* INTERFACE ROUTINES
* ExecutorStart()
* ExecutorRun()
+ * ExecutorFinish()
* ExecutorEnd()
*
- * The old ExecutorMain() has been replaced by ExecutorStart(),
- * ExecutorRun() and ExecutorEnd()
- *
- * These three procedures are the external interfaces to the executor.
+ * These four procedures are the external interface to the executor.
* In each case, the query descriptor is required as an argument.
*
- * ExecutorStart() must be called at the beginning of execution of any
- * query plan and ExecutorEnd() should always be called at the end of
- * execution of a plan.
+ * ExecutorStart must be called at the beginning of execution of any
+ * query plan and ExecutorEnd must always be called at the end of
+ * execution of a plan (unless it is aborted due to error).
*
* ExecutorRun accepts direction and count arguments that specify whether
* the plan is to be executed forwards, backwards, and for how many tuples.
+ * In some cases ExecutorRun may be called multiple times to process all
+ * the tuples for a plan. It is also acceptable to stop short of executing
+ * the whole plan (but only if it is a SELECT).
+ *
+ * ExecutorFinish must be called after the final ExecutorRun call and
+ * before ExecutorEnd. This can be omitted only in case of EXPLAIN,
+ * which should also omit ExecutorRun.
*
- * Portions Copyright (c) 1996-2007, PostgreSQL Global Development Group
+ * Portions Copyright (c) 1996-2011, PostgreSQL Global Development Group
* Portions Copyright (c) 1994, Regents of the University of California
*
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/backend/executor/execMain.c,v 1.291 2007/03/25 19:45:14 tgl Exp $
+ * src/backend/executor/execMain.c
*
*-------------------------------------------------------------------------
*/
#include "postgres.h"
-#include "access/heapam.h"
#include "access/reloptions.h"
+#include "access/sysattr.h"
#include "access/transam.h"
#include "access/xact.h"
#include "catalog/heap.h"
#include "commands/trigger.h"
#include "executor/execdebug.h"
#include "executor/instrument.h"
-#include "executor/nodeSubplan.h"
#include "miscadmin.h"
#include "optimizer/clauses.h"
#include "parser/parse_clause.h"
#include "parser/parsetree.h"
+#include "storage/bufmgr.h"
+#include "storage/lmgr.h"
#include "storage/smgr.h"
+#include "tcop/utility.h"
#include "utils/acl.h"
#include "utils/lsyscache.h"
#include "utils/memutils.h"
+#include "utils/snapmgr.h"
+#include "utils/tqual.h"
-typedef struct evalPlanQual
-{
- Index rti;
- EState *estate;
- PlanState *planstate;
- struct evalPlanQual *next; /* stack of active PlanQual plans */
- struct evalPlanQual *free; /* list of free PlanQual plans */
-} evalPlanQual;
+/* Hooks for plugins to get control in ExecutorStart/Run/Finish/End */
+ExecutorStart_hook_type ExecutorStart_hook = NULL;
+ExecutorRun_hook_type ExecutorRun_hook = NULL;
+ExecutorFinish_hook_type ExecutorFinish_hook = NULL;
+ExecutorEnd_hook_type ExecutorEnd_hook = NULL;
+
+/* Hook for plugin to get control in ExecCheckRTPerms() */
+ExecutorCheckPerms_hook_type ExecutorCheckPerms_hook = NULL;
/* decls for local routines only used within this module */
static void InitPlan(QueryDesc *queryDesc, int eflags);
-static void initResultRelInfo(ResultRelInfo *resultRelInfo,
- Index resultRelationIndex,
- List *rangeTable,
- CmdType operation,
- bool doInstrument);
+static void CheckValidRowMarkRel(Relation rel, RowMarkType markType);
+static void ExecPostprocessPlan(EState *estate);
static void ExecEndPlan(PlanState *planstate, EState *estate);
-static TupleTableSlot *ExecutePlan(EState *estate, PlanState *planstate,
+static void ExecutePlan(EState *estate, PlanState *planstate,
CmdType operation,
+ bool sendTuples,
long numberTuples,
ScanDirection direction,
DestReceiver *dest);
-static void ExecSelect(TupleTableSlot *slot,
- DestReceiver *dest, EState *estate);
-static void ExecInsert(TupleTableSlot *slot, ItemPointer tupleid,
- TupleTableSlot *planSlot,
- DestReceiver *dest, EState *estate);
-static void ExecDelete(ItemPointer tupleid,
- TupleTableSlot *planSlot,
- DestReceiver *dest, EState *estate);
-static void ExecUpdate(TupleTableSlot *slot, ItemPointer tupleid,
- TupleTableSlot *planSlot,
- DestReceiver *dest, EState *estate);
-static void ExecProcessReturning(ProjectionInfo *projectReturning,
- TupleTableSlot *tupleSlot,
- TupleTableSlot *planSlot,
- DestReceiver *dest);
-static TupleTableSlot *EvalPlanQualNext(EState *estate);
-static void EndEvalPlanQual(EState *estate);
-static void ExecCheckRTPerms(List *rangeTable);
-static void ExecCheckRTEPerms(RangeTblEntry *rte);
+static bool ExecCheckRTEPerms(RangeTblEntry *rte);
static void ExecCheckXactReadOnly(PlannedStmt *plannedstmt);
-static void EvalPlanQualStart(evalPlanQual *epq, EState *estate,
- evalPlanQual *priorepq);
-static void EvalPlanQualStop(evalPlanQual *epq);
+static void EvalPlanQualStart(EPQState *epqstate, EState *parentestate,
+ Plan *planTree);
static void OpenIntoRel(QueryDesc *queryDesc);
static void CloseIntoRel(QueryDesc *queryDesc);
static void intorel_startup(DestReceiver *self, int operation, TupleDesc typeinfo);
* This routine must be called at the beginning of any execution of any
* query plan
*
- * Takes a QueryDesc previously created by CreateQueryDesc (it's not real
- * clear why we bother to separate the two functions, but...). The tupDesc
+ * Takes a QueryDesc previously created by CreateQueryDesc (which is separate
+ * only because some places use QueryDescs for utility commands). The tupDesc
* field of the QueryDesc is filled in to describe the tuples that will be
* returned, and the internal fields (estate and planstate) are set up.
*
*
* NB: the CurrentMemoryContext when this is called will become the parent
* of the per-query context used for this Executor invocation.
+ *
+ * We provide a function hook variable that lets loadable plugins
+ * get control when ExecutorStart is called. Such a plugin would
+ * normally call standard_ExecutorStart().
+ *
* ----------------------------------------------------------------
*/
void
ExecutorStart(QueryDesc *queryDesc, int eflags)
{
+ if (ExecutorStart_hook)
+ (*ExecutorStart_hook) (queryDesc, eflags);
+ else
+ standard_ExecutorStart(queryDesc, eflags);
+}
+
+void
+standard_ExecutorStart(QueryDesc *queryDesc, int eflags)
+{
EState *estate;
MemoryContext oldcontext;
oldcontext = MemoryContextSwitchTo(estate->es_query_cxt);
/*
- * Fill in parameters, if any, from queryDesc
+ * Fill in external parameters, if any, from queryDesc; and allocate
+ * workspace for internal parameters
*/
estate->es_param_list_info = queryDesc->params;
palloc0(queryDesc->plannedstmt->nParamExec * sizeof(ParamExecData));
/*
+ * If non-read-only query, set the command ID to mark output tuples with
+ */
+ switch (queryDesc->operation)
+ {
+ case CMD_SELECT:
+
+ /*
+ * SELECT INTO, SELECT FOR UPDATE/SHARE and modifying CTEs need to
+ * mark tuples
+ */
+ if (queryDesc->plannedstmt->intoClause != NULL ||
+ queryDesc->plannedstmt->rowMarks != NIL ||
+ queryDesc->plannedstmt->hasModifyingCTE)
+ estate->es_output_cid = GetCurrentCommandId(true);
+
+ /*
+ * A SELECT without modifying CTEs can't possibly queue triggers,
+ * so force skip-triggers mode. This is just a marginal efficiency
+ * hack, since AfterTriggerBeginQuery/AfterTriggerEndQuery aren't
+ * all that expensive, but we might as well do it.
+ */
+ if (!queryDesc->plannedstmt->hasModifyingCTE)
+ eflags |= EXEC_FLAG_SKIP_TRIGGERS;
+ break;
+
+ case CMD_INSERT:
+ case CMD_DELETE:
+ case CMD_UPDATE:
+ estate->es_output_cid = GetCurrentCommandId(true);
+ break;
+
+ default:
+ elog(ERROR, "unrecognized operation code: %d",
+ (int) queryDesc->operation);
+ break;
+ }
+
+ /*
* Copy other important information into the EState
*/
- estate->es_snapshot = queryDesc->snapshot;
- estate->es_crosscheck_snapshot = queryDesc->crosscheck_snapshot;
- estate->es_instrument = queryDesc->doInstrument;
+ estate->es_snapshot = RegisterSnapshot(queryDesc->snapshot);
+ estate->es_crosscheck_snapshot = RegisterSnapshot(queryDesc->crosscheck_snapshot);
+ estate->es_top_eflags = eflags;
+ estate->es_instrument = queryDesc->instrument_options;
/*
* Initialize the plan state tree
*/
InitPlan(queryDesc, eflags);
+ /*
+ * Set up an AFTER-trigger statement context, unless told not to, or
+ * unless it's EXPLAIN-only mode (when ExecutorFinish won't be called).
+ */
+ if (!(eflags & (EXEC_FLAG_SKIP_TRIGGERS | EXEC_FLAG_EXPLAIN_ONLY)))
+ AfterTriggerBeginQuery();
+
MemoryContextSwitchTo(oldcontext);
}
* Note: count = 0 is interpreted as no portal limit, i.e., run to
* completion.
*
+ * There is no return value, but output tuples (if any) are sent to
+ * the destination receiver specified in the QueryDesc; and the number
+ * of tuples processed at the top level can be found in
+ * estate->es_processed.
+ *
+ * We provide a function hook variable that lets loadable plugins
+ * get control when ExecutorRun is called. Such a plugin would
+ * normally call standard_ExecutorRun().
+ *
* ----------------------------------------------------------------
*/
-TupleTableSlot *
+void
ExecutorRun(QueryDesc *queryDesc,
ScanDirection direction, long count)
{
+ if (ExecutorRun_hook)
+ (*ExecutorRun_hook) (queryDesc, direction, count);
+ else
+ standard_ExecutorRun(queryDesc, direction, count);
+}
+
+void
+standard_ExecutorRun(QueryDesc *queryDesc,
+ ScanDirection direction, long count)
+{
EState *estate;
CmdType operation;
DestReceiver *dest;
bool sendTuples;
- TupleTableSlot *result;
MemoryContext oldcontext;
/* sanity checks */
estate = queryDesc->estate;
Assert(estate != NULL);
+ Assert(!(estate->es_top_eflags & EXEC_FLAG_EXPLAIN_ONLY));
/*
* Switch into per-query memory context
*/
oldcontext = MemoryContextSwitchTo(estate->es_query_cxt);
+ /* Allow instrumentation of Executor overall runtime */
+ if (queryDesc->totaltime)
+ InstrStartNode(queryDesc->totaltime);
+
/*
* extract information from the query descriptor and the query feature.
*/
estate->es_lastoid = InvalidOid;
sendTuples = (operation == CMD_SELECT ||
- queryDesc->plannedstmt->returningLists);
+ queryDesc->plannedstmt->hasReturning);
if (sendTuples)
(*dest->rStartup) (dest, operation, queryDesc->tupDesc);
/*
* run plan
*/
- if (ScanDirectionIsNoMovement(direction))
- result = NULL;
- else
- result = ExecutePlan(estate,
- queryDesc->planstate,
- operation,
- count,
- direction,
- dest);
+ if (!ScanDirectionIsNoMovement(direction))
+ ExecutePlan(estate,
+ queryDesc->planstate,
+ operation,
+ sendTuples,
+ count,
+ direction,
+ dest);
/*
* shutdown tuple receiver, if we started it
if (sendTuples)
(*dest->rShutdown) (dest);
+ if (queryDesc->totaltime)
+ InstrStopNode(queryDesc->totaltime, estate->es_processed);
+
MemoryContextSwitchTo(oldcontext);
+}
- return result;
+/* ----------------------------------------------------------------
+ * ExecutorFinish
+ *
+ * This routine must be called after the last ExecutorRun call.
+ * It performs cleanup such as firing AFTER triggers. It is
+ * separate from ExecutorEnd because EXPLAIN ANALYZE needs to
+ * include these actions in the total runtime.
+ *
+ * We provide a function hook variable that lets loadable plugins
+ * get control when ExecutorFinish is called. Such a plugin would
+ * normally call standard_ExecutorFinish().
+ *
+ * ----------------------------------------------------------------
+ */
+void
+ExecutorFinish(QueryDesc *queryDesc)
+{
+ if (ExecutorFinish_hook)
+ (*ExecutorFinish_hook) (queryDesc);
+ else
+ standard_ExecutorFinish(queryDesc);
+}
+
+void
+standard_ExecutorFinish(QueryDesc *queryDesc)
+{
+ EState *estate;
+ MemoryContext oldcontext;
+
+ /* sanity checks */
+ Assert(queryDesc != NULL);
+
+ estate = queryDesc->estate;
+
+ Assert(estate != NULL);
+ Assert(!(estate->es_top_eflags & EXEC_FLAG_EXPLAIN_ONLY));
+
+ /* This should be run once and only once per Executor instance */
+ Assert(!estate->es_finished);
+
+ /* Switch into per-query memory context */
+ oldcontext = MemoryContextSwitchTo(estate->es_query_cxt);
+
+ /* Allow instrumentation of Executor overall runtime */
+ if (queryDesc->totaltime)
+ InstrStartNode(queryDesc->totaltime);
+
+ /* Run ModifyTable nodes to completion */
+ ExecPostprocessPlan(estate);
+
+ /* Execute queued AFTER triggers, unless told not to */
+ if (!(estate->es_top_eflags & EXEC_FLAG_SKIP_TRIGGERS))
+ AfterTriggerEndQuery(estate);
+
+ if (queryDesc->totaltime)
+ InstrStopNode(queryDesc->totaltime, 0);
+
+ MemoryContextSwitchTo(oldcontext);
+
+ estate->es_finished = true;
}
/* ----------------------------------------------------------------
*
* This routine must be called at the end of execution of any
* query plan
+ *
+ * We provide a function hook variable that lets loadable plugins
+ * get control when ExecutorEnd is called. Such a plugin would
+ * normally call standard_ExecutorEnd().
+ *
* ----------------------------------------------------------------
*/
void
ExecutorEnd(QueryDesc *queryDesc)
{
+ if (ExecutorEnd_hook)
+ (*ExecutorEnd_hook) (queryDesc);
+ else
+ standard_ExecutorEnd(queryDesc);
+}
+
+void
+standard_ExecutorEnd(QueryDesc *queryDesc)
+{
EState *estate;
MemoryContext oldcontext;
Assert(estate != NULL);
/*
+ * Check that ExecutorFinish was called, unless in EXPLAIN-only mode. This
+ * Assert is needed because ExecutorFinish is new as of 9.1, and callers
+ * might forget to call it.
