execPartition.c File Reference

#include "postgres.h"
#include "access/table.h"
#include "access/tableam.h"
#include "catalog/index.h"
#include "catalog/partition.h"
#include "executor/execPartition.h"
#include "executor/executor.h"
#include "executor/nodeModifyTable.h"
#include "foreign/fdwapi.h"
#include "mb/pg_wchar.h"
#include "miscadmin.h"
#include "partitioning/partbounds.h"
#include "partitioning/partdesc.h"
#include "partitioning/partprune.h"
#include "rewrite/rewriteManip.h"
#include "utils/acl.h"
#include "utils/lsyscache.h"
#include "utils/partcache.h"
#include "utils/rls.h"
#include "utils/ruleutils.h"

Include dependency graph for execPartition.c:

Go to the source code of this file.

Data Structures
struct	PartitionTupleRouting
struct	PartitionDispatchData

Macros
#define	PARTITION_CACHED_FIND_THRESHOLD   16

Typedefs
typedef struct PartitionDispatchData	PartitionDispatchData

Functions
static ResultRelInfo *	ExecInitPartitionInfo (ModifyTableState *mtstate, EState *estate, PartitionTupleRouting *proute, PartitionDispatch dispatch, ResultRelInfo *rootResultRelInfo, int partidx)
static void	ExecInitRoutingInfo (ModifyTableState *mtstate, EState *estate, PartitionTupleRouting *proute, PartitionDispatch dispatch, ResultRelInfo *partRelInfo, int partidx, bool is_borrowed_rel)
static PartitionDispatch	ExecInitPartitionDispatchInfo (EState *estate, PartitionTupleRouting *proute, Oid partoid, PartitionDispatch parent_pd, int partidx, ResultRelInfo *rootResultRelInfo)
static void	FormPartitionKeyDatum (PartitionDispatch pd, TupleTableSlot *slot, EState *estate, Datum *values, bool *isnull)
static int	get_partition_for_tuple (PartitionDispatch pd, const Datum *values, const bool *isnull)
static char *	ExecBuildSlotPartitionKeyDescription (Relation rel, const Datum *values, const bool *isnull, int maxfieldlen)
static List *	adjust_partition_colnos (List *colnos, ResultRelInfo *leaf_part_rri)
static List *	adjust_partition_colnos_using_map (List *colnos, AttrMap *attrMap)
static PartitionPruneState *	CreatePartitionPruneState (EState *estate, PartitionPruneInfo *pruneinfo, Bitmapset **all_leafpart_rtis)
static void	InitPartitionPruneContext (PartitionPruneContext *context, List *pruning_steps, PartitionDesc partdesc, PartitionKey partkey, PlanState *planstate, ExprContext *econtext)
static void	InitExecPartitionPruneContexts (PartitionPruneState *prunestate, PlanState *parent_plan, Bitmapset *initially_valid_subplans, int n_total_subplans)
static void	find_matching_subplans_recurse (PartitionPruningData *prunedata, PartitionedRelPruningData *pprune, bool initial_prune, Bitmapset **validsubplans, Bitmapset **validsubplan_rtis)
PartitionTupleRouting *	ExecSetupPartitionTupleRouting (EState *estate, Relation rel)
ResultRelInfo *	ExecFindPartition (ModifyTableState *mtstate, ResultRelInfo *rootResultRelInfo, PartitionTupleRouting *proute, TupleTableSlot *slot, EState *estate)
static bool	IsIndexCompatibleAsArbiter (Relation arbiterIndexRelation, IndexInfo *arbiterIndexInfo, Relation indexRelation, IndexInfo *indexInfo)
void	ExecCleanupTupleRouting (ModifyTableState *mtstate, PartitionTupleRouting *proute)
void	ExecDoInitialPruning (EState *estate)
PartitionPruneState *	ExecInitPartitionExecPruning (PlanState *planstate, int n_total_subplans, int part_prune_index, Bitmapset *relids, Bitmapset **initially_valid_subplans)
Bitmapset *	ExecFindMatchingSubPlans (PartitionPruneState *prunestate, bool initial_prune, Bitmapset **validsubplan_rtis)

Macro Definition Documentation

PARTITION_CACHED_FIND_THRESHOLD

#define PARTITION_CACHED_FIND_THRESHOLD   16

Definition at line 1489 of file execPartition.c.

Typedef Documentation

PartitionDispatchData

typedef struct PartitionDispatchData PartitionDispatchData

Function Documentation

adjust_partition_colnos()

static List * adjust_partition_colnos ( List *  colnos, ResultRelInfo *  leaf_part_rri  )

static

Definition at line 1840 of file execPartition.c.

{
    TupleConversionMap *map = ExecGetChildToRootMap(leaf_part_rri);
 
    Assert(map != NULL);
 
    return adjust_partition_colnos_using_map(colnos, map->attrMap);
}

References adjust_partition_colnos_using_map(), Assert(), TupleConversionMap::attrMap, and ExecGetChildToRootMap().

Referenced by ExecInitPartitionInfo().

adjust_partition_colnos_using_map()

static List * adjust_partition_colnos_using_map ( List *  colnos, AttrMap *  attrMap  )

static

Definition at line 1857 of file execPartition.c.

{
    List       *new_colnos = NIL;
    ListCell   *lc;
 
    Assert(attrMap != NULL);    /* else we shouldn't be here */
 
    foreach(lc, colnos)
    {
        AttrNumber  parentattrno = lfirst_int(lc);
 
        if (parentattrno <= 0 ||
            parentattrno > attrMap->maplen ||
            attrMap->attnums[parentattrno - 1] == 0)
            elog(ERROR, "unexpected attno %d in target column list",
                 parentattrno);
        new_colnos = lappend_int(new_colnos,
                                 attrMap->attnums[parentattrno - 1]);
    }
 
    return new_colnos;
}

References Assert(), AttrMap::attnums, elog, ERROR, lappend_int(), lfirst_int, AttrMap::maplen, and NIL.

Referenced by adjust_partition_colnos(), and ExecInitPartitionInfo().

CreatePartitionPruneState()

static PartitionPruneState * CreatePartitionPruneState ( EState *  estate, PartitionPruneInfo *  pruneinfo, Bitmapset **  all_leafpart_rtis  )

static

Definition at line 2107 of file execPartition.c.

{
    PartitionPruneState *prunestate;
    int         n_part_hierarchies;
    ListCell   *lc;
    int         i;
 
    /*
     * Expression context that will be used by partkey_datum_from_expr() to
     * evaluate expressions for comparison against partition bounds.
     */
    ExprContext *econtext = CreateExprContext(estate);
 
    /* For data reading, executor always includes detached partitions */
    if (estate->es_partition_directory == NULL)
        estate->es_partition_directory =
            CreatePartitionDirectory(estate->es_query_cxt, false);
 
    n_part_hierarchies = list_length(pruneinfo->prune_infos);
    Assert(n_part_hierarchies > 0);
 
    /*
     * Allocate the data structure
     */
    prunestate = (PartitionPruneState *)
        palloc(offsetof(PartitionPruneState, partprunedata) +
               sizeof(PartitionPruningData *) * n_part_hierarchies);
 
    /* Save ExprContext for use during InitExecPartitionPruneContexts(). */
    prunestate->econtext = econtext;
    prunestate->execparamids = NULL;
    /* other_subplans can change at runtime, so we need our own copy */
    prunestate->other_subplans = bms_copy(pruneinfo->other_subplans);
    prunestate->do_initial_prune = false;   /* may be set below */
    prunestate->do_exec_prune = false;  /* may be set below */
    prunestate->num_partprunedata = n_part_hierarchies;
 
    /*
     * Create a short-term memory context which we'll use when making calls to
     * the partition pruning functions.  This avoids possible memory leaks,
     * since the pruning functions call comparison functions that aren't under
     * our control.
     */
    prunestate->prune_context =
        AllocSetContextCreate(CurrentMemoryContext,
                              "Partition Prune",
                              ALLOCSET_DEFAULT_SIZES);
 
    i = 0;
    foreach(lc, pruneinfo->prune_infos)
    {
        List       *partrelpruneinfos = lfirst_node(List, lc);
        int         npartrelpruneinfos = list_length(partrelpruneinfos);
        PartitionPruningData *prunedata;
        ListCell   *lc2;
        int         j;
 
        prunedata = (PartitionPruningData *)
            palloc(offsetof(PartitionPruningData, partrelprunedata) +
                   npartrelpruneinfos * sizeof(PartitionedRelPruningData));
        prunestate->partprunedata[i] = prunedata;
        prunedata->num_partrelprunedata = npartrelpruneinfos;
 
        j = 0;
        foreach(lc2, partrelpruneinfos)
        {
            PartitionedRelPruneInfo *pinfo = lfirst_node(PartitionedRelPruneInfo, lc2);
            PartitionedRelPruningData *pprune = &prunedata->partrelprunedata[j];
            Relation    partrel;
            PartitionDesc partdesc;
            PartitionKey partkey;
 
            /*
             * We can rely on the copies of the partitioned table's partition
             * key and partition descriptor appearing in its relcache entry,
             * because that entry will be held open and locked for the
             * duration of this executor run.
             */
            partrel = ExecGetRangeTableRelation(estate, pinfo->rtindex, false);
 
            /* Remember for InitExecPartitionPruneContexts(). */
            pprune->partrel = partrel;
 
            partkey = RelationGetPartitionKey(partrel);
            partdesc = PartitionDirectoryLookup(estate->es_partition_directory,
                                                partrel);
 
            /*
             * Initialize the subplan_map and subpart_map.
             *
             * The set of partitions that exist now might not be the same that
             * existed when the plan was made.  The normal case is that it is;
             * optimize for that case with a quick comparison, and just copy
             * the subplan_map and make subpart_map, leafpart_rti_map point to
             * the ones in PruneInfo.
             *
             * For the case where they aren't identical, we could have more
             * partitions on either side; or even exactly the same number of
             * them on both but the set of OIDs doesn't match fully.  Handle
             * this by creating new subplan_map and subpart_map arrays that
             * corresponds to the ones in the PruneInfo where the new
             * partition descriptor's OIDs match.  Any that don't match can be
             * set to -1, as if they were pruned.  By construction, both
             * arrays are in partition bounds order.
             */
            pprune->nparts = partdesc->nparts;
            pprune->subplan_map = palloc(sizeof(int) * partdesc->nparts);
 
            if (partdesc->nparts == pinfo->nparts &&
                memcmp(partdesc->oids, pinfo->relid_map,
                       sizeof(int) * partdesc->nparts) == 0)
            {
                pprune->subpart_map = pinfo->subpart_map;
                pprune->leafpart_rti_map = pinfo->leafpart_rti_map;
                memcpy(pprune->subplan_map, pinfo->subplan_map,
                       sizeof(int) * pinfo->nparts);
            }
            else
            {
                int         pd_idx = 0;
                int         pp_idx;
 
