[CALCITE-6038] Remove 'ORDER BY ... LIMIT n' when input has at most o…

…ne row, n >= 1, and there is no 'OFFSET' clause

core/src/main/java/org/apache/calcite/rel/rules/SortRemoveRedundantRule.java

@@ -26,11 +26,12 @@
 
 import java.util.Optional;
 
-/** Rule that removes redundant {@code Order By} or {@code Limit} when its input RelNode's
- * max row count is less than or equal to specified row count.All of them
- * are represented by {@link Sort}
+/**
+ * Rule that removes redundant {@code Order By} or {@code Limit}
+ * when its input RelNode's max row count is less than or equal to specified row count.
+ * All of them are represented by {@link Sort}
  *
- * <p> If a {@code Sort} is pure order by,and its offset is null,when its input RelNode's
+ * <p> If a {@code Sort} is order by,and its offset is null,when its input RelNode's
  * max row count is less than or equal to 1,then we could remove the redundant sort.
  *
  * <p> For example:
@@ -43,9 +44,19 @@
  *  select max(totalprice) from orders}
  *  </pre></blockquote>
  *
+ * <p> For example:
+ * <blockquote><pre>{@code
+ *  SELECT count(*) FROM orders ORDER BY 1 LIMIT 10 }
+ *  </pre></blockquote>
+ *
+ * <p> could be converted to
+ * <blockquote><pre>{@code
+ *  SELECT count(*) FROM orders}
+ *  </pre></blockquote>
+ *
  * <p> If a {@code Sort} is pure limit,and its offset is null, when its input
  * RelNode's max row count is less than or equal to the limit's fetch,then we could
- * remove the redundant sort.
+ * remove the redundant {@code Limit}.
  *
  * <p> For example:
  * <blockquote><pre>{@code
@@ -81,8 +92,10 @@ protected SortRemoveRedundantRule(final SortRemoveRedundantRule.Config config) {
     final Double inputMaxRowCount = call.getMetadataQuery().getMaxRowCount(sort.getInput());
 
     // Get the target max row count with sort's semantics.
-    // If sort is pure order by, the target max row count is 1.
+    // If sort is 'order by x' or 'order by x limit n', the target max row count is 1.
     // If sort is pure limit, the target max row count is the limit's fetch.
+    // If the limit's fetch is 0, we could use CoreRules.SORT_FETCH_ZERO_INSTANCE to deal with it,
+    // so we don't need to deal with it in this rule.
     final Optional<Double> targetMaxRowCount = getSortInputSpecificMaxRowCount(sort);
 
     if (!targetMaxRowCount.isPresent()) {
@@ -97,13 +110,15 @@ protected SortRemoveRedundantRule(final SortRemoveRedundantRule.Config config) {
   }
 
   private Optional<Double> getSortInputSpecificMaxRowCount(Sort sort) {
-    // If the sort is pure limit, the specific max row count is limit's fetch.
-    if (RelOptUtil.isPureLimit(sort)) {
-      final double limit =
-          sort.fetch instanceof RexLiteral ? RexLiteral.intValue(sort.fetch) : -1D;
-      return Optional.of(limit);
+    if (RelOptUtil.isLimit(sort)) {
+      final double fetch =
+          sort.fetch instanceof RexLiteral ? RexLiteral.intValue(sort.fetch) : 0D;
+      if (fetch == 0D) {
+        return Optional.empty();
+      }
+      // If sort is 'order by x limit n', the target max row count is 1.
+      return RelOptUtil.isOrder(sort) ? Optional.of(1D) : Optional.of(fetch);
     } else if (RelOptUtil.isPureOrder(sort)) {
-      // If the sort is pure order by, the specific max row count is 1.
       return Optional.of(1D);
     }
     return Optional.empty();

core/src/test/java/org/apache/calcite/test/RelOptRulesTest.java

@@ -1368,6 +1368,30 @@ private void checkSemiOrAntiJoinProjectTranspose(JoinRelType type) {
         .checkUnchanged();
   }
 