+ */
+ Assert(estate->es_finished ||
+ (estate->es_top_eflags & EXEC_FLAG_EXPLAIN_ONLY));
+
+ /*
* Switch into per-query memory context to run ExecEndPlan
*/
oldcontext = MemoryContextSwitchTo(estate->es_query_cxt);
if (estate->es_select_into)
CloseIntoRel(queryDesc);
+ /* do away with our snapshots */
+ UnregisterSnapshot(estate->es_snapshot);
+ UnregisterSnapshot(estate->es_crosscheck_snapshot);
+
/*
* Must switch out of context before destroying it
*/
queryDesc->tupDesc = NULL;
queryDesc->estate = NULL;
queryDesc->planstate = NULL;
+ queryDesc->totaltime = NULL;
}
/* ----------------------------------------------------------------
/*
* rescan plan
*/
- ExecReScan(queryDesc->planstate, NULL);
+ ExecReScan(queryDesc->planstate);
MemoryContextSwitchTo(oldcontext);
}
/*
* ExecCheckRTPerms
* Check access permissions for all relations listed in a range table.
+ *
+ * Returns true if permissions are adequate. Otherwise, throws an appropriate
+ * error if ereport_on_violation is true, or simply returns false otherwise.
*/
-static void
-ExecCheckRTPerms(List *rangeTable)
+bool
+ExecCheckRTPerms(List *rangeTable, bool ereport_on_violation)
{
ListCell *l;
+ bool result = true;
foreach(l, rangeTable)
{
- ExecCheckRTEPerms((RangeTblEntry *) lfirst(l));
+ RangeTblEntry *rte = (RangeTblEntry *) lfirst(l);
+
+ result = ExecCheckRTEPerms(rte);
+ if (!result)
+ {
+ Assert(rte->rtekind == RTE_RELATION);
+ if (ereport_on_violation)
+ aclcheck_error(ACLCHECK_NO_PRIV, ACL_KIND_CLASS,
+ get_rel_name(rte->relid));
+ return false;
+ }
}
+
+ if (ExecutorCheckPerms_hook)
+ result = (*ExecutorCheckPerms_hook) (rangeTable,
+ ereport_on_violation);
+ return result;
}
/*
* ExecCheckRTEPerms
* Check access permissions for a single RTE.
*/
-static void
+static bool
ExecCheckRTEPerms(RangeTblEntry *rte)
{
AclMode requiredPerms;
+ AclMode relPerms;
+ AclMode remainingPerms;
Oid relOid;
Oid userid;
+ Bitmapset *tmpset;
+ int col;
/*
* Only plain-relation RTEs need to be checked here. Function RTEs are
* Join, subquery, and special RTEs need no checks.
*/
if (rte->rtekind != RTE_RELATION)
- return;
+ return true;
/*
* No work if requiredPerms is empty.
*/
requiredPerms = rte->requiredPerms;
if (requiredPerms == 0)
- return;
+ return true;
relOid = rte->relid;
userid = rte->checkAsUser ? rte->checkAsUser : GetUserId();
/*
- * We must have *all* the requiredPerms bits, so use aclmask not aclcheck.
+ * We must have *all* the requiredPerms bits, but some of the bits can be
+ * satisfied from column-level rather than relation-level permissions.
+ * First, remove any bits that are satisfied by relation permissions.
*/
- if (pg_class_aclmask(relOid, userid, requiredPerms, ACLMASK_ALL)
- != requiredPerms)
- aclcheck_error(ACLCHECK_NO_PRIV, ACL_KIND_CLASS,
- get_rel_name(relOid));
+ relPerms = pg_class_aclmask(relOid, userid, requiredPerms, ACLMASK_ALL);
+ remainingPerms = requiredPerms & ~relPerms;
+ if (remainingPerms != 0)
+ {
+ /*
+ * If we lack any permissions that exist only as relation permissions,
+ * we can fail straight away.
+ */
+ if (remainingPerms & ~(ACL_SELECT | ACL_INSERT | ACL_UPDATE))
+ return false;
+
+ /*
+ * Check to see if we have the needed privileges at column level.
+ *
+ * Note: failures just report a table-level error; it would be nicer
+ * to report a column-level error if we have some but not all of the
+ * column privileges.
+ */
+ if (remainingPerms & ACL_SELECT)
+ {
+ /*
+ * When the query doesn't explicitly reference any columns (for
+ * example, SELECT COUNT(*) FROM table), allow the query if we
+ * have SELECT on any column of the rel, as per SQL spec.
+ */
+ if (bms_is_empty(rte->selectedCols))
+ {
+ if (pg_attribute_aclcheck_all(relOid, userid, ACL_SELECT,
+ ACLMASK_ANY) != ACLCHECK_OK)
+ return false;
+ }
+
+ tmpset = bms_copy(rte->selectedCols);
+ while ((col = bms_first_member(tmpset)) >= 0)
+ {
+ /* remove the column number offset */
+ col += FirstLowInvalidHeapAttributeNumber;
+ if (col == InvalidAttrNumber)
+ {
+ /* Whole-row reference, must have priv on all cols */
+ if (pg_attribute_aclcheck_all(relOid, userid, ACL_SELECT,
+ ACLMASK_ALL) != ACLCHECK_OK)
+ return false;
+ }
+ else
+ {
+ if (pg_attribute_aclcheck(relOid, col, userid,
+ ACL_SELECT) != ACLCHECK_OK)
+ return false;
+ }
+ }
+ bms_free(tmpset);
+ }
+
+ /*
+ * Basically the same for the mod columns, with either INSERT or
+ * UPDATE privilege as specified by remainingPerms.
+ */
+ remainingPerms &= ~ACL_SELECT;
+ if (remainingPerms != 0)
+ {
+ /*
+ * When the query doesn't explicitly change any columns, allow the
+ * query if we have permission on any column of the rel. This is
+ * to handle SELECT FOR UPDATE as well as possible corner cases in
+ * INSERT and UPDATE.
+ */
+ if (bms_is_empty(rte->modifiedCols))
+ {
+ if (pg_attribute_aclcheck_all(relOid, userid, remainingPerms,
+ ACLMASK_ANY) != ACLCHECK_OK)
+ return false;
+ }
+
+ tmpset = bms_copy(rte->modifiedCols);
+ while ((col = bms_first_member(tmpset)) >= 0)
+ {
+ /* remove the column number offset */
+ col += FirstLowInvalidHeapAttributeNumber;
+ if (col == InvalidAttrNumber)
+ {
+ /* whole-row reference can't happen here */
+ elog(ERROR, "whole-row update is not implemented");
+ }
+ else
+ {
+ if (pg_attribute_aclcheck(relOid, col, userid,
+ remainingPerms) != ACLCHECK_OK)
+ return false;
+ }
+ }
+ bms_free(tmpset);
+ }
+ }
+ return true;
}
/*
* Check that the query does not imply any writes to non-temp tables.
+ *
+ * Note: in a Hot Standby slave this would need to reject writes to temp
+ * tables as well; but an HS slave can't have created any temp tables
+ * in the first place, so no need to check that.
*/
static void
ExecCheckXactReadOnly(PlannedStmt *plannedstmt)
/*
* CREATE TABLE AS or SELECT INTO?
*
- * XXX should we allow this if the destination is temp?
+ * XXX should we allow this if the destination is temp? Considering that
+ * it would still require catalog changes, probably not.
*/
- if (plannedstmt->into != NULL)
- goto fail;
+ if (plannedstmt->intoClause != NULL)
+ PreventCommandIfReadOnly(CreateCommandTag((Node *) plannedstmt));
/* Fail if write permissions are requested on any non-temp table */
foreach(l, plannedstmt->rtable)
if (isTempNamespace(get_rel_namespace(rte->relid)))
continue;
- goto fail;
+ PreventCommandIfReadOnly(CreateCommandTag((Node *) plannedstmt));
}
-
- return;
-
-fail:
- ereport(ERROR,
- (errcode(ERRCODE_READ_ONLY_SQL_TRANSACTION),
- errmsg("transaction is read-only")));
}
/*
* Do permissions checks
*/
- ExecCheckRTPerms(rangeTable);
+ ExecCheckRTPerms(rangeTable, true);
/*
* initialize the node's execution state
*/
estate->es_range_table = rangeTable;
+ estate->es_plannedstmt = plannedstmt;
/*
- * initialize result relation stuff
+ * initialize result relation stuff, and open/lock the result rels.
+ *
+ * We must do this before initializing the plan tree, else we might try to
+ * do a lock upgrade if a result rel is also a source rel.
*/
if (plannedstmt->resultRelations)
{
resultRelInfo = resultRelInfos;
foreach(l, resultRelations)
{
- initResultRelInfo(resultRelInfo,
- lfirst_int(l),
- rangeTable,
- operation,
+ Index resultRelationIndex = lfirst_int(l);
+ Oid resultRelationOid;
+ Relation resultRelation;
+
+ resultRelationOid = getrelid(resultRelationIndex, rangeTable);
+ resultRelation = heap_open(resultRelationOid, RowExclusiveLock);
+ InitResultRelInfo(resultRelInfo,
+ resultRelation,
+ resultRelationIndex,
estate->es_instrument);
resultRelInfo++;
}
estate->es_result_relations = resultRelInfos;
estate->es_num_result_relations = numResultRelations;
- /* Initialize to first or only result rel */
- estate->es_result_relation_info = resultRelInfos;
+ /* es_result_relation_info is NULL except when within ModifyTable */
+ estate->es_result_relation_info = NULL;
}
else
{
}
/*
- * Detect whether we're doing SELECT INTO. If so, set the es_into_oids
- * flag appropriately so that the plan tree will be initialized with the
- * correct tuple descriptors. (Other SELECT INTO stuff comes later.)
- */
- estate->es_select_into = false;
- if (operation == CMD_SELECT && plannedstmt->into != NULL)
- {
- estate->es_select_into = true;
- estate->es_into_oids = interpretOidsOption(plannedstmt->into->options);
- }
-
- /*
- * Have to lock relations selected FOR UPDATE/FOR SHARE before we
- * initialize the plan tree, else we'd be doing a lock upgrade.
+ * Similarly, we have to lock relations selected FOR UPDATE/FOR SHARE
+ * before we initialize the plan tree, else we'd be risking lock upgrades.
* While we are at it, build the ExecRowMark list.
*/
estate->es_rowMarks = NIL;
foreach(l, plannedstmt->rowMarks)
{
- RowMarkClause *rc = (RowMarkClause *) lfirst(l);
- Oid relid = getrelid(rc->rti, rangeTable);
+ PlanRowMark *rc = (PlanRowMark *) lfirst(l);
+ Oid relid;
Relation relation;
ExecRowMark *erm;
- relation = heap_open(relid, RowShareLock);
+ /* ignore "parent" rowmarks; they are irrelevant at runtime */
+ if (rc->isParent)
+ continue;
+
+ switch (rc->markType)
+ {
+ case ROW_MARK_EXCLUSIVE:
+ case ROW_MARK_SHARE:
+ relid = getrelid(rc->rti, rangeTable);
+ relation = heap_open(relid, RowShareLock);
+ break;
+ case ROW_MARK_REFERENCE:
+ relid = getrelid(rc->rti, rangeTable);
+ relation = heap_open(relid, AccessShareLock);
+ break;
+ case ROW_MARK_COPY:
+ /* there's no real table here ... */
+ relation = NULL;
+ break;
+ default:
+ elog(ERROR, "unrecognized markType: %d", rc->markType);
+ relation = NULL; /* keep compiler quiet */
+ break;
+ }
+
+ /* Check that relation is a legal target for marking */
+ if (relation)
+ CheckValidRowMarkRel(relation, rc->markType);
+
erm = (ExecRowMark *) palloc(sizeof(ExecRowMark));
erm->relation = relation;
erm->rti = rc->rti;
- erm->forUpdate = rc->forUpdate;
+ erm->prti = rc->prti;
+ erm->rowmarkId = rc->rowmarkId;
+ erm->markType = rc->markType;
erm->noWait = rc->noWait;
- /* We'll set up ctidAttno below */
- erm->ctidAttNo = InvalidAttrNumber;
+ ItemPointerSetInvalid(&(erm->curCtid));
estate->es_rowMarks = lappend(estate->es_rowMarks, erm);
}
/*
- * Initialize the executor "tuple" table. We need slots for all the plan
- * nodes, plus possibly output slots for the junkfilter(s). At this point
- * we aren't sure if we need junkfilters, so just add slots for them
- * unconditionally. Also, if it's not a SELECT, set up a slot for use for
- * trigger output tuples. Also, one for RETURNING-list evaluation.
+ * Detect whether we're doing SELECT INTO. If so, set the es_into_oids
+ * flag appropriately so that the plan tree will be initialized with the
+ * correct tuple descriptors. (Other SELECT INTO stuff comes later.)
*/
+ estate->es_select_into = false;
+ if (operation == CMD_SELECT && plannedstmt->intoClause != NULL)
{
- int nSlots;
-
- /* Slots for the main plan tree */
- nSlots = ExecCountSlotsNode(plan);
- /* Add slots for subplans and initplans */
- foreach(l, plannedstmt->subplans)
- {
- Plan *subplan = (Plan *) lfirst(l);
-
- nSlots += ExecCountSlotsNode(subplan);
- }
- /* Add slots for junkfilter(s) */
- if (plannedstmt->resultRelations != NIL)
- nSlots += list_length(plannedstmt->resultRelations);
- else
- nSlots += 1;
- if (operation != CMD_SELECT)
- nSlots++; /* for es_trig_tuple_slot */
- if (plannedstmt->returningLists)
- nSlots++; /* for RETURNING projection */
-
- estate->es_tupleTable = ExecCreateTupleTable(nSlots);
-
- if (operation != CMD_SELECT)
- estate->es_trig_tuple_slot =
- ExecAllocTableSlot(estate->es_tupleTable);
+ estate->es_select_into = true;
+ estate->es_into_oids = interpretOidsOption(plannedstmt->intoClause->options);
}
+ /*
+ * Initialize the executor's tuple table to empty.
+ */
+ estate->es_tupleTable = NIL;
+ estate->es_trig_tuple_slot = NULL;
+ estate->es_trig_oldtup_slot = NULL;
+
/* mark EvalPlanQual not active */
- estate->es_plannedstmt = plannedstmt;
- estate->es_evalPlanQual = NULL;
- estate->es_evTupleNull = NULL;
- estate->es_evTuple = NULL;
- estate->es_useEvalPlan = false;
+ estate->es_epqTuple = NULL;
+ estate->es_epqTupleSet = NULL;
+ estate->es_epqScanDone = NULL;
/*
- * Initialize private state information for each SubPlan. We must do
- * this before running ExecInitNode on the main query tree, since
+ * Initialize private state information for each SubPlan. We must do this
+ * before running ExecInitNode on the main query tree, since
* ExecInitSubPlan expects to be able to find these entries.