                /*
                 * When the partition arrays are not identical, there could be
                 * some new ones but it's also possible that one was removed;
                 * we cope with both situations by walking the arrays and
                 * discarding those that don't match.
                 *
                 * If the number of partitions on both sides match, it's still
                 * possible that one partition has been detached and another
                 * attached.  Cope with that by creating a map that skips any
                 * mismatches.
                 */
                pprune->subpart_map = palloc(sizeof(int) * partdesc->nparts);
                pprune->leafpart_rti_map = palloc(sizeof(int) * partdesc->nparts);
 
                for (pp_idx = 0; pp_idx < partdesc->nparts; pp_idx++)
                {
                    /* Skip any InvalidOid relid_map entries */
                    while (pd_idx < pinfo->nparts &&
                           !OidIsValid(pinfo->relid_map[pd_idx]))
                        pd_idx++;
 
            recheck:
                    if (pd_idx < pinfo->nparts &&
                        pinfo->relid_map[pd_idx] == partdesc->oids[pp_idx])
                    {
                        /* match... */
                        pprune->subplan_map[pp_idx] =
                            pinfo->subplan_map[pd_idx];
                        pprune->subpart_map[pp_idx] =
                            pinfo->subpart_map[pd_idx];
                        pprune->leafpart_rti_map[pp_idx] =
                            pinfo->leafpart_rti_map[pd_idx];
                        pd_idx++;
                        continue;
                    }
 
                    /*
                     * There isn't an exact match in the corresponding
                     * positions of both arrays.  Peek ahead in
                     * pinfo->relid_map to see if we have a match for the
                     * current partition in partdesc.  Normally if a match
                     * exists it's just one element ahead, and it means the
                     * planner saw one extra partition that we no longer see
                     * now (its concurrent detach finished just in between);
                     * so we skip that one by updating pd_idx to the new
                     * location and jumping above.  We can then continue to
                     * match the rest of the elements after skipping the OID
                     * with no match; no future matches are tried for the
                     * element that was skipped, because we know the arrays to
                     * be in the same order.
                     *
                     * If we don't see a match anywhere in the rest of the
                     * pinfo->relid_map array, that means we see an element
                     * now that the planner didn't see, so mark that one as
                     * pruned and move on.
                     */
                    for (int pd_idx2 = pd_idx + 1; pd_idx2 < pinfo->nparts; pd_idx2++)
                    {
                        if (pd_idx2 >= pinfo->nparts)
                            break;
                        if (pinfo->relid_map[pd_idx2] == partdesc->oids[pp_idx])
                        {
                            pd_idx = pd_idx2;
                            goto recheck;
                        }
                    }
 
                    pprune->subpart_map[pp_idx] = -1;
                    pprune->subplan_map[pp_idx] = -1;
                    pprune->leafpart_rti_map[pp_idx] = 0;
                }
            }
 
            /* present_parts is also subject to later modification */
            pprune->present_parts = bms_copy(pinfo->present_parts);
 
            /*
             * Only initial_context is initialized here.  exec_context is
             * initialized during ExecInitPartitionExecPruning() when the
             * parent plan's PlanState is available.
             *
             * Note that we must skip execution-time (both "init" and "exec")
             * partition pruning in EXPLAIN (GENERIC_PLAN), since parameter
             * values may be missing.
             */
            pprune->initial_pruning_steps = pinfo->initial_pruning_steps;
            if (pinfo->initial_pruning_steps &&
                !(econtext->ecxt_estate->es_top_eflags & EXEC_FLAG_EXPLAIN_GENERIC))
            {
                InitPartitionPruneContext(&pprune->initial_context,
                                          pprune->initial_pruning_steps,
                                          partdesc, partkey, NULL,
                                          econtext);
                /* Record whether initial pruning is needed at any level */
                prunestate->do_initial_prune = true;
            }
            pprune->exec_pruning_steps = pinfo->exec_pruning_steps;
            if (pinfo->exec_pruning_steps &&
                !(econtext->ecxt_estate->es_top_eflags & EXEC_FLAG_EXPLAIN_GENERIC))
            {
                /* Record whether exec pruning is needed at any level */
                prunestate->do_exec_prune = true;
            }
 
            /*
             * Accumulate the IDs of all PARAM_EXEC Params affecting the
             * partitioning decisions at this plan node.
             */
            prunestate->execparamids = bms_add_members(prunestate->execparamids,
                                                       pinfo->execparamids);
 
            /*
             * Return all leaf partition indexes if we're skipping pruning in
             * the EXPLAIN (GENERIC_PLAN) case.
             */
            if (pinfo->initial_pruning_steps && !prunestate->do_initial_prune)
            {
                int         part_index = -1;
 
                while ((part_index = bms_next_member(pprune->present_parts,
                                                     part_index)) >= 0)
                {
                    Index       rtindex = pprune->leafpart_rti_map[part_index];
 
                    if (rtindex)
                        *all_leafpart_rtis = bms_add_member(*all_leafpart_rtis,
                                                            rtindex);
                }
            }
 
            j++;
        }
        i++;
    }
 
    return prunestate;
}

References ALLOCSET_DEFAULT_SIZES, AllocSetContextCreate, Assert(), bms_add_member(), bms_add_members(), bms_copy(), bms_next_member(), CreateExprContext(), CreatePartitionDirectory(), CurrentMemoryContext, PartitionPruneState::do_exec_prune, PartitionPruneState::do_initial_prune, PartitionPruneState::econtext, ExprContext::ecxt_estate, EState::es_partition_directory, EState::es_query_cxt, EState::es_top_eflags, EXEC_FLAG_EXPLAIN_GENERIC, PartitionedRelPruningData::exec_pruning_steps, PartitionedRelPruneInfo::exec_pruning_steps, ExecGetRangeTableRelation(), PartitionPruneState::execparamids, PartitionedRelPruneInfo::execparamids, i, PartitionedRelPruningData::initial_context, PartitionedRelPruningData::initial_pruning_steps, PartitionedRelPruneInfo::initial_pruning_steps, InitPartitionPruneContext(), j, PartitionedRelPruningData::leafpart_rti_map, lfirst_node, list_length(), PartitionedRelPruningData::nparts, PartitionedRelPruneInfo::nparts, PartitionDescData::nparts, PartitionPruneState::num_partprunedata, PartitionPruningData::num_partrelprunedata, OidIsValid, PartitionDescData::oids, PartitionPruneState::other_subplans, PartitionPruneInfo::other_subplans, palloc(), PartitionDirectoryLookup(), PartitionPruneState::partprunedata, PartitionedRelPruningData::partrel, PartitionPruningData::partrelprunedata, PartitionedRelPruningData::present_parts, PartitionedRelPruneInfo::present_parts, PartitionPruneState::prune_context, PartitionPruneInfo::prune_infos, RelationGetPartitionKey(), PartitionedRelPruneInfo::rtindex, PartitionedRelPruningData::subpart_map, and PartitionedRelPruningData::subplan_map.

Referenced by ExecDoInitialPruning().

ExecBuildSlotPartitionKeyDescription()

static char * ExecBuildSlotPartitionKeyDescription ( Relation  rel, const Datum *  values, const bool *  isnull, int  maxfieldlen  )

static

Definition at line 1752 of file execPartition.c.

{
    StringInfoData buf;
    PartitionKey key = RelationGetPartitionKey(rel);
    int         partnatts = get_partition_natts(key);
    int         i;
    Oid         relid = RelationGetRelid(rel);
    AclResult   aclresult;
 
    if (check_enable_rls(relid, InvalidOid, true) == RLS_ENABLED)
        return NULL;
 
    /* If the user has table-level access, just go build the description. */
    aclresult = pg_class_aclcheck(relid, GetUserId(), ACL_SELECT);
    if (aclresult != ACLCHECK_OK)
    {
        /*
         * Step through the columns of the partition key and make sure the
         * user has SELECT rights on all of them.
         */
        for (i = 0; i < partnatts; i++)
        {
            AttrNumber  attnum = get_partition_col_attnum(key, i);
 
            /*
             * If this partition key column is an expression, we return no
             * detail rather than try to figure out what column(s) the
             * expression includes and if the user has SELECT rights on them.
             */
            if (attnum == InvalidAttrNumber ||
                pg_attribute_aclcheck(relid, attnum, GetUserId(),
                                      ACL_SELECT) != ACLCHECK_OK)
                return NULL;
        }
    }
 
    initStringInfo(&buf);
    appendStringInfo(&buf, "(%s) = (",
                     pg_get_partkeydef_columns(relid, true));
 
    for (i = 0; i < partnatts; i++)
    {
        char       *val;
        int         vallen;
 
        if (isnull[i])
            val = "null";
        else
        {
            Oid         foutoid;
            bool        typisvarlena;
 
            getTypeOutputInfo(get_partition_col_typid(key, i),
                              &foutoid, &typisvarlena);
            val = OidOutputFunctionCall(foutoid, values[i]);
        }
 
        if (i > 0)
            appendStringInfoString(&buf, ", ");
 
        /* truncate if needed */
        vallen = strlen(val);
        if (vallen <= maxfieldlen)
            appendBinaryStringInfo(&buf, val, vallen);
        else
        {
            vallen = pg_mbcliplen(val, vallen, maxfieldlen);
            appendBinaryStringInfo(&buf, val, vallen);
            appendStringInfoString(&buf, "...");
        }
    }
 
    appendStringInfoChar(&buf, ')');
 
    return buf.data;
}

References ACL_SELECT, ACLCHECK_OK, appendBinaryStringInfo(), appendStringInfo(), appendStringInfoChar(), appendStringInfoString(), attnum, buf, check_enable_rls(), get_partition_col_attnum(), get_partition_col_typid(), get_partition_natts(), getTypeOutputInfo(), GetUserId(), i, initStringInfo(), InvalidAttrNumber, InvalidOid, sort-test::key, OidOutputFunctionCall(), pg_attribute_aclcheck(), pg_class_aclcheck(), pg_get_partkeydef_columns(), pg_mbcliplen(), RelationGetPartitionKey(), RelationGetRelid, RLS_ENABLED, val, and values.

Referenced by ExecFindPartition().

ExecCleanupTupleRouting()

void ExecCleanupTupleRouting	(	ModifyTableState *	mtstate,
		PartitionTupleRouting *	proute
	)

Definition at line 1374 of file execPartition.c.

