MERGE post-commit review

Review comments from Andres Freund

* Consolidate code into AfterTriggerGetTransitionTable()
* Rename nodeMerge.c to execMerge.c
* Rename nodeMerge.h to execMerge.h
* Move MERGE handling in ExecInitModifyTable()
  into a execMerge.c ExecInitMerge()
* Move mt_merge_subcommands flags into execMerge.h
* Rename opt_and_condition to opt_merge_when_and_condition
* Wordsmith various comments

Author: Pavan Deolasee
Reviewer: Simon Riggs

src/backend/commands/trigger.c

@@ -96,6 +96,12 @@ static HeapTuple ExecCallTriggerFunc(TriggerData *trigdata,
 					FmgrInfo *finfo,
 					Instrumentation *instr,
 					MemoryContext per_tuple_context);
+static Tuplestorestate *AfterTriggerGetTransitionTable(int event,
+					HeapTuple oldtup,
+					HeapTuple newtup,
+					TransitionCaptureState *transition_capture);
+static void TransitionTableAddTuple(HeapTuple heaptup, Tuplestorestate *tuplestore,
+					TupleConversionMap *map);
 static void AfterTriggerSaveEvent(EState *estate, ResultRelInfo *relinfo,
 					  int event, bool row_trigger,
 					  HeapTuple oldtup, HeapTuple newtup,
@@ -3846,6 +3852,14 @@ struct AfterTriggersTableData
 	bool		before_trig_done;	/* did we already queue BS triggers? */
 	bool		after_trig_done;	/* did we already queue AS triggers? */
 	AfterTriggerEventList after_trig_events;	/* if so, saved list pointer */
+
+	/*
+	 * We maintain separate transaction tables for UPDATE/INSERT/DELETE since
+	 * MERGE can run all three actions in a single statement. Note that UPDATE
+	 * needs both old and new transition tables whereas INSERT needs only new
+	 * and DELETE needs only old.
+	 */
+
 	/* "old" transition table for UPDATE, if any */
 	Tuplestorestate *old_upd_tuplestore;
 	/* "new" transition table for UPDATE, if any */
@@ -5716,6 +5730,84 @@ AfterTriggerPendingOnRel(Oid relid)
 	return false;
 }
 
+/*
+ * Get the transition table for the given event and depending on whether we are
+ * processing the old or the new tuple.
+ */
+static Tuplestorestate *
+AfterTriggerGetTransitionTable(int event,
+							   HeapTuple oldtup,
+							   HeapTuple newtup,
+							   TransitionCaptureState *transition_capture)
+{
+	Tuplestorestate	*tuplestore = NULL;
+	bool			delete_old_table = transition_capture->tcs_delete_old_table;
+	bool			update_old_table = transition_capture->tcs_update_old_table;
+	bool			update_new_table = transition_capture->tcs_update_new_table;
+	bool			insert_new_table = transition_capture->tcs_insert_new_table;;
+
+	/*
+	 * For INSERT events newtup should be non-NULL, for DELETE events
+	 * oldtup should be non-NULL, whereas for UPDATE events normally both
+	 * oldtup and newtup are non-NULL.  But for UPDATE events fired for
+	 * capturing transition tuples during UPDATE partition-key row
+	 * movement, oldtup is NULL when the event is for a row being inserted,
+	 * whereas newtup is NULL when the event is for a row being deleted.
+	 */
+	Assert(!(event == TRIGGER_EVENT_DELETE && delete_old_table &&
+				oldtup == NULL));
+	Assert(!(event == TRIGGER_EVENT_INSERT && insert_new_table &&
+				newtup == NULL));
+
+	/*
+	 * We're called either for the newtup or the oldtup, but not both at the
+	 * same time.
+	 */
+	Assert((oldtup != NULL) ^ (newtup != NULL));
+
+	if (oldtup != NULL)
+	{
+		if (event == TRIGGER_EVENT_DELETE && delete_old_table)
+			tuplestore = transition_capture->tcs_private->old_del_tuplestore;
+		else if (event == TRIGGER_EVENT_UPDATE && update_old_table)
+			tuplestore = transition_capture->tcs_private->old_upd_tuplestore;
+	}
+
+	if (newtup != NULL)
+	{
+		if (event == TRIGGER_EVENT_INSERT && insert_new_table)
+			tuplestore = transition_capture->tcs_private->new_ins_tuplestore;
+		else if (event == TRIGGER_EVENT_UPDATE && update_new_table)
+			tuplestore = transition_capture->tcs_private->new_upd_tuplestore;
+	}
+
+	return tuplestore;
+}
+
+/*
+ * Add the given heap tuple to the given tuplestore, applying the conversion
+ * map if necessary.
+ */
+static void
+TransitionTableAddTuple(HeapTuple heaptup, Tuplestorestate *tuplestore,
+						TupleConversionMap *map)
+{
+	/*
+	 * Nothing needs to be done if we don't have a tuplestore.
+	 */
+	if (tuplestore == NULL)
+		return;
+
+	if (map != NULL)
+	{
+		HeapTuple	converted = do_convert_tuple(heaptup, map);
+
+		tuplestore_puttuple(tuplestore, converted);
+		pfree(converted);
+	}
+	else
+		tuplestore_puttuple(tuplestore, heaptup);
+}
 