+  /** Test case for
+   * <a href="https://issues.apache.org/jira/browse/CALCITE-6038">[CALCITE-6038]
+   * Remove 'ORDER BY ... LIMIT n' when input has at most one row, n >= 1,
+   * and there is no 'OFFSET' clause</a>. */
+  @Test void testSortRemoveWhenIsOrderAndLimit() {
+    final String sql = "SELECT count(*) FROM sales.emp ORDER BY 1 LIMIT 10";
+    sql(sql)
+        .withRule(CoreRules.SORT_REMOVE_REDUNDANT)
+        .check();
+  }
+
+  /** Test case for
+   * <a href="https://issues.apache.org/jira/browse/CALCITE-6038">[CALCITE-6038]
+   * Remove 'ORDER BY ... LIMIT n' when input has at most one row, n >= 1,
+   * and there is no 'OFFSET' clause</a>. */
+  @Test void testSortNotRemoveWhenIsOrderAndLimit() {
+    final String sql = "select * from\n"
+        + "(SELECT * FROM sales.emp limit 100)\n"
+        + "ORDER BY 1 LIMIT 10";
+    sql(sql)
+        .withRule(CoreRules.SORT_REMOVE_REDUNDANT)
+        .checkUnchanged();
+  }
+
   /** Tests that an {@link EnumerableLimit} and {@link EnumerableSort} are
    * replaced by an {@link EnumerableLimitSort}, per
    * <a href="https://issues.apache.org/jira/browse/CALCITE-3920">[CALCITE-3920]

core/src/test/resources/org/apache/calcite/test/RelOptRulesTest.xml

@@ -14100,6 +14100,22 @@ LogicalProject(DEPTNO=[$0], EMPNO=[$2])
     LogicalJoin(condition=[=($0, $9)], joinType=[left])
       LogicalTableScan(table=[[CATALOG, SALES, DEPT]])
       LogicalTableScan(table=[[CATALOG, SALES, EMP]])
+]]>
+    </Resource>
+  </TestCase>
+  <TestCase name="testSortNotRemoveWhenIsOrderAndLimit">
+    <Resource name="sql">
+      <![CDATA[select * from
+(SELECT * FROM sales.emp limit 100)
+ORDER BY 1 LIMIT 10]]>
+    </Resource>
+    <Resource name="planBefore">
+      <![CDATA[
+LogicalSort(sort0=[$0], dir0=[ASC], fetch=[10])
+  LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4], SAL=[$5], COMM=[$6], DEPTNO=[$7], SLACKER=[$8])
+    LogicalSort(fetch=[100])
+      LogicalProject(EMPNO=[$0], ENAME=[$1], JOB=[$2], MGR=[$3], HIREDATE=[$4], SAL=[$5], COMM=[$6], DEPTNO=[$7], SLACKER=[$8])
+        LogicalTableScan(table=[[CATALOG, SALES, EMP]])
 ]]>
     </Resource>
   </TestCase>
@@ -14295,6 +14311,26 @@ LogicalSort(fetch=[10])
       <![CDATA[
 LogicalProject(T1=[$0])
   LogicalValues(tuples=[[{ 1 }, { 2 }, { 3 }, { 4 }, { 5 }, { 6 }]])
+]]>
+    </Resource>
+  </TestCase>
+  <TestCase name="testSortRemoveWhenIsOrderAndLimit">
+    <Resource name="sql">
+      <![CDATA[SELECT count(*) FROM sales.emp ORDER BY 1 LIMIT 10]]>
+    </Resource>
+    <Resource name="planBefore">
+      <![CDATA[
+LogicalSort(sort0=[$0], dir0=[ASC], fetch=[10])
+  LogicalAggregate(group=[{}], EXPR$0=[COUNT()])
+    LogicalProject($f0=[0])
+      LogicalTableScan(table=[[CATALOG, SALES, EMP]])
+]]>
+    </Resource>
+    <Resource name="planAfter">
+      <![CDATA[
+LogicalAggregate(group=[{}], EXPR$0=[COUNT()])
+  LogicalProject($f0=[0])
+    LogicalTableScan(table=[[CATALOG, SALES, EMP]])
 ]]>
     </Resource>
   </TestCase>