fix tests

Signed-off-by: Kai Huang <ahkcs@amazon.com>

Author: ahkcs

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

@@ -210,8 +210,8 @@ public void testMultisearchWithDifferentIndicesSchemaMerge() throws IOException
         executeQuery(
             String.format(
                 "| multisearch [search source=%s | where age > 35 | fields account_number,"
-                    + " firstname, age, balance] [search source=%s | where age > 35 | fields"
-                    + " account_number, balance, age] | stats count() as total_count",
+                    + " firstname, balance] [search source=%s | where age > 35 | fields"
+                    + " account_number, balance] | stats count() as total_count",
                 TEST_INDEX_ACCOUNT, TEST_INDEX_BANK));
 
     verifySchema(result, schema("total_count", null, "bigint"));
@@ -345,18 +345,23 @@ public void testMultisearchCrossIndexFieldSelection() throws IOException {
   }
 
   @Test
-  public void testMultisearchTypeConflictWithStats() throws IOException {
-    JSONObject result =
-        executeQuery(
-            String.format(
-                "| multisearch "
-                    + "[search source=%s | fields age] "
-                    + "[search source=%s | fields age] "
-                    + "| stats count() as total",
-                TEST_INDEX_ACCOUNT, TEST_INDEX_LOCATIONS_TYPE_CONFLICT));
-
-    verifySchema(result, schema("total", null, "bigint"));
+  public void testMultisearchTypeConflictWithStats() {
+    Exception exception =
+        assertThrows(
+            ResponseException.class,
+            () ->
+                executeQuery(
+                    String.format(
+                        "| multisearch "
+                            + "[search source=%s | fields age] "
+                            + "[search source=%s | fields age] "
+                            + "| stats count() as total",
+                        TEST_INDEX_ACCOUNT, TEST_INDEX_LOCATIONS_TYPE_CONFLICT)));
 
-    verifyDataRows(result, rows(1010L));
+    assertTrue(
+        "Error message should indicate type conflict",
+        exception
+            .getMessage()
+            .contains("Schema unification failed: field 'age' has conflicting types"));
   }
 }

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

@@ -9,6 +9,7 @@
 import java.util.List;
 import org.apache.calcite.rel.RelNode;
 import org.apache.calcite.test.CalciteAssert;
+import org.junit.Assert;
 import org.junit.Test;
 
 public class CalcitePPLAppendTest extends CalcitePPLAbstractTest {
@@ -71,15 +72,16 @@ public void testAppendEmptySearchCommand() {
   @Test
   public void testAppendNested() {
     String ppl =
-        "source=EMP | append [ | where DEPTNO = 10 | append [ source=EMP | where DEPTNO = 20 ] ]";
+        "source=EMP | fields ENAME, SAL | append [ | append [ source=EMP | where DEPTNO = 20 ] ]";
     RelNode root = getRelNode(ppl);
     String expectedLogical =
         "LogicalUnion(all=[true])\n"
-            + "  LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4],"
-            + " SAL=[$5], COMM=[$6], DEPTNO=[$7], EMPNO0=[null:SMALLINT])\n"
+            + "  LogicalProject(ENAME=[$1], SAL=[$5], EMPNO=[null:SMALLINT], JOB=[null:VARCHAR(9)],"
+            + " MGR=[null:SMALLINT], HIREDATE=[null:DATE], COMM=[null:DECIMAL(7, 2)],"
+            + " DEPTNO=[null:TINYINT])\n"
             + "    LogicalTableScan(table=[[scott, EMP]])\n"
-            + "  LogicalProject(EMPNO=[null:SMALLINT], ENAME=[$1], JOB=[$2], MGR=[$3],"
-            + " HIREDATE=[$4], SAL=[$5], COMM=[$6], DEPTNO=[$7], EMPNO0=[$0])\n"
+            + "  LogicalProject(ENAME=[$1], SAL=[$5], EMPNO=[$0], JOB=[$2], MGR=[$3],"
+            + " HIREDATE=[$4], COMM=[$6], DEPTNO=[$7])\n"
             + "    LogicalUnion(all=[true])\n"
             + "      LogicalValues(tuples=[[]])\n"
             + "      LogicalFilter(condition=[=($7, 20)])\n"
@@ -88,12 +90,12 @@ public void testAppendNested() {
     verifyResultCount(root, 19); // 14 original table rows + 5 filtered subquery rows
 