{
    int         i;
 
    /*
     * Remember, proute->partition_dispatch_info[0] corresponds to the root
     * partitioned table, which we must not try to close, because it is the
     * main target table of the query that will be closed by callers such as
     * ExecEndPlan() or DoCopy(). Also, tupslot is NULL for the root
     * partitioned table.
     */
    for (i = 1; i < proute->num_dispatch; i++)
    {
        PartitionDispatch pd = proute->partition_dispatch_info[i];
 
        table_close(pd->reldesc, NoLock);
 
        if (pd->tupslot)
            ExecDropSingleTupleTableSlot(pd->tupslot);
    }
 
    for (i = 0; i < proute->num_partitions; i++)
    {
        ResultRelInfo *resultRelInfo = proute->partitions[i];
 
        /* Allow any FDWs to shut down */
        if (resultRelInfo->ri_FdwRoutine != NULL &&
            resultRelInfo->ri_FdwRoutine->EndForeignInsert != NULL)
            resultRelInfo->ri_FdwRoutine->EndForeignInsert(mtstate->ps.state,
                                                           resultRelInfo);
 
        /*
         * Close it if it's not one of the result relations borrowed from the
         * owning ModifyTableState; those will be closed by ExecEndPlan().
         */
        if (proute->is_borrowed_rel[i])
            continue;
 
        ExecCloseIndices(resultRelInfo);
        table_close(resultRelInfo->ri_RelationDesc, NoLock);
    }
}

References FdwRoutine::EndForeignInsert, ExecCloseIndices(), ExecDropSingleTupleTableSlot(), i, PartitionTupleRouting::is_borrowed_rel, NoLock, PartitionTupleRouting::num_dispatch, PartitionTupleRouting::num_partitions, PartitionTupleRouting::partition_dispatch_info, PartitionTupleRouting::partitions, ModifyTableState::ps, PartitionDispatchData::reldesc, ResultRelInfo::ri_FdwRoutine, ResultRelInfo::ri_RelationDesc, PlanState::state, table_close(), and PartitionDispatchData::tupslot.

Referenced by CopyFrom(), ExecEndModifyTable(), and finish_edata().

ExecDoInitialPruning()

void ExecDoInitialPruning

(

EState *

estate

)

Definition at line 1957 of file execPartition.c.

{
    ListCell   *lc;
 
    foreach(lc, estate->es_part_prune_infos)
    {
        PartitionPruneInfo *pruneinfo = lfirst_node(PartitionPruneInfo, lc);
        PartitionPruneState *prunestate;
        Bitmapset  *validsubplans = NULL;
        Bitmapset  *all_leafpart_rtis = NULL;
        Bitmapset  *validsubplan_rtis = NULL;
 
        /* Create and save the PartitionPruneState. */
        prunestate = CreatePartitionPruneState(estate, pruneinfo,
                                               &all_leafpart_rtis);
        estate->es_part_prune_states = lappend(estate->es_part_prune_states,
                                               prunestate);
 
        /*
         * Perform initial pruning steps, if any, and save the result
         * bitmapset or NULL as described in the header comment.
         */
        if (prunestate->do_initial_prune)
            validsubplans = ExecFindMatchingSubPlans(prunestate, true,
                                                     &validsubplan_rtis);
        else
            validsubplan_rtis = all_leafpart_rtis;
 
        estate->es_unpruned_relids = bms_add_members(estate->es_unpruned_relids,
                                                     validsubplan_rtis);
        estate->es_part_prune_results = lappend(estate->es_part_prune_results,
                                                validsubplans);
    }
}

References bms_add_members(), CreatePartitionPruneState(), PartitionPruneState::do_initial_prune, EState::es_part_prune_infos, EState::es_part_prune_results, EState::es_part_prune_states, EState::es_unpruned_relids, ExecFindMatchingSubPlans(), lappend(), and lfirst_node.

Referenced by InitPlan().

ExecFindMatchingSubPlans()

Bitmapset * ExecFindMatchingSubPlans	(	PartitionPruneState *	prunestate,
		bool	initial_prune,
		Bitmapset **	validsubplan_rtis
	)

Definition at line 2631 of file execPartition.c.

{
    Bitmapset  *result = NULL;
    MemoryContext oldcontext;
    int         i;
 
    /*
     * Either we're here on the initial prune done during pruning
     * initialization, or we're at a point where PARAM_EXEC Params can be
     * evaluated *and* there are steps in which to do so.
     */
    Assert(initial_prune || prunestate->do_exec_prune);
    Assert(validsubplan_rtis != NULL || !initial_prune);
 
    /*
     * Switch to a temp context to avoid leaking memory in the executor's
     * query-lifespan memory context.
     */
    oldcontext = MemoryContextSwitchTo(prunestate->prune_context);
 
    /*
     * For each hierarchy, do the pruning tests, and add nondeletable
     * subplans' indexes to "result".
     */
    for (i = 0; i < prunestate->num_partprunedata; i++)
    {
        PartitionPruningData *prunedata = prunestate->partprunedata[i];
        PartitionedRelPruningData *pprune;
 
        /*
         * We pass the zeroth item, belonging to the root table of the
         * hierarchy, and find_matching_subplans_recurse() takes care of
         * recursing to other (lower-level) parents as needed.
         */
        pprune = &prunedata->partrelprunedata[0];
        find_matching_subplans_recurse(prunedata, pprune, initial_prune,
                                       &result, validsubplan_rtis);
 
        /*
         * Expression eval may have used space in ExprContext too. Avoid
         * accessing exec_context during initial pruning, as it is not valid
         * at that stage.
         */
        if (!initial_prune && pprune->exec_pruning_steps)
            ResetExprContext(pprune->exec_context.exprcontext);
    }
 
    /* Add in any subplans that partition pruning didn't account for */
    result = bms_add_members(result, prunestate->other_subplans);
 
    MemoryContextSwitchTo(oldcontext);
 
    /* Copy result out of the temp context before we reset it */
    result = bms_copy(result);
    if (validsubplan_rtis)
        *validsubplan_rtis = bms_copy(*validsubplan_rtis);
 
    MemoryContextReset(prunestate->prune_context);
 
    return result;
}

References Assert(), bms_add_members(), bms_copy(), PartitionPruneState::do_exec_prune, PartitionedRelPruningData::exec_context, PartitionedRelPruningData::exec_pruning_steps, PartitionPruneContext::exprcontext, find_matching_subplans_recurse(), i, MemoryContextReset(), MemoryContextSwitchTo(), PartitionPruneState::num_partprunedata, PartitionPruneState::other_subplans, PartitionPruneState::partprunedata, PartitionPruningData::partrelprunedata, PartitionPruneState::prune_context, and ResetExprContext.

Referenced by choose_next_subplan_for_leader(), choose_next_subplan_for_worker(), choose_next_subplan_locally(), ExecAppendAsyncBegin(), ExecDoInitialPruning(), and ExecMergeAppend().

ExecFindPartition()

ResultRelInfo * ExecFindPartition	(	ModifyTableState *	mtstate,
		ResultRelInfo *	rootResultRelInfo,
		PartitionTupleRouting *	proute,
		TupleTableSlot *	slot,
		EState *	estate
	)

Definition at line 266 of file execPartition.c.

{
    PartitionDispatch *pd = proute->partition_dispatch_info;
    Datum       values[PARTITION_MAX_KEYS];
    bool        isnull[PARTITION_MAX_KEYS];
    Relation    rel;
    PartitionDispatch dispatch;
    PartitionDesc partdesc;
    ExprContext *ecxt = GetPerTupleExprContext(estate);
    TupleTableSlot *ecxt_scantuple_saved = ecxt->ecxt_scantuple;
    TupleTableSlot *rootslot = slot;
    TupleTableSlot *myslot = NULL;
    MemoryContext oldcxt;
    ResultRelInfo *rri = NULL;
 
    /* use per-tuple context here to avoid leaking memory */
    oldcxt = MemoryContextSwitchTo(GetPerTupleMemoryContext(estate));
 
    /*
     * First check the root table's partition constraint, if any.  No point in
     * routing the tuple if it doesn't belong in the root table itself.
     */
    if (rootResultRelInfo->ri_RelationDesc->rd_rel->relispartition)
        ExecPartitionCheck(rootResultRelInfo, slot, estate, true);
 
    /* start with the root partitioned table */
    dispatch = pd[0];
    while (dispatch != NULL)
    {
        int         partidx = -1;
        bool        is_leaf;
 
        CHECK_FOR_INTERRUPTS();
 
        rel = dispatch->reldesc;
        partdesc = dispatch->partdesc;
 
        /*
         * Extract partition key from tuple. Expression evaluation machinery
         * that FormPartitionKeyDatum() invokes expects ecxt_scantuple to
         * point to the correct tuple slot.  The slot might have changed from
         * what was used for the parent table if the table of the current
         * partitioning level has different tuple descriptor from the parent.
         * So update ecxt_scantuple accordingly.
         */
        ecxt->ecxt_scantuple = slot;
        FormPartitionKeyDatum(dispatch, slot, estate, values, isnull);
 
        /*
         * If this partitioned table has no partitions or no partition for
         * these values, error out.
         */
        if (partdesc->nparts == 0 ||
            (partidx = get_partition_for_tuple(dispatch, values, isnull)) < 0)
        {
            char       *val_desc;
 
            val_desc = ExecBuildSlotPartitionKeyDescription(rel,
                                                            values, isnull, 64);
            Assert(OidIsValid(RelationGetRelid(rel)));
            ereport(ERROR,
                    (errcode(ERRCODE_CHECK_VIOLATION),
                     errmsg("no partition of relation \"%s\" found for row",
                            RelationGetRelationName(rel)),
                     val_desc ?
                     errdetail("Partition key of the failing row contains %s.",
                               val_desc) : 0,
                     errtable(rel)));
        }
 
        is_leaf = partdesc->is_leaf[partidx];
        if (is_leaf)
        {
            /*
             * We've reached the leaf -- hurray, we're done.  Look to see if
             * we've already got a ResultRelInfo for this partition.
             */
            if (likely(dispatch->indexes[partidx] >= 0))
            {
                /* ResultRelInfo already built */
                Assert(dispatch->indexes[partidx] < proute->num_partitions);
                rri = proute->partitions[dispatch->indexes[partidx]];
            }
            else
            {
                /*
                 * If the partition is known in the owning ModifyTableState
                 * node, we can re-use that ResultRelInfo instead of creating
                 * a new one with ExecInitPartitionInfo().
                 */
                rri = ExecLookupResultRelByOid(mtstate,
                                               partdesc->oids[partidx],
                                               true, false);
                if (rri)
                {
                    ModifyTable *node = (ModifyTable *) mtstate->ps.plan;
 
                    /* Verify this ResultRelInfo allows INSERTs */
                    CheckValidResultRel(rri, CMD_INSERT,
                                        node ? node->onConflictAction : ONCONFLICT_NONE,
                                        NIL);
 
                    /*
                     * Initialize information needed to insert this and
                     * subsequent tuples routed to this partition.
                     */
                    ExecInitRoutingInfo(mtstate, estate, proute, dispatch,
                                        rri, partidx, true);
                }
                else
                {
                    /* We need to create a new one. */
                    rri = ExecInitPartitionInfo(mtstate, estate, proute,
                                                dispatch,
                                                rootResultRelInfo, partidx);
                }
            }
            Assert(rri != NULL);
 
            /* Signal to terminate the loop */
            dispatch = NULL;
        }
        else
        {
            /*
             * Partition is a sub-partitioned table; get the PartitionDispatch
             */
            if (likely(dispatch->indexes[partidx] >= 0))
            {
                /* Already built. */
                Assert(dispatch->indexes[partidx] < proute->num_dispatch);
 
                rri = proute->nonleaf_partitions[dispatch->indexes[partidx]];
 