 /* ----------
  * AfterTriggerSaveEvent()
@@ -5777,95 +5869,37 @@ AfterTriggerSaveEvent(EState *estate, ResultRelInfo *relinfo,
 	{
 		HeapTuple	original_insert_tuple = transition_capture->tcs_original_insert_tuple;
 		TupleConversionMap *map = transition_capture->tcs_map;
-		bool		delete_old_table = transition_capture->tcs_delete_old_table;
-		bool		update_old_table = transition_capture->tcs_update_old_table;
-		bool		update_new_table = transition_capture->tcs_update_new_table;
-		bool		insert_new_table = transition_capture->tcs_insert_new_table;;
 
 		/*
-		 * For INSERT events newtup should be non-NULL, for DELETE events
+		 * Capture the old tuple in the appropriate transition table based on
-		 * oldtup should be non-NULL, whereas for UPDATE events normally both
+		 * the event.
-		 * oldtup and newtup are non-NULL.  But for UPDATE events fired for
-		 * capturing transition tuples during UPDATE partition-key row
-		 * movement, oldtup is NULL when the event is for a row being inserted,
-		 * whereas newtup is NULL when the event is for a row being deleted.
 		 */
-		Assert(!(event == TRIGGER_EVENT_DELETE && delete_old_table &&
+		if (oldtup != NULL)
-				 oldtup == NULL));
-		Assert(!(event == TRIGGER_EVENT_INSERT && insert_new_table &&
-				 newtup == NULL));
-
-		if (oldtup != NULL &&
-			(event == TRIGGER_EVENT_DELETE && delete_old_table))
 		{
-			Tuplestorestate *old_tuplestore;
+			Tuplestorestate *tuplestore =
-
+				AfterTriggerGetTransitionTable(event,
-			old_tuplestore = transition_capture->tcs_private->old_del_tuplestore;
+											   oldtup,
-
+											   NULL,
-			if (map != NULL)
+											   transition_capture);
-			{
+			TransitionTableAddTuple(oldtup, tuplestore, map);
-				HeapTuple	converted = do_convert_tuple(oldtup, map);
-
-				tuplestore_puttuple(old_tuplestore, converted);
-				pfree(converted);
 		}
-			else
+
-				tuplestore_puttuple(old_tuplestore, oldtup);
+		/*
-		}
+		 * Capture the new tuple in the appropriate transition table based on
-		if (oldtup != NULL &&
+		 * the event.
-			 (event == TRIGGER_EVENT_UPDATE && update_old_table))
+		 */
+		if (newtup != NULL)
 		{
-			Tuplestorestate *old_tuplestore;
+			Tuplestorestate *tuplestore =
-
+				AfterTriggerGetTransitionTable(event,
-			old_tuplestore = transition_capture->tcs_private->old_upd_tuplestore;
+											   NULL,
-
+											   newtup,
-			if (map != NULL)
+											   transition_capture);
-			{
-				HeapTuple	converted = do_convert_tuple(oldtup, map);
-
-				tuplestore_puttuple(old_tuplestore, converted);
-				pfree(converted);
-			}
-			else
-				tuplestore_puttuple(old_tuplestore, oldtup);
-		}
-		if (newtup != NULL &&
-			(event == TRIGGER_EVENT_INSERT && insert_new_table))
-		{
-			Tuplestorestate *new_tuplestore;
-
-			new_tuplestore = transition_capture->tcs_private->new_ins_tuplestore;
 
 			if (original_insert_tuple != NULL)
-				tuplestore_puttuple(new_tuplestore, original_insert_tuple);
+				tuplestore_puttuple(tuplestore, original_insert_tuple);
-			else if (map != NULL)
-			{
-				HeapTuple	converted = do_convert_tuple(newtup, map);
-
-				tuplestore_puttuple(new_tuplestore, converted);
-				pfree(converted);
-			}
 			else
-				tuplestore_puttuple(new_tuplestore, newtup);
+				TransitionTableAddTuple(newtup, tuplestore, map);
-		}
-		if (newtup != NULL &&
-			(event == TRIGGER_EVENT_UPDATE && update_new_table))
-		{
-			Tuplestorestate *new_tuplestore;
-
-			new_tuplestore = transition_capture->tcs_private->new_upd_tuplestore;
-
-			if (original_insert_tuple != NULL)
-				tuplestore_puttuple(new_tuplestore, original_insert_tuple);
-			else if (map != NULL)
-			{
-				HeapTuple	converted = do_convert_tuple(newtup, map);
-
-				tuplestore_puttuple(new_tuplestore, converted);
-				pfree(converted);
-			}
-			else
-				tuplestore_puttuple(new_tuplestore, newtup);
 		}
 
