nodesubplan.c

PostgreSQL 8.1.4的源码 适用于Linux下的开源数据库系统

/*-------------------------------------------------------------------------
 *
 * nodeSubplan.c
 *	  routines to support subselects
 *
 * Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
 * Portions Copyright (c) 1994, Regents of the University of California
 *
 * IDENTIFICATION
 *	  $PostgreSQL: pgsql/src/backend/executor/nodeSubplan.c,v 1.70.2.1 2005/11/22 18:23:09 momjian Exp $
 *
 *-------------------------------------------------------------------------
 */
/*
 *	 INTERFACE ROUTINES
 *		ExecSubPlan  - process a subselect
 *		ExecInitSubPlan - initialize a subselect
 *		ExecEndSubPlan	- shut down a subselect
 */
#include "postgres.h"

#include "access/heapam.h"
#include "executor/executor.h"
#include "executor/nodeSubplan.h"
#include "nodes/makefuncs.h"
#include "parser/parse_expr.h"
#include "utils/array.h"
#include "utils/datum.h"
#include "utils/lsyscache.h"
#include "utils/memutils.h"


static Datum ExecHashSubPlan(SubPlanState *node,
				ExprContext *econtext,
				bool *isNull);
static Datum ExecScanSubPlan(SubPlanState *node,
				ExprContext *econtext,
				bool *isNull);
static void buildSubPlanHash(SubPlanState *node);
static bool findPartialMatch(TupleHashTable hashtable, TupleTableSlot *slot);
static bool slotAllNulls(TupleTableSlot *slot);
static bool slotNoNulls(TupleTableSlot *slot);


/* ----------------------------------------------------------------
 *		ExecSubPlan
 * ----------------------------------------------------------------
 */
Datum
ExecSubPlan(SubPlanState *node,
			ExprContext *econtext,
			bool *isNull,
			ExprDoneCond *isDone)
{
	SubPlan    *subplan = (SubPlan *) node->xprstate.expr;

	/* Set default values for result flags: non-null, not a set result */
	*isNull = false;
	if (isDone)
		*isDone = ExprSingleResult;

	if (subplan->setParam != NIL)
		elog(ERROR, "cannot set parent params from subquery");

	if (subplan->useHashTable)
		return ExecHashSubPlan(node, econtext, isNull);
	else
		return ExecScanSubPlan(node, econtext, isNull);
}

/*
 * ExecHashSubPlan: store subselect result in an in-memory hash table
 */
static Datum
ExecHashSubPlan(SubPlanState *node,
				ExprContext *econtext,
				bool *isNull)
{
	SubPlan    *subplan = (SubPlan *) node->xprstate.expr;
	PlanState  *planstate = node->planstate;
	ExprContext *innerecontext = node->innerecontext;
	TupleTableSlot *slot;

	/* Shouldn't have any direct correlation Vars */
	if (subplan->parParam != NIL || node->args != NIL)
		elog(ERROR, "hashed subplan with direct correlation not supported");

	/*
	 * If first time through or we need to rescan the subplan, build the hash
	 * table.
	 */
	if (node->hashtable == NULL || planstate->chgParam != NULL)
		buildSubPlanHash(node);

	/*
	 * The result for an empty subplan is always FALSE; no need to evaluate
	 * lefthand side.
	 */
	*isNull = false;
	if (!node->havehashrows && !node->havenullrows)
		return BoolGetDatum(false);

	/*
	 * Evaluate lefthand expressions and form a projection tuple. First we
	 * have to set the econtext to use (hack alert!).
	 */
	node->projLeft->pi_exprContext = econtext;
	slot = ExecProject(node->projLeft, NULL);

	/*
	 * Note: because we are typically called in a per-tuple context, we have
	 * to explicitly clear the projected tuple before returning. Otherwise,
	 * we'll have a double-free situation: the per-tuple context will probably
	 * be reset before we're called again, and then the tuple slot will think
	 * it still needs to free the tuple.
	 */

	/*
	 * Since the hashtable routines will use innerecontext's per-tuple memory
	 * as working memory, be sure to reset it for each tuple.
	 */
	ResetExprContext(innerecontext);