*/
Assert(estate->es_subplanstates == NIL);
i = 1; /* subplan indices count from 1 */
foreach(l, plannedstmt->subplans)
{
- Plan *subplan = (Plan *) lfirst(l);
- PlanState *subplanstate;
- int sp_eflags;
+ Plan *subplan = (Plan *) lfirst(l);
+ PlanState *subplanstate;
+ int sp_eflags;
/*
- * A subplan will never need to do BACKWARD scan nor MARK/RESTORE.
- * If it is a parameterless subplan (not initplan), we suggest that it
- * be prepared to handle REWIND efficiently; otherwise there is no
- * need.
+ * A subplan will never need to do BACKWARD scan nor MARK/RESTORE. If
+ * it is a parameterless subplan (not initplan), we suggest that it be
+ * prepared to handle REWIND efficiently; otherwise there is no need.
*/
sp_eflags = eflags & EXEC_FLAG_EXPLAIN_ONLY;
if (bms_is_member(i, plannedstmt->rewindPlanIDs))
tupType = ExecGetResultType(planstate);
/*
- * Initialize the junk filter if needed. SELECT and INSERT queries need a
- * filter if there are any junk attrs in the tlist. INSERT and SELECT
- * INTO also need a filter if the plan may return raw disk tuples (else
- * heap_insert will be scribbling on the source relation!). UPDATE and
- * DELETE always need a filter, since there's always a junk 'ctid'
- * attribute present --- no need to look first.
+ * Initialize the junk filter if needed. SELECT queries need a filter if
+ * there are any junk attrs in the top-level tlist.
*/
+ if (operation == CMD_SELECT)
{
bool junk_filter_needed = false;
ListCell *tlist;
- switch (operation)
+ foreach(tlist, plan->targetlist)
{
- case CMD_SELECT:
- case CMD_INSERT:
- foreach(tlist, plan->targetlist)
- {
- TargetEntry *tle = (TargetEntry *) lfirst(tlist);
+ TargetEntry *tle = (TargetEntry *) lfirst(tlist);
- if (tle->resjunk)
- {
- junk_filter_needed = true;
- break;
- }
- }
- if (!junk_filter_needed &&
- (operation == CMD_INSERT || estate->es_select_into) &&
- ExecMayReturnRawTuples(planstate))
- junk_filter_needed = true;
- break;
- case CMD_UPDATE:
- case CMD_DELETE:
+ if (tle->resjunk)
+ {
junk_filter_needed = true;
break;
- default:
- break;
+ }
}
if (junk_filter_needed)
{
- /*
- * If there are multiple result relations, each one needs its own
- * junk filter. Note this is only possible for UPDATE/DELETE, so
- * we can't be fooled by some needing a filter and some not.
- */
- if (list_length(plannedstmt->resultRelations) > 1)
- {
- PlanState **appendplans;
- int as_nplans;
- ResultRelInfo *resultRelInfo;
-
- /* Top plan had better be an Append here. */
- Assert(IsA(plan, Append));
- Assert(((Append *) plan)->isTarget);
- Assert(IsA(planstate, AppendState));
- appendplans = ((AppendState *) planstate)->appendplans;
- as_nplans = ((AppendState *) planstate)->as_nplans;
- Assert(as_nplans == estate->es_num_result_relations);
- resultRelInfo = estate->es_result_relations;
- for (i = 0; i < as_nplans; i++)
- {
- PlanState *subplan = appendplans[i];
- JunkFilter *j;
-
- j = ExecInitJunkFilter(subplan->plan->targetlist,
- resultRelInfo->ri_RelationDesc->rd_att->tdhasoid,
- ExecAllocTableSlot(estate->es_tupleTable));
- /*
- * Since it must be UPDATE/DELETE, there had better be
- * a "ctid" junk attribute in the tlist ... but ctid could
- * be at a different resno for each result relation.
- * We look up the ctid resnos now and save them in the
- * junkfilters.
- */
- j->jf_junkAttNo = ExecFindJunkAttribute(j, "ctid");
- if (!AttributeNumberIsValid(j->jf_junkAttNo))
- elog(ERROR, "could not find junk ctid column");
- resultRelInfo->ri_junkFilter = j;
- resultRelInfo++;
- }
-
- /*
- * Set active junkfilter too; at this point ExecInitAppend has
- * already selected an active result relation...
- */
- estate->es_junkFilter =
- estate->es_result_relation_info->ri_junkFilter;
- }
- else
- {
- /* Normal case with just one JunkFilter */
- JunkFilter *j;
-
- j = ExecInitJunkFilter(planstate->plan->targetlist,
- tupType->tdhasoid,
- ExecAllocTableSlot(estate->es_tupleTable));
- estate->es_junkFilter = j;
- if (estate->es_result_relation_info)
- estate->es_result_relation_info->ri_junkFilter = j;
-
- if (operation == CMD_SELECT)
- {
- /* For SELECT, want to return the cleaned tuple type */
- tupType = j->jf_cleanTupType;
- /* For SELECT FOR UPDATE/SHARE, find the ctid attrs now */
- foreach(l, estate->es_rowMarks)
- {
- ExecRowMark *erm = (ExecRowMark *) lfirst(l);
- char resname[32];
-
- snprintf(resname, sizeof(resname), "ctid%u", erm->rti);
- erm->ctidAttNo = ExecFindJunkAttribute(j, resname);
- if (!AttributeNumberIsValid(erm->ctidAttNo))
- elog(ERROR, "could not find junk \"%s\" column",
- resname);
- }
- }
- else if (operation == CMD_UPDATE || operation == CMD_DELETE)
- {
- /* For UPDATE/DELETE, find the ctid junk attr now */
- j->jf_junkAttNo = ExecFindJunkAttribute(j, "ctid");
- if (!AttributeNumberIsValid(j->jf_junkAttNo))
- elog(ERROR, "could not find junk ctid column");
- }
- }
- }
- else
- estate->es_junkFilter = NULL;
- }
-
- /*
- * Initialize RETURNING projections if needed.
- */
- if (plannedstmt->returningLists)
- {
- TupleTableSlot *slot;
- ExprContext *econtext;
- ResultRelInfo *resultRelInfo;
-
- /*
- * We set QueryDesc.tupDesc to be the RETURNING rowtype in this case.
- * We assume all the sublists will generate the same output tupdesc.
- */
- tupType = ExecTypeFromTL((List *) linitial(plannedstmt->returningLists),
- false);
-
- /* Set up a slot for the output of the RETURNING projection(s) */
- slot = ExecAllocTableSlot(estate->es_tupleTable);
- ExecSetSlotDescriptor(slot, tupType);
- /* Need an econtext too */
- econtext = CreateExprContext(estate);
+ JunkFilter *j;
- /*
- * Build a projection for each result rel. Note that any SubPlans in
- * the RETURNING lists get attached to the topmost plan node.
- */
- Assert(list_length(plannedstmt->returningLists) == estate->es_num_result_relations);
- resultRelInfo = estate->es_result_relations;
- foreach(l, plannedstmt->returningLists)
- {
- List *rlist = (List *) lfirst(l);
- List *rliststate;
+ j = ExecInitJunkFilter(planstate->plan->targetlist,
+ tupType->tdhasoid,
+ ExecInitExtraTupleSlot(estate));
+ estate->es_junkFilter = j;
- rliststate = (List *) ExecInitExpr((Expr *) rlist, planstate);
- resultRelInfo->ri_projectReturning =
- ExecBuildProjectionInfo(rliststate, econtext, slot,
- resultRelInfo->ri_RelationDesc->rd_att);
- resultRelInfo++;
+ /* Want to return the cleaned tuple type */
+ tupType = j->jf_cleanTupType;
}
}
}
/*
- * Initialize ResultRelInfo data for one result relation
+ * Check that a proposed result relation is a legal target for the operation
+ *
+ * In most cases parser and/or planner should have noticed this already, but
+ * let's make sure. In the view case we do need a test here, because if the
+ * view wasn't rewritten by a rule, it had better have an INSTEAD trigger.
+ *
+ * Note: when changing this function, you probably also need to look at
+ * CheckValidRowMarkRel.
*/
-static void
-initResultRelInfo(ResultRelInfo *resultRelInfo,
- Index resultRelationIndex,
- List *rangeTable,
- CmdType operation,
- bool doInstrument)
+void
+CheckValidResultRel(Relation resultRel, CmdType operation)
{
- Oid resultRelationOid;
- Relation resultRelationDesc;
+ TriggerDesc *trigDesc = resultRel->trigdesc;
- resultRelationOid = getrelid(resultRelationIndex, rangeTable);
- resultRelationDesc = heap_open(resultRelationOid, RowExclusiveLock);
-
- switch (resultRelationDesc->rd_rel->relkind)
+ switch (resultRel->rd_rel->relkind)
{
+ case RELKIND_RELATION:
+ /* OK */
+ break;
case RELKIND_SEQUENCE:
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("cannot change sequence \"%s\"",
- RelationGetRelationName(resultRelationDesc))));
+ RelationGetRelationName(resultRel))));
break;
case RELKIND_TOASTVALUE:
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("cannot change TOAST relation \"%s\"",
- RelationGetRelationName(resultRelationDesc))));
+ RelationGetRelationName(resultRel))));
break;
case RELKIND_VIEW:
+ switch (operation)
+ {
+ case CMD_INSERT:
+ if (!trigDesc || !trigDesc->trig_insert_instead_row)
+ ereport(ERROR,
+ (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
+ errmsg("cannot insert into view \"%s\"",
+ RelationGetRelationName(resultRel)),
+ errhint("You need an unconditional ON INSERT DO INSTEAD rule or an INSTEAD OF INSERT trigger.")));
+ break;
+ case CMD_UPDATE:
+ if (!trigDesc || !trigDesc->trig_update_instead_row)
+ ereport(ERROR,
+ (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
+ errmsg("cannot update view \"%s\"",
+ RelationGetRelationName(resultRel)),
+ errhint("You need an unconditional ON UPDATE DO INSTEAD rule or an INSTEAD OF UPDATE trigger.")));
+ break;
+ case CMD_DELETE:
+ if (!trigDesc || !trigDesc->trig_delete_instead_row)
+ ereport(ERROR,
+ (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
+ errmsg("cannot delete from view \"%s\"",
+ RelationGetRelationName(resultRel)),
+ errhint("You need an unconditional ON DELETE DO INSTEAD rule or an INSTEAD OF DELETE trigger.")));
+ break;
+ default:
+ elog(ERROR, "unrecognized CmdType: %d", (int) operation);
+ break;
+ }
+ break;
+ case RELKIND_FOREIGN_TABLE:
+ ereport(ERROR,
+ (errcode(ERRCODE_WRONG_OBJECT_TYPE),
+ errmsg("cannot change foreign table \"%s\"",
+ RelationGetRelationName(resultRel))));
+ break;
+ default:
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
- errmsg("cannot change view \"%s\"",
- RelationGetRelationName(resultRelationDesc))));
+ errmsg("cannot change relation \"%s\"",
+ RelationGetRelationName(resultRel))));
break;
}
+}
- MemSet(resultRelInfo, 0, sizeof(ResultRelInfo));
- resultRelInfo->type = T_ResultRelInfo;
- resultRelInfo->ri_RangeTableIndex = resultRelationIndex;
- resultRelInfo->ri_RelationDesc = resultRelationDesc;
- resultRelInfo->ri_NumIndices = 0;
- resultRelInfo->ri_IndexRelationDescs = NULL;
- resultRelInfo->ri_IndexRelationInfo = NULL;
- /* make a copy so as not to depend on relcache info not changing... */
- resultRelInfo->ri_TrigDesc = CopyTriggerDesc(resultRelationDesc->trigdesc);
- if (resultRelInfo->ri_TrigDesc)
- {
- int n = resultRelInfo->ri_TrigDesc->numtriggers;
+/*
+ * Check that a proposed rowmark target relation is a legal target
+ *
+ * In most cases parser and/or planner should have noticed this already, but
+ * they don't cover all cases.
+ */
+static void
+CheckValidRowMarkRel(Relation rel, RowMarkType markType)
+{
+ switch (rel->rd_rel->relkind)
+ {
+ case RELKIND_RELATION:
+ /* OK */
+ break;
+ case RELKIND_SEQUENCE:
+ /* Must disallow this because we don't vacuum sequences */
+ ereport(ERROR,
+ (errcode(ERRCODE_WRONG_OBJECT_TYPE),
+ errmsg("cannot lock rows in sequence \"%s\"",
+ RelationGetRelationName(rel))));
+ break;
+ case RELKIND_TOASTVALUE:
+ /* We could allow this, but there seems no good reason to */
+ ereport(ERROR,
+ (errcode(ERRCODE_WRONG_OBJECT_TYPE),
+ errmsg("cannot lock rows in TOAST relation \"%s\"",
+ RelationGetRelationName(rel))));
+ break;
+ case RELKIND_VIEW:
+ /* Should not get here */
+ ereport(ERROR,
+ (errcode(ERRCODE_WRONG_OBJECT_TYPE),
+ errmsg("cannot lock rows in view \"%s\"",
+ RelationGetRelationName(rel))));
+ break;
+ case RELKIND_FOREIGN_TABLE:
+ /* Perhaps we can support this someday, but not today */
+ ereport(ERROR,
+ (errcode(ERRCODE_WRONG_OBJECT_TYPE),
+ errmsg("cannot lock rows in foreign table \"%s\"",
+ RelationGetRelationName(rel))));
+ break;
+ default:
+ ereport(ERROR,
+ (errcode(ERRCODE_WRONG_OBJECT_TYPE),
+ errmsg("cannot lock rows in relation \"%s\"",
+ RelationGetRelationName(rel))));
+ break;
+ }
+}
+
+/*
+ * Initialize ResultRelInfo data for one result relation
+ *
+ * Caution: before Postgres 9.1, this function included the relkind checking
+ * that's now in CheckValidResultRel, and it also did ExecOpenIndices if
+ * appropriate. Be sure callers cover those needs.
+ */
+void
+InitResultRelInfo(ResultRelInfo *resultRelInfo,
+ Relation resultRelationDesc,
+ Index resultRelationIndex,
+ int instrument_options)
+{
+ MemSet(resultRelInfo, 0, sizeof(ResultRelInfo));
+ resultRelInfo->type = T_ResultRelInfo;
+ resultRelInfo->ri_RangeTableIndex = resultRelationIndex;
+ resultRelInfo->ri_RelationDesc = resultRelationDesc;
+ resultRelInfo->ri_NumIndices = 0;
+ resultRelInfo->ri_IndexRelationDescs = NULL;
+ resultRelInfo->ri_IndexRelationInfo = NULL;
+ /* make a copy so as not to depend on relcache info not changing... */
+ resultRelInfo->ri_TrigDesc = CopyTriggerDesc(resultRelationDesc->trigdesc);
+ if (resultRelInfo->ri_TrigDesc)
+ {
+ int n = resultRelInfo->ri_TrigDesc->numtriggers;
resultRelInfo->ri_TrigFunctions = (FmgrInfo *)
palloc0(n * sizeof(FmgrInfo));
- if (doInstrument)
- resultRelInfo->ri_TrigInstrument = InstrAlloc(n);
- else
- resultRelInfo->ri_TrigInstrument = NULL;
+ resultRelInfo->ri_TrigWhenExprs = (List **)
+ palloc0(n * sizeof(List *));
+ if (instrument_options)
+ resultRelInfo->ri_TrigInstrument = InstrAlloc(n, instrument_options);
}
else
{
resultRelInfo->ri_TrigFunctions = NULL;
+ resultRelInfo->ri_TrigWhenExprs = NULL;
resultRelInfo->ri_TrigInstrument = NULL;
}
resultRelInfo->ri_ConstraintExprs = NULL;
resultRelInfo->ri_junkFilter = NULL;
resultRelInfo->ri_projectReturning = NULL;
+}
+
+/*
+ * ExecGetTriggerResultRel
+ *
+ * Get a ResultRelInfo for a trigger target relation. Most of the time,
+ * triggers are fired on one of the result relations of the query, and so
+ * we can just return a member of the es_result_relations array. (Note: in
+ * self-join situations there might be multiple members with the same OID;
+ * if so it doesn't matter which one we pick.) However, it is sometimes
+ * necessary to fire triggers on other relations; this happens mainly when an
+ * RI update trigger queues additional triggers on other relations, which will
+ * be processed in the context of the outer query. For efficiency's sake,
+ * we want to have a ResultRelInfo for those triggers too; that can avoid
+ * repeated re-opening of the relation. (It also provides a way for EXPLAIN
+ * ANALYZE to report the runtimes of such triggers.) So we make additional
+ * ResultRelInfo's as needed, and save them in es_trig_target_relations.