 		/*

src/backend/executor/Makefile

@@ -14,7 +14,7 @@ include $(top_builddir)/src/Makefile.global
 
 OBJS = execAmi.o execCurrent.o execExpr.o execExprInterp.o \
        execGrouping.o execIndexing.o execJunk.o \
-       execMain.o execParallel.o execPartition.o execProcnode.o \
+       execMain.o execMerge.o execParallel.o execPartition.o execProcnode.o \
        execReplication.o execScan.o execSRF.o execTuples.o \
        execUtils.o functions.o instrument.o nodeAppend.o nodeAgg.o \
        nodeBitmapAnd.o nodeBitmapOr.o \
@@ -22,7 +22,7 @@ OBJS = execAmi.o execCurrent.o execExpr.o execExprInterp.o \
        nodeCustom.o nodeFunctionscan.o nodeGather.o \
        nodeHash.o nodeHashjoin.o nodeIndexscan.o nodeIndexonlyscan.o \
        nodeLimit.o nodeLockRows.o nodeGatherMerge.o \
-       nodeMaterial.o nodeMergeAppend.o nodeMergejoin.o nodeMerge.o nodeModifyTable.o \
+       nodeMaterial.o nodeMergeAppend.o nodeMergejoin.o nodeModifyTable.o \
        nodeNestloop.o nodeProjectSet.o nodeRecursiveunion.o nodeResult.o \
        nodeSamplescan.o nodeSeqscan.o nodeSetOp.o nodeSort.o nodeUnique.o \
        nodeValuesscan.o \

src/backend/executor/README

@@ -39,13 +39,14 @@ ModifyTable node visits each of those rows and marks the row deleted.
 
 MERGE runs one generic plan that returns candidate target rows. Each row
 consists of a super-row that contains all the columns needed by any of the
-individual actions, plus a CTID and a TABLEOID junk columns. The CTID column is
+individual actions, plus CTID and TABLEOID junk columns. The CTID column is
 required to know if a matching target row was found or not and the TABLEOID
 column is needed to find the underlying target partition, in case when the
-target table is a partition table. If the CTID column is set we attempt to
+target table is a partition table. When a matching target tuple is found, the
-activate WHEN MATCHED actions, or if it is NULL then we will attempt to
+CTID column identifies the matching target tuple and we attempt to activate
-activate WHEN NOT MATCHED actions. Once we know which action is activated we
+WHEN MATCHED actions. If a matching tuple is not found, then CTID column is
-form the final result row and apply only those changes.
+NULL and we attempt to activate WHEN NOT MATCHED actions. Once we know which
+action is activated we form the final result row and apply only those changes.
 
 XXX a great deal more documentation needs to be written here...
 

src/backend/executor/execMerge.c

@@ -1,6 +1,6 @@
 /*-------------------------------------------------------------------------
  *
- * nodeMerge.c
+ * execMerge.c
  *	  routines to handle Merge nodes relating to the MERGE command
  *
  * Portions Copyright (c) 1996-2018, PostgreSQL Global Development Group
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  src/backend/executor/nodeMerge.c
+ *	  src/backend/executor/execMerge.c
  *
  *-------------------------------------------------------------------------
  */
@@ -22,7 +22,7 @@
 #include "executor/execPartition.h"
 #include "executor/executor.h"
 #include "executor/nodeModifyTable.h"
-#include "executor/nodeMerge.h"
+#include "executor/execMerge.h"
 #include "miscadmin.h"
 #include "nodes/nodeFuncs.h"
 #include "storage/bufmgr.h"
@@ -32,6 +32,110 @@
 #include "utils/rel.h"
 #include "utils/tqual.h"
 
