Allow PPD when subject is not a column with grouping sets present (Zh…

…ihua Deng, reviewed by Jesus Camacho Rodriguez)

Closes #1255

ql/src/java/org/apache/hadoop/hive/ql/ppd/OpProcFactory.java

@@ -792,40 +792,40 @@ public Object process(Node nd, Stack<Node> stack, NodeProcessorCtx procCtx,
       return null;
     }
 
+    private void getGBYKeyPosFromExpr(ExprNodeDesc expr, List<ExprNodeDesc> groupByKeys,
+        List<Integer> gbyKeyPos) {
+      for (int i = 0; i < groupByKeys.size(); i++) {
+        if (groupByKeys.get(i).isSame(expr)) {
+          gbyKeyPos.add(i);
+          return;
+        }
+      }
+      if (expr.getChildren() != null) {
+        for (int i = 0; i < expr.getChildren().size(); i++) {
+          getGBYKeyPosFromExpr(expr.getChildren().get(i), groupByKeys, gbyKeyPos);
+        }
+      }
+    }
+
     private boolean canPredPushdown(ExprNodeDesc expr, List<ExprNodeDesc> groupByKeys,
         FastBitSet[] bitSets, int groupingSetPosition) {
-      List<ExprNodeDesc> columns = new ArrayList<ExprNodeDesc>();
-      extractCols(expr, columns);
-      for (ExprNodeDesc col : columns) {
-        int index = groupByKeys.indexOf(col);
-        assert index >= 0;
+      List<Integer> gbyKeyPos = new ArrayList<Integer>();
+      getGBYKeyPosFromExpr(expr, groupByKeys, gbyKeyPos);
+      // gbyKeysInExpr can be empty, maybe the expr is a boolean constant, let the expr push down
+      for (Integer pos : gbyKeyPos) {
         for (FastBitSet bitset : bitSets) {
           int keyPos = bitset.nextClearBit(0);
-          while (keyPos < groupingSetPosition && keyPos != index) {
+          while (keyPos < groupingSetPosition && keyPos != pos) {
             keyPos = bitset.nextClearBit(keyPos + 1);
           }
-          // If the column has not be found in grouping sets, the expr should not be pushed down
-          if (keyPos != index) {
+          // If the gbyKey has not be found in grouping sets, the expr should not be pushed down
+          if (keyPos != pos) {
             return false;
           }
         }
       }
       return true;
     }
-
-    // Extract columns from expression
-    private void extractCols(ExprNodeDesc expr, List<ExprNodeDesc> columns) {
-      if (expr instanceof ExprNodeColumnDesc) {
-        columns.add(expr);
-      }
-
-      if (expr instanceof ExprNodeGenericFuncDesc) {
-        List<ExprNodeDesc> children = expr.getChildren();
-        for (int i = 0; i < children.size(); ++i) {
-          extractCols(children.get(i), columns);
-        }
-      }
-    }
   }
 
   /**

ql/src/test/queries/clientpositive/groupby_grouping_sets_pushdown1.q

@@ -39,4 +39,56 @@ SELECT * FROM (
 SELECT a, b, sum(s)
 FROM T1
 GROUP BY a, b GROUPING SETS ((a), (a, b))
-) t WHERE b IS NULL;
+) t WHERE b IS NULL;
+
+EXPLAIN EXTENDED SELECT * FROM (
+SELECT upper(a) x, b, sum(s)
+FROM T1
+GROUP BY a, b GROUPING SETS ((a), (a, b))
+) t WHERE x in ("AAA", "BBB");
+
+SELECT * FROM (
+SELECT upper(a) x, b, sum(s)
+FROM T1
+GROUP BY a, b GROUPING SETS ((a), (a, b))
+) t WHERE x in ('AAA', 'BBB');
+
+EXPLAIN EXTENDED SELECT a, b, sum(s)
+FROM T1
+GROUP BY a, b GROUPING SETS ((a), (a, b))
+HAVING upper(a) = 'AAA' AND 1 != 1;
+
+SELECT a, b, sum(s)
+FROM T1
+GROUP BY a, b GROUPING SETS ((a), (a, b))
+HAVING upper(a) = 'AAA' AND 1 != 1;
+
+EXPLAIN EXTENDED SELECT a, b, sum(s)
+FROM T1
+GROUP BY a, b GROUPING SETS ((), (a), (a, b))
+HAVING upper(a) = 'AAA' AND sum(s) > 100;
+
+SELECT a, b, sum(s)
+FROM T1
+GROUP BY a, b GROUPING SETS ((), (a), (a, b))
+HAVING upper(a) = 'AAA' AND sum(s) > 100;
+
+EXPLAIN EXTENDED SELECT upper(a), b, sum(s)
+FROM T1
+GROUP BY upper(a), b GROUPING SETS ((upper(a)), (upper(a), b))
+HAVING upper(a) = 'AAA' AND sum(s) > 100;
+
+SELECT upper(a), b, sum(s)
+FROM T1
+GROUP BY upper(a), b GROUPING SETS ((upper(a)), (upper(a), b))
+HAVING upper(a) = 'AAA' AND sum(s) > 100;
+
+EXPLAIN EXTENDED SELECT a, b, sum(s)
+FROM T1
+GROUP BY a, b GROUPING SETS ((b), (a, b))
+HAVING sum(s) > 100 and a IS NOT NULL AND upper(b) = 'BBB';
+
+SELECT a, b, sum(s)
+FROM T1
+GROUP BY a, b GROUPING SETS ((b), (a, b))
+HAVING sum(s) > 100 and a IS NOT NULL AND upper(b) = 'BBB';