+ */
+ResultRelInfo *
+ExecGetTriggerResultRel(EState *estate, Oid relid)
+{
+ ResultRelInfo *rInfo;
+ int nr;
+ ListCell *l;
+ Relation rel;
+ MemoryContext oldcontext;
+
+ /* First, search through the query result relations */
+ rInfo = estate->es_result_relations;
+ nr = estate->es_num_result_relations;
+ while (nr > 0)
+ {
+ if (RelationGetRelid(rInfo->ri_RelationDesc) == relid)
+ return rInfo;
+ rInfo++;
+ nr--;
+ }
+ /* Nope, but maybe we already made an extra ResultRelInfo for it */
+ foreach(l, estate->es_trig_target_relations)
+ {
+ rInfo = (ResultRelInfo *) lfirst(l);
+ if (RelationGetRelid(rInfo->ri_RelationDesc) == relid)
+ return rInfo;
+ }
+ /* Nope, so we need a new one */
+
+ /*
+ * Open the target relation's relcache entry. We assume that an
+ * appropriate lock is still held by the backend from whenever the trigger
+ * event got queued, so we need take no new lock here. Also, we need not
+ * recheck the relkind, so no need for CheckValidResultRel.
+ */
+ rel = heap_open(relid, NoLock);
/*
- * If there are indices on the result relation, open them and save
- * descriptors in the result relation info, so that we can add new index
- * entries for the tuples we add/update. We need not do this for a
- * DELETE, however, since deletion doesn't affect indexes.
+ * Make the new entry in the right context.
*/
- if (resultRelationDesc->rd_rel->relhasindex &&
- operation != CMD_DELETE)
- ExecOpenIndices(resultRelInfo);
+ oldcontext = MemoryContextSwitchTo(estate->es_query_cxt);
+ rInfo = makeNode(ResultRelInfo);
+ InitResultRelInfo(rInfo,
+ rel,
+ 0, /* dummy rangetable index */
+ estate->es_instrument);
+ estate->es_trig_target_relations =
+ lappend(estate->es_trig_target_relations, rInfo);
+ MemoryContextSwitchTo(oldcontext);
+
+ /*
+ * Currently, we don't need any index information in ResultRelInfos used
+ * only for triggers, so no need to call ExecOpenIndices.
+ */
+
+ return rInfo;
}
/*
* recognize how far down the requirement really goes, but for now we just
* make all plan nodes do the same thing if the top level forces the choice.
*
- * We assume that estate->es_result_relation_info is already set up to
- * describe the target relation. Note that in an UPDATE that spans an
- * inheritance tree, some of the target relations may have OIDs and some not.
- * We have to make the decisions on a per-relation basis as we initialize
- * each of the child plans of the topmost Append plan.
+ * We assume that if we are generating tuples for INSERT or UPDATE,
+ * estate->es_result_relation_info is already set up to describe the target
+ * relation. Note that in an UPDATE that spans an inheritance tree, some of
+ * the target relations may have OIDs and some not. We have to make the
+ * decisions on a per-relation basis as we initialize each of the subplans of
+ * the ModifyTable node, so ModifyTable has to set es_result_relation_info
+ * while initializing each subplan.
*
* SELECT INTO is even uglier, because we don't have the INTO relation's
* descriptor available when this code runs; we have to look aside at a
bool
ExecContextForcesOids(PlanState *planstate, bool *hasoids)
{
+ ResultRelInfo *ri = planstate->state->es_result_relation_info;
+
+ if (ri != NULL)
+ {
+ Relation rel = ri->ri_RelationDesc;
+
+ if (rel != NULL)
+ {
+ *hasoids = rel->rd_rel->relhasoids;
+ return true;
+ }
+ }
+
if (planstate->state->es_select_into)
{
*hasoids = planstate->state->es_into_oids;
return true;
}
- else
+
+ return false;
+}
+
+/* ----------------------------------------------------------------
+ * ExecPostprocessPlan
+ *
+ * Give plan nodes a final chance to execute before shutdown
+ * ----------------------------------------------------------------
+ */
+static void
+ExecPostprocessPlan(EState *estate)
+{
+ ListCell *lc;
+
+ /*
+ * Make sure nodes run forward.
+ */
+ estate->es_direction = ForwardScanDirection;
+
+ /*
+ * Run any secondary ModifyTable nodes to completion, in case the main
+ * query did not fetch all rows from them. (We do this to ensure that
+ * such nodes have predictable results.)
+ */
+ foreach(lc, estate->es_auxmodifytables)
{
- ResultRelInfo *ri = planstate->state->es_result_relation_info;
+ PlanState *ps = (PlanState *) lfirst(lc);
- if (ri != NULL)
+ for (;;)
{
- Relation rel = ri->ri_RelationDesc;
+ TupleTableSlot *slot;
- if (rel != NULL)
- {
- *hasoids = rel->rd_rel->relhasoids;
- return true;
- }
+ /* Reset the per-output-tuple exprcontext each time */
+ ResetPerTupleExprContext(estate);
+
+ slot = ExecProcNode(ps);
+
+ if (TupIsNull(slot))
+ break;
}
}
-
- return false;
}
/* ----------------------------------------------------------------
ListCell *l;
/*
- * shut down any PlanQual processing we were doing
- */
- if (estate->es_evalPlanQual != NULL)
- EndEvalPlanQual(estate);
-
- /*
* shut down the node-type-specific query processing
*/
ExecEndNode(planstate);
*/
foreach(l, estate->es_subplanstates)
{
- PlanState *subplanstate = (PlanState *) lfirst(l);
+ PlanState *subplanstate = (PlanState *) lfirst(l);
ExecEndNode(subplanstate);
}
/*
- * destroy the executor "tuple" table.
+ * destroy the executor's tuple table. Actually we only care about
+ * releasing buffer pins and tupdesc refcounts; there's no need to pfree
+ * the TupleTableSlots, since the containing memory context is about to go
+ * away anyway.
*/
- ExecDropTupleTable(estate->es_tupleTable, true);
- estate->es_tupleTable = NULL;
+ ExecResetTupleTable(estate->es_tupleTable, false);
/*
* close the result relation(s) if any, but hold locks until xact commit.
}
/*
+ * likewise close any trigger target relations
+ */
+ foreach(l, estate->es_trig_target_relations)
+ {
+ resultRelInfo = (ResultRelInfo *) lfirst(l);
+ /* Close indices and then the relation itself */
+ ExecCloseIndices(resultRelInfo);
+ heap_close(resultRelInfo->ri_RelationDesc, NoLock);
+ }
+
+ /*
* close any relations selected FOR UPDATE/FOR SHARE, again keeping locks
*/
foreach(l, estate->es_rowMarks)
{
- ExecRowMark *erm = lfirst(l);
+ ExecRowMark *erm = (ExecRowMark *) lfirst(l);
- heap_close(erm->relation, NoLock);
+ if (erm->relation)
+ heap_close(erm->relation, NoLock);
}
}
/* ----------------------------------------------------------------
* ExecutePlan
*
- * processes the query plan to retrieve 'numberTuples' tuples in the
- * direction specified.
- *
- * Retrieves all tuples if numberTuples is 0
+ * Processes the query plan until we have processed 'numberTuples' tuples,
+ * moving in the specified direction.
*
- * result is either a slot containing the last tuple in the case
- * of a SELECT or NULL otherwise.
+ * Runs to completion if numberTuples is 0
*
* Note: the ctid attribute is a 'junk' attribute that is removed before the
* user can see it
* ----------------------------------------------------------------
*/
-static TupleTableSlot *
+static void
ExecutePlan(EState *estate,
PlanState *planstate,
CmdType operation,
+ bool sendTuples,
long numberTuples,
ScanDirection direction,
DestReceiver *dest)
{
- JunkFilter *junkfilter;
- TupleTableSlot *planSlot;
TupleTableSlot *slot;
- ItemPointer tupleid = NULL;
- ItemPointerData tuple_ctid;
long current_tuple_count;
- TupleTableSlot *result;
/*
* initialize local variables
*/
current_tuple_count = 0;
- result = NULL;
/*
* Set the direction.
estate->es_direction = direction;
/*
- * Process BEFORE EACH STATEMENT triggers
- */
- switch (operation)
- {
- case CMD_UPDATE:
- ExecBSUpdateTriggers(estate, estate->es_result_relation_info);
- break;
- case CMD_DELETE:
- ExecBSDeleteTriggers(estate, estate->es_result_relation_info);
- break;
- case CMD_INSERT:
- ExecBSInsertTriggers(estate, estate->es_result_relation_info);
- break;
- default:
- /* do nothing */
- break;
- }
-
- /*
* Loop until we've processed the proper number of tuples from the plan.
*/
-
for (;;)
{
/* Reset the per-output-tuple exprcontext */
/*
* Execute the plan and obtain a tuple
*/
-lnext: ;
- if (estate->es_useEvalPlan)
- {
- planSlot = EvalPlanQualNext(estate);
- if (TupIsNull(planSlot))
- planSlot = ExecProcNode(planstate);
- }
- else
- planSlot = ExecProcNode(planstate);
+ slot = ExecProcNode(planstate);
/*
* if the tuple is null, then we assume there is nothing more to
- * process so we just return null...
+ * process so we just end the loop...
*/
- if (TupIsNull(planSlot))
- {
- result = NULL;
+ if (TupIsNull(slot))
break;
- }
- slot = planSlot;
/*
- * if we have a junk filter, then project a new tuple with the junk
+ * If we have a junk filter, then project a new tuple with the junk
* removed.
*
* Store this new "clean" tuple in the junkfilter's resultSlot.
* (Formerly, we stored it back over the "dirty" tuple, which is WRONG
* because that tuple slot has the wrong descriptor.)
- *
- * Also, extract all the junk information we need.
*/
- if ((junkfilter = estate->es_junkFilter) != NULL)
- {
- Datum datum;
- bool isNull;
-
- /*
- * extract the 'ctid' junk attribute.
- */
- if (operation == CMD_UPDATE || operation == CMD_DELETE)
- {
- datum = ExecGetJunkAttribute(slot, junkfilter->jf_junkAttNo,
- &isNull);
- /* shouldn't ever get a null result... */
- if (isNull)
- elog(ERROR, "ctid is NULL");
-
- tupleid = (ItemPointer) DatumGetPointer(datum);
- tuple_ctid = *tupleid; /* make sure we don't free the ctid!! */
- tupleid = &tuple_ctid;
- }
-
- /*
- * Process any FOR UPDATE or FOR SHARE locking requested.
- */
- else if (estate->es_rowMarks != NIL)
- {
- ListCell *l;
-
- lmark: ;
- foreach(l, estate->es_rowMarks)
- {
- ExecRowMark *erm = lfirst(l);
- HeapTupleData tuple;
- Buffer buffer;
- ItemPointerData update_ctid;
- TransactionId update_xmax;
- TupleTableSlot *newSlot;
- LockTupleMode lockmode;
- HTSU_Result test;
-
- datum = ExecGetJunkAttribute(slot,
- erm->ctidAttNo,
- &isNull);
- /* shouldn't ever get a null result... */
- if (isNull)
- elog(ERROR, "ctid is NULL");
-
- tuple.t_self = *((ItemPointer) DatumGetPointer(datum));
-
- if (erm->forUpdate)
- lockmode = LockTupleExclusive;
- else
- lockmode = LockTupleShared;
-
- test = heap_lock_tuple(erm->relation, &tuple, &buffer,
- &update_ctid, &update_xmax,
- estate->es_snapshot->curcid,
- lockmode, erm->noWait);
- ReleaseBuffer(buffer);
- switch (test)
- {
- case HeapTupleSelfUpdated:
- /* treat it as deleted; do not process */
- goto lnext;
-
- case HeapTupleMayBeUpdated:
- break;
-
- case HeapTupleUpdated:
- if (IsXactIsoLevelSerializable)
- ereport(ERROR,
- (errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
- errmsg("could not serialize access due to concurrent update")));
- if (!ItemPointerEquals(&update_ctid,
- &tuple.t_self))
- {
- /* updated, so look at updated version */
- newSlot = EvalPlanQual(estate,
- erm->rti,
- &update_ctid,
- update_xmax,
- estate->es_snapshot->curcid);
- if (!TupIsNull(newSlot))
- {
- slot = planSlot = newSlot;
- estate->es_useEvalPlan = true;
- goto lmark;
- }
- }
-
- /*
- * if tuple was deleted or PlanQual failed for
- * updated tuple - we must not return this tuple!
- */
- goto lnext;
-
- default:
- elog(ERROR, "unrecognized heap_lock_tuple status: %u",
- test);
- return NULL;
- }
- }
- }
-
- /*
- * Create a new "clean" tuple with all junk attributes removed. We
- * don't need to do this for DELETE, however (there will in fact
- * be no non-junk attributes in a DELETE!)
- */
- if (operation != CMD_DELETE)
- slot = ExecFilterJunk(junkfilter, slot);
- }
+ if (estate->es_junkFilter != NULL)
+ slot = ExecFilterJunk(estate->es_junkFilter, slot);
/*
- * now that we have a tuple, do the appropriate thing with it.. either
- * return it to the user, add it to a relation someplace, delete it
- * from a relation, or modify some of its attributes.
+ * If we are supposed to send the tuple somewhere, do so. (In
+ * practice, this is probably always the case at this point.)
*/
- switch (operation)
- {
- case CMD_SELECT:
- ExecSelect(slot, dest, estate);
- result = slot;
- break;
-
- case CMD_INSERT:
- ExecInsert(slot, tupleid, planSlot, dest, estate);
- result = NULL;
- break;
-
- case CMD_DELETE:
- ExecDelete(tupleid, planSlot, dest, estate);
- result = NULL;
- break;
+ if (sendTuples)
+ (*dest->receiveSlot) (slot, dest);
- case CMD_UPDATE:
- ExecUpdate(slot, tupleid, planSlot, dest, estate);
- result = NULL;
- break;
-
- default:
- elog(ERROR, "unrecognized operation code: %d",
- (int) operation);
- result = NULL;
- break;
- }
+ /*
+ * Count tuples processed, if this is a SELECT. (For other operation
+ * types, the ModifyTable plan node must count the appropriate
+ * events.)