+static void ExecMergeNotMatched(ModifyTableState *mtstate, EState *estate,
+								TupleTableSlot *slot);
+static bool ExecMergeMatched(ModifyTableState *mtstate, EState *estate,
+							 TupleTableSlot *slot, JunkFilter *junkfilter,
+							 ItemPointer tupleid);
+/*
+ * Perform MERGE.
+ */
+void
+ExecMerge(ModifyTableState *mtstate, EState *estate, TupleTableSlot *slot,
+		  JunkFilter *junkfilter, ResultRelInfo *resultRelInfo)
+{
+	ExprContext *econtext = mtstate->ps.ps_ExprContext;
+	ItemPointer tupleid;
+	ItemPointerData tuple_ctid;
+	bool		matched = false;
+	char		relkind;
+	Datum		datum;
+	bool		isNull;
+
+	relkind = resultRelInfo->ri_RelationDesc->rd_rel->relkind;
+	Assert(relkind == RELKIND_RELATION ||
+		   relkind == RELKIND_PARTITIONED_TABLE);
+
+	/*
+	 * Reset per-tuple memory context to free any expression evaluation
+	 * storage allocated in the previous cycle.
+	 */
+	ResetExprContext(econtext);
+
+	/*
+	 * We run a JOIN between the target relation and the source relation to
+	 * find a set of candidate source rows that has matching row in the target
+	 * table and a set of candidate source rows that does not have matching
+	 * row in the target table. If the join returns us a tuple with target
+	 * relation's tid set, that implies that the join found a matching row for
+	 * the given source tuple. This case triggers the WHEN MATCHED clause of
+	 * the MERGE. Whereas a NULL in the target relation's ctid column
+	 * indicates a NOT MATCHED case.
+	 */
+	datum = ExecGetJunkAttribute(slot, junkfilter->jf_junkAttNo, &isNull);
+
+	if (!isNull)
+	{
+		matched = true;
+		tupleid = (ItemPointer) DatumGetPointer(datum);
+		tuple_ctid = *tupleid;	/* be sure we don't free ctid!! */
+		tupleid = &tuple_ctid;
+	}
+	else
+	{
+		matched = false;
+		tupleid = NULL;			/* we don't need it for INSERT actions */
+	}
+
+	/*
+	 * If we are dealing with a WHEN MATCHED case, we execute the first action
+	 * for which the additional WHEN MATCHED AND quals pass. If an action
+	 * without quals is found, that action is executed.
+	 *
+	 * Similarly, if we are dealing with WHEN NOT MATCHED case, we look at the
+	 * given WHEN NOT MATCHED actions in sequence until one passes.
+	 *
+	 * Things get interesting in case of concurrent update/delete of the
+	 * target tuple. Such concurrent update/delete is detected while we are
+	 * executing a WHEN MATCHED action.
+	 *
+	 * A concurrent update can:
+	 *
+	 * 1. modify the target tuple so that it no longer satisfies the
+	 * additional quals attached to the current WHEN MATCHED action OR
+	 *
+	 * In this case, we are still dealing with a WHEN MATCHED case, but
+	 * we should recheck the list of WHEN MATCHED actions and choose the first
+	 * one that satisfies the new target tuple.
+	 *
+	 * 2. modify the target tuple so that the join quals no longer pass and
+	 * hence the source tuple no longer has a match.
+	 *
+	 * In the second case, the source tuple no longer matches the target tuple,
+	 * so we now instead find a qualifying WHEN NOT MATCHED action to execute.
+	 *
+	 * A concurrent delete, changes a WHEN MATCHED case to WHEN NOT MATCHED.
+	 *
+	 * ExecMergeMatched takes care of following the update chain and
+	 * re-finding the qualifying WHEN MATCHED action, as long as the updated
+	 * target tuple still satisfies the join quals i.e. it still remains a
+	 * WHEN MATCHED case. If the tuple gets deleted or the join quals fail, it
+	 * returns and we try ExecMergeNotMatched. Given that ExecMergeMatched
+	 * always make progress by following the update chain and we never switch
+	 * from ExecMergeNotMatched to ExecMergeMatched, there is no risk of a
+	 * livelock.
+	 */
+	if (matched)
+		matched = ExecMergeMatched(mtstate, estate, slot, junkfilter, tupleid);
+
+	/*
+	 * Either we were dealing with a NOT MATCHED tuple or ExecMergeNotMatched()
+	 * returned "false", indicating the previously MATCHED tuple is no longer a
+	 * matching tuple.
+	 */
+	if (!matched)
+		ExecMergeNotMatched(mtstate, estate, slot);
+}
 
 /*
  * Check and execute the first qualifying MATCHED action. The current target
@@ -248,8 +352,9 @@ lmerge_matched:;
 
 					/*
 					 * This state should never be reached since the underlying
-					 * JOIN runs with a MVCC snapshot and should only return
+					 * JOIN runs with a MVCC snapshot and EvalPlanQual runs
-					 * rows visible to us.
+					 * with a dirty snapshot. So such a row should have never
+					 * been returned for MERGE.
 					 */
 					elog(ERROR, "unexpected invisible tuple");
 					break;
@@ -392,10 +497,10 @@ ExecMergeNotMatched(ModifyTableState *mtstate, EState *estate,
 	TupleTableSlot	*myslot;
 
 	/*
-	 * We are dealing with NOT MATCHED tuple. Since for MERGE, partition tree
+	 * We are dealing with NOT MATCHED tuple. Since for MERGE, the partition
-	 * is not expanded for the result relation, we continue to work with the
+	 * tree is not expanded for the result relation, we continue to work with
-	 * currently active result relation, which should be of the root of the
+	 * the currently active result relation, which corresponds to the root
-	 * partition tree.
+	 * of the partition tree.
 	 */
 	resultRelInfo = mtstate->resultRelInfo;
 
@@ -474,102 +579,105 @@ ExecMergeNotMatched(ModifyTableState *mtstate, EState *estate,
 	}
 }
 