                /*
                 * Move down to the next partition level and search again
                 * until we find a leaf partition that matches this tuple
                 */
                dispatch = pd[dispatch->indexes[partidx]];
            }
            else
            {
                /* Not yet built. Do that now. */
                PartitionDispatch subdispatch;
 
                /*
                 * Create the new PartitionDispatch.  We pass the current one
                 * in as the parent PartitionDispatch
                 */
                subdispatch = ExecInitPartitionDispatchInfo(estate,
                                                            proute,
                                                            partdesc->oids[partidx],
                                                            dispatch, partidx,
                                                            mtstate->rootResultRelInfo);
                Assert(dispatch->indexes[partidx] >= 0 &&
                       dispatch->indexes[partidx] < proute->num_dispatch);
 
                rri = proute->nonleaf_partitions[dispatch->indexes[partidx]];
                dispatch = subdispatch;
            }
 
            /*
             * Convert the tuple to the new parent's layout, if different from
             * the previous parent.
             */
            if (dispatch->tupslot)
            {
                AttrMap    *map = dispatch->tupmap;
                TupleTableSlot *tempslot = myslot;
 
                myslot = dispatch->tupslot;
                slot = execute_attr_map_slot(map, slot, myslot);
 
                if (tempslot != NULL)
                    ExecClearTuple(tempslot);
            }
        }
 
        /*
         * If this partition is the default one, we must check its partition
         * constraint now, which may have changed concurrently due to
         * partitions being added to the parent.
         *
         * (We do this here, and do not rely on ExecInsert doing it, because
         * we don't want to miss doing it for non-leaf partitions.)
         */
        if (partidx == partdesc->boundinfo->default_index)
        {
            /*
             * The tuple must match the partition's layout for the constraint
             * expression to be evaluated successfully.  If the partition is
             * sub-partitioned, that would already be the case due to the code
             * above, but for a leaf partition the tuple still matches the
             * parent's layout.
             *
             * Note that we have a map to convert from root to current
             * partition, but not from immediate parent to current partition.
             * So if we have to convert, do it from the root slot; if not, use
             * the root slot as-is.
             */
            if (is_leaf)
            {
                TupleConversionMap *map = ExecGetRootToChildMap(rri, estate);
 
                if (map)
                    slot = execute_attr_map_slot(map->attrMap, rootslot,
                                                 rri->ri_PartitionTupleSlot);
                else
                    slot = rootslot;
            }
 
            ExecPartitionCheck(rri, slot, estate, true);
        }
    }
 
    /* Release the tuple in the lowest parent's dedicated slot. */
    if (myslot != NULL)
        ExecClearTuple(myslot);
    /* and restore ecxt's scantuple */
    ecxt->ecxt_scantuple = ecxt_scantuple_saved;
    MemoryContextSwitchTo(oldcxt);
 
    return rri;
}

References Assert(), TupleConversionMap::attrMap, PartitionDescData::boundinfo, CHECK_FOR_INTERRUPTS, CheckValidResultRel(), CMD_INSERT, PartitionBoundInfoData::default_index, ExprContext::ecxt_scantuple, ereport, errcode(), errdetail(), errmsg(), ERROR, errtable(), ExecBuildSlotPartitionKeyDescription(), ExecClearTuple(), ExecGetRootToChildMap(), ExecInitPartitionDispatchInfo(), ExecInitPartitionInfo(), ExecInitRoutingInfo(), ExecLookupResultRelByOid(), ExecPartitionCheck(), execute_attr_map_slot(), FormPartitionKeyDatum(), get_partition_for_tuple(), GetPerTupleExprContext, GetPerTupleMemoryContext, PartitionDispatchData::indexes, PartitionDescData::is_leaf, likely, MemoryContextSwitchTo(), NIL, PartitionTupleRouting::nonleaf_partitions, PartitionDescData::nparts, PartitionTupleRouting::num_dispatch, PartitionTupleRouting::num_partitions, OidIsValid, PartitionDescData::oids, ONCONFLICT_NONE, ModifyTable::onConflictAction, PartitionDispatchData::partdesc, PartitionTupleRouting::partition_dispatch_info, PARTITION_MAX_KEYS, PartitionTupleRouting::partitions, PlanState::plan, ModifyTableState::ps, RelationData::rd_rel, RelationGetRelationName, RelationGetRelid, PartitionDispatchData::reldesc, ResultRelInfo::ri_PartitionTupleSlot, ResultRelInfo::ri_RelationDesc, ModifyTableState::rootResultRelInfo, PartitionDispatchData::tupmap, PartitionDispatchData::tupslot, and values.

Referenced by apply_handle_tuple_routing(), CopyFrom(), and ExecPrepareTupleRouting().

ExecInitPartitionDispatchInfo()

static PartitionDispatch ExecInitPartitionDispatchInfo ( EState *  estate, PartitionTupleRouting *  proute, Oid  partoid, PartitionDispatch  parent_pd, int  partidx, ResultRelInfo *  rootResultRelInfo  )

static

Definition at line 1235 of file execPartition.c.

{
    Relation    rel;
    PartitionDesc partdesc;
    PartitionDispatch pd;
    int         dispatchidx;
    MemoryContext oldcxt;
 
    /*
     * For data modification, it is better that executor does not include
     * partitions being detached, except when running in snapshot-isolation
     * mode.  This means that a read-committed transaction immediately gets a
     * "no partition for tuple" error when a tuple is inserted into a
     * partition that's being detached concurrently, but a transaction in
     * repeatable-read mode can still use such a partition.
     */
    if (estate->es_partition_directory == NULL)
        estate->es_partition_directory =
            CreatePartitionDirectory(estate->es_query_cxt,
                                     !IsolationUsesXactSnapshot());
 
    oldcxt = MemoryContextSwitchTo(proute->memcxt);
 
    /*
     * Only sub-partitioned tables need to be locked here.  The root
     * partitioned table will already have been locked as it's referenced in
     * the query's rtable.
     */
    if (partoid != RelationGetRelid(proute->partition_root))
        rel = table_open(partoid, RowExclusiveLock);
    else
        rel = proute->partition_root;
    partdesc = PartitionDirectoryLookup(estate->es_partition_directory, rel);
 
    pd = (PartitionDispatch) palloc(offsetof(PartitionDispatchData, indexes) +
                                    partdesc->nparts * sizeof(int));
    pd->reldesc = rel;
    pd->key = RelationGetPartitionKey(rel);
    pd->keystate = NIL;
    pd->partdesc = partdesc;
    if (parent_pd != NULL)
    {
        TupleDesc   tupdesc = RelationGetDescr(rel);
 
        /*
         * For sub-partitioned tables where the column order differs from its
         * direct parent partitioned table, we must store a tuple table slot
         * initialized with its tuple descriptor and a tuple conversion map to
         * convert a tuple from its parent's rowtype to its own.  This is to
         * make sure that we are looking at the correct row using the correct
         * tuple descriptor when computing its partition key for tuple
         * routing.
         */
        pd->tupmap = build_attrmap_by_name_if_req(RelationGetDescr(parent_pd->reldesc),
                                                  tupdesc,
                                                  false);
        pd->tupslot = pd->tupmap ?
            MakeSingleTupleTableSlot(tupdesc, &TTSOpsVirtual) : NULL;
    }
    else
    {
        /* Not required for the root partitioned table */
        pd->tupmap = NULL;
        pd->tupslot = NULL;
    }
 
    /*
     * Initialize with -1 to signify that the corresponding partition's
     * ResultRelInfo or PartitionDispatch has not been created yet.
     */
    memset(pd->indexes, -1, sizeof(int) * partdesc->nparts);
 
    /* Track in PartitionTupleRouting for later use */
    dispatchidx = proute->num_dispatch++;
 
    /* Allocate or enlarge the array, as needed */
    if (proute->num_dispatch >= proute->max_dispatch)
    {
        if (proute->max_dispatch == 0)
        {
            proute->max_dispatch = 4;
            proute->partition_dispatch_info = (PartitionDispatch *)
                palloc(sizeof(PartitionDispatch) * proute->max_dispatch);
            proute->nonleaf_partitions = (ResultRelInfo **)
                palloc(sizeof(ResultRelInfo *) * proute->max_dispatch);
        }
        else
        {
            proute->max_dispatch *= 2;
            proute->partition_dispatch_info = (PartitionDispatch *)
                repalloc(proute->partition_dispatch_info,
                         sizeof(PartitionDispatch) * proute->max_dispatch);
            proute->nonleaf_partitions = (ResultRelInfo **)
                repalloc(proute->nonleaf_partitions,
                         sizeof(ResultRelInfo *) * proute->max_dispatch);
        }
    }
    proute->partition_dispatch_info[dispatchidx] = pd;
 
    /*
     * If setting up a PartitionDispatch for a sub-partitioned table, we may
     * also need a minimally valid ResultRelInfo for checking the partition
     * constraint later; set that up now.
     */
    if (parent_pd)
    {
        ResultRelInfo *rri = makeNode(ResultRelInfo);
 
        InitResultRelInfo(rri, rel, 0, rootResultRelInfo, 0);
        proute->nonleaf_partitions[dispatchidx] = rri;
    }
    else
        proute->nonleaf_partitions[dispatchidx] = NULL;
 
    /*
     * Finally, if setting up a PartitionDispatch for a sub-partitioned table,
     * install a downlink in the parent to allow quick descent.
     */
    if (parent_pd)
    {
        Assert(parent_pd->indexes[partidx] == -1);
        parent_pd->indexes[partidx] = dispatchidx;
    }
 
    MemoryContextSwitchTo(oldcxt);
 
    return pd;
}

References Assert(), build_attrmap_by_name_if_req(), CreatePartitionDirectory(), EState::es_partition_directory, EState::es_query_cxt, PartitionDispatchData::indexes, InitResultRelInfo(), IsolationUsesXactSnapshot, PartitionDispatchData::key, PartitionDispatchData::keystate, makeNode, MakeSingleTupleTableSlot(), PartitionTupleRouting::max_dispatch, PartitionTupleRouting::memcxt, MemoryContextSwitchTo(), NIL, PartitionTupleRouting::nonleaf_partitions, PartitionDescData::nparts, PartitionTupleRouting::num_dispatch, palloc(), PartitionDispatchData::partdesc, PartitionTupleRouting::partition_dispatch_info, PartitionTupleRouting::partition_root, PartitionDirectoryLookup(), RelationGetDescr, RelationGetPartitionKey(), RelationGetRelid, PartitionDispatchData::reldesc, repalloc(), RowExclusiveLock, table_open(), TTSOpsVirtual, PartitionDispatchData::tupmap, and PartitionDispatchData::tupslot.

Referenced by ExecFindPartition(), and ExecSetupPartitionTupleRouting().