     String expectedSparkSql =
-        "SELECT `EMPNO`, `ENAME`, `JOB`, `MGR`, `HIREDATE`, `SAL`, `COMM`, `DEPTNO`, CAST(NULL AS"
-            + " SMALLINT) `EMPNO0`\n"
+        "SELECT `ENAME`, `SAL`, CAST(NULL AS SMALLINT) `EMPNO`, CAST(NULL AS STRING) `JOB`,"
+            + " CAST(NULL AS SMALLINT) `MGR`, CAST(NULL AS DATE) `HIREDATE`, CAST(NULL AS"
+            + " DECIMAL(7, 2)) `COMM`, CAST(NULL AS TINYINT) `DEPTNO`\n"
             + "FROM `scott`.`EMP`\n"
             + "UNION ALL\n"
-            + "SELECT CAST(NULL AS SMALLINT) `EMPNO`, `ENAME`, `JOB`, `MGR`, `HIREDATE`, `SAL`,"
-            + " `COMM`, `DEPTNO`, `EMPNO` `EMPNO0`\n"
+            + "SELECT `ENAME`, `SAL`, `EMPNO`, `JOB`, `MGR`, `HIREDATE`, `COMM`, `DEPTNO`\n"
             + "FROM (SELECT *\n"
             + "FROM (VALUES (NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL)) `t` (`EMPNO`,"
             + " `ENAME`, `JOB`, `MGR`, `HIREDATE`, `SAL`, `COMM`, `DEPTNO`)\n"
@@ -109,61 +111,63 @@ public void testAppendNested() {
   public void testAppendEmptySourceWithJoin() {
     List<String> emptySourceWithEmptySourceJoinPPLs =
         Arrays.asList(
-            "source=EMP | append [ | where DEPTNO = 10 | join on ENAME = DNAME DEPT ]",
-            "source=EMP | append [ | where DEPTNO = 10 | cross join on ENAME = DNAME DEPT ]",
-            "source=EMP | append [ | where DEPTNO = 10 | left join on ENAME = DNAME DEPT ]",
-            "source=EMP | append [ | where DEPTNO = 10 | semi join on ENAME = DNAME DEPT ]",
-            "source=EMP | append [ | where DEPTNO = 10 | anti join on ENAME = DNAME DEPT ]");
+            "source=EMP | fields EMPNO, ENAME, JOB | append [ | where DEPTNO = 10 | join on ENAME"
+                + " = DNAME DEPT ]",
+            "source=EMP | fields EMPNO, ENAME, JOB | append [ | where DEPTNO = 10 | cross join on"
+                + " ENAME = DNAME DEPT ]",
+            "source=EMP | fields EMPNO, ENAME, JOB | append [ | where DEPTNO = 10 | left join on"
+                + " ENAME = DNAME DEPT ]",
+            "source=EMP | fields EMPNO, ENAME, JOB | append [ | where DEPTNO = 10 | semi join on"
+                + " ENAME = DNAME DEPT ]",
+            "source=EMP | fields EMPNO, ENAME, JOB | append [ | where DEPTNO = 10 | anti join on"
+                + " ENAME = DNAME DEPT ]");
 
     for (String ppl : emptySourceWithEmptySourceJoinPPLs) {
       RelNode root = getRelNode(ppl);
       String expectedLogical =
           "LogicalUnion(all=[true])\n"
-              + "  LogicalTableScan(table=[[scott, EMP]])\n"
+              + "  LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2])\n"
+              + "    LogicalTableScan(table=[[scott, EMP]])\n"
               + "  LogicalValues(tuples=[[]])\n";
       verifyLogical(root, expectedLogical);
       verifyResultCount(root, 14);
 