-/*
- * Perform MERGE.
- */
 void
-ExecMerge(ModifyTableState *mtstate, EState *estate, TupleTableSlot *slot,
+ExecInitMerge(ModifyTableState *mtstate, EState *estate,
-		  JunkFilter *junkfilter, ResultRelInfo *resultRelInfo)
+			  ResultRelInfo *resultRelInfo)
 {
-	ExprContext *econtext = mtstate->ps.ps_ExprContext;
+	ListCell   *l;
-	ItemPointer tupleid;
+	ExprContext *econtext;
-	ItemPointerData tuple_ctid;
+	List	   *mergeMatchedActionStates = NIL;
-	bool		matched = false;
+	List	   *mergeNotMatchedActionStates = NIL;
-	char		relkind;
+	TupleDesc	relationDesc = resultRelInfo->ri_RelationDesc->rd_att;
-	Datum		datum;
+	ModifyTable *node = (ModifyTable *) mtstate->ps.plan;
-	bool		isNull;
 
-	relkind = resultRelInfo->ri_RelationDesc->rd_rel->relkind;
+	if (node->mergeActionList == NIL)
-	Assert(relkind == RELKIND_RELATION ||
+		return;
-		   relkind == RELKIND_PARTITIONED_TABLE);
+
+	mtstate->mt_merge_subcommands = 0;
+
+	if (mtstate->ps.ps_ExprContext == NULL)
+		ExecAssignExprContext(estate, &mtstate->ps);
+
+	econtext = mtstate->ps.ps_ExprContext;
+
+	/* initialize slot for the existing tuple */
+	Assert(mtstate->mt_existing == NULL);
+	mtstate->mt_existing =
+		ExecInitExtraTupleSlot(mtstate->ps.state,
+							   mtstate->mt_partition_tuple_routing ?
+							   NULL : relationDesc);
+
+	/* initialize slot for merge actions */
+	Assert(mtstate->mt_mergeproj == NULL);
+	mtstate->mt_mergeproj =
+		ExecInitExtraTupleSlot(mtstate->ps.state,
+							   mtstate->mt_partition_tuple_routing ?
+							   NULL : relationDesc);
 
 	/*
-	 * Reset per-tuple memory context to free any expression evaluation
+	 * Create a MergeActionState for each action on the mergeActionList
-	 * storage allocated in the previous cycle.
+	 * and add it to either a list of matched actions or not-matched
+	 * actions.
 	 */
-	ResetExprContext(econtext);
+	foreach(l, node->mergeActionList)
-
-	/*
-	 * We run a JOIN between the target relation and the source relation to
-	 * find a set of candidate source rows that has matching row in the target
-	 * table and a set of candidate source rows that does not have matching
-	 * row in the target table. If the join returns us a tuple with target
-	 * relation's tid set, that implies that the join found a matching row for
-	 * the given source tuple. This case triggers the WHEN MATCHED clause of
-	 * the MERGE. Whereas a NULL in the target relation's ctid column
-	 * indicates a NOT MATCHED case.
-	 */
-	datum = ExecGetJunkAttribute(slot, junkfilter->jf_junkAttNo, &isNull);
-
-	if (!isNull)
 	{
-		matched = true;
+		MergeAction *action = (MergeAction *) lfirst(l);
-		tupleid = (ItemPointer) DatumGetPointer(datum);
+		MergeActionState *action_state = makeNode(MergeActionState);
-		tuple_ctid = *tupleid;	/* be sure we don't free ctid!! */
+		TupleDesc	tupDesc;
-		tupleid = &tuple_ctid;
+
-	}
+		action_state->matched = action->matched;
+		action_state->commandType = action->commandType;
+		action_state->whenqual = ExecInitQual((List *) action->qual,
+				&mtstate->ps);
+
+		/* create target slot for this action's projection */
+		tupDesc = ExecTypeFromTL((List *) action->targetList,
+				resultRelInfo->ri_RelationDesc->rd_rel->relhasoids);
+		action_state->tupDesc = tupDesc;
+
+		/* build action projection state */
+		action_state->proj =
+			ExecBuildProjectionInfo(action->targetList, econtext,
+					mtstate->mt_mergeproj, &mtstate->ps,
+					resultRelInfo->ri_RelationDesc->rd_att);
+
+		/*
+		 * We create two lists - one for WHEN MATCHED actions and one
+		 * for WHEN NOT MATCHED actions - and stick the
+		 * MergeActionState into the appropriate list.
+		 */
+		if (action_state->matched)
+			mergeMatchedActionStates =
+				lappend(mergeMatchedActionStates, action_state);
 		else
+			mergeNotMatchedActionStates =
+				lappend(mergeNotMatchedActionStates, action_state);
+
+		switch (action->commandType)
 		{
-		matched = false;
+			case CMD_INSERT:
-		tupleid = NULL;			/* we don't need it for INSERT actions */
+				ExecCheckPlanOutput(resultRelInfo->ri_RelationDesc,
+									action->targetList);
+				mtstate->mt_merge_subcommands |= MERGE_INSERT;
+				break;
+			case CMD_UPDATE:
+				ExecCheckPlanOutput(resultRelInfo->ri_RelationDesc,
+									action->targetList);
+				mtstate->mt_merge_subcommands |= MERGE_UPDATE;
+				break;
+			case CMD_DELETE:
+				mtstate->mt_merge_subcommands |= MERGE_DELETE;
+				break;
+			case CMD_NOTHING:
+				break;
+			default:
+				elog(ERROR, "unknown operation");
+				break;
 		}
 