ExecInitPartitionExecPruning()

PartitionPruneState * ExecInitPartitionExecPruning	(	PlanState *	planstate,
		int	n_total_subplans,
		int	part_prune_index,
		Bitmapset *	relids,
		Bitmapset **	initially_valid_subplans
	)

Definition at line 2013 of file execPartition.c.

{
    PartitionPruneState *prunestate;
    EState     *estate = planstate->state;
    PartitionPruneInfo *pruneinfo;
 
    /* Obtain the pruneinfo we need. */
    pruneinfo = list_nth_node(PartitionPruneInfo, estate->es_part_prune_infos,
                              part_prune_index);
 
    /* Its relids better match the plan node's or the planner messed up. */
    if (!bms_equal(relids, pruneinfo->relids))
        elog(ERROR, "wrong pruneinfo with relids=%s found at part_prune_index=%d contained in plan node with relids=%s",
             bmsToString(pruneinfo->relids), part_prune_index,
             bmsToString(relids));
 
    /*
     * The PartitionPruneState would have been created by
     * ExecDoInitialPruning() and stored as the part_prune_index'th element of
     * EState.es_part_prune_states.
     */
    prunestate = list_nth(estate->es_part_prune_states, part_prune_index);
    Assert(prunestate != NULL);
 
    /* Use the result of initial pruning done by ExecDoInitialPruning(). */
    if (prunestate->do_initial_prune)
        *initially_valid_subplans = list_nth_node(Bitmapset,
                                                  estate->es_part_prune_results,
                                                  part_prune_index);
    else
    {
        /* No pruning, so we'll need to initialize all subplans */
        Assert(n_total_subplans > 0);
        *initially_valid_subplans = bms_add_range(NULL, 0,
                                                  n_total_subplans - 1);
    }
 
    /*
     * The exec pruning state must also be initialized, if needed, before it
     * can be used for pruning during execution.
     *
     * This also re-sequences subplan indexes contained in prunestate to
     * account for any that were removed due to initial pruning; refer to the
     * condition in InitExecPartitionPruneContexts() that is used to determine
     * whether to do this.  If no exec pruning needs to be done, we would thus
     * leave the maps to be in an invalid state, but that's ok since that data
     * won't be consulted again (cf initial Assert in
     * ExecFindMatchingSubPlans).
     */
    if (prunestate->do_exec_prune)
        InitExecPartitionPruneContexts(prunestate, planstate,
                                       *initially_valid_subplans,
                                       n_total_subplans);
 
    return prunestate;
}

References Assert(), bms_add_range(), bms_equal(), bmsToString(), PartitionPruneState::do_exec_prune, PartitionPruneState::do_initial_prune, elog, ERROR, EState::es_part_prune_infos, EState::es_part_prune_results, EState::es_part_prune_states, InitExecPartitionPruneContexts(), list_nth(), list_nth_node, PartitionPruneInfo::relids, and PlanState::state.

Referenced by ExecInitAppend(), and ExecInitMergeAppend().

ExecInitPartitionInfo()

static ResultRelInfo * ExecInitPartitionInfo ( ModifyTableState *  mtstate, EState *  estate, PartitionTupleRouting *  proute, PartitionDispatch  dispatch, ResultRelInfo *  rootResultRelInfo, int  partidx  )

static

Definition at line 562 of file execPartition.c.

{
    ModifyTable *node = (ModifyTable *) mtstate->ps.plan;
    Oid         partOid = dispatch->partdesc->oids[partidx];
    Relation    partrel;
    int         firstVarno = mtstate->resultRelInfo[0].ri_RangeTableIndex;
    Relation    firstResultRel = mtstate->resultRelInfo[0].ri_RelationDesc;
    ResultRelInfo *leaf_part_rri;
    MemoryContext oldcxt;
    AttrMap    *part_attmap = NULL;
    bool        found_whole_row;
 
    oldcxt = MemoryContextSwitchTo(proute->memcxt);
 
    partrel = table_open(partOid, RowExclusiveLock);
 
    leaf_part_rri = makeNode(ResultRelInfo);
    InitResultRelInfo(leaf_part_rri,
                      partrel,
                      0,
                      rootResultRelInfo,
                      estate->es_instrument);
 
    /*
     * Verify result relation is a valid target for an INSERT.  An UPDATE of a
     * partition-key becomes a DELETE+INSERT operation, so this check is still
     * required when the operation is CMD_UPDATE.
     */
    CheckValidResultRel(leaf_part_rri, CMD_INSERT,
                        node ? node->onConflictAction : ONCONFLICT_NONE, NIL);
 
    /*
     * Open partition indices.  The user may have asked to check for conflicts
     * within this leaf partition and do "nothing" instead of throwing an
     * error.  Be prepared in that case by initializing the index information
     * needed by ExecInsert() to perform speculative insertions.
     */
    if (partrel->rd_rel->relhasindex &&
        leaf_part_rri->ri_IndexRelationDescs == NULL)
        ExecOpenIndices(leaf_part_rri,
                        (node != NULL &&
                         node->onConflictAction != ONCONFLICT_NONE));
 
    /*
     * Build WITH CHECK OPTION constraints for the partition.  Note that we
     * didn't build the withCheckOptionList for partitions within the planner,
     * but simple translation of varattnos will suffice.  This only occurs for
     * the INSERT case or in the case of UPDATE/MERGE tuple routing where we
     * didn't find a result rel to reuse.
     */
    if (node && node->withCheckOptionLists != NIL)
    {
        List       *wcoList;
        List       *wcoExprs = NIL;
        ListCell   *ll;
 
        /*
         * In the case of INSERT on a partitioned table, there is only one
         * plan.  Likewise, there is only one WCO list, not one per partition.
         * For UPDATE/MERGE, there are as many WCO lists as there are plans.
         */
        Assert((node->operation == CMD_INSERT &&
                list_length(node->withCheckOptionLists) == 1 &&
                list_length(node->resultRelations) == 1) ||
               (node->operation == CMD_UPDATE &&
                list_length(node->withCheckOptionLists) ==
                list_length(node->resultRelations)) ||
               (node->operation == CMD_MERGE &&
                list_length(node->withCheckOptionLists) ==
                list_length(node->resultRelations)));
 
        /*
         * Use the WCO list of the first plan as a reference to calculate
         * attno's for the WCO list of this partition.  In the INSERT case,
         * that refers to the root partitioned table, whereas in the UPDATE
         * tuple routing case, that refers to the first partition in the
         * mtstate->resultRelInfo array.  In any case, both that relation and
         * this partition should have the same columns, so we should be able
         * to map attributes successfully.
         */
        wcoList = linitial(node->withCheckOptionLists);
 
        /*
         * Convert Vars in it to contain this partition's attribute numbers.
         */
        part_attmap =
            build_attrmap_by_name(RelationGetDescr(partrel),
                                  RelationGetDescr(firstResultRel),
                                  false);
        wcoList = (List *)
            map_variable_attnos((Node *) wcoList,
                                firstVarno, 0,
                                part_attmap,
                                RelationGetForm(partrel)->reltype,
                                &found_whole_row);
        /* We ignore the value of found_whole_row. */
 
        foreach(ll, wcoList)
        {
            WithCheckOption *wco = lfirst_node(WithCheckOption, ll);
            ExprState  *wcoExpr = ExecInitQual(castNode(List, wco->qual),
                                               &mtstate->ps);
 
            wcoExprs = lappend(wcoExprs, wcoExpr);
        }
 
        leaf_part_rri->ri_WithCheckOptions = wcoList;
        leaf_part_rri->ri_WithCheckOptionExprs = wcoExprs;
    }
 
    /*
     * Build the RETURNING projection for the partition.  Note that we didn't
     * build the returningList for partitions within the planner, but simple
     * translation of varattnos will suffice.  This only occurs for the INSERT
     * case or in the case of UPDATE/MERGE tuple routing where we didn't find
     * a result rel to reuse.
     */
    if (node && node->returningLists != NIL)
    {
        TupleTableSlot *slot;
        ExprContext *econtext;
        List       *returningList;
 
        /* See the comment above for WCO lists. */
        Assert((node->operation == CMD_INSERT &&
                list_length(node->returningLists) == 1 &&
                list_length(node->resultRelations) == 1) ||
               (node->operation == CMD_UPDATE &&
                list_length(node->returningLists) ==
                list_length(node->resultRelations)) ||
               (node->operation == CMD_MERGE &&
                list_length(node->returningLists) ==
                list_length(node->resultRelations)));
 
        /*
         * Use the RETURNING list of the first plan as a reference to
         * calculate attno's for the RETURNING list of this partition.  See
         * the comment above for WCO lists for more details on why this is
         * okay.
         */
        returningList = linitial(node->returningLists);
 
        /*
         * Convert Vars in it to contain this partition's attribute numbers.
         */
        if (part_attmap == NULL)
            part_attmap =
                build_attrmap_by_name(RelationGetDescr(partrel),
                                      RelationGetDescr(firstResultRel),
                                      false);
        returningList = (List *)
            map_variable_attnos((Node *) returningList,
                                firstVarno, 0,
                                part_attmap,
                                RelationGetForm(partrel)->reltype,
                                &found_whole_row);
        /* We ignore the value of found_whole_row. */
 
        leaf_part_rri->ri_returningList = returningList;
 
        /*
         * Initialize the projection itself.
         *
         * Use the slot and the expression context that would have been set up
         * in ExecInitModifyTable() for projection's output.
         */
        Assert(mtstate->ps.ps_ResultTupleSlot != NULL);
        slot = mtstate->ps.ps_ResultTupleSlot;
        Assert(mtstate->ps.ps_ExprContext != NULL);
        econtext = mtstate->ps.ps_ExprContext;
        leaf_part_rri->ri_projectReturning =
            ExecBuildProjectionInfo(returningList, econtext, slot,
                                    &mtstate->ps, RelationGetDescr(partrel));
    }
 
    /* Set up information needed for routing tuples to the partition. */
    ExecInitRoutingInfo(mtstate, estate, proute, dispatch,
                        leaf_part_rri, partidx, false);
 
    /*
     * If there is an ON CONFLICT clause, initialize state for it.
     */
    if (node && node->onConflictAction != ONCONFLICT_NONE)
    {
        TupleDesc   partrelDesc = RelationGetDescr(partrel);
        ExprContext *econtext = mtstate->ps.ps_ExprContext;
        List       *arbiterIndexes = NIL;
        int         additional_arbiters = 0;
 
        /*
         * If there is a list of arbiter indexes, map it to a list of indexes
         * in the partition.  We also add any "identical indexes" to any of
         * those, to cover the case where one of them is concurrently being
         * reindexed.
         */
        if (rootResultRelInfo->ri_onConflictArbiterIndexes != NIL)
        {
            List       *unparented_idxs = NIL,
                       *arbiters_listidxs = NIL;
 
            for (int listidx = 0; listidx < leaf_part_rri->ri_NumIndices; listidx++)
            {
                Oid         indexoid;
                List       *ancestors;
 
