HIVE-29368: only increment ndv by one inextractNDVGroupingColumns() if it is "known"

Author: konstantinb

ql/src/java/org/apache/hadoop/hive/ql/stats/StatsUtils.java

@@ -2087,7 +2087,8 @@ private static List<Long> extractNDVGroupingColumns(List<ColStatistics> colStats
     for (ColStatistics cs : colStats) {
       if (cs != null) {
         long ndv = cs.getCountDistint();
-        if (cs.getNumNulls() > 0) {
+        // Only increment ndv value if it is "known"
+        if (ndv > 0 && cs.getNumNulls() > 0) {
           ndv = StatsUtils.safeAdd(ndv, 1);
         }
         ndvValues.add(ndv);

ql/src/test/results/clientpositive/llap/subquery_notin.q.out

@@ -1536,13 +1536,13 @@ STAGE PLANS:
                         minReductionHashAggr: 0.99
                         mode: hash
                         outputColumnNames: _col0
-                        Statistics: Num rows: 1 Data size: 87 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 83 Data size: 87 Basic stats: COMPLETE Column stats: COMPLETE
                         Reduce Output Operator
                           key expressions: _col0 (type: string)
                           null sort order: z
                           sort order: +
                           Map-reduce partition columns: _col0 (type: string)
-                          Statistics: Num rows: 1 Data size: 87 Basic stats: COMPLETE Column stats: COMPLETE
+                          Statistics: Num rows: 83 Data size: 87 Basic stats: COMPLETE Column stats: COMPLETE
             Execution mode: vectorized, llap
             LLAP IO: all inputs
         Reducer 2
@@ -1555,11 +1555,11 @@ STAGE PLANS:
                   0 _col0 (type: string)
                   1 _col0 (type: string)
                 outputColumnNames: _col0, _col2
-                Statistics: Num rows: 167 Data size: 14537 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 230 Data size: 20270 Basic stats: COMPLETE Column stats: COMPLETE
                 Reduce Output Operator
                   null sort order:
                   sort order:
-                  Statistics: Num rows: 167 Data size: 14537 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 230 Data size: 20270 Basic stats: COMPLETE Column stats: COMPLETE
                   value expressions: _col0 (type: string), _col2 (type: boolean)
         Reducer 3
             Execution mode: llap
@@ -1571,21 +1571,21 @@ STAGE PLANS:
                   0
                   1
                 outputColumnNames: _col0, _col2, _col3, _col4
-                Statistics: Num rows: 167 Data size: 17209 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 230 Data size: 23950 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: _col0 (type: string), _col3 (type: bigint), _col4 (type: bigint), _col2 (type: boolean)
                   outputColumnNames: _col0, _col1, _col2, _col4
-                  Statistics: Num rows: 167 Data size: 17209 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 230 Data size: 23950 Basic stats: COMPLETE Column stats: COMPLETE
                   Filter Operator
                     predicate: ((_col1 = 0L) or (_col4 is null and (_col2 >= _col1) and _col0 is not null)) (type: boolean)
-                    Statistics: Num rows: 167 Data size: 17209 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 230 Data size: 23950 Basic stats: COMPLETE Column stats: COMPLETE
                     Select Operator
                       expressions: _col0 (type: string)
                       outputColumnNames: _col0
-                      Statistics: Num rows: 167 Data size: 14529 Basic stats: COMPLETE Column stats: COMPLETE
+                      Statistics: Num rows: 230 Data size: 20010 Basic stats: COMPLETE Column stats: COMPLETE
                       File Output Operator
                         compressed: false
-                        Statistics: Num rows: 167 Data size: 14529 Basic stats: COMPLETE Column stats: COMPLETE
+                        Statistics: Num rows: 230 Data size: 20010 Basic stats: COMPLETE Column stats: COMPLETE
                         table:
                             input format: org.apache.hadoop.mapred.SequenceFileInputFormat
                             output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
@@ -1610,17 +1610,17 @@ STAGE PLANS:
                 keys: KEY._col0 (type: string)
                 mode: mergepartial
                 outputColumnNames: _col0
-                Statistics: Num rows: 1 Data size: 87 Basic stats: COMPLETE Column stats: COMPLETE
+                Statistics: Num rows: 41 Data size: 87 Basic stats: COMPLETE Column stats: COMPLETE
                 Select Operator
                   expressions: _col0 (type: string), true (type: boolean)
                   outputColumnNames: _col0, _col1
-                  Statistics: Num rows: 1 Data size: 91 Basic stats: COMPLETE Column stats: COMPLETE
+                  Statistics: Num rows: 41 Data size: 251 Basic stats: COMPLETE Column stats: COMPLETE
                   Reduce Output Operator
                     key expressions: _col0 (type: string)
                     null sort order: z
                     sort order: +
                     Map-reduce partition columns: _col0 (type: string)
-                    Statistics: Num rows: 1 Data size: 91 Basic stats: COMPLETE Column stats: COMPLETE
+                    Statistics: Num rows: 41 Data size: 251 Basic stats: COMPLETE Column stats: COMPLETE
                     value expressions: _col1 (type: boolean)
 
   Stage: Stage-0