       String expectedSparkSql =
-          "SELECT *\n"
+          "SELECT `EMPNO`, `ENAME`, `JOB`\n"
               + "FROM `scott`.`EMP`\n"
               + "UNION ALL\n"
               + "SELECT *\n"
-              + "FROM (VALUES (NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL)) `t` (`EMPNO`,"
-              + " `ENAME`, `JOB`, `MGR`, `HIREDATE`, `SAL`, `COMM`, `DEPTNO`)\n"
+              + "FROM (VALUES (NULL, NULL, NULL)) `t` (`EMPNO`, `ENAME`, `JOB`)\n"
               + "WHERE 1 = 0";
       verifyPPLToSparkSQL(root, expectedSparkSql);
     }
 
     List<String> emptySourceWithRightOrFullJoinPPLs =
         Arrays.asList(
-            "source=EMP | append [ | where DEPTNO = 10 | right join on ENAME = DNAME DEPT ]",
-            "source=EMP | append [ | where DEPTNO = 10 | full join on ENAME = DNAME DEPT ]");
+            "source=EMP | fields EMPNO, ENAME, JOB | append [ | where DEPTNO = 10 | right join on"
+                + " ENAME = DNAME DEPT ]",
+            "source=EMP | fields EMPNO, ENAME, JOB | append [ | where DEPTNO = 10 | full join on"
+                + " ENAME = DNAME DEPT ]");
 
     for (String ppl : emptySourceWithRightOrFullJoinPPLs) {
       RelNode root = getRelNode(ppl);
       String expectedLogical =
           "LogicalUnion(all=[true])\n"
-              + "  LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4],"
-              + " SAL=[$5], COMM=[$6], DEPTNO=[$7], DEPTNO0=[null:TINYINT],"
+              + "  LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], DEPTNO=[null:TINYINT],"
               + " DNAME=[null:VARCHAR(14)], LOC=[null:VARCHAR(13)])\n"
               + "    LogicalTableScan(table=[[scott, EMP]])\n"
               + "  LogicalProject(EMPNO=[null:SMALLINT], ENAME=[null:VARCHAR(10)],"
-              + " JOB=[null:VARCHAR(9)], MGR=[null:SMALLINT], HIREDATE=[null:DATE],"
-              + " SAL=[null:DECIMAL(7, 2)], COMM=[null:DECIMAL(7, 2)], DEPTNO=[null:TINYINT],"
-              + " DEPTNO0=[$0], DNAME=[$1], LOC=[$2])\n"
+              + " JOB=[null:VARCHAR(9)], DEPTNO=[$0], DNAME=[$1], LOC=[$2])\n"
               + "    LogicalTableScan(table=[[scott, DEPT]])\n";
       verifyLogical(root, expectedLogical);
 
