HIVE-24167: Compilation fails due to equivalence mapping violation when CTE materialization is enabled

ql/src/java/org/apache/hadoop/hive/ql/parse/TezCompiler.java

@@ -26,7 +26,6 @@
 import java.util.HashMap;
 import java.util.HashSet;
 import java.util.IdentityHashMap;
-import java.util.Iterator;
 import java.util.LinkedHashMap;
 import java.util.LinkedHashSet;
 import java.util.LinkedList;
@@ -100,7 +99,6 @@
 import org.apache.hadoop.hive.ql.optimizer.SharedWorkOptimizer;
 import org.apache.hadoop.hive.ql.optimizer.SortedDynPartitionOptimizer;
 import org.apache.hadoop.hive.ql.optimizer.topnkey.TopNKeyProcessor;
-import org.apache.hadoop.hive.ql.optimizer.calcite.reloperators.HiveFilter;
 import org.apache.hadoop.hive.ql.optimizer.correlation.ReduceSinkDeDuplication;
 import org.apache.hadoop.hive.ql.optimizer.topnkey.TopNKeyPushdownProcessor;
 import org.apache.hadoop.hive.ql.optimizer.correlation.ReduceSinkJoinDeDuplication;
@@ -117,7 +115,6 @@
 import org.apache.hadoop.hive.ql.optimizer.physical.SerializeFilter;
 import org.apache.hadoop.hive.ql.optimizer.physical.StageIDsRearranger;
 import org.apache.hadoop.hive.ql.optimizer.physical.Vectorizer;
-import org.apache.hadoop.hive.ql.optimizer.signature.OpTreeSignature;
 import org.apache.hadoop.hive.ql.optimizer.stats.annotation.AnnotateWithStatistics;
 import org.apache.hadoop.hive.ql.plan.AggregationDesc;
 import org.apache.hadoop.hive.ql.plan.AppMasterEventDesc;
@@ -139,7 +136,6 @@
 import org.apache.hadoop.hive.ql.plan.TezWork;
 import org.apache.hadoop.hive.ql.plan.mapper.AuxOpTreeSignature;
 import org.apache.hadoop.hive.ql.plan.mapper.PlanMapper;
-import org.apache.hadoop.hive.ql.plan.mapper.PlanMapper.EquivGroup;
 import org.apache.hadoop.hive.ql.session.SessionState;
 import org.apache.hadoop.hive.ql.session.SessionState.LogHelper;
 import org.apache.hadoop.hive.ql.stats.OperatorStats;
@@ -991,35 +987,6 @@ private void removeSemiJoinIfNoStats(OptimizeTezProcContext procCtx)
     ogw.startWalking(topNodes, null);
   }
 
-  private static class CollectAll implements SemanticNodeProcessor {
-    private PlanMapper planMapper;
-
-    @Override
-    public Object process(Node nd, Stack<Node> stack, NodeProcessorCtx procCtx, Object... nodeOutputs)
-        throws SemanticException {
-      ParseContext pCtx = ((OptimizeTezProcContext) procCtx).parseContext;
-      planMapper = pCtx.getContext().getPlanMapper();
-      FilterOperator fop = (FilterOperator) nd;
-      OpTreeSignature sig = planMapper.getSignatureOf(fop);
-      List<EquivGroup> ar = getGroups(planMapper, HiveFilter.class);
-
-
-      return nd;
-    }
-
-    private List<EquivGroup> getGroups(PlanMapper planMapper2, Class<HiveFilter> class1) {
-      Iterator<EquivGroup> it = planMapper.iterateGroups();
-      List<EquivGroup> ret = new ArrayList<PlanMapper.EquivGroup>();
-      while (it.hasNext()) {
-        EquivGroup g = it.next();
-        if (g.getAll(class1).size() > 0) {
-          ret.add(g);
-        }
-      }
-      return ret;
-    }
-  }
-
   private static class MarkRuntimeStatsAsIncorrect implements SemanticNodeProcessor {
 
     private PlanMapper planMapper;

ql/src/java/org/apache/hadoop/hive/ql/plan/mapper/PlanMapper.java

@@ -37,13 +37,17 @@
 import org.apache.hadoop.hive.ql.optimizer.signature.OpTreeSignatureFactory;
 import com.google.common.annotations.VisibleForTesting;
 import com.google.common.collect.Sets;
+import org.apache.hadoop.hive.ql.stats.OperatorStats;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
 
 /**
  * Enables to connect related objects to eachother.
  *
  * Most importantly it aids to connect Operators to OperatorStats and probably RelNodes.
  */
 public class PlanMapper {
+  private static final Logger LOG = LoggerFactory.getLogger(PlanMapper.class);
 
   Set<EquivGroup> groups = new HashSet<>();
   private Map<Object, EquivGroup> objectMap = new CompositeMap<>(OpTreeSignature.class, AuxOpTreeSignature.class);
@@ -217,7 +221,9 @@ private void link(Object o1, Object o2, boolean mayMerge) {
     }
     if (mGroups.size() > 1) {
       if (!mayMerge) {
-        throw new RuntimeException("equivalence mapping violation");
+        LOG.warn("Illegally linking {} and {}", o1, o2);
+        mGroups.forEach(g -> g.add(new OperatorStats.IncorrectRuntimeStatsMarker()));
+        return;
       }
       EquivGroup newGrp = new EquivGroup();
       newGrp.add(o1);
@@ -256,20 +262,15 @@ public <T> List<T> getAll(Class<T> clazz) {
     return ret;
   }
 
-  public void runMapper(GroupTransformer mapper) {
-    for (EquivGroup equivGroup : groups) {
-      mapper.map(equivGroup);
-    }
-  }
-
-  public <T> List<T> lookupAll(Class<T> clazz, Object key) {
+  private <T> List<T> lookupAll(Class<T> clazz, Object key) {
     EquivGroup group = objectMap.get(key);
     if (group == null) {
       throw new NoSuchElementException(Objects.toString(key));
     }
     return group.getAll(clazz);
   }
 
+  @VisibleForTesting
   public <T> T lookup(Class<T> clazz, Object key) {
     List<T> all = lookupAll(clazz, key);
     if (all.size() != 1) {
@@ -279,7 +280,6 @@ public <T> T lookup(Class<T> clazz, Object key) {
     return all.get(0);
   }
 
-  @VisibleForTesting
   public Iterator<EquivGroup> iterateGroups() {
     return groups.iterator();
 

ql/src/test/queries/clientpositive/perf/cbo_query14.q

@@ -1,5 +1,4 @@
 set hive.optimize.cte.suggester.class=org.apache.hadoop.hive.ql.optimizer.calcite.CommonTableExpressionPrintSuggester;
---! qt:disabled:HIVE-24167
 set hive.mapred.mode=nonstrict;
 -- start query 1 in stream 0 using template query14.tpl and seed 1819994127
 explain cbo

ql/src/test/queries/clientpositive/perf/query14.q

@@ -1,4 +1,3 @@
---! qt:disabled:HIVE-24167
 set hive.mapred.mode=nonstrict;
 -- start query 1 in stream 0 using template query14.tpl and seed 1819994127
 explain