-	/*
+		resultRelInfo->ri_mergeState->matchedActionStates =
-	 * If we are dealing with a WHEN MATCHED case, we execute the first action
+					mergeMatchedActionStates;
-	 * for which the additional WHEN MATCHED AND quals pass. If an action
+		resultRelInfo->ri_mergeState->notMatchedActionStates =
-	 * without quals is found, that action is executed.
+					mergeNotMatchedActionStates;
-	 *
+	}
-	 * Similarly, if we are dealing with WHEN NOT MATCHED case, we look at the
-	 * given WHEN NOT MATCHED actions in sequence until one passes.
-	 *
-	 * Things get interesting in case of concurrent update/delete of the
-	 * target tuple. Such concurrent update/delete is detected while we are
-	 * executing a WHEN MATCHED action.
-	 *
-	 * A concurrent update can:
-	 *
-	 * 1. modify the target tuple so that it no longer satisfies the
-	 * additional quals attached to the current WHEN MATCHED action OR
-	 *
-	 * In this case, we are still dealing with a WHEN MATCHED case, but
-	 * we should recheck the list of WHEN MATCHED actions and choose the first
-	 * one that satisfies the new target tuple.
-	 *
-	 * 2. modify the target tuple so that the join quals no longer pass and
-	 * hence the source tuple no longer has a match.
-	 *
-	 * In the second case, the source tuple no longer matches the target tuple,
-	 * so we now instead find a qualifying WHEN NOT MATCHED action to execute.
-	 *
-	 * A concurrent delete, changes a WHEN MATCHED case to WHEN NOT MATCHED.
-	 *
-	 * ExecMergeMatched takes care of following the update chain and
-	 * re-finding the qualifying WHEN MATCHED action, as long as the updated
-	 * target tuple still satisfies the join quals i.e. it still remains a
-	 * WHEN MATCHED case. If the tuple gets deleted or the join quals fail, it
-	 * returns and we try ExecMergeNotMatched. Given that ExecMergeMatched
-	 * always make progress by following the update chain and we never switch
-	 * from ExecMergeNotMatched to ExecMergeMatched, there is no risk of a
-	 * livelock.
-	 */
-	if (matched)
-		matched = ExecMergeMatched(mtstate, estate, slot, junkfilter, tupleid);
-
-	/*
-	 * Either we were dealing with a NOT MATCHED tuple or ExecMergeNotMatched()
-	 * returned "false", indicating the previously MATCHED tuple is no longer a
-	 * matching tuple.
-	 */
-	if (!matched)
-		ExecMergeNotMatched(mtstate, estate, slot);
 }

src/backend/executor/execPartition.c

@@ -313,6 +313,10 @@ ExecFindPartition(ResultRelInfo *resultRelInfo, PartitionDispatch *pd,
 /*
  * Given OID of the partition leaf, return the index of the leaf in the
  * partition hierarchy.
+ *
+ * NB: This is an O(N) operation. Unfortunately, there are many other problem
+ * areas with more than a handful partitions, so we don't try to optimise this
+ * code right now.
  */
 int
 ExecFindPartitionByOid(PartitionTupleRouting *proute, Oid partoid)
@@ -325,7 +329,10 @@ ExecFindPartitionByOid(PartitionTupleRouting *proute, Oid partoid)
 			break;
 	}
 
-	Assert(i < proute->num_partitions);
+	if (i >= proute->num_partitions)
+		ereport(ERROR,
+				(errcode(ERRCODE_INTERNAL_ERROR),
+				 errmsg("no partition found for OID %u", partoid)));
 	return i;
 }
 

src/backend/executor/nodeModifyTable.c

@@ -42,7 +42,7 @@
 #include "commands/trigger.h"
 #include "executor/execPartition.h"
 #include "executor/executor.h"
-#include "executor/nodeMerge.h"
+#include "executor/execMerge.h"
 #include "executor/nodeModifyTable.h"
 #include "foreign/fdwapi.h"
 #include "miscadmin.h"
@@ -69,11 +69,6 @@ static void ExecSetupChildParentMapForSubplan(ModifyTableState *mtstate);
 static TupleConversionMap *tupconv_map_for_subplan(ModifyTableState *node,
 						int whichplan);
 