+ */
+ if (operation == CMD_SELECT)
+ (estate->es_processed)++;
/*
* check our tuple count.. if we've processed the proper number then
if (numberTuples && numberTuples == current_tuple_count)
break;
}
-
- /*
- * Process AFTER EACH STATEMENT triggers
- */
- switch (operation)
- {
- case CMD_UPDATE:
- ExecASUpdateTriggers(estate, estate->es_result_relation_info);
- break;
- case CMD_DELETE:
- ExecASDeleteTriggers(estate, estate->es_result_relation_info);
- break;
- case CMD_INSERT:
- ExecASInsertTriggers(estate, estate->es_result_relation_info);
- break;
- default:
- /* do nothing */
- break;
- }
-
- /*
- * here, result is either a slot containing a tuple in the case of a
- * SELECT or NULL otherwise.
- */
- return result;
-}
-
-/* ----------------------------------------------------------------
- * ExecSelect
- *
- * SELECTs are easy.. we just pass the tuple to the appropriate
- * output function.
- * ----------------------------------------------------------------
- */
-static void
-ExecSelect(TupleTableSlot *slot,
- DestReceiver *dest,
- EState *estate)
-{
- (*dest->receiveSlot) (slot, dest);
- IncrRetrieved();
- (estate->es_processed)++;
-}
-
-/* ----------------------------------------------------------------
- * ExecInsert
- *
- * INSERTs are trickier.. we have to insert the tuple into
- * the base relation and insert appropriate tuples into the
- * index relations.
- * ----------------------------------------------------------------
- */
-static void
-ExecInsert(TupleTableSlot *slot,
- ItemPointer tupleid,
- TupleTableSlot *planSlot,
- DestReceiver *dest,
- EState *estate)
-{
- HeapTuple tuple;
- ResultRelInfo *resultRelInfo;
- Relation resultRelationDesc;
- Oid newId;
-
- /*
- * get the heap tuple out of the tuple table slot, making sure we have a
- * writable copy
- */
- tuple = ExecMaterializeSlot(slot);
-
- /*
- * get information on the (current) result relation
- */
- resultRelInfo = estate->es_result_relation_info;
- resultRelationDesc = resultRelInfo->ri_RelationDesc;
-
- /* BEFORE ROW INSERT Triggers */
- if (resultRelInfo->ri_TrigDesc &&
- resultRelInfo->ri_TrigDesc->n_before_row[TRIGGER_EVENT_INSERT] > 0)
- {
- HeapTuple newtuple;
-
- newtuple = ExecBRInsertTriggers(estate, resultRelInfo, tuple);
-
- if (newtuple == NULL) /* "do nothing" */
- return;
-
- if (newtuple != tuple) /* modified by Trigger(s) */
- {
- /*
- * Put the modified tuple into a slot for convenience of routines
- * below. We assume the tuple was allocated in per-tuple memory
- * context, and therefore will go away by itself. The tuple table
- * slot should not try to clear it.
- */
- TupleTableSlot *newslot = estate->es_trig_tuple_slot;
-
- if (newslot->tts_tupleDescriptor != slot->tts_tupleDescriptor)
- ExecSetSlotDescriptor(newslot, slot->tts_tupleDescriptor);
- ExecStoreTuple(newtuple, newslot, InvalidBuffer, false);
- slot = newslot;
- tuple = newtuple;
- }
- }
-
- /*
- * Check the constraints of the tuple
- */
- if (resultRelationDesc->rd_att->constr)
- ExecConstraints(resultRelInfo, slot, estate);
-
- /*
- * insert the tuple
- *
- * Note: heap_insert returns the tid (location) of the new tuple in the
- * t_self field.
- */
- newId = heap_insert(resultRelationDesc, tuple,
- estate->es_snapshot->curcid,
- true, true);
-
- IncrAppended();
- (estate->es_processed)++;
- estate->es_lastoid = newId;
- setLastTid(&(tuple->t_self));
-
- /*
- * insert index entries for tuple
- */
- if (resultRelInfo->ri_NumIndices > 0)
- ExecInsertIndexTuples(slot, &(tuple->t_self), estate, false);
-
- /* AFTER ROW INSERT Triggers */
- ExecARInsertTriggers(estate, resultRelInfo, tuple);
-
- /* Process RETURNING if present */
- if (resultRelInfo->ri_projectReturning)
- ExecProcessReturning(resultRelInfo->ri_projectReturning,
- slot, planSlot, dest);
-}
-
-/* ----------------------------------------------------------------
- * ExecDelete
- *
- * DELETE is like UPDATE, except that we delete the tuple and no
- * index modifications are needed
- * ----------------------------------------------------------------
- */
-static void
-ExecDelete(ItemPointer tupleid,
- TupleTableSlot *planSlot,
- DestReceiver *dest,
- EState *estate)
-{
- ResultRelInfo *resultRelInfo;
- Relation resultRelationDesc;
- HTSU_Result result;
- ItemPointerData update_ctid;
- TransactionId update_xmax;
-
- /*
- * get information on the (current) result relation
- */
- resultRelInfo = estate->es_result_relation_info;
- resultRelationDesc = resultRelInfo->ri_RelationDesc;
-
- /* BEFORE ROW DELETE Triggers */
- if (resultRelInfo->ri_TrigDesc &&
- resultRelInfo->ri_TrigDesc->n_before_row[TRIGGER_EVENT_DELETE] > 0)
- {
- bool dodelete;
-
- dodelete = ExecBRDeleteTriggers(estate, resultRelInfo, tupleid,
- estate->es_snapshot->curcid);
-
- if (!dodelete) /* "do nothing" */
- return;
- }
-
- /*
- * delete the tuple
- *
- * Note: if es_crosscheck_snapshot isn't InvalidSnapshot, we check that
- * the row to be deleted is visible to that snapshot, and throw a can't-
- * serialize error if not. This is a special-case behavior needed for
- * referential integrity updates in serializable transactions.
- */
-ldelete:;
- result = heap_delete(resultRelationDesc, tupleid,
- &update_ctid, &update_xmax,
- estate->es_snapshot->curcid,
- estate->es_crosscheck_snapshot,
- true /* wait for commit */ );
- switch (result)
- {
- case HeapTupleSelfUpdated:
- /* already deleted by self; nothing to do */
- return;
-
- case HeapTupleMayBeUpdated:
- break;
-
- case HeapTupleUpdated:
- if (IsXactIsoLevelSerializable)
- ereport(ERROR,
- (errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
- errmsg("could not serialize access due to concurrent update")));
- else if (!ItemPointerEquals(tupleid, &update_ctid))
- {
- TupleTableSlot *epqslot;
-
- epqslot = EvalPlanQual(estate,
- resultRelInfo->ri_RangeTableIndex,
- &update_ctid,
- update_xmax,
- estate->es_snapshot->curcid);
- if (!TupIsNull(epqslot))
- {
- *tupleid = update_ctid;
- goto ldelete;
- }
- }
- /* tuple already deleted; nothing to do */
- return;
-
- default:
- elog(ERROR, "unrecognized heap_delete status: %u", result);
- return;
- }
-
- IncrDeleted();
- (estate->es_processed)++;
-
- /*
- * Note: Normally one would think that we have to delete index tuples
- * associated with the heap tuple now...
- *
- * ... but in POSTGRES, we have no need to do this because VACUUM will
- * take care of it later. We can't delete index tuples immediately
- * anyway, since the tuple is still visible to other transactions.
- */
-
- /* AFTER ROW DELETE Triggers */
- ExecARDeleteTriggers(estate, resultRelInfo, tupleid);
-
- /* Process RETURNING if present */
- if (resultRelInfo->ri_projectReturning)
- {
- /*
- * We have to put the target tuple into a slot, which means first we
- * gotta fetch it. We can use the trigger tuple slot.
- */
- TupleTableSlot *slot = estate->es_trig_tuple_slot;
- HeapTupleData deltuple;
- Buffer delbuffer;
-
- deltuple.t_self = *tupleid;
- if (!heap_fetch(resultRelationDesc, SnapshotAny,
- &deltuple, &delbuffer, false, NULL))
- elog(ERROR, "failed to fetch deleted tuple for DELETE RETURNING");
-
- if (slot->tts_tupleDescriptor != RelationGetDescr(resultRelationDesc))
- ExecSetSlotDescriptor(slot, RelationGetDescr(resultRelationDesc));
- ExecStoreTuple(&deltuple, slot, InvalidBuffer, false);
-
- ExecProcessReturning(resultRelInfo->ri_projectReturning,
- slot, planSlot, dest);
-
- ExecClearTuple(slot);
- ReleaseBuffer(delbuffer);
- }
}
-/* ----------------------------------------------------------------
- * ExecUpdate
- *
- * note: we can't run UPDATE queries with transactions
- * off because UPDATEs are actually INSERTs and our
- * scan will mistakenly loop forever, updating the tuple
- * it just inserted.. This should be fixed but until it
- * is, we don't want to get stuck in an infinite loop
- * which corrupts your database..
- * ----------------------------------------------------------------
- */
-static void
-ExecUpdate(TupleTableSlot *slot,
- ItemPointer tupleid,
- TupleTableSlot *planSlot,
- DestReceiver *dest,
- EState *estate)
-{
- HeapTuple tuple;
- ResultRelInfo *resultRelInfo;
- Relation resultRelationDesc;
- HTSU_Result result;
- ItemPointerData update_ctid;
- TransactionId update_xmax;
-
- /*
- * abort the operation if not running transactions
- */
- if (IsBootstrapProcessingMode())
- elog(ERROR, "cannot UPDATE during bootstrap");
-
- /*
- * get the heap tuple out of the tuple table slot, making sure we have a
- * writable copy
- */
- tuple = ExecMaterializeSlot(slot);
-
- /*
- * get information on the (current) result relation
- */
- resultRelInfo = estate->es_result_relation_info;
- resultRelationDesc = resultRelInfo->ri_RelationDesc;
-
- /* BEFORE ROW UPDATE Triggers */
- if (resultRelInfo->ri_TrigDesc &&
- resultRelInfo->ri_TrigDesc->n_before_row[TRIGGER_EVENT_UPDATE] > 0)
- {
- HeapTuple newtuple;
-
- newtuple = ExecBRUpdateTriggers(estate, resultRelInfo,
- tupleid, tuple,
- estate->es_snapshot->curcid);
-
- if (newtuple == NULL) /* "do nothing" */
- return;
-
- if (newtuple != tuple) /* modified by Trigger(s) */
- {
- /*
- * Put the modified tuple into a slot for convenience of routines
- * below. We assume the tuple was allocated in per-tuple memory
- * context, and therefore will go away by itself. The tuple table
- * slot should not try to clear it.
- */
- TupleTableSlot *newslot = estate->es_trig_tuple_slot;
-
- if (newslot->tts_tupleDescriptor != slot->tts_tupleDescriptor)
- ExecSetSlotDescriptor(newslot, slot->tts_tupleDescriptor);
- ExecStoreTuple(newtuple, newslot, InvalidBuffer, false);
- slot = newslot;
- tuple = newtuple;
- }
- }
-
- /*
- * Check the constraints of the tuple
- *
- * If we generate a new candidate tuple after EvalPlanQual testing, we
- * must loop back here and recheck constraints. (We don't need to redo
- * triggers, however. If there are any BEFORE triggers then trigger.c
- * will have done heap_lock_tuple to lock the correct tuple, so there's no
- * need to do them again.)
- */
-lreplace:;
- if (resultRelationDesc->rd_att->constr)
- ExecConstraints(resultRelInfo, slot, estate);
-
- /*
- * replace the heap tuple
- *
- * Note: if es_crosscheck_snapshot isn't InvalidSnapshot, we check that
- * the row to be updated is visible to that snapshot, and throw a can't-
- * serialize error if not. This is a special-case behavior needed for
- * referential integrity updates in serializable transactions.
- */
- result = heap_update(resultRelationDesc, tupleid, tuple,
- &update_ctid, &update_xmax,
- estate->es_snapshot->curcid,
- estate->es_crosscheck_snapshot,
- true /* wait for commit */ );
- switch (result)
- {
- case HeapTupleSelfUpdated:
- /* already deleted by self; nothing to do */
- return;
-
- case HeapTupleMayBeUpdated:
- break;
-
- case HeapTupleUpdated:
- if (IsXactIsoLevelSerializable)
- ereport(ERROR,
- (errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
- errmsg("could not serialize access due to concurrent update")));
- else if (!ItemPointerEquals(tupleid, &update_ctid))
- {
- TupleTableSlot *epqslot;
-
- epqslot = EvalPlanQual(estate,
- resultRelInfo->ri_RangeTableIndex,
- &update_ctid,
- update_xmax,
- estate->es_snapshot->curcid);
- if (!TupIsNull(epqslot))
- {
- *tupleid = update_ctid;
- slot = ExecFilterJunk(estate->es_junkFilter, epqslot);
- tuple = ExecMaterializeSlot(slot);
- goto lreplace;
- }
- }
- /* tuple already deleted; nothing to do */
- return;
-
- default:
- elog(ERROR, "unrecognized heap_update status: %u", result);
- return;
- }
-
- IncrReplaced();
- (estate->es_processed)++;
-
- /*
- * Note: instead of having to update the old index tuples associated with
- * the heap tuple, all we do is form and insert new index tuples. This is
- * because UPDATEs are actually DELETEs and INSERTs, and index tuple
- * deletion is done later by VACUUM (see notes in ExecDelete). All we do
- * here is insert new index tuples. -cim 9/27/89
- */
-
- /*
- * insert index entries for tuple
- *
- * Note: heap_update returns the tid (location) of the new tuple in the
- * t_self field.
- */
- if (resultRelInfo->ri_NumIndices > 0)
- ExecInsertIndexTuples(slot, &(tuple->t_self), estate, false);
-
- /* AFTER ROW UPDATE Triggers */
- ExecARUpdateTriggers(estate, resultRelInfo, tupleid, tuple);
-
- /* Process RETURNING if present */
- if (resultRelInfo->ri_projectReturning)
- ExecProcessReturning(resultRelInfo->ri_projectReturning,
- slot, planSlot, dest);
-}
/*
* ExecRelCheck --- check that tuple meets constraints for result relation
}
}
+
/*
- * ExecProcessReturning --- evaluate a RETURNING list and send to dest
+ * ExecFindRowMark -- find the ExecRowMark struct for given rangetable index
+ */
+ExecRowMark *
+ExecFindRowMark(EState *estate, Index rti)
+{
+ ListCell *lc;
+
+ foreach(lc, estate->es_rowMarks)
+ {
+ ExecRowMark *erm = (ExecRowMark *) lfirst(lc);
+
+ if (erm->rti == rti)
+ return erm;
+ }
+ elog(ERROR, "failed to find ExecRowMark for rangetable index %u", rti);
+ return NULL; /* keep compiler quiet */
+}
+
+/*
+ * ExecBuildAuxRowMark -- create an ExecAuxRowMark struct
*
- * projectReturning: RETURNING projection info for current result rel
- * tupleSlot: slot holding tuple actually inserted/updated/deleted
- * planSlot: slot holding tuple returned by top plan node
- * dest: where to send the output
+ * Inputs are the underlying ExecRowMark struct and the targetlist of the
+ * input plan node (not planstate node!). We need the latter to find out
+ * the column numbers of the resjunk columns.