	/*
	 * If the LHS is all non-null, probe for an exact match in the main hash
	 * table.  If we find one, the result is TRUE. Otherwise, scan the
	 * partly-null table to see if there are any rows that aren't provably
	 * unequal to the LHS; if so, the result is UNKNOWN.  (We skip that part
	 * if we don't care about UNKNOWN.) Otherwise, the result is FALSE.
	 *
	 * Note: the reason we can avoid a full scan of the main hash table is
	 * that the combining operators are assumed never to yield NULL when both
	 * inputs are non-null.  If they were to do so, we might need to produce
	 * UNKNOWN instead of FALSE because of an UNKNOWN result in comparing the
	 * LHS to some main-table entry --- which is a comparison we will not even
	 * make, unless there's a chance match of hash keys.
	 */
	if (slotNoNulls(slot))
	{
		if (node->havehashrows &&
			LookupTupleHashEntry(node->hashtable, slot, NULL) != NULL)
		{
			ExecClearTuple(slot);
			return BoolGetDatum(true);
		}
		if (node->havenullrows &&
			findPartialMatch(node->hashnulls, slot))
		{
			ExecClearTuple(slot);
			*isNull = true;
			return BoolGetDatum(false);
		}
		ExecClearTuple(slot);
		return BoolGetDatum(false);
	}

	/*
	 * When the LHS is partly or wholly NULL, we can never return TRUE. If we
	 * don't care about UNKNOWN, just return FALSE.  Otherwise, if the LHS is
	 * wholly NULL, immediately return UNKNOWN.  (Since the combining
	 * operators are strict, the result could only be FALSE if the sub-select
	 * were empty, but we already handled that case.) Otherwise, we must scan
	 * both the main and partly-null tables to see if there are any rows that
	 * aren't provably unequal to the LHS; if so, the result is UNKNOWN.
	 * Otherwise, the result is FALSE.
	 */
	if (node->hashnulls == NULL)
	{
		ExecClearTuple(slot);
		return BoolGetDatum(false);
	}
	if (slotAllNulls(slot))
	{
		ExecClearTuple(slot);
		*isNull = true;
		return BoolGetDatum(false);
	}
	/* Scan partly-null table first, since more likely to get a match */
	if (node->havenullrows &&
		findPartialMatch(node->hashnulls, slot))
	{
		ExecClearTuple(slot);
		*isNull = true;
		return BoolGetDatum(false);
	}
	if (node->havehashrows &&
		findPartialMatch(node->hashtable, slot))
	{
		ExecClearTuple(slot);
		*isNull = true;
		return BoolGetDatum(false);
	}
	ExecClearTuple(slot);
	return BoolGetDatum(false);
}

/*
 * ExecScanSubPlan: default case where we have to rescan subplan each time
 */
static Datum
ExecScanSubPlan(SubPlanState *node,
				ExprContext *econtext,
				bool *isNull)
{
	SubPlan    *subplan = (SubPlan *) node->xprstate.expr;
	PlanState  *planstate = node->planstate;
	SubLinkType subLinkType = subplan->subLinkType;
	bool		useOr = subplan->useOr;
	MemoryContext oldcontext;
	TupleTableSlot *slot;
	Datum		result;
	bool		found = false;	/* TRUE if got at least one subplan tuple */
	ListCell   *pvar;
	ListCell   *l;
	ArrayBuildState *astate = NULL;

	/*
	 * We are probably in a short-lived expression-evaluation context. Switch
	 * to the child plan's per-query context for manipulating its chgParam,
	 * calling ExecProcNode on it, etc.
	 */
	oldcontext = MemoryContextSwitchTo(node->sub_estate->es_query_cxt);

	/*
	 * Set Params of this plan from parent plan correlation values. (Any
	 * calculation we have to do is done in the parent econtext, since the
	 * Param values don't need to have per-query lifetime.)
	 */
	Assert(list_length(subplan->parParam) == list_length(node->args));

	forboth(l, subplan->parParam, pvar, node->args)
	{
		int			paramid = lfirst_int(l);
		ParamExecData *prm = &(econtext->ecxt_param_exec_vals[paramid]);

		prm->value = ExecEvalExprSwitchContext((ExprState *) lfirst(pvar),
											   econtext,
											   &(prm->isnull),
											   NULL);
		planstate->chgParam = bms_add_member(planstate->chgParam, paramid);
	}

	ExecReScan(planstate, NULL);