       String expectedSparkSql =
-          "SELECT `EMPNO`, `ENAME`, `JOB`, `MGR`, `HIREDATE`, `SAL`, `COMM`, `DEPTNO`, CAST(NULL AS"
-              + " TINYINT) `DEPTNO0`, CAST(NULL AS STRING) `DNAME`, CAST(NULL AS STRING) `LOC`\n"
+          "SELECT `EMPNO`, `ENAME`, `JOB`, CAST(NULL AS TINYINT) `DEPTNO`, CAST(NULL AS STRING)"
+              + " `DNAME`, CAST(NULL AS STRING) `LOC`\n"
               + "FROM `scott`.`EMP`\n"
               + "UNION ALL\n"
               + "SELECT CAST(NULL AS SMALLINT) `EMPNO`, CAST(NULL AS STRING) `ENAME`, CAST(NULL AS"
-              + " STRING) `JOB`, CAST(NULL AS SMALLINT) `MGR`, CAST(NULL AS DATE) `HIREDATE`,"
-              + " CAST(NULL AS DECIMAL(7, 2)) `SAL`, CAST(NULL AS DECIMAL(7, 2)) `COMM`, CAST(NULL"
-              + " AS TINYINT) `DEPTNO`, `DEPTNO` `DEPTNO0`, `DNAME`, `LOC`\n"
+              + " STRING) `JOB`, `DEPTNO`, `DNAME`, `LOC`\n"
               + "FROM `scott`.`DEPT`";
       verifyPPLToSparkSQL(root, expectedSparkSql);
     }
@@ -172,27 +176,27 @@ public void testAppendEmptySourceWithJoin() {
   @Test
   public void testAppendDifferentIndex() {
     String ppl =
-        "source=EMP | fields EMPNO, DEPTNO | append [ source=DEPT | fields DEPTNO, DNAME | where"
+        "source=EMP | fields EMPNO, ENAME | append [ source=DEPT | fields DEPTNO, DNAME | where"
             + " DEPTNO = 20 ]";
     RelNode root = getRelNode(ppl);
     String expectedLogical =
         "LogicalUnion(all=[true])\n"
-            + "  LogicalProject(EMPNO=[$0], DEPTNO=[$7], DEPTNO0=[null:TINYINT],"
+            + "  LogicalProject(EMPNO=[$0], ENAME=[$1], DEPTNO=[null:TINYINT],"
             + " DNAME=[null:VARCHAR(14)])\n"
             + "    LogicalTableScan(table=[[scott, EMP]])\n"
-            + "  LogicalProject(EMPNO=[null:SMALLINT], DEPTNO=[null:TINYINT], DEPTNO0=[$0],"
+            + "  LogicalProject(EMPNO=[null:SMALLINT], ENAME=[null:VARCHAR(10)], DEPTNO=[$0],"
             + " DNAME=[$1])\n"
             + "    LogicalFilter(condition=[=($0, 20)])\n"
             + "      LogicalProject(DEPTNO=[$0], DNAME=[$1])\n"
             + "        LogicalTableScan(table=[[scott, DEPT]])\n";
     verifyLogical(root, expectedLogical);
 
     String expectedSparkSql =
-        "SELECT `EMPNO`, `DEPTNO`, CAST(NULL AS TINYINT) `DEPTNO0`, CAST(NULL AS STRING) `DNAME`\n"
+        "SELECT `EMPNO`, `ENAME`, CAST(NULL AS TINYINT) `DEPTNO`, CAST(NULL AS STRING) `DNAME`\n"
             + "FROM `scott`.`EMP`\n"
             + "UNION ALL\n"
-            + "SELECT CAST(NULL AS SMALLINT) `EMPNO`, CAST(NULL AS TINYINT) `DEPTNO`, `DEPTNO`"
-            + " `DEPTNO0`, `DNAME`\n"
+            + "SELECT CAST(NULL AS SMALLINT) `EMPNO`, CAST(NULL AS STRING) `ENAME`, `DEPTNO`,"
+            + " `DNAME`\n"
             + "FROM (SELECT `DEPTNO`, `DNAME`\n"
             + "FROM `scott`.`DEPT`) `t0`\n"
             + "WHERE `DEPTNO` = 20";
@@ -227,22 +231,9 @@ public void testAppendWithMergedColumns() {
   public void testAppendWithConflictTypeColumn() {
     String ppl =
         "source=EMP | fields DEPTNO | append [ source=EMP | fields DEPTNO | eval DEPTNO = 20 ]";
-    RelNode root = getRelNode(ppl);
-    String expectedLogical =
-        "LogicalUnion(all=[true])\n"
-            + "  LogicalProject(DEPTNO=[$7], DEPTNO0=[null:INTEGER])\n"
-            + "    LogicalTableScan(table=[[scott, EMP]])\n"
-            + "  LogicalProject(DEPTNO=[null:TINYINT], DEPTNO0=[20])\n"
-            + "    LogicalTableScan(table=[[scott, EMP]])\n";
-    verifyLogical(root, expectedLogical);
-    verifyResultCount(root, 28);
-
-    String expectedSparkSql =
-        "SELECT `DEPTNO`, CAST(NULL AS INTEGER) `DEPTNO0`\n"
-            + "FROM `scott`.`EMP`\n"
-            + "UNION ALL\n"
-            + "SELECT CAST(NULL AS TINYINT) `DEPTNO`, 20 `DEPTNO0`\n"
-            + "FROM `scott`.`EMP`";
-    verifyPPLToSparkSQL(root, expectedSparkSql);
+    Exception exception =
+        Assert.assertThrows(IllegalArgumentException.class, () -> getRelNode(ppl));
+    verifyErrorMessageContains(
+        exception, "Schema unification failed: field 'DEPTNO' has conflicting types");
   }
 }