*/
-static void
-ExecProcessReturning(ProjectionInfo *projectReturning,
- TupleTableSlot *tupleSlot,
- TupleTableSlot *planSlot,
- DestReceiver *dest)
+ExecAuxRowMark *
+ExecBuildAuxRowMark(ExecRowMark *erm, List *targetlist)
{
- ExprContext *econtext = projectReturning->pi_exprContext;
- TupleTableSlot *retSlot;
+ ExecAuxRowMark *aerm = (ExecAuxRowMark *) palloc0(sizeof(ExecAuxRowMark));
+ char resname[32];
- /*
- * Reset per-tuple memory context to free any expression evaluation
- * storage allocated in the previous cycle.
- */
- ResetExprContext(econtext);
+ aerm->rowmark = erm;
- /* Make tuple and any needed join variables available to ExecProject */
- econtext->ecxt_scantuple = tupleSlot;
- econtext->ecxt_outertuple = planSlot;
+ /* Look up the resjunk columns associated with this rowmark */
+ if (erm->relation)
+ {
+ Assert(erm->markType != ROW_MARK_COPY);
- /* Compute the RETURNING expressions */
- retSlot = ExecProject(projectReturning, NULL);
+ /* if child rel, need tableoid */
+ if (erm->rti != erm->prti)
+ {
+ snprintf(resname, sizeof(resname), "tableoid%u", erm->rowmarkId);
+ aerm->toidAttNo = ExecFindJunkAttributeInTlist(targetlist,
+ resname);
+ if (!AttributeNumberIsValid(aerm->toidAttNo))
+ elog(ERROR, "could not find junk %s column", resname);
+ }
- /* Send to dest */
- (*dest->receiveSlot) (retSlot, dest);
+ /* always need ctid for real relations */
+ snprintf(resname, sizeof(resname), "ctid%u", erm->rowmarkId);
+ aerm->ctidAttNo = ExecFindJunkAttributeInTlist(targetlist,
+ resname);
+ if (!AttributeNumberIsValid(aerm->ctidAttNo))
+ elog(ERROR, "could not find junk %s column", resname);
+ }
+ else
+ {
+ Assert(erm->markType == ROW_MARK_COPY);
+
+ snprintf(resname, sizeof(resname), "wholerow%u", erm->rowmarkId);
+ aerm->wholeAttNo = ExecFindJunkAttributeInTlist(targetlist,
+ resname);
+ if (!AttributeNumberIsValid(aerm->wholeAttNo))
+ elog(ERROR, "could not find junk %s column", resname);
+ }
+
+ return aerm;
+}
+
+
+/*
+ * EvalPlanQual logic --- recheck modified tuple(s) to see if we want to
+ * process the updated version under READ COMMITTED rules.
+ *
+ * See backend/executor/README for some info about how this works.
+ */
- ExecClearTuple(retSlot);
-}
/*
* Check a modified tuple to see if we want to process its updated version
* under READ COMMITTED rules.
*
- * See backend/executor/README for some info about how this works.
- *
- * estate - executor state data
+ * estate - outer executor state data
+ * epqstate - state for EvalPlanQual rechecking
+ * relation - table containing tuple
* rti - rangetable index of table containing tuple
* *tid - t_ctid from the outdated tuple (ie, next updated version)
* priorXmax - t_xmax from the outdated tuple
- * curCid - command ID of current command of my transaction
*
* *tid is also an output parameter: it's modified to hold the TID of the
* latest version of the tuple (note this may be changed even on failure)
* NULL if we determine we shouldn't process the row.
*/
TupleTableSlot *
-EvalPlanQual(EState *estate, Index rti,
- ItemPointer tid, TransactionId priorXmax, CommandId curCid)
+EvalPlanQual(EState *estate, EPQState *epqstate,
+ Relation relation, Index rti,
+ ItemPointer tid, TransactionId priorXmax)
{
- evalPlanQual *epq;
- EState *epqstate;
- Relation relation;
- HeapTupleData tuple;
- HeapTuple copyTuple = NULL;
- SnapshotData SnapshotDirty;
- bool endNode;
+ TupleTableSlot *slot;
+ HeapTuple copyTuple;
- Assert(rti != 0);
+ Assert(rti > 0);
/*
- * find relation containing target tuple
+ * Get and lock the updated version of the row; if fail, return NULL.
*/
- if (estate->es_result_relation_info != NULL &&
- estate->es_result_relation_info->ri_RangeTableIndex == rti)
- relation = estate->es_result_relation_info->ri_RelationDesc;
- else
- {
- ListCell *l;
+ copyTuple = EvalPlanQualFetch(estate, relation, LockTupleExclusive,
+ tid, priorXmax);
- relation = NULL;
- foreach(l, estate->es_rowMarks)
- {
- if (((ExecRowMark *) lfirst(l))->rti == rti)
- {
- relation = ((ExecRowMark *) lfirst(l))->relation;
- break;
- }
- }
- if (relation == NULL)
- elog(ERROR, "could not find RowMark for RT index %u", rti);
- }
+ if (copyTuple == NULL)
+ return NULL;
+
+ /*
+ * For UPDATE/DELETE we have to return tid of actual row we're executing
+ * PQ for.
+ */
+ *tid = copyTuple->t_self;
+
+ /*
+ * Need to run a recheck subquery. Initialize or reinitialize EPQ state.
+ */
+ EvalPlanQualBegin(epqstate, estate);
+
+ /*
+ * Free old test tuple, if any, and store new tuple where relation's scan
+ * node will see it
+ */
+ EvalPlanQualSetTuple(epqstate, rti, copyTuple);
+
+ /*
+ * Fetch any non-locked source rows
+ */
+ EvalPlanQualFetchRowMarks(epqstate);
+
+ /*
+ * Run the EPQ query. We assume it will return at most one tuple.
+ */
+ slot = EvalPlanQualNext(epqstate);
+
+ /*
+ * If we got a tuple, force the slot to materialize the tuple so that it
+ * is not dependent on any local state in the EPQ query (in particular,
+ * it's highly likely that the slot contains references to any pass-by-ref
+ * datums that may be present in copyTuple). As with the next step, this
+ * is to guard against early re-use of the EPQ query.
+ */
+ if (!TupIsNull(slot))
+ (void) ExecMaterializeSlot(slot);
+
+ /*
+ * Clear out the test tuple. This is needed in case the EPQ query is
+ * re-used to test a tuple for a different relation. (Not clear that can
+ * really happen, but let's be safe.)
+ */
+ EvalPlanQualSetTuple(epqstate, rti, NULL);
+
+ return slot;
+}
+
+/*
+ * Fetch a copy of the newest version of an outdated tuple
+ *
+ * estate - executor state data
+ * relation - table containing tuple
+ * lockmode - requested tuple lock mode
+ * *tid - t_ctid from the outdated tuple (ie, next updated version)
+ * priorXmax - t_xmax from the outdated tuple
+ *
+ * Returns a palloc'd copy of the newest tuple version, or NULL if we find
+ * that there is no newest version (ie, the row was deleted not updated).
+ * If successful, we have locked the newest tuple version, so caller does not
+ * need to worry about it changing anymore.
+ *
+ * Note: properly, lockmode should be declared as enum LockTupleMode,
+ * but we use "int" to avoid having to include heapam.h in executor.h.
+ */
+HeapTuple
+EvalPlanQualFetch(EState *estate, Relation relation, int lockmode,
+ ItemPointer tid, TransactionId priorXmax)
+{
+ HeapTuple copyTuple = NULL;
+ HeapTupleData tuple;
+ SnapshotData SnapshotDirty;
/*
- * fetch tid tuple
+ * fetch target tuple
*
* Loop here to deal with updated or busy tuples
*/
if (heap_fetch(relation, &SnapshotDirty, &tuple, &buffer, true, NULL))
{
+ HTSU_Result test;
+ ItemPointerData update_ctid;
+ TransactionId update_xmax;
+
/*
* If xmin isn't what we're expecting, the slot must have been
* recycled and reused for an unrelated tuple. This implies that
/*
* If tuple was inserted by our own transaction, we have to check
- * cmin against curCid: cmin >= curCid means our command cannot
- * see the tuple, so we should ignore it. Without this we are
- * open to the "Halloween problem" of indefinitely re-updating the
- * same tuple. (We need not check cmax because
+ * cmin against es_output_cid: cmin >= current CID means our
+ * command cannot see the tuple, so we should ignore it. Without
+ * this we are open to the "Halloween problem" of indefinitely
+ * re-updating the same tuple. (We need not check cmax because
* HeapTupleSatisfiesDirty will consider a tuple deleted by our
* transaction dead, regardless of cmax.) We just checked that
* priorXmax == xmin, so we can test that variable instead of
* doing HeapTupleHeaderGetXmin again.
*/
if (TransactionIdIsCurrentTransactionId(priorXmax) &&
- HeapTupleHeaderGetCmin(tuple.t_data) >= curCid)
+ HeapTupleHeaderGetCmin(tuple.t_data) >= estate->es_output_cid)
{
ReleaseBuffer(buffer);
return NULL;
}
/*
+ * This is a live tuple, so now try to lock it.
+ */
+ test = heap_lock_tuple(relation, &tuple, &buffer,
+ &update_ctid, &update_xmax,
+ estate->es_output_cid,
+ lockmode, false);
+ /* We now have two pins on the buffer, get rid of one */
+ ReleaseBuffer(buffer);
+
+ switch (test)
+ {
+ case HeapTupleSelfUpdated:
+ /* treat it as deleted; do not process */
+ ReleaseBuffer(buffer);
+ return NULL;
+
+ case HeapTupleMayBeUpdated:
+ /* successfully locked */
+ break;
+
+ case HeapTupleUpdated:
+ ReleaseBuffer(buffer);
+ if (IsolationUsesXactSnapshot())
+ ereport(ERROR,
+ (errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
+ errmsg("could not serialize access due to concurrent update")));
+ if (!ItemPointerEquals(&update_ctid, &tuple.t_self))
+ {
+ /* it was updated, so look at the updated version */
+ tuple.t_self = update_ctid;
+ /* updated row should have xmin matching this xmax */
+ priorXmax = update_xmax;
+ continue;
+ }
+ /* tuple was deleted, so give up */
+ return NULL;
+
+ default:
+ ReleaseBuffer(buffer);
+ elog(ERROR, "unrecognized heap_lock_tuple status: %u",
+ test);
+ return NULL; /* keep compiler quiet */
+ }
+
+ /*
* We got tuple - now copy it for use by recheck query.
*/
copyTuple = heap_copytuple(&tuple);
* mean that the row was updated or deleted by either a committed xact
* or our own xact. If it was deleted, we can ignore it; if it was
* updated then chain up to the next version and repeat the whole
- * test.
+ * process.
*
* As above, it should be safe to examine xmax and t_ctid without the
* buffer content lock, because they can't be changing.
}
/*
- * For UPDATE/DELETE we have to return tid of actual row we're executing
- * PQ for.
+ * Return the copied tuple
*/
- *tid = tuple.t_self;
+ return copyTuple;
+}
- /*
- * Need to run a recheck subquery. Find or create a PQ stack entry.
- */
- epq = estate->es_evalPlanQual;
- endNode = true;
+/*
+ * EvalPlanQualInit -- initialize during creation of a plan state node
+ * that might need to invoke EPQ processing.
+ *
+ * Note: subplan/auxrowmarks can be NULL/NIL if they will be set later
+ * with EvalPlanQualSetPlan.
+ */
+void
+EvalPlanQualInit(EPQState *epqstate, EState *estate,
+ Plan *subplan, List *auxrowmarks, int epqParam)
+{
+ /* Mark the EPQ state inactive */
+ epqstate->estate = NULL;
+ epqstate->planstate = NULL;
+ epqstate->origslot = NULL;
+ /* ... and remember data that EvalPlanQualBegin will need */
+ epqstate->plan = subplan;
+ epqstate->arowMarks = auxrowmarks;
+ epqstate->epqParam = epqParam;
+}
- if (epq != NULL && epq->rti == 0)
- {
- /* Top PQ stack entry is idle, so re-use it */
- Assert(!(estate->es_useEvalPlan) && epq->next == NULL);
- epq->rti = rti;
- endNode = false;
- }
+/*
+ * EvalPlanQualSetPlan -- set or change subplan of an EPQState.
+ *
+ * We need this so that ModifyTuple can deal with multiple subplans.
+ */
+void
+EvalPlanQualSetPlan(EPQState *epqstate, Plan *subplan, List *auxrowmarks)
+{
+ /* If we have a live EPQ query, shut it down */
+ EvalPlanQualEnd(epqstate);
+ /* And set/change the plan pointer */
+ epqstate->plan = subplan;
+ /* The rowmarks depend on the plan, too */
+ epqstate->arowMarks = auxrowmarks;
+}
- /*
- * If this is request for another RTE - Ra, - then we have to check wasn't
- * PlanQual requested for Ra already and if so then Ra' row was updated
- * again and we have to re-start old execution for Ra and forget all what
- * we done after Ra was suspended. Cool? -:))
- */
- if (epq != NULL && epq->rti != rti &&
- epq->estate->es_evTuple[rti - 1] != NULL)
- {
- do
- {
- evalPlanQual *oldepq;
-
- /* stop execution */
- EvalPlanQualStop(epq);
- /* pop previous PlanQual from the stack */
- oldepq = epq->next;
- Assert(oldepq && oldepq->rti != 0);
- /* push current PQ to freePQ stack */
- oldepq->free = epq;
- epq = oldepq;
- estate->es_evalPlanQual = epq;
- } while (epq->rti != rti);
- }
+/*
+ * Install one test tuple into EPQ state, or clear test tuple if tuple == NULL
+ *
+ * NB: passed tuple must be palloc'd; it may get freed later
+ */
+void
+EvalPlanQualSetTuple(EPQState *epqstate, Index rti, HeapTuple tuple)
+{
+ EState *estate = epqstate->estate;
+
+ Assert(rti > 0);
/*
- * If we are requested for another RTE then we have to suspend execution
- * of current PlanQual and start execution for new one.