                /*
                 * If one of this index's ancestors is in the root's arbiter
                 * list, then use this index as arbiter for this partition.
                 * Otherwise, if this index has no parent, track it for later,
                 * in case REINDEX CONCURRENTLY is working on one of the
                 * arbiters.
                 *
                 * XXX get_partition_ancestors is slow: it scans pg_inherits
                 * each time.  Consider a syscache or some other way to cache?
                 */
                indexoid = RelationGetRelid(leaf_part_rri->ri_IndexRelationDescs[listidx]);
                ancestors = get_partition_ancestors(indexoid);
                if (ancestors != NIL)
                {
                    foreach_oid(parent_idx, rootResultRelInfo->ri_onConflictArbiterIndexes)
                    {
                        if (list_member_oid(ancestors, parent_idx))
                        {
                            arbiterIndexes = lappend_oid(arbiterIndexes, indexoid);
                            arbiters_listidxs = lappend_int(arbiters_listidxs, listidx);
                            break;
                        }
                    }
                }
                else
                    unparented_idxs = lappend_int(unparented_idxs, listidx);
                list_free(ancestors);
            }
 
            /*
             * If we found any indexes with no ancestors, it's possible that
             * some arbiter index is undergoing concurrent reindex.  Match all
             * unparented indexes against arbiters; add unparented matching
             * ones as "additional arbiters".
             *
             * This is critical so that all concurrent transactions use the
             * same set as arbiters during REINDEX CONCURRENTLY, to avoid
             * spurious "duplicate key" errors.
             */
            if (unparented_idxs && arbiterIndexes)
            {
                foreach_int(unparented_i, unparented_idxs)
                {
                    Relation    unparented_rel;
                    IndexInfo  *unparenred_ii;
 
                    unparented_rel = leaf_part_rri->ri_IndexRelationDescs[unparented_i];
                    unparenred_ii = leaf_part_rri->ri_IndexRelationInfo[unparented_i];
 
                    Assert(!list_member_oid(arbiterIndexes,
                                            unparented_rel->rd_index->indexrelid));
 
                    /* Ignore indexes not ready */
                    if (!unparenred_ii->ii_ReadyForInserts)
                        continue;
 
                    foreach_int(arbiter_i, arbiters_listidxs)
                    {
                        Relation    arbiter_rel;
                        IndexInfo  *arbiter_ii;
 
                        arbiter_rel = leaf_part_rri->ri_IndexRelationDescs[arbiter_i];
                        arbiter_ii = leaf_part_rri->ri_IndexRelationInfo[arbiter_i];
 
                        /*
                         * If the non-ancestor index is compatible with the
                         * arbiter, use the non-ancestor as arbiter too.
                         */
                        if (IsIndexCompatibleAsArbiter(arbiter_rel,
                                                       arbiter_ii,
                                                       unparented_rel,
                                                       unparenred_ii))
                        {
                            arbiterIndexes = lappend_oid(arbiterIndexes,
                                                         unparented_rel->rd_index->indexrelid);
                            additional_arbiters++;
                            break;
                        }
                    }
                }
            }
            list_free(unparented_idxs);
            list_free(arbiters_listidxs);
        }
 
        /*
         * We expect to find as many arbiter indexes on this partition as the
         * root has, plus however many "additional arbiters" (to wit: those
         * being concurrently rebuilt) we found.
         */
        if (list_length(rootResultRelInfo->ri_onConflictArbiterIndexes) !=
            list_length(arbiterIndexes) - additional_arbiters)
            elog(ERROR, "invalid arbiter index list");
        leaf_part_rri->ri_onConflictArbiterIndexes = arbiterIndexes;
 
        /*
         * In the DO UPDATE case, we have some more state to initialize.
         */
        if (node->onConflictAction == ONCONFLICT_UPDATE)
        {
            OnConflictSetState *onconfl = makeNode(OnConflictSetState);
            TupleConversionMap *map;
 
            map = ExecGetRootToChildMap(leaf_part_rri, estate);
 
            Assert(node->onConflictSet != NIL);
            Assert(rootResultRelInfo->ri_onConflict != NULL);
 
            leaf_part_rri->ri_onConflict = onconfl;
 
            /*
             * Need a separate existing slot for each partition, as the
             * partition could be of a different AM, even if the tuple
             * descriptors match.
             */
            onconfl->oc_Existing =
                table_slot_create(leaf_part_rri->ri_RelationDesc,
                                  &mtstate->ps.state->es_tupleTable);
 
            /*
             * If the partition's tuple descriptor matches exactly the root
             * parent (the common case), we can re-use most of the parent's ON
             * CONFLICT SET state, skipping a bunch of work.  Otherwise, we
             * need to create state specific to this partition.
             */
            if (map == NULL)
            {
                /*
                 * It's safe to reuse these from the partition root, as we
                 * only process one tuple at a time (therefore we won't
                 * overwrite needed data in slots), and the results of
                 * projections are independent of the underlying storage.
                 * Projections and where clauses themselves don't store state
                 * / are independent of the underlying storage.
                 */
                onconfl->oc_ProjSlot =
                    rootResultRelInfo->ri_onConflict->oc_ProjSlot;
                onconfl->oc_ProjInfo =
                    rootResultRelInfo->ri_onConflict->oc_ProjInfo;
                onconfl->oc_WhereClause =
                    rootResultRelInfo->ri_onConflict->oc_WhereClause;
            }
            else
            {
                List       *onconflset;
                List       *onconflcols;
 
                /*
                 * Translate expressions in onConflictSet to account for
                 * different attribute numbers.  For that, map partition
                 * varattnos twice: first to catch the EXCLUDED
                 * pseudo-relation (INNER_VAR), and second to handle the main
                 * target relation (firstVarno).
                 */
                onconflset = copyObject(node->onConflictSet);
                if (part_attmap == NULL)
                    part_attmap =
                        build_attrmap_by_name(RelationGetDescr(partrel),
                                              RelationGetDescr(firstResultRel),
                                              false);
                onconflset = (List *)
                    map_variable_attnos((Node *) onconflset,
                                        INNER_VAR, 0,
                                        part_attmap,
                                        RelationGetForm(partrel)->reltype,
                                        &found_whole_row);
                /* We ignore the value of found_whole_row. */
                onconflset = (List *)
                    map_variable_attnos((Node *) onconflset,
                                        firstVarno, 0,
                                        part_attmap,
                                        RelationGetForm(partrel)->reltype,
                                        &found_whole_row);
                /* We ignore the value of found_whole_row. */
 
                /* Finally, adjust the target colnos to match the partition. */
                onconflcols = adjust_partition_colnos(node->onConflictCols,
                                                      leaf_part_rri);
 
                /* create the tuple slot for the UPDATE SET projection */
                onconfl->oc_ProjSlot =
                    table_slot_create(partrel,
                                      &mtstate->ps.state->es_tupleTable);
 
                /* build UPDATE SET projection state */
                onconfl->oc_ProjInfo =
                    ExecBuildUpdateProjection(onconflset,
                                              true,
                                              onconflcols,
                                              partrelDesc,
                                              econtext,
                                              onconfl->oc_ProjSlot,
                                              &mtstate->ps);
 
                /*
                 * If there is a WHERE clause, initialize state where it will
                 * be evaluated, mapping the attribute numbers appropriately.
                 * As with onConflictSet, we need to map partition varattnos
                 * to the partition's tupdesc.
                 */
                if (node->onConflictWhere)
                {
                    List       *clause;
 
                    clause = copyObject((List *) node->onConflictWhere);
                    clause = (List *)
                        map_variable_attnos((Node *) clause,
                                            INNER_VAR, 0,
                                            part_attmap,
                                            RelationGetForm(partrel)->reltype,
                                            &found_whole_row);
                    /* We ignore the value of found_whole_row. */
                    clause = (List *)
                        map_variable_attnos((Node *) clause,
                                            firstVarno, 0,
                                            part_attmap,
                                            RelationGetForm(partrel)->reltype,
                                            &found_whole_row);
                    /* We ignore the value of found_whole_row. */
                    onconfl->oc_WhereClause =
                        ExecInitQual(clause, &mtstate->ps);
                }
            }
        }
    }
 
    /*
     * Since we've just initialized this ResultRelInfo, it's not in any list
     * attached to the estate as yet.  Add it, so that it can be found later.
     *
     * Note that the entries in this list appear in no predetermined order,
     * because partition result rels are initialized as and when they're
     * needed.
     */
    MemoryContextSwitchTo(estate->es_query_cxt);
    estate->es_tuple_routing_result_relations =
        lappend(estate->es_tuple_routing_result_relations,
                leaf_part_rri);
 
    /*
     * Initialize information about this partition that's needed to handle
     * MERGE.  We take the "first" result relation's mergeActionList as
     * reference and make copy for this relation, converting stuff that
     * references attribute numbers to match this relation's.
     *
     * This duplicates much of the logic in ExecInitMerge(), so if something
     * changes there, look here too.
     */
    if (node && node->operation == CMD_MERGE)
    {
        List       *firstMergeActionList = linitial(node->mergeActionLists);
        ListCell   *lc;
        ExprContext *econtext = mtstate->ps.ps_ExprContext;
        Node       *joinCondition;
 
        if (part_attmap == NULL)
            part_attmap =
                build_attrmap_by_name(RelationGetDescr(partrel),
                                      RelationGetDescr(firstResultRel),
                                      false);
 
        if (unlikely(!leaf_part_rri->ri_projectNewInfoValid))
            ExecInitMergeTupleSlots(mtstate, leaf_part_rri);
 
        /* Initialize state for join condition checking. */
        joinCondition =
            map_variable_attnos(linitial(node->mergeJoinConditions),
                                firstVarno, 0,
                                part_attmap,
                                RelationGetForm(partrel)->reltype,
                                &found_whole_row);
        /* We ignore the value of found_whole_row. */
        leaf_part_rri->ri_MergeJoinCondition =
            ExecInitQual((List *) joinCondition, &mtstate->ps);
 
        foreach(lc, firstMergeActionList)
        {
            /* Make a copy for this relation to be safe.  */
            MergeAction *action = copyObject(lfirst(lc));
            MergeActionState *action_state;
 
            /* Generate the action's state for this relation */
            action_state = makeNode(MergeActionState);
            action_state->mas_action = action;
 
            /* And put the action in the appropriate list */
            leaf_part_rri->ri_MergeActions[action->matchKind] =
                lappend(leaf_part_rri->ri_MergeActions[action->matchKind],
                        action_state);
 
            switch (action->commandType)
            {
                case CMD_INSERT:
 
                    /*
                     * ExecCheckPlanOutput() already done on the targetlist
                     * when "first" result relation initialized and it is same
                     * for all result relations.
                     */
                    action_state->mas_proj =
                        ExecBuildProjectionInfo(action->targetList, econtext,
                                                leaf_part_rri->ri_newTupleSlot,
                                                &mtstate->ps,
                                                RelationGetDescr(partrel));
                    break;
                case CMD_UPDATE:
 