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

@@ -142,30 +142,28 @@ public void testMultisearchCrossIndices() {
     // Test multisearch with different tables (indices)
     String ppl =
         "| multisearch [search source=EMP | where DEPTNO = 10 | fields EMPNO, ENAME,"
-            + " DEPTNO] [search source=DEPT | where DEPTNO = 10 | fields DEPTNO, DNAME | eval EMPNO"
-            + " = DEPTNO, ENAME = DNAME]";
+            + " JOB] [search source=DEPT | where DEPTNO = 10 | fields DEPTNO, DNAME, LOC]";
     RelNode root = getRelNode(ppl);
     String expectedLogical =
         "LogicalUnion(all=[true])\n"
-            + "  LogicalProject(EMPNO=[$0], ENAME=[$1], DEPTNO=[$7], DEPTNO0=[null:TINYINT],"
-            + " DNAME=[null:VARCHAR(14)], EMPNO0=[null:TINYINT], ENAME0=[null:VARCHAR(14)])\n"
+            + "  LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], DEPTNO=[null:TINYINT],"
+            + " DNAME=[null:VARCHAR(14)], LOC=[null:VARCHAR(13)])\n"
             + "    LogicalFilter(condition=[=($7, 10)])\n"
             + "      LogicalTableScan(table=[[scott, EMP]])\n"
             + "  LogicalProject(EMPNO=[null:SMALLINT], ENAME=[null:VARCHAR(10)],"
-            + " DEPTNO=[null:TINYINT], DEPTNO0=[$0], DNAME=[$1], EMPNO0=[$0], ENAME0=[$1])\n"
+            + " JOB=[null:VARCHAR(9)], DEPTNO=[$0], DNAME=[$1], LOC=[$2])\n"
             + "    LogicalFilter(condition=[=($0, 10)])\n"
             + "      LogicalTableScan(table=[[scott, DEPT]])\n";
     verifyLogical(root, expectedLogical);
 
     String expectedSparkSql =
-        "SELECT `EMPNO`, `ENAME`, `DEPTNO`, CAST(NULL AS TINYINT) `DEPTNO0`, CAST(NULL AS STRING)"
-            + " `DNAME`, CAST(NULL AS TINYINT) `EMPNO0`, CAST(NULL AS STRING) `ENAME0`\n"
+        "SELECT `EMPNO`, `ENAME`, `JOB`, CAST(NULL AS TINYINT) `DEPTNO`, CAST(NULL AS STRING)"
+            + " `DNAME`, CAST(NULL AS STRING) `LOC`\n"
             + "FROM `scott`.`EMP`\n"
             + "WHERE `DEPTNO` = 10\n"
             + "UNION ALL\n"
             + "SELECT CAST(NULL AS SMALLINT) `EMPNO`, CAST(NULL AS STRING) `ENAME`, CAST(NULL AS"
-            + " TINYINT) `DEPTNO`, `DEPTNO` `DEPTNO0`, `DNAME`, `DEPTNO` `EMPNO0`, `DNAME`"
-            + " `ENAME0`\n"
+            + " STRING) `JOB`, `DEPTNO`, `DNAME`, `LOC`\n"
             + "FROM `scott`.`DEPT`\n"
             + "WHERE `DEPTNO` = 10";
     verifyPPLToSparkSQL(root, expectedSparkSql);