+ * free old test tuple, if any, and store new tuple where relation's scan
+ * node will see it
*/
- if (epq == NULL || epq->rti != rti)
- {
- /* try to reuse plan used previously */
- evalPlanQual *newepq = (epq != NULL) ? epq->free : NULL;
+ if (estate->es_epqTuple[rti - 1] != NULL)
+ heap_freetuple(estate->es_epqTuple[rti - 1]);
+ estate->es_epqTuple[rti - 1] = tuple;
+ estate->es_epqTupleSet[rti - 1] = true;
+}
- if (newepq == NULL) /* first call or freePQ stack is empty */
- {
- newepq = (evalPlanQual *) palloc0(sizeof(evalPlanQual));
- newepq->free = NULL;
- newepq->estate = NULL;
- newepq->planstate = NULL;
- }
- else
- {
- /* recycle previously used PlanQual */
- Assert(newepq->estate == NULL);
- epq->free = NULL;
- }
- /* push current PQ to the stack */
- newepq->next = epq;
- epq = newepq;
- estate->es_evalPlanQual = epq;
- epq->rti = rti;
- endNode = false;
- }
+/*
+ * Fetch back the current test tuple (if any) for the specified RTI
+ */
+HeapTuple
+EvalPlanQualGetTuple(EPQState *epqstate, Index rti)
+{
+ EState *estate = epqstate->estate;
- Assert(epq->rti == rti);
+ Assert(rti > 0);
- /*
- * Ok - we're requested for the same RTE. Unfortunately we still have to
- * end and restart execution of the plan, because ExecReScan wouldn't
- * ensure that upper plan nodes would reset themselves. We could make
- * that work if insertion of the target tuple were integrated with the
- * Param mechanism somehow, so that the upper plan nodes know that their
- * children's outputs have changed.
- *
- * Note that the stack of free evalPlanQual nodes is quite useless at the
- * moment, since it only saves us from pallocing/releasing the
- * evalPlanQual nodes themselves. But it will be useful once we implement
- * ReScan instead of end/restart for re-using PlanQual nodes.
- */
- if (endNode)
+ return estate->es_epqTuple[rti - 1];
+}
+
+/*
+ * Fetch the current row values for any non-locked relations that need
+ * to be scanned by an EvalPlanQual operation. origslot must have been set
+ * to contain the current result row (top-level row) that we need to recheck.
+ */
+void
+EvalPlanQualFetchRowMarks(EPQState *epqstate)
+{
+ ListCell *l;
+
+ Assert(epqstate->origslot != NULL);
+
+ foreach(l, epqstate->arowMarks)
{
- /* stop execution */
- EvalPlanQualStop(epq);
- }
+ ExecAuxRowMark *aerm = (ExecAuxRowMark *) lfirst(l);
+ ExecRowMark *erm = aerm->rowmark;
+ Datum datum;
+ bool isNull;
+ HeapTupleData tuple;
- /*
- * Initialize new recheck query.
- *
- * Note: if we were re-using PlanQual plans via ExecReScan, we'd need to
- * instead copy down changeable state from the top plan (including
- * es_result_relation_info, es_junkFilter) and reset locally changeable
- * state in the epq (including es_param_exec_vals, es_evTupleNull).
- */
- EvalPlanQualStart(epq, estate, epq->next);
+ if (RowMarkRequiresRowShareLock(erm->markType))
+ elog(ERROR, "EvalPlanQual doesn't support locking rowmarks");
- /*
- * free old RTE' tuple, if any, and store target tuple where relation's
- * scan node will see it
- */
- epqstate = epq->estate;
- if (epqstate->es_evTuple[rti - 1] != NULL)
- heap_freetuple(epqstate->es_evTuple[rti - 1]);
- epqstate->es_evTuple[rti - 1] = copyTuple;
+ /* clear any leftover test tuple for this rel */
+ EvalPlanQualSetTuple(epqstate, erm->rti, NULL);
- return EvalPlanQualNext(estate);
-}
+ if (erm->relation)
+ {
+ Buffer buffer;
-static TupleTableSlot *
-EvalPlanQualNext(EState *estate)
-{
- evalPlanQual *epq = estate->es_evalPlanQual;
- MemoryContext oldcontext;
- TupleTableSlot *slot;
+ Assert(erm->markType == ROW_MARK_REFERENCE);
- Assert(epq->rti != 0);
+ /* if child rel, must check whether it produced this row */
+ if (erm->rti != erm->prti)
+ {
+ Oid tableoid;
-lpqnext:;
- oldcontext = MemoryContextSwitchTo(epq->estate->es_query_cxt);
- slot = ExecProcNode(epq->planstate);
- MemoryContextSwitchTo(oldcontext);
+ datum = ExecGetJunkAttribute(epqstate->origslot,
+ aerm->toidAttNo,
+ &isNull);
+ /* non-locked rels could be on the inside of outer joins */
+ if (isNull)
+ continue;
+ tableoid = DatumGetObjectId(datum);
- /*
- * No more tuples for this PQ. Continue previous one.
- */
- if (TupIsNull(slot))
- {
- evalPlanQual *oldepq;
+ if (tableoid != RelationGetRelid(erm->relation))
+ {
+ /* this child is inactive right now */
+ continue;
+ }
+ }
- /* stop execution */
- EvalPlanQualStop(epq);
- /* pop old PQ from the stack */
- oldepq = epq->next;
- if (oldepq == NULL)
+ /* fetch the tuple's ctid */
+ datum = ExecGetJunkAttribute(epqstate->origslot,
+ aerm->ctidAttNo,
+ &isNull);
+ /* non-locked rels could be on the inside of outer joins */
+ if (isNull)
+ continue;
+ tuple.t_self = *((ItemPointer) DatumGetPointer(datum));
+
+ /* okay, fetch the tuple */
+ if (!heap_fetch(erm->relation, SnapshotAny, &tuple, &buffer,
+ false, NULL))
+ elog(ERROR, "failed to fetch tuple for EvalPlanQual recheck");
+
+ /* successful, copy and store tuple */
+ EvalPlanQualSetTuple(epqstate, erm->rti,
+ heap_copytuple(&tuple));
+ ReleaseBuffer(buffer);
+ }
+ else
{
- /* this is the first (oldest) PQ - mark as free */
- epq->rti = 0;
- estate->es_useEvalPlan = false;
- /* and continue Query execution */
- return NULL;
+ HeapTupleHeader td;
+
+ Assert(erm->markType == ROW_MARK_COPY);
+
+ /* fetch the whole-row Var for the relation */
+ datum = ExecGetJunkAttribute(epqstate->origslot,
+ aerm->wholeAttNo,
+ &isNull);
+ /* non-locked rels could be on the inside of outer joins */
+ if (isNull)
+ continue;
+ td = DatumGetHeapTupleHeader(datum);
+
+ /* build a temporary HeapTuple control structure */
+ tuple.t_len = HeapTupleHeaderGetDatumLength(td);
+ ItemPointerSetInvalid(&(tuple.t_self));
+ tuple.t_tableOid = InvalidOid;
+ tuple.t_data = td;
+
+ /* copy and store tuple */
+ EvalPlanQualSetTuple(epqstate, erm->rti,
+ heap_copytuple(&tuple));
}
- Assert(oldepq->rti != 0);
- /* push current PQ to freePQ stack */
- oldepq->free = epq;
- epq = oldepq;
- estate->es_evalPlanQual = epq;
- goto lpqnext;
}
+}
+
+/*
+ * Fetch the next row (if any) from EvalPlanQual testing
+ *
+ * (In practice, there should never be more than one row...)
+ */
+TupleTableSlot *
+EvalPlanQualNext(EPQState *epqstate)
+{
+ MemoryContext oldcontext;
+ TupleTableSlot *slot;
+
+ oldcontext = MemoryContextSwitchTo(epqstate->estate->es_query_cxt);
+ slot = ExecProcNode(epqstate->planstate);
+ MemoryContextSwitchTo(oldcontext);
return slot;
}
-static void
-EndEvalPlanQual(EState *estate)
+/*
+ * Initialize or reset an EvalPlanQual state tree
+ */
+void
+EvalPlanQualBegin(EPQState *epqstate, EState *parentestate)
{
- evalPlanQual *epq = estate->es_evalPlanQual;
+ EState *estate = epqstate->estate;
- if (epq->rti == 0) /* plans already shutdowned */
+ if (estate == NULL)
{
- Assert(epq->next == NULL);
- return;
+ /* First time through, so create a child EState */
+ EvalPlanQualStart(epqstate, parentestate, epqstate->plan);
}
-
- for (;;)
+ else
{
- evalPlanQual *oldepq;
+ /*
+ * We already have a suitable child EPQ tree, so just reset it.
+ */
+ int rtsize = list_length(parentestate->es_range_table);
+ PlanState *planstate = epqstate->planstate;
+
+ MemSet(estate->es_epqScanDone, 0, rtsize * sizeof(bool));
- /* stop execution */
- EvalPlanQualStop(epq);
- /* pop old PQ from the stack */
- oldepq = epq->next;
- if (oldepq == NULL)
+ /* Recopy current values of parent parameters */
+ if (parentestate->es_plannedstmt->nParamExec > 0)
{
- /* this is the first (oldest) PQ - mark as free */
- epq->rti = 0;
- estate->es_useEvalPlan = false;
- break;
+ int i = parentestate->es_plannedstmt->nParamExec;
+
+ while (--i >= 0)
+ {
+ /* copy value if any, but not execPlan link */
+ estate->es_param_exec_vals[i].value =
+ parentestate->es_param_exec_vals[i].value;
+ estate->es_param_exec_vals[i].isnull =
+ parentestate->es_param_exec_vals[i].isnull;
+ }
}
- Assert(oldepq->rti != 0);
- /* push current PQ to freePQ stack */
- oldepq->free = epq;
- epq = oldepq;
- estate->es_evalPlanQual = epq;
+
+ /*
+ * Mark child plan tree as needing rescan at all scan nodes. The
+ * first ExecProcNode will take care of actually doing the rescan.
+ */
+ planstate->chgParam = bms_add_member(planstate->chgParam,
+ epqstate->epqParam);
}
}
/*
- * Start execution of one level of PlanQual.
+ * Start execution of an EvalPlanQual plan tree.
*
* This is a cut-down version of ExecutorStart(): we copy some state from
* the top-level estate rather than initializing it fresh.
*/
static void
-EvalPlanQualStart(evalPlanQual *epq, EState *estate, evalPlanQual *priorepq)
+EvalPlanQualStart(EPQState *epqstate, EState *parentestate, Plan *planTree)
{
- EState *epqstate;
+ EState *estate;
int rtsize;
MemoryContext oldcontext;
ListCell *l;
- rtsize = list_length(estate->es_range_table);
+ rtsize = list_length(parentestate->es_range_table);
- epq->estate = epqstate = CreateExecutorState();
+ epqstate->estate = estate = CreateExecutorState();
- oldcontext = MemoryContextSwitchTo(epqstate->es_query_cxt);
+ oldcontext = MemoryContextSwitchTo(estate->es_query_cxt);
/*
- * The epqstates share the top query's copy of unchanging state such as
+ * Child EPQ EStates share the parent's copy of unchanging state such as
* the snapshot, rangetable, result-rel info, and external Param info.
* They need their own copies of local state, including a tuple table,
* es_param_exec_vals, etc.
*/
- epqstate->es_direction = ForwardScanDirection;
- epqstate->es_snapshot = estate->es_snapshot;
- epqstate->es_crosscheck_snapshot = estate->es_crosscheck_snapshot;
- epqstate->es_range_table = estate->es_range_table;
- epqstate->es_result_relations = estate->es_result_relations;
- epqstate->es_num_result_relations = estate->es_num_result_relations;
- epqstate->es_result_relation_info = estate->es_result_relation_info;
- epqstate->es_junkFilter = estate->es_junkFilter;
- epqstate->es_into_relation_descriptor = estate->es_into_relation_descriptor;
- epqstate->es_into_relation_use_wal = estate->es_into_relation_use_wal;
- epqstate->es_param_list_info = estate->es_param_list_info;
- if (estate->es_plannedstmt->nParamExec > 0)
- epqstate->es_param_exec_vals = (ParamExecData *)
- palloc0(estate->es_plannedstmt->nParamExec * sizeof(ParamExecData));
- epqstate->es_rowMarks = estate->es_rowMarks;
- epqstate->es_instrument = estate->es_instrument;
- epqstate->es_select_into = estate->es_select_into;
- epqstate->es_into_oids = estate->es_into_oids;
- epqstate->es_plannedstmt = estate->es_plannedstmt;
-
- /*
- * Each epqstate must have its own es_evTupleNull state, but all the stack
- * entries share es_evTuple state. This allows sub-rechecks to inherit
- * the value being examined by an outer recheck.
- */
- epqstate->es_evTupleNull = (bool *) palloc0(rtsize * sizeof(bool));
- if (priorepq == NULL)
- /* first PQ stack entry */
- epqstate->es_evTuple = (HeapTuple *)
- palloc0(rtsize * sizeof(HeapTuple));
+ estate->es_direction = ForwardScanDirection;
+ estate->es_snapshot = parentestate->es_snapshot;
+ estate->es_crosscheck_snapshot = parentestate->es_crosscheck_snapshot;
+ estate->es_range_table = parentestate->es_range_table;
+ estate->es_plannedstmt = parentestate->es_plannedstmt;
+ estate->es_junkFilter = parentestate->es_junkFilter;
+ estate->es_output_cid = parentestate->es_output_cid;
+ estate->es_result_relations = parentestate->es_result_relations;
+ estate->es_num_result_relations = parentestate->es_num_result_relations;
+ estate->es_result_relation_info = parentestate->es_result_relation_info;
+ /* es_trig_target_relations must NOT be copied */
+ estate->es_rowMarks = parentestate->es_rowMarks;
+ estate->es_top_eflags = parentestate->es_top_eflags;
+ estate->es_instrument = parentestate->es_instrument;
+ estate->es_select_into = parentestate->es_select_into;
+ estate->es_into_oids = parentestate->es_into_oids;
+ /* es_auxmodifytables must NOT be copied */
+
+ /*
+ * The external param list is simply shared from parent. The internal
+ * param workspace has to be local state, but we copy the initial values
+ * from the parent, so as to have access to any param values that were
+ * already set from other parts of the parent's plan tree.
+ */
+ estate->es_param_list_info = parentestate->es_param_list_info;
+ if (parentestate->es_plannedstmt->nParamExec > 0)
+ {
+ int i = parentestate->es_plannedstmt->nParamExec;
+
+ estate->es_param_exec_vals = (ParamExecData *)
+ palloc0(i * sizeof(ParamExecData));
+ while (--i >= 0)
+ {
+ /* copy value if any, but not execPlan link */
+ estate->es_param_exec_vals[i].value =
+ parentestate->es_param_exec_vals[i].value;
+ estate->es_param_exec_vals[i].isnull =
+ parentestate->es_param_exec_vals[i].isnull;
+ }
+ }
+
+ /*
+ * Each EState must have its own es_epqScanDone state, but if we have
+ * nested EPQ checks they should share es_epqTuple arrays. This allows
+ * sub-rechecks to inherit the values being examined by an outer recheck.
+ */
+ estate->es_epqScanDone = (bool *) palloc0(rtsize * sizeof(bool));
+ if (parentestate->es_epqTuple != NULL)
+ {
+ estate->es_epqTuple = parentestate->es_epqTuple;
+ estate->es_epqTupleSet = parentestate->es_epqTupleSet;
+ }
else
- /* later stack entries share the same storage */
- epqstate->es_evTuple = priorepq->estate->es_evTuple;
+ {
+ estate->es_epqTuple = (HeapTuple *)
+ palloc0(rtsize * sizeof(HeapTuple));
+ estate->es_epqTupleSet = (bool *)
+ palloc0(rtsize * sizeof(bool));
+ }
/*
- * Create sub-tuple-table; we needn't redo the CountSlots work though.