                    /*
                     * Convert updateColnos from "first" result relation
                     * attribute numbers to this result rel's.
                     */
                    if (part_attmap)
                        action->updateColnos =
                            adjust_partition_colnos_using_map(action->updateColnos,
                                                              part_attmap);
                    action_state->mas_proj =
                        ExecBuildUpdateProjection(action->targetList,
                                                  true,
                                                  action->updateColnos,
                                                  RelationGetDescr(leaf_part_rri->ri_RelationDesc),
                                                  econtext,
                                                  leaf_part_rri->ri_newTupleSlot,
                                                  NULL);
                    break;
                case CMD_DELETE:
                case CMD_NOTHING:
                    /* Nothing to do */
                    break;
 
                default:
                    elog(ERROR, "unknown action in MERGE WHEN clause");
            }
 
            /* found_whole_row intentionally ignored. */
            action->qual =
                map_variable_attnos(action->qual,
                                    firstVarno, 0,
                                    part_attmap,
                                    RelationGetForm(partrel)->reltype,
                                    &found_whole_row);
            action_state->mas_whenqual =
                ExecInitQual((List *) action->qual, &mtstate->ps);
        }
    }
    MemoryContextSwitchTo(oldcxt);
 
    return leaf_part_rri;
}

References generate_unaccent_rules::action, adjust_partition_colnos(), adjust_partition_colnos_using_map(), Assert(), build_attrmap_by_name(), castNode, CheckValidResultRel(), CMD_DELETE, CMD_INSERT, CMD_MERGE, CMD_NOTHING, CMD_UPDATE, copyObject, elog, ERROR, EState::es_instrument, EState::es_query_cxt, EState::es_tuple_routing_result_relations, EState::es_tupleTable, ExecBuildProjectionInfo(), ExecBuildUpdateProjection(), ExecGetRootToChildMap(), ExecInitMergeTupleSlots(), ExecInitQual(), ExecInitRoutingInfo(), ExecOpenIndices(), foreach_int, foreach_oid, get_partition_ancestors(), IndexInfo::ii_ReadyForInserts, InitResultRelInfo(), INNER_VAR, IsIndexCompatibleAsArbiter(), lappend(), lappend_int(), lappend_oid(), lfirst, lfirst_node, linitial, list_free(), list_length(), list_member_oid(), makeNode, map_variable_attnos(), MergeActionState::mas_action, MergeActionState::mas_proj, MergeActionState::mas_whenqual, PartitionTupleRouting::memcxt, MemoryContextSwitchTo(), ModifyTable::mergeActionLists, ModifyTable::mergeJoinConditions, NIL, OnConflictSetState::oc_Existing, OnConflictSetState::oc_ProjInfo, OnConflictSetState::oc_ProjSlot, OnConflictSetState::oc_WhereClause, PartitionDescData::oids, ONCONFLICT_NONE, ONCONFLICT_UPDATE, ModifyTable::onConflictAction, ModifyTable::onConflictCols, ModifyTable::onConflictSet, ModifyTable::onConflictWhere, ModifyTable::operation, PartitionDispatchData::partdesc, PlanState::plan, ModifyTableState::ps, PlanState::ps_ExprContext, PlanState::ps_ResultTupleSlot, WithCheckOption::qual, RelationData::rd_index, RelationGetDescr, RelationGetForm, RelationGetRelid, ModifyTable::resultRelations, ModifyTableState::resultRelInfo, ModifyTable::returningLists, ResultRelInfo::ri_onConflict, ResultRelInfo::ri_onConflictArbiterIndexes, ResultRelInfo::ri_RangeTableIndex, ResultRelInfo::ri_RelationDesc, RowExclusiveLock, PlanState::state, table_open(), table_slot_create(), unlikely, and ModifyTable::withCheckOptionLists.

Referenced by ExecFindPartition().

ExecInitRoutingInfo()

static void ExecInitRoutingInfo ( ModifyTableState *  mtstate, EState *  estate, PartitionTupleRouting *  proute, PartitionDispatch  dispatch, ResultRelInfo *  partRelInfo, int  partidx, bool  is_borrowed_rel  )

static

Definition at line 1127 of file execPartition.c.

{
    MemoryContext oldcxt;
    int         rri_index;
 
    oldcxt = MemoryContextSwitchTo(proute->memcxt);
 
    /*
     * Set up tuple conversion between root parent and the partition if the
     * two have different rowtypes.  If conversion is indeed required, also
     * initialize a slot dedicated to storing this partition's converted
     * tuples.  Various operations that are applied to tuples after routing,
     * such as checking constraints, will refer to this slot.
     */
    if (ExecGetRootToChildMap(partRelInfo, estate) != NULL)
    {
        Relation    partrel = partRelInfo->ri_RelationDesc;
 
        /*
         * This pins the partition's TupleDesc, which will be released at the
         * end of the command.
         */
        partRelInfo->ri_PartitionTupleSlot =
            table_slot_create(partrel, &estate->es_tupleTable);
    }
    else
        partRelInfo->ri_PartitionTupleSlot = NULL;
 
    /*
     * If the partition is a foreign table, let the FDW init itself for
     * routing tuples to the partition.
     */
    if (partRelInfo->ri_FdwRoutine != NULL &&
        partRelInfo->ri_FdwRoutine->BeginForeignInsert != NULL)
        partRelInfo->ri_FdwRoutine->BeginForeignInsert(mtstate, partRelInfo);
 
    /*
     * Determine if the FDW supports batch insert and determine the batch size
     * (a FDW may support batching, but it may be disabled for the
     * server/table or for this particular query).
     *
     * If the FDW does not support batching, we set the batch size to 1.
     */
    if (partRelInfo->ri_FdwRoutine != NULL &&
        partRelInfo->ri_FdwRoutine->GetForeignModifyBatchSize &&
        partRelInfo->ri_FdwRoutine->ExecForeignBatchInsert)
        partRelInfo->ri_BatchSize =
            partRelInfo->ri_FdwRoutine->GetForeignModifyBatchSize(partRelInfo);
    else
        partRelInfo->ri_BatchSize = 1;
 
    Assert(partRelInfo->ri_BatchSize >= 1);
 
    partRelInfo->ri_CopyMultiInsertBuffer = NULL;
 
    /*
     * Keep track of it in the PartitionTupleRouting->partitions array.
     */
    Assert(dispatch->indexes[partidx] == -1);
 
    rri_index = proute->num_partitions++;
 
    /* Allocate or enlarge the array, as needed */
    if (proute->num_partitions >= proute->max_partitions)
    {
        if (proute->max_partitions == 0)
        {
            proute->max_partitions = 8;
            proute->partitions = (ResultRelInfo **)
                palloc(sizeof(ResultRelInfo *) * proute->max_partitions);
            proute->is_borrowed_rel = (bool *)
                palloc(sizeof(bool) * proute->max_partitions);
        }
        else
        {
            proute->max_partitions *= 2;
            proute->partitions = (ResultRelInfo **)
                repalloc(proute->partitions, sizeof(ResultRelInfo *) *
                         proute->max_partitions);
            proute->is_borrowed_rel = (bool *)
                repalloc(proute->is_borrowed_rel, sizeof(bool) *
                         proute->max_partitions);
        }
    }
 
    proute->partitions[rri_index] = partRelInfo;
    proute->is_borrowed_rel[rri_index] = is_borrowed_rel;
    dispatch->indexes[partidx] = rri_index;
 
    MemoryContextSwitchTo(oldcxt);
}

References Assert(), FdwRoutine::BeginForeignInsert, EState::es_tupleTable, FdwRoutine::ExecForeignBatchInsert, ExecGetRootToChildMap(), FdwRoutine::GetForeignModifyBatchSize, PartitionDispatchData::indexes, PartitionTupleRouting::is_borrowed_rel, PartitionTupleRouting::max_partitions, PartitionTupleRouting::memcxt, MemoryContextSwitchTo(), PartitionTupleRouting::num_partitions, palloc(), PartitionTupleRouting::partitions, repalloc(), ResultRelInfo::ri_BatchSize, ResultRelInfo::ri_CopyMultiInsertBuffer, ResultRelInfo::ri_FdwRoutine, ResultRelInfo::ri_PartitionTupleSlot, ResultRelInfo::ri_RelationDesc, and table_slot_create().

Referenced by ExecFindPartition(), and ExecInitPartitionInfo().

ExecSetupPartitionTupleRouting()

PartitionTupleRouting * ExecSetupPartitionTupleRouting	(	EState *	estate,
		Relation	rel
	)

Definition at line 219 of file execPartition.c.

{
    PartitionTupleRouting *proute;
 
    /*
     * Here we attempt to expend as little effort as possible in setting up
     * the PartitionTupleRouting.  Each partition's ResultRelInfo is built on
     * demand, only when we actually need to route a tuple to that partition.
     * The reason for this is that a common case is for INSERT to insert a
     * single tuple into a partitioned table and this must be fast.
     */
    proute = (PartitionTupleRouting *) palloc0(sizeof(PartitionTupleRouting));
    proute->partition_root = rel;
    proute->memcxt = CurrentMemoryContext;
    /* Rest of members initialized by zeroing */
 
    /*
     * Initialize this table's PartitionDispatch object.  Here we pass in the
     * parent as NULL as we don't need to care about any parent of the target
     * partitioned table.
     */
    ExecInitPartitionDispatchInfo(estate, proute, RelationGetRelid(rel),
                                  NULL, 0, NULL);
 
    return proute;
}

References CurrentMemoryContext, ExecInitPartitionDispatchInfo(), PartitionTupleRouting::memcxt, palloc0(), PartitionTupleRouting::partition_root, and RelationGetRelid.

Referenced by apply_handle_tuple_routing(), CopyFrom(), ExecCrossPartitionUpdate(), ExecInitMerge(), and ExecInitModifyTable().

find_matching_subplans_recurse()

static void find_matching_subplans_recurse ( PartitionPruningData *  prunedata, PartitionedRelPruningData *  pprune, bool  initial_prune, Bitmapset **  validsubplans, Bitmapset **  validsubplan_rtis  )

static

Definition at line 2704 of file execPartition.c.

