[Backport 2.19-dev] Fix function identify problem in converting to sql dialect

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

@@ -856,8 +856,8 @@ 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 → 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 → 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.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_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_min.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)])

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

@@ -0,0 +1,9 @@
+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, $8, 'banana':VARCHAR)])
+        CalciteLogicalIndexScan(table=[[OpenSearch, opensearch-sql_test_index_account]])
+  physical: |
+    EnumerableLimit(fetch=[10000])
+      EnumerableCalc(expr#0..16=[{inputs}], expr#17=[1], expr#18=[2], expr#19=[3], expr#20=['banana':VARCHAR], expr#21=[SCALAR_MAX($t17, $t18, $t19, $t8, $t20)], proj#0..10=[{exprs}], new=[$t21])
+        CalciteEnumerableIndexScan(table=[[OpenSearch, opensearch-sql_test_index_account]])

integ-test/src/test/resources/expectedOutput/calcite_no_pushdown/explain_eval_min.yaml

@@ -0,0 +1,9 @@
+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, $8, 'banana':VARCHAR)])
+        CalciteLogicalIndexScan(table=[[OpenSearch, opensearch-sql_test_index_account]])
+  physical: |
+    EnumerableLimit(fetch=[10000])
+      EnumerableCalc(expr#0..16=[{inputs}], expr#17=[1], expr#18=[2], expr#19=[3], expr#20=['banana':VARCHAR], expr#21=[SCALAR_MIN($t17, $t18, $t19, $t8, $t20)], proj#0..10=[{exprs}], new=[$t21])
+        CalciteEnumerableIndexScan(table=[[OpenSearch, opensearch-sql_test_index_account]])

ppl/src/test/java/org/opensearch/sql/ppl/calcite/CalcitePPLAggregationTest.java

@@ -447,9 +447,9 @@ public void testAvgBySpanAndFields() {
 
     String expectedSparkSql =
         ""
-            + "SELECT AVG(`SAL`) `avg(SAL)`, `SPAN`(`EMPNO`, 500, NULL) `empno_span`, `DEPTNO`\n"
+            + "SELECT AVG(`SAL`) `avg(SAL)`, SPAN(`EMPNO`, 500, NULL) `empno_span`, `DEPTNO`\n"
             + "FROM `scott`.`EMP`\n"
-            + "GROUP BY `DEPTNO`, `SPAN`(`EMPNO`, 500, NULL)\n"
+            + "GROUP BY `DEPTNO`, SPAN(`EMPNO`, 500, NULL)\n"
             + "ORDER BY `DEPTNO`, 2";
     verifyPPLToSparkSQL(root, expectedSparkSql);
   }
@@ -470,10 +470,10 @@ public void testAvgByTimeSpanAndFields() {
     verifyLogical(root, expectedLogical);
 
     String expectedSparkSql =
-        "SELECT AVG(`SAL`) `avg(SAL)`, `SPAN`(`HIREDATE`, 1, 'y') `hiredate_span`, `DEPTNO`\n"
+        "SELECT AVG(`SAL`) `avg(SAL)`, SPAN(`HIREDATE`, 1, 'y') `hiredate_span`, `DEPTNO`\n"
             + "FROM `scott`.`EMP`\n"
             + "WHERE `HIREDATE` IS NOT NULL\n"
-            + "GROUP BY `DEPTNO`, `SPAN`(`HIREDATE`, 1, 'y')\n"
+            + "GROUP BY `DEPTNO`, SPAN(`HIREDATE`, 1, 'y')\n"
             + "ORDER BY `DEPTNO`, 2";
     verifyPPLToSparkSQL(root, expectedSparkSql);
   }
@@ -596,9 +596,9 @@ public void testStddevSampByFieldWithAlias() {
 
     String expectedSparkSql =
         ""
-            + "SELECT STDDEV_SAMP(`SAL`) `samp`, `SPAN`(`EMPNO`, 100, NULL) `empno_span`\n"
+            + "SELECT STDDEV_SAMP(`SAL`) `samp`, SPAN(`EMPNO`, 100, NULL) `empno_span`\n"
             + "FROM `scott`.`EMP`\n"
-            + "GROUP BY `SPAN`(`EMPNO`, 100, NULL)";
+            + "GROUP BY SPAN(`EMPNO`, 100, NULL)";
     verifyPPLToSparkSQL(root, expectedSparkSql);
   }
 
@@ -982,10 +982,10 @@ public void testSortAggregationMetrics2() {
     verifyLogical(root, expectedLogical);
 
