new MinFunction().toUDF("MIN");
+  public static final SqlOperator SCALAR_MAX = new ScalarMaxFunction().toUDF("SCALAR_MAX");
+  public static final SqlOperator SCALAR_MIN = new ScalarMinFunction().toUDF("SCALAR_MIN");
 
   public static final SqlOperator COSH =
       adaptMathFunctionToUDF(

core/src/main/java/org/opensearch/sql/expression/function/PPLFuncImpTable.java

@@ -859,8 +859,8 @@ void populate() {
       registerOperator(INTERNAL_TRANSLATE3, SqlLibraryOperators.TRANSLATE3);
 
       // Register eval functions for PPL max() and min() calls
-      registerOperator(MAX, PPLBuiltinOperators.MAX);
-      registerOperator(MIN, PPLBuiltinOperators.MIN);
+      registerOperator(MAX, PPLBuiltinOperators.SCALAR_MAX);
+      registerOperator(MIN, PPLBuiltinOperators.SCALAR_MIN);
 
       // Register PPL UDF operator
       registerOperator(COSH, PPLBuiltinOperators.COSH);

core/src/main/java/org/opensearch/sql/expression/function/UserDefinedFunctionBuilder.java

@@ -59,6 +59,13 @@ default SqlUserDefinedFunction toUDF(String functionName, boolean isDeterministi
       public boolean isDeterministic() {
         return isDeterministic;
       }
+
+      @Override
+      public SqlIdentifier getSqlIdentifier() {
+        // to avoid convert to sql dialog as identifier, use keyword instead
+        // check the code SqlUtil.unparseFunctionSyntax()
+        return null;
+      }
     };
   }
 }

core/src/main/java/org/opensearch/sql/expression/function/udf/math/MaxFunction.java renamed to core/src/main/java/org/opensearch/sql/expression/function/udf/math/ScalarMaxFunction.java

@@ -24,9 +24,9 @@
  * MAX(value1, value2, ...) returns the maximum value from the arguments. For mixed types, strings
  * have higher precedence than numbers.
  */
-public class MaxFunction extends ImplementorUDF {
+public class ScalarMaxFunction extends ImplementorUDF {
 
-  public MaxFunction() {
+  public ScalarMaxFunction() {
     super(new MaxImplementor(), NullPolicy.ALL);
   }
 

core/src/main/java/org/opensearch/sql/expression/function/udf/math/MinFunction.java renamed to core/src/main/java/org/opensearch/sql/expression/function/udf/math/ScalarMinFunction.java

@@ -24,9 +24,9 @@
  * MIN(value1, value2, ...) returns the minimum value from the arguments. For mixed types, numbers
  * have higher precedence than strings.
  */
-public class MinFunction extends ImplementorUDF {
+public class ScalarMinFunction extends ImplementorUDF {
 
-  public MinFunction() {
+  public ScalarMinFunction() {
     super(new MinImplementor(), NullPolicy.ALL);
   }
 

integ-test/src/test/java/org/opensearch/sql/calcite/remote/CalciteExplainIT.java

@@ -1300,19 +1300,19 @@ public void testExplainSortOnMeasureMultiBucketsNotMultiTermsNotPushDown() throw
 
   @Test
   public void testExplainEvalMax() throws IOException {
-    String expected = loadExpectedPlan("explain_eval_max.json");
-    assertJsonEqualsIgnoreId(
+    String expected = loadExpectedPlan("explain_eval_max.yaml");
+    assertYamlEqualsIgnoreId(
         expected,
-        explainQueryToString(
+        explainQueryYaml(
             "source=opensearch-sql_test_index_account | eval new = max(1, 2, 3, age, 'banana')"));
   }
 
   @Test
   public void testExplainEvalMin() throws IOException {
-    String expected = loadExpectedPlan("explain_eval_min.json");
-    assertJsonEqualsIgnoreId(
+    String expected = loadExpectedPlan("explain_eval_min.yaml");
+    assertYamlEqualsIgnoreId(
         expected,
-        explainQueryToString(
+        explainQueryYaml(
             "source=opensearch-sql_test_index_account | eval new = min(1, 2, 3, age, 'banana')"));
   }
 

integ-test/src/test/resources/expectedOutput/calcite/explain_eval_max.json

@@ -0,0 +1,8 @@
+calcite:
+  logical: |
+    LogicalSystemLimit(fetch=[10000], type=[QUERY_SIZE_LIMIT])
+      LogicalProject(account_number=[$0], firstname=[$1], address=[$2], balance=[$3], gender=[$4], city=[$5], employer=[$6], state=[$7], age=[$8], email=[$9], lastname=[$10], new=[SCALAR_MAX(1, 2, 3, $8, 'banana':VARCHAR)])
+        CalciteLogicalIndexScan(table=[[OpenSearch, opensearch-sql_test_index_account]])
+  physical: |
+    EnumerableCalc(expr#0..10=[{inputs}], expr#11=[1], expr#12=[2], expr#13=[3], expr#14=['banana':VARCHAR], expr#15=[SCALAR_MAX($t11, $t12, $t13, $t8, $t14)], proj#0..10=[{exprs}], $f11=[$t15])
+      CalciteEnumerableIndexScan(table=[[OpenSearch, opensearch-sql_test_index_account]], PushDownContext=[[PROJECT->[account_number, firstname, address, balance, gender, city, employer, state, age, email, lastname], LIMIT->10000], OpenSearchRequestBuilder(sourceBuilder={"from":0,"size":10000,"timeout":"1m","_source":{"includes":["account_number","firstname","address","balance","gender","city","employer","state","age","email","lastname"],"excludes":[]}}, requestedTotalSize=10000, pageSize=null, startFrom=0)])

integ-test/src/test/resources/expectedOutput/calcite/explain_eval_max.yaml

@@ -0,0 +1,8 @@
+calcite:
+  logical: |
+    LogicalSystemLimit(fetch=[10000], type=[QUERY_SIZE_LIMIT])
+      LogicalProject(account_number=[$0], firstname=[$1], address=[$2], balance=[$3], gender=[$4], city=[$5], employer=[$6], state=[$7], age=[$8], email=[$9], lastname=[$10], new=[SCALAR_MIN(1, 2, 3, $8, 'banana':VARCHAR)])
+        CalciteLogicalIndexScan(table=[[OpenSearch, opensearch-sql_test_index_account]])
+  physical: |
+    EnumerableCalc(expr#0..10=[{inputs}], expr#11=[1], expr#12=[2], expr#13=[3], expr#14=['banana':VARCHAR], expr#15=[SCALAR_MIN($t11, $t12, $t13, $t8, $t14)], proj#0..10=[{exprs}], $f11=[$t15])
+      CalciteEnumerableIndexScan(table=[[OpenSearch, opensearch-sql_test_index_account]], PushDownContext=[[PROJECT->[account_number, firstname, address, balance, gender, city, employer, state, age, email, lastname], LIMIT->10000], OpenSearchRequestBuilder(sourceBuilder={"from":0,"size":10000,"timeout":"1m","_source":{"includes":["account_number","firstname","address","balance","gender","city","employer","state","age","email","lastname"],"excludes":[]}}, requestedTotalSize=10000, pageSize=null, startFrom=0)])