-/* flags for mt_merge_subcommands */
-#define MERGE_INSERT	0x01
-#define MERGE_UPDATE	0x02
-#define MERGE_DELETE	0x04
-
 /*
  * Verify that the tuples to be produced by INSERT or UPDATE match the
  * target relation's rowtype
@@ -86,7 +81,7 @@ static TupleConversionMap *tupconv_map_for_subplan(ModifyTableState *node,
  * The plan output is represented by its targetlist, because that makes
  * handling the dropped-column case easier.
  */
-static void
+void
 ExecCheckPlanOutput(Relation resultRel, List *targetList)
 {
 	TupleDesc	resultDesc = RelationGetDescr(resultRel);
@@ -2660,104 +2655,8 @@ ExecInitModifyTable(ModifyTable *node, EState *estate, int eflags)
 	}
 
 	resultRelInfo = mtstate->resultRelInfo;
-
+	if (mtstate->operation == CMD_MERGE)
-	if (node->mergeActionList)
+		ExecInitMerge(mtstate, estate, resultRelInfo);
-	{
-		ListCell   *l;
-		ExprContext *econtext;
-		List	   *mergeMatchedActionStates = NIL;
-		List	   *mergeNotMatchedActionStates = NIL;
-		TupleDesc	relationDesc = resultRelInfo->ri_RelationDesc->rd_att;
-
-		mtstate->mt_merge_subcommands = 0;
-
-		if (mtstate->ps.ps_ExprContext == NULL)
-			ExecAssignExprContext(estate, &mtstate->ps);
-
-		econtext = mtstate->ps.ps_ExprContext;
-
-		/* initialize slot for the existing tuple */
-		Assert(mtstate->mt_existing == NULL);
-		mtstate->mt_existing =
-			ExecInitExtraTupleSlot(mtstate->ps.state,
-								   mtstate->mt_partition_tuple_routing ?
-								   NULL : relationDesc);
-
-		/* initialize slot for merge actions */
-		Assert(mtstate->mt_mergeproj == NULL);
-		mtstate->mt_mergeproj =
-			ExecInitExtraTupleSlot(mtstate->ps.state,
-								   mtstate->mt_partition_tuple_routing ?
-								   NULL : relationDesc);
-
-		/*
-		 * Create a MergeActionState for each action on the mergeActionList
-		 * and add it to either a list of matched actions or not-matched
-		 * actions.
-		 */
-		foreach(l, node->mergeActionList)
-		{
-			MergeAction *action = (MergeAction *) lfirst(l);
-			MergeActionState *action_state = makeNode(MergeActionState);
-			TupleDesc	tupDesc;
-
-			action_state->matched = action->matched;
-			action_state->commandType = action->commandType;
-			action_state->whenqual = ExecInitQual((List *) action->qual,
-					&mtstate->ps);
-
-			/* create target slot for this action's projection */
-			tupDesc = ExecTypeFromTL((List *) action->targetList,
-					resultRelInfo->ri_RelationDesc->rd_rel->relhasoids);
-			action_state->tupDesc = tupDesc;
-
-			/* build action projection state */
-			action_state->proj =
-				ExecBuildProjectionInfo(action->targetList, econtext,
-						mtstate->mt_mergeproj, &mtstate->ps,
-						resultRelInfo->ri_RelationDesc->rd_att);
-
-			/*
-			 * We create two lists - one for WHEN MATCHED actions and one
-			 * for WHEN NOT MATCHED actions - and stick the
-			 * MergeActionState into the appropriate list.
-			 */
-			if (action_state->matched)
-				mergeMatchedActionStates =
-					lappend(mergeMatchedActionStates, action_state);
-			else
-				mergeNotMatchedActionStates =
-					lappend(mergeNotMatchedActionStates, action_state);
-
-			switch (action->commandType)
-			{
-				case CMD_INSERT:
-					ExecCheckPlanOutput(resultRelInfo->ri_RelationDesc,
-										action->targetList);
-					mtstate->mt_merge_subcommands |= MERGE_INSERT;
-					break;
-				case CMD_UPDATE:
-					ExecCheckPlanOutput(resultRelInfo->ri_RelationDesc,
-										action->targetList);
-					mtstate->mt_merge_subcommands |= MERGE_UPDATE;
-					break;
-				case CMD_DELETE:
-					mtstate->mt_merge_subcommands |= MERGE_DELETE;
-					break;
-				case CMD_NOTHING:
-					break;
-				default:
-					elog(ERROR, "unknown operation");
-					break;
-			}
-
-			resultRelInfo->ri_mergeState->matchedActionStates =
-						mergeMatchedActionStates;
-			resultRelInfo->ri_mergeState->notMatchedActionStates =
-						mergeNotMatchedActionStates;
-
-		}
-	}
 
 	/* select first subplan */
 	mtstate->mt_whichplan = 0;

src/backend/optimizer/plan/setrefs.c

@@ -852,14 +852,14 @@ set_plan_refs(PlannerInfo *root, Plan *plan, int rtoffset)
 				}
 
 				/*
-				 * The MERGE produces the target rows by performing a right
+				 * The MERGE statement produces the target rows by performing a
-				 * join between the target relation and the source relation
+				 * right join between the target relation and the source
-				 * (which could be a plain relation or a subquery). The INSERT
+				 * relation (which could be a plain relation or a subquery).
-				 * and UPDATE actions of the MERGE requires access to the
+				 * The INSERT and UPDATE actions of the MERGE statement
-				 * columns from the source relation. We arrange things so that
+				 * requires access to the columns from the source relation. We
-				 * the source relation attributes are available as INNER_VAR
+				 * arrange things so that the source relation attributes are
-				 * and the target relation attributes are available from the
+				 * available as INNER_VAR and the target relation attributes
-				 * scan tuple.
+				 * are available from the scan tuple.
 				 */
 				if (splan->mergeActionList != NIL)
 				{

src/backend/parser/gram.y

@@ -585,7 +585,7 @@ static Node *makeRecursiveViewSelect(char *relname, List *aliases, Node *query);
 %type <list>		hash_partbound partbound_datum_list range_datum_list
 %type <defelt>		hash_partbound_elem
 
-%type <node>	merge_when_clause opt_and_condition
+%type <node>	merge_when_clause opt_merge_when_and_condition
 %type <list>	merge_when_list
 %type <node>	merge_update merge_delete merge_insert
 
@@ -11129,7 +11129,7 @@ merge_when_list:
 			;
 
 merge_when_clause:
-			WHEN MATCHED opt_and_condition THEN merge_update
+			WHEN MATCHED opt_merge_when_and_condition THEN merge_update
 				{
 					MergeAction *m = makeNode(MergeAction);
 
@@ -11140,7 +11140,7 @@ merge_when_clause:
 
 					$$ = (Node *)m;
 				}
-			| WHEN MATCHED opt_and_condition THEN merge_delete
+			| WHEN MATCHED opt_merge_when_and_condition THEN merge_delete
 				{
 					MergeAction *m = makeNode(MergeAction);
 
@@ -11151,7 +11151,7 @@ merge_when_clause:
 
 					$$ = (Node *)m;
 				}
-			| WHEN NOT MATCHED opt_and_condition THEN merge_insert
+			| WHEN NOT MATCHED opt_merge_when_and_condition THEN merge_insert
 				{
 					MergeAction *m = makeNode(MergeAction);
 
@@ -11162,7 +11162,7 @@ merge_when_clause:
 
 					$$ = (Node *)m;
 				}
-			| WHEN NOT MATCHED opt_and_condition THEN DO NOTHING
+			| WHEN NOT MATCHED opt_merge_when_and_condition THEN DO NOTHING
 				{
 					MergeAction *m = makeNode(MergeAction);
 
@@ -11175,7 +11175,7 @@ merge_when_clause:
 				}
 			;
 
-opt_and_condition:
+opt_merge_when_and_condition:
 			AND a_expr 				{ $$ = $2; }
 			| 			 			{ $$ = NULL; }
 			;

src/include/executor/execMerge.h

@@ -0,0 +1,31 @@
+/*-------------------------------------------------------------------------
+ *
+ * execMerge.h
+ *
+ *
+ * Portions Copyright (c) 1996-2018, PostgreSQL Global Development Group
+ * Portions Copyright (c) 1994, Regents of the University of California
+ *
+ * src/include/executor/execMerge.h
+ *
+ *-------------------------------------------------------------------------
+ */
+#ifndef EXECMERGE_H
+#define EXECMERGE_H
+
+#include "nodes/execnodes.h"
+
+/* flags for mt_merge_subcommands */
+#define MERGE_INSERT	0x01
+#define MERGE_UPDATE	0x02
+#define MERGE_DELETE	0x04
+
+extern void ExecMerge(ModifyTableState *mtstate, EState *estate,
+					  TupleTableSlot *slot, JunkFilter *junkfilter,
+					  ResultRelInfo *resultRelInfo);
+
+extern void ExecInitMerge(ModifyTableState *mtstate,
+						  EState *estate,
+						  ResultRelInfo *resultRelInfo);
+
+#endif							/* NODEMERGE_H */

src/include/executor/nodeMerge.h

@@ -1,22 +0,0 @@
-/*-------------------------------------------------------------------------
- *
- * nodeMerge.h
- *
- *
- * Portions Copyright (c) 1996-2018, PostgreSQL Global Development Group
- * Portions Copyright (c) 1994, Regents of the University of California
- *
- * src/include/executor/nodeMerge.h
- *
- *-------------------------------------------------------------------------
- */
-#ifndef NODEMERGE_H
-#define NODEMERGE_H
-
-#include "nodes/execnodes.h"
-
-extern void
-ExecMerge(ModifyTableState *mtstate, EState *estate, TupleTableSlot *slot,
-		  JunkFilter *junkfilter, ResultRelInfo *resultRelInfo);
-
-#endif							/* NODEMERGE_H */

src/include/executor/nodeModifyTable.h

@@ -39,5 +39,6 @@ extern TupleTableSlot *ExecInsert(ModifyTableState *mtstate,
 		   EState *estate,
 		   MergeActionState *actionState,
 		   bool canSetTag);
+extern void ExecCheckPlanOutput(Relation resultRel, List *targetList);
 
 #endif							/* NODEMODIFYTABLE_H */