{
    Bitmapset  *partset;
    int         i;
 
    /* Guard against stack overflow due to overly deep partition hierarchy. */
    check_stack_depth();
 
    /*
     * Prune as appropriate, if we have pruning steps matching the current
     * execution context.  Otherwise just include all partitions at this
     * level.
     */
    if (initial_prune && pprune->initial_pruning_steps)
        partset = get_matching_partitions(&pprune->initial_context,
                                          pprune->initial_pruning_steps);
    else if (!initial_prune && pprune->exec_pruning_steps)
        partset = get_matching_partitions(&pprune->exec_context,
                                          pprune->exec_pruning_steps);
    else
        partset = pprune->present_parts;
 
    /* Translate partset into subplan indexes */
    i = -1;
    while ((i = bms_next_member(partset, i)) >= 0)
    {
        if (pprune->subplan_map[i] >= 0)
        {
            *validsubplans = bms_add_member(*validsubplans,
                                            pprune->subplan_map[i]);
 
            /*
             * Only report leaf partitions. Non-leaf partitions may appear
             * here when they use an unflattened Append or MergeAppend.
             */
            if (validsubplan_rtis && pprune->leafpart_rti_map[i])
                *validsubplan_rtis = bms_add_member(*validsubplan_rtis,
                                                    pprune->leafpart_rti_map[i]);
        }
        else
        {
            int         partidx = pprune->subpart_map[i];
 
            if (partidx >= 0)
                find_matching_subplans_recurse(prunedata,
                                               &prunedata->partrelprunedata[partidx],
                                               initial_prune, validsubplans,
                                               validsubplan_rtis);
            else
            {
                /*
                 * We get here if the planner already pruned all the sub-
                 * partitions for this partition.  Silently ignore this
                 * partition in this case.  The end result is the same: we
                 * would have pruned all partitions just the same, but we
                 * don't have any pruning steps to execute to verify this.
                 */
            }
        }
    }
}

References bms_add_member(), bms_next_member(), check_stack_depth(), PartitionedRelPruningData::exec_context, PartitionedRelPruningData::exec_pruning_steps, find_matching_subplans_recurse(), get_matching_partitions(), i, PartitionedRelPruningData::initial_context, PartitionedRelPruningData::initial_pruning_steps, PartitionedRelPruningData::leafpart_rti_map, PartitionPruningData::partrelprunedata, PartitionedRelPruningData::present_parts, PartitionedRelPruningData::subpart_map, and PartitionedRelPruningData::subplan_map.

Referenced by ExecFindMatchingSubPlans(), and find_matching_subplans_recurse().

FormPartitionKeyDatum()

static void FormPartitionKeyDatum ( PartitionDispatch  pd, TupleTableSlot *  slot, EState *  estate, Datum *  values, bool *  isnull  )

static

Definition at line 1435 of file execPartition.c.

{
    ListCell   *partexpr_item;
    int         i;
 
    if (pd->key->partexprs != NIL && pd->keystate == NIL)
    {
        /* Check caller has set up context correctly */
        Assert(estate != NULL &&
               GetPerTupleExprContext(estate)->ecxt_scantuple == slot);
 
        /* First time through, set up expression evaluation state */
        pd->keystate = ExecPrepareExprList(pd->key->partexprs, estate);
    }
 
    partexpr_item = list_head(pd->keystate);
    for (i = 0; i < pd->key->partnatts; i++)
    {
        AttrNumber  keycol = pd->key->partattrs[i];
        Datum       datum;
        bool        isNull;
 
        if (keycol != 0)
        {
            /* Plain column; get the value directly from the heap tuple */
            datum = slot_getattr(slot, keycol, &isNull);
        }
        else
        {
            /* Expression; need to evaluate it */
            if (partexpr_item == NULL)
                elog(ERROR, "wrong number of partition key expressions");
            datum = ExecEvalExprSwitchContext((ExprState *) lfirst(partexpr_item),
                                              GetPerTupleExprContext(estate),
                                              &isNull);
            partexpr_item = lnext(pd->keystate, partexpr_item);
        }
        values[i] = datum;
        isnull[i] = isNull;
    }
 
    if (partexpr_item != NULL)
        elog(ERROR, "wrong number of partition key expressions");
}

References Assert(), elog, ERROR, ExecEvalExprSwitchContext(), ExecPrepareExprList(), GetPerTupleExprContext, i, PartitionDispatchData::key, PartitionDispatchData::keystate, lfirst, list_head(), lnext(), NIL, PartitionKeyData::partattrs, PartitionKeyData::partexprs, PartitionKeyData::partnatts, slot_getattr(), and values.

Referenced by ExecFindPartition().

get_partition_for_tuple()

static int get_partition_for_tuple ( PartitionDispatch  pd, const Datum *  values, const bool *  isnull  )

static

Definition at line 1532 of file execPartition.c.

{
    int         bound_offset = -1;
    int         part_index = -1;
    PartitionKey key = pd->key;
    PartitionDesc partdesc = pd->partdesc;
    PartitionBoundInfo boundinfo = partdesc->boundinfo;
 
    /*
     * In the switch statement below, when we perform a cached lookup for
     * RANGE and LIST partitioned tables, if we find that the last found
     * partition matches the 'values', we return the partition index right
     * away.  We do this instead of breaking out of the switch as we don't
     * want to execute the code about the DEFAULT partition or do any updates
     * for any of the cache-related fields.  That would be a waste of effort
     * as we already know it's not the DEFAULT partition and have no need to
     * increment the number of times we found the same partition any higher
     * than PARTITION_CACHED_FIND_THRESHOLD.
     */
 
    /* Route as appropriate based on partitioning strategy. */
    switch (key->strategy)
    {
        case PARTITION_STRATEGY_HASH:
            {
                uint64      rowHash;
 
                /* hash partitioning is too cheap to bother caching */
                rowHash = compute_partition_hash_value(key->partnatts,
                                                       key->partsupfunc,
                                                       key->partcollation,
                                                       values, isnull);
 
                /*
                 * HASH partitions can't have a DEFAULT partition and we don't
                 * do any caching work for them, so just return the part index
                 */
                return boundinfo->indexes[rowHash % boundinfo->nindexes];
            }
 
        case PARTITION_STRATEGY_LIST:
            if (isnull[0])
            {
                /* this is far too cheap to bother doing any caching */
                if (partition_bound_accepts_nulls(boundinfo))
                {
                    /*
                     * When there is a NULL partition we just return that
                     * directly.  We don't have a bound_offset so it's not
                     * valid to drop into the code after the switch which
                     * checks and updates the cache fields.  We perhaps should
                     * be invalidating the details of the last cached
                     * partition but there's no real need to.  Keeping those
                     * fields set gives a chance at matching to the cached
                     * partition on the next lookup.
                     */
                    return boundinfo->null_index;
                }
            }
            else
            {
                bool        equal;
 
                if (partdesc->last_found_count >= PARTITION_CACHED_FIND_THRESHOLD)
                {
                    int         last_datum_offset = partdesc->last_found_datum_index;
                    Datum       lastDatum = boundinfo->datums[last_datum_offset][0];
                    int32       cmpval;
 
                    /* does the last found datum index match this datum? */
                    cmpval = DatumGetInt32(FunctionCall2Coll(&key->partsupfunc[0],
                                                             key->partcollation[0],
                                                             lastDatum,
                                                             values[0]));
 
                    if (cmpval == 0)
                        return boundinfo->indexes[last_datum_offset];
 
                    /* fall-through and do a manual lookup */
                }
 
                bound_offset = partition_list_bsearch(key->partsupfunc,
                                                      key->partcollation,
                                                      boundinfo,
                                                      values[0], &equal);
                if (bound_offset >= 0 && equal)
                    part_index = boundinfo->indexes[bound_offset];
            }
            break;
 
        case PARTITION_STRATEGY_RANGE:
            {
                bool        equal = false,
                            range_partkey_has_null = false;
                int         i;
 
                /*
                 * No range includes NULL, so this will be accepted by the
                 * default partition if there is one, and otherwise rejected.
                 */
                for (i = 0; i < key->partnatts; i++)
                {
                    if (isnull[i])
                    {
                        range_partkey_has_null = true;
                        break;
                    }
                }
 
                /* NULLs belong in the DEFAULT partition */
                if (range_partkey_has_null)
                    break;
 
                if (partdesc->last_found_count >= PARTITION_CACHED_FIND_THRESHOLD)
                {
                    int         last_datum_offset = partdesc->last_found_datum_index;
                    Datum      *lastDatums = boundinfo->datums[last_datum_offset];
                    PartitionRangeDatumKind *kind = boundinfo->kind[last_datum_offset];
                    int32       cmpval;
 
                    /* check if the value is >= to the lower bound */
                    cmpval = partition_rbound_datum_cmp(key->partsupfunc,
                                                        key->partcollation,
                                                        lastDatums,
                                                        kind,
                                                        values,
                                                        key->partnatts);
 
                    /*
                     * If it's equal to the lower bound then no need to check
                     * the upper bound.
                     */
                    if (cmpval == 0)
                        return boundinfo->indexes[last_datum_offset + 1];
 
                    if (cmpval < 0 && last_datum_offset + 1 < boundinfo->ndatums)
                    {
                        /* check if the value is below the upper bound */
                        lastDatums = boundinfo->datums[last_datum_offset + 1];
                        kind = boundinfo->kind[last_datum_offset + 1];
                        cmpval = partition_rbound_datum_cmp(key->partsupfunc,
                                                            key->partcollation,
                                                            lastDatums,
                                                            kind,
                                                            values,
                                                            key->partnatts);
 
                        if (cmpval > 0)
                            return boundinfo->indexes[last_datum_offset + 1];
                    }
                    /* fall-through and do a manual lookup */
                }
 
                bound_offset = partition_range_datum_bsearch(key->partsupfunc,
                                                             key->partcollation,
                                                             boundinfo,
                                                             key->partnatts,
                                                             values,
                                                             &equal);
 
                /*
                 * The bound at bound_offset is less than or equal to the
                 * tuple value, so the bound at offset+1 is the upper bound of
                 * the partition we're looking for, if there actually exists
                 * one.
                 */
                part_index = boundinfo->indexes[bound_offset + 1];
            }
            break;
 
        default:
            elog(ERROR, "unexpected partition strategy: %d",
                 (int) key->strategy);
    }
 
    /*
     * part_index < 0 means we failed to find a partition of this parent. Use
     * the default partition, if there is one.
     */
    if (part_index < 0)
    {
        /*
         * No need to reset the cache fields here.  The next set of values
         * might end up belonging to the cached partition, so leaving the
         * cache alone improves the chances of a cache hit on the next lookup.
         */
        return boundinfo->default_index;
    }
 
    /* we should only make it here when the code above set bound_offset */
    Assert(bound_offset >= 0);
 
    /*
     * Attend to the cache fields.  If the bound_offset matches the last
     * cached bound offset then we've found the same partition as last time,
     * so bump the count by one.  If all goes well, we'll eventually reach
     * PARTITION_CACHED_FIND_THRESHOLD and try the cache path next time
     * around.  Otherwise, we'll reset the cache count back to 1 to mark that
     * we've found this partition for the first time.
     */
    if (bound_offset == partdesc->last_found_datum_index)
        partdesc->last_found_count++;
    else
    {
        partdesc->last_found_count = 1;
        partdesc->last_found_part_index = part_index;
        partdesc->last_found_datum_index = bound_offset;
    }
 
    return part_index;
}

References Assert(), PartitionDescData::boundinfo, compute_partition_hash_value(), DatumGetInt32(), PartitionBoundInfoData::datums, PartitionBoundInfoData::default_index, elog, equal(), ERROR