     String expectedSparkSql =
-        "SELECT AVG(`SAL`) `avg`, `SPAN`(`HIREDATE`, 1, 'y') `hiredate_span`\n"
+        "SELECT AVG(`SAL`) `avg`, SPAN(`HIREDATE`, 1, 'y') `hiredate_span`\n"
             + "FROM `scott`.`EMP`\n"
             + "WHERE `HIREDATE` IS NOT NULL\n"
-            + "GROUP BY `SPAN`(`HIREDATE`, 1, 'y')\n"
+            + "GROUP BY SPAN(`HIREDATE`, 1, 'y')\n"
             + "ORDER BY 1";
     verifyPPLToSparkSQL(root, expectedSparkSql);
   }

ppl/src/test/java/org/opensearch/sql/ppl/calcite/CalcitePPLArrayFunctionTest.java

@@ -33,7 +33,7 @@ public void testMvjoinWithStringArray() {
     verifyResult(root, expectedResult);
 
     String expectedSparkSql =
-        "SELECT ARRAY_JOIN(`array`('a', 'b', 'c'), ',') `joined`\n"
+        "SELECT ARRAY_JOIN(ARRAY('a', 'b', 'c'), ',') `joined`\n"
             + "FROM `scott`.`EMP`\n"
             + "LIMIT 1";
     verifyPPLToSparkSQL(root, expectedSparkSql);
@@ -59,7 +59,7 @@ public void testMvjoinWithDifferentDelimiter() {
     verifyResult(root, expectedResult);
 
     String expectedSparkSql =
-        "SELECT ARRAY_JOIN(`array`('apple', 'banana', 'cherry'), ' | ') `joined`\n"
+        "SELECT ARRAY_JOIN(ARRAY('apple', 'banana', 'cherry'), ' | ') `joined`\n"
             + "FROM `scott`.`EMP`\n"
             + "LIMIT 1";
     verifyPPLToSparkSQL(root, expectedSparkSql);
@@ -82,7 +82,7 @@ public void testMvjoinWithEmptyArray() {
     verifyResult(root, expectedResult);
 
     String expectedSparkSql =
-        "SELECT ARRAY_JOIN(`array`(), ',') `joined`\n" + "FROM `scott`.`EMP`\n" + "LIMIT 1";
+        "SELECT ARRAY_JOIN(ARRAY(), ',') `joined`\n" + "FROM `scott`.`EMP`\n" + "LIMIT 1";
     verifyPPLToSparkSQL(root, expectedSparkSql);
   }
 
@@ -101,7 +101,7 @@ public void testMvjoinWithFieldReference() {
     verifyLogical(root, expectedLogical);
 
     String expectedSparkSql =
-        "SELECT ARRAY_JOIN(`array`(`ENAME`, `JOB`), '-') `joined`\n"
+        "SELECT ARRAY_JOIN(ARRAY(`ENAME`, `JOB`), '-') `joined`\n"
             + "FROM `scott`.`EMP`\n"
             + "LIMIT 1";
     verifyPPLToSparkSQL(root, expectedSparkSql);

ppl/src/test/java/org/opensearch/sql/ppl/calcite/CalcitePPLBasicTest.java

@@ -493,11 +493,11 @@ public void testDecimalLiteral() {
     verifyLogical(root, expectedLogical);
 
     String expectedSparkSql =
-        "SELECT `DIVIDE`(22, 7.0) `r1`, `DIVIDE`(22, 7.0E0) `r2`, `DIVIDE`(22.0, 7) `r3`,"
-            + " `DIVIDE`(2.20E1, 7) `r4`, 0.1 * 0.2 `r5`, 1E-1 * 2E-1 `r6`, 0.1 + 0.2 `r7`, 1E-1 +"
-            + " 2E-1 `r8`, 0.06 - 0.01 `r9`, 6E-2 - 1E-2 `r10`, `DIVIDE`(0.1, 0.3) * 0.3 `r11`,"
-            + " `DIVIDE`(1E-1, 3E-1) * 3E-1 `r12`, POWER(POWER(2.0, 5E-1), 2) `r13`,"
-            + " POWER(POWER(2.0E0, 5E-1), 2) `r14`, `DIVIDE`(7.0, 0) `r15`, `DIVIDE`(7, 0.0)"
+        "SELECT DIVIDE(22, 7.0) `r1`, DIVIDE(22, 7.0E0) `r2`, DIVIDE(22.0, 7) `r3`,"
+            + " DIVIDE(2.20E1, 7) `r4`, 0.1 * 0.2 `r5`, 1E-1 * 2E-1 `r6`, 0.1 + 0.2 `r7`, 1E-1 +"
+            + " 2E-1 `r8`, 0.06 - 0.01 `r9`, 6E-2 - 1E-2 `r10`, DIVIDE(0.1, 0.3) * 0.3 `r11`,"
+            + " DIVIDE(1E-1, 3E-1) * 3E-1 `r12`, POWER(POWER(2.0, 5E-1), 2) `r13`,"
+            + " POWER(POWER(2.0E0, 5E-1), 2) `r14`, DIVIDE(7.0, 0) `r15`, DIVIDE(7, 0.0)"
             + " `r16`\n"
             + "FROM `scott`.`EMP`\n"
             + "LIMIT 1";