+ * Each estate also has its own tuple table.
*/
- epqstate->es_tupleTable =
- ExecCreateTupleTable(estate->es_tupleTable->size);
+ estate->es_tupleTable = NIL;
/*
- * Initialize private state information for each SubPlan. We must do
- * this before running ExecInitNode on the main query tree, since
- * ExecInitSubPlan expects to be able to find these entries.
+ * Initialize private state information for each SubPlan. We must do this
+ * before running ExecInitNode on the main query tree, since
+ * ExecInitSubPlan expects to be able to find these entries. Some of the
+ * SubPlans might not be used in the part of the plan tree we intend to
+ * run, but since it's not easy to tell which, we just initialize them
+ * all. (However, if the subplan is headed by a ModifyTable node, then it
+ * must be a data-modifying CTE, which we will certainly not need to
+ * re-run, so we can skip initializing it. This is just an efficiency
+ * hack; it won't skip data-modifying CTEs for which the ModifyTable node
+ * is not at the top.)
*/
- Assert(epqstate->es_subplanstates == NIL);
- foreach(l, estate->es_plannedstmt->subplans)
+ Assert(estate->es_subplanstates == NIL);
+ foreach(l, parentestate->es_plannedstmt->subplans)
{
- Plan *subplan = (Plan *) lfirst(l);
- PlanState *subplanstate;
+ Plan *subplan = (Plan *) lfirst(l);
+ PlanState *subplanstate;
- subplanstate = ExecInitNode(subplan, epqstate, 0);
+ /* Don't initialize ModifyTable subplans, per comment above */
+ if (IsA(subplan, ModifyTable))
+ subplanstate = NULL;
+ else
+ subplanstate = ExecInitNode(subplan, estate, 0);
- epqstate->es_subplanstates = lappend(epqstate->es_subplanstates,
- subplanstate);
+ estate->es_subplanstates = lappend(estate->es_subplanstates,
+ subplanstate);
}
/*
- * Initialize the private state information for all the nodes in the query
- * tree. This opens files, allocates storage and leaves us ready to start
- * processing tuples.
+ * Initialize the private state information for all the nodes in the part
+ * of the plan tree we need to run. This opens files, allocates storage
+ * and leaves us ready to start processing tuples.
*/
- epq->planstate = ExecInitNode(estate->es_plannedstmt->planTree, epqstate, 0);
+ epqstate->planstate = ExecInitNode(planTree, estate, 0);
MemoryContextSwitchTo(oldcontext);
}
/*
- * End execution of one level of PlanQual.
+ * EvalPlanQualEnd -- shut down at termination of parent plan state node,
+ * or if we are done with the current EPQ child.
*
* This is a cut-down version of ExecutorEnd(); basically we want to do most
* of the normal cleanup, but *not* close result relations (which we are
- * just sharing from the outer query).
+ * just sharing from the outer query). We do, however, have to close any
+ * trigger target relations that got opened, since those are not shared.
+ * (There probably shouldn't be any of the latter, but just in case...)
*/
-static void
-EvalPlanQualStop(evalPlanQual *epq)
+void
+EvalPlanQualEnd(EPQState *epqstate)
{
- EState *epqstate = epq->estate;
+ EState *estate = epqstate->estate;
MemoryContext oldcontext;
ListCell *l;
- oldcontext = MemoryContextSwitchTo(epqstate->es_query_cxt);
+ if (estate == NULL)
+ return; /* idle, so nothing to do */
- ExecEndNode(epq->planstate);
+ oldcontext = MemoryContextSwitchTo(estate->es_query_cxt);
+
+ ExecEndNode(epqstate->planstate);
- foreach(l, epqstate->es_subplanstates)
+ foreach(l, estate->es_subplanstates)
{
- PlanState *subplanstate = (PlanState *) lfirst(l);
+ PlanState *subplanstate = (PlanState *) lfirst(l);
ExecEndNode(subplanstate);
}
- ExecDropTupleTable(epqstate->es_tupleTable, true);
- epqstate->es_tupleTable = NULL;
+ /* throw away the per-estate tuple table */
+ ExecResetTupleTable(estate->es_tupleTable, false);
- if (epqstate->es_evTuple[epq->rti - 1] != NULL)
+ /* close any trigger target relations attached to this EState */
+ foreach(l, estate->es_trig_target_relations)
{
- heap_freetuple(epqstate->es_evTuple[epq->rti - 1]);
- epqstate->es_evTuple[epq->rti - 1] = NULL;
+ ResultRelInfo *resultRelInfo = (ResultRelInfo *) lfirst(l);
+
+ /* Close indices and then the relation itself */
+ ExecCloseIndices(resultRelInfo);
+ heap_close(resultRelInfo->ri_RelationDesc, NoLock);
}
MemoryContextSwitchTo(oldcontext);
- FreeExecutorState(epqstate);
+ FreeExecutorState(estate);
- epq->estate = NULL;
- epq->planstate = NULL;
+ /* Mark EPQState idle */
+ epqstate->estate = NULL;
+ epqstate->planstate = NULL;
+ epqstate->origslot = NULL;
}
*
* We implement SELECT INTO by diverting SELECT's normal output with
* a specialized DestReceiver type.
- *
- * TODO: remove some of the INTO-specific cruft from EState, and keep
- * it in the DestReceiver instead.
*/
typedef struct
{
DestReceiver pub; /* publicly-known function pointers */
EState *estate; /* EState we are working with */
+ Relation rel; /* Relation to write to */
+ int hi_options; /* heap_insert performance options */
+ BulkInsertState bistate; /* bulk insert state */
} DR_intorel;
/*
static void
OpenIntoRel(QueryDesc *queryDesc)
{
- IntoClause *into = queryDesc->plannedstmt->into;
+ IntoClause *into = queryDesc->plannedstmt->intoClause;
EState *estate = queryDesc->estate;
Relation intoRelationDesc;
char *intoName;
Oid namespaceId;
Oid tablespaceId;
Datum reloptions;
- AclResult aclresult;
Oid intoRelationId;
- TupleDesc tupdesc;
DR_intorel *myState;
+ static char *validnsps[] = HEAP_RELOPT_NAMESPACES;
Assert(into);
/*
+ * XXX This code needs to be kept in sync with DefineRelation(). Maybe we
+ * should try to use that function instead.
+ */
+
+ /*
* Check consistency of arguments
*/
- if (into->onCommit != ONCOMMIT_NOOP && !into->rel->istemp)
+ if (into->onCommit != ONCOMMIT_NOOP
+ && into->rel->relpersistence != RELPERSISTENCE_TEMP)
ereport(ERROR,
(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
errmsg("ON COMMIT can only be used on temporary tables")));
* Find namespace to create in, check its permissions
*/
intoName = into->rel->relname;
- namespaceId = RangeVarGetCreationNamespace(into->rel);
+ namespaceId = RangeVarGetAndCheckCreationNamespace(into->rel);
+ RangeVarAdjustRelationPersistence(into->rel, namespaceId);
- aclresult = pg_namespace_aclcheck(namespaceId, GetUserId(),
- ACL_CREATE);
- if (aclresult != ACLCHECK_OK)
- aclcheck_error(aclresult, ACL_KIND_NAMESPACE,
- get_namespace_name(namespaceId));
+ /*
+ * Security check: disallow creating temp tables from security-restricted
+ * code. This is needed because calling code might not expect untrusted
+ * tables to appear in pg_temp at the front of its search path.
+ */
+ if (into->rel->relpersistence == RELPERSISTENCE_TEMP
+ && InSecurityRestrictedOperation())
+ ereport(ERROR,
+ (errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
+ errmsg("cannot create temporary table within security-restricted operation")));
/*
- * Select tablespace to use. If not specified, use default_tablespace
+ * Select tablespace to use. If not specified, use default tablespace
* (which may in turn default to database's default).
*/
if (into->tableSpaceName)
{
- tablespaceId = get_tablespace_oid(into->tableSpaceName);
- if (!OidIsValid(tablespaceId))
- ereport(ERROR,
- (errcode(ERRCODE_UNDEFINED_OBJECT),
- errmsg("tablespace \"%s\" does not exist",
- into->tableSpaceName)));
+ tablespaceId = get_tablespace_oid(into->tableSpaceName, false);
}
else
{
- tablespaceId = GetDefaultTablespace();
+ tablespaceId = GetDefaultTablespace(into->rel->relpersistence);
/* note InvalidOid is OK in this case */
}
/* Check permissions except when using the database's default space */
- if (OidIsValid(tablespaceId))
+ if (OidIsValid(tablespaceId) && tablespaceId != MyDatabaseTableSpace)
{
AclResult aclresult;
/* Parse and validate any reloptions */
reloptions = transformRelOptions((Datum) 0,
into->options,
+ NULL,
+ validnsps,
true,
false);
(void) heap_reloptions(RELKIND_RELATION, reloptions, true);
- /* have to copy the actual tupdesc to get rid of any constraints */
- tupdesc = CreateTupleDescCopy(queryDesc->tupDesc);
-
/* Now we can actually create the new relation */
intoRelationId = heap_create_with_catalog(intoName,
namespaceId,
tablespaceId,
InvalidOid,
+ InvalidOid,
+ InvalidOid,
GetUserId(),
- tupdesc,
+ queryDesc->tupDesc,
+ NIL,
RELKIND_RELATION,
+ into->rel->relpersistence,
+ false,
false,
true,
0,
into->onCommit,
reloptions,
+ true,
allowSystemTableMods);
-
- FreeTupleDesc(tupdesc);
+ Assert(intoRelationId != InvalidOid);
/*
* Advance command counter so that the newly-created relation's catalog
* AlterTableCreateToastTable ends with CommandCounterIncrement(), so that
* the TOAST table will be visible for insertion.
*/
- AlterTableCreateToastTable(intoRelationId);
+ reloptions = transformRelOptions((Datum) 0,
+ into->options,
+ "toast",
+ validnsps,
+ true,
+ false);
- /*
- * And open the constructed table for writing.
- */
- intoRelationDesc = heap_open(intoRelationId, AccessExclusiveLock);
+ (void) heap_reloptions(RELKIND_TOASTVALUE, reloptions, true);
- /* use_wal off requires rd_targblock be initially invalid */
- Assert(intoRelationDesc->rd_targblock == InvalidBlockNumber);
+ AlterTableCreateToastTable(intoRelationId, reloptions);
/*
- * We can skip WAL-logging the insertions, unless PITR is in use.
- *
- * Note that for a non-temp INTO table, this is safe only because we know
- * that the catalog changes above will have been WAL-logged, and so
- * RecordTransactionCommit will think it needs to WAL-log the eventual
- * transaction commit. Else the commit might be lost, even though all the
- * data is safely fsync'd ...
+ * And open the constructed table for writing.
*/
- estate->es_into_relation_use_wal = XLogArchivingActive();
- estate->es_into_relation_descriptor = intoRelationDesc;
+ intoRelationDesc = heap_open(intoRelationId, AccessExclusiveLock);
/*
* Now replace the query's DestReceiver with one for SELECT INTO
*/
- queryDesc->dest = CreateDestReceiver(DestIntoRel, NULL);
+ queryDesc->dest = CreateDestReceiver(DestIntoRel);
myState = (DR_intorel *) queryDesc->dest;
Assert(myState->pub.mydest == DestIntoRel);
myState->estate = estate;
+ myState->rel = intoRelationDesc;
+
+ /*
+ * We can skip WAL-logging the insertions, unless PITR or streaming
+ * replication is in use. We can skip the FSM in any case.
+ */
+ myState->hi_options = HEAP_INSERT_SKIP_FSM |
+ (XLogIsNeeded() ? 0 : HEAP_INSERT_SKIP_WAL);
+ myState->bistate = GetBulkInsertState();
+
+ /* Not using WAL requires smgr_targblock be initially invalid */
+ Assert(RelationGetTargetBlock(intoRelationDesc) == InvalidBlockNumber);
}
/*
static void
CloseIntoRel(QueryDesc *queryDesc)
{
- EState *estate = queryDesc->estate;
+ DR_intorel *myState = (DR_intorel *) queryDesc->dest;
/* OpenIntoRel might never have gotten called */
- if (estate->es_into_relation_descriptor)
+ if (myState && myState->pub.mydest == DestIntoRel && myState->rel)
{
- /*
- * If we skipped using WAL, and it's not a temp relation, we must
- * force the relation down to disk before it's safe to commit the
- * transaction. This requires forcing out any dirty buffers and then
- * doing a forced fsync.
- */
- if (!estate->es_into_relation_use_wal &&
- !estate->es_into_relation_descriptor->rd_istemp)
- heap_sync(estate->es_into_relation_descriptor);
+ FreeBulkInsertState(myState->bistate);
+
+ /* If we skipped using WAL, must heap_sync before commit */
+ if (myState->hi_options & HEAP_INSERT_SKIP_WAL)
+ heap_sync(myState->rel);
/* close rel, but keep lock until commit */
- heap_close(estate->es_into_relation_descriptor, NoLock);
+ heap_close(myState->rel, NoLock);
- estate->es_into_relation_descriptor = NULL;
+ myState->rel = NULL;
}
}
/*
* CreateIntoRelDestReceiver -- create a suitable DestReceiver object
- *
- * Since CreateDestReceiver doesn't accept the parameters we'd need,
- * we just leave the private fields empty here. OpenIntoRel will
- * fill them in.
*/
DestReceiver *
CreateIntoRelDestReceiver(void)
{
- DR_intorel *self = (DR_intorel *) palloc(sizeof(DR_intorel));
+ DR_intorel *self = (DR_intorel *) palloc0(sizeof(DR_intorel));
self->pub.receiveSlot = intorel_receive;
self->pub.rStartup = intorel_startup;
self->pub.rDestroy = intorel_destroy;
self->pub.mydest = DestIntoRel;
- self->estate = NULL;
+ /* private fields will be set by OpenIntoRel */
return (DestReceiver *) self;
}
intorel_receive(TupleTableSlot *slot, DestReceiver *self)
{
DR_intorel *myState = (DR_intorel *) self;
- EState *estate = myState->estate;
HeapTuple tuple;
- tuple = ExecCopySlotTuple(slot);
+ /*
+ * get the heap tuple out of the tuple table slot, making sure we have a
+ * writable copy
+ */
+ tuple = ExecMaterializeSlot(slot);
+
+ /*
+ * force assignment of new OID (see comments in ExecInsert)
+ */
+ if (myState->rel->rd_rel->relhasoids)
+ HeapTupleSetOid(tuple, InvalidOid);
- heap_insert(estate->es_into_relation_descriptor,
+ heap_insert(myState->rel,
tuple,
- estate->es_snapshot->curcid,
- estate->es_into_relation_use_wal,
- false); /* never any point in using FSM */
+ myState->estate->es_output_cid,
+ myState->hi_options,
+ myState->bistate);
/* We know this is a newly created relation, so there are no indexes */
-
- heap_freetuple(tuple);
-
- IncrAppended();
}
/*