	/*
	 * For all sublink types except EXPR_SUBLINK and ARRAY_SUBLINK, the result
	 * is boolean as are the results of the combining operators. We combine
	 * results within a tuple (if there are multiple columns) using OR
	 * semantics if "useOr" is true, AND semantics if not. We then combine
	 * results across tuples (if the subplan produces more than one) using OR
	 * semantics for ANY_SUBLINK or AND semantics for ALL_SUBLINK.
	 * (MULTIEXPR_SUBLINK doesn't allow multiple tuples from the subplan.)
	 * NULL results from the combining operators are handled according to the
	 * usual SQL semantics for OR and AND.	The result for no input tuples is
	 * FALSE for ANY_SUBLINK, TRUE for ALL_SUBLINK, NULL for
	 * MULTIEXPR_SUBLINK.
	 *
	 * For EXPR_SUBLINK we require the subplan to produce no more than one
	 * tuple, else an error is raised. For ARRAY_SUBLINK we allow the subplan
	 * to produce more than one tuple. In either case, if zero tuples are
	 * produced, we return NULL. Assuming we get a tuple, we just use its
	 * first column (there can be only one non-junk column in this case).
	 */
	result = BoolGetDatum(subLinkType == ALL_SUBLINK);
	*isNull = false;

	for (slot = ExecProcNode(planstate);
		 !TupIsNull(slot);
		 slot = ExecProcNode(planstate))
	{
		TupleDesc	tdesc = slot->tts_tupleDescriptor;
		Datum		rowresult = BoolGetDatum(!useOr);
		bool		rownull = false;
		int			col = 1;
		ListCell   *plst;

		if (subLinkType == EXISTS_SUBLINK)
		{
			found = true;
			result = BoolGetDatum(true);
			break;
		}

		if (subLinkType == EXPR_SUBLINK)
		{
			/* cannot allow multiple input tuples for EXPR sublink */
			if (found)
				ereport(ERROR,
						(errcode(ERRCODE_CARDINALITY_VIOLATION),
						 errmsg("more than one row returned by a subquery used as an expression")));
			found = true;

			/*
			 * We need to copy the subplan's tuple in case the result is of
			 * pass-by-ref type --- our return value will point into this
			 * copied tuple!  Can't use the subplan's instance of the tuple
			 * since it won't still be valid after next ExecProcNode() call.
			 * node->curTuple keeps track of the copied tuple for eventual
			 * freeing.
			 */
			MemoryContextSwitchTo(econtext->ecxt_per_query_memory);
			if (node->curTuple)
				heap_freetuple(node->curTuple);
			node->curTuple = ExecCopySlotTuple(slot);
			MemoryContextSwitchTo(node->sub_estate->es_query_cxt);

			result = heap_getattr(node->curTuple, col, tdesc, isNull);
			/* keep scanning subplan to make sure there's only one tuple */
			continue;
		}

		if (subLinkType == ARRAY_SUBLINK)
		{
			Datum		dvalue;
			bool		disnull;

			found = true;
			/* stash away current value */
			dvalue = slot_getattr(slot, 1, &disnull);
			astate = accumArrayResult(astate, dvalue, disnull,
									  tdesc->attrs[0]->atttypid,
									  oldcontext);
			/* keep scanning subplan to collect all values */
			continue;
		}

		/* cannot allow multiple input tuples for MULTIEXPR sublink either */
		if (subLinkType == MULTIEXPR_SUBLINK && found)
			ereport(ERROR,
					(errcode(ERRCODE_CARDINALITY_VIOLATION),
					 errmsg("more than one row returned by a subquery used as an expression")));

		found = true;

		/*
		 * For ALL, ANY, and MULTIEXPR sublinks, iterate over combining
		 * operators for columns of tuple.
		 */
		Assert(list_length(node->exprs) == list_length(subplan->paramIds));

		forboth(l, node->exprs, plst, subplan->paramIds)
		{
			ExprState  *exprstate = (ExprState *) lfirst(l);
			int			paramid = lfirst_int(plst);
			ParamExecData *prmdata;
			Datum		expresult;
			bool		expnull;

			/*
			 * Load up the Param representing this column of the sub-select.
			 */
			prmdata = &(econtext->ecxt_param_exec_vals[paramid]);
			Assert(prmdata->execPlan == NULL);
			prmdata->value = slot_getattr(slot, col,
										  &(prmdata->isnull));

			/*
			 * Now we can eval the combining operator for this column.
			 */
			expresult = ExecEvalExprSwitchContext(exprstate, econtext,
												  &expnull, NULL);

			/*
			 * Combine the result into the row result as appropriate.
			 */
			if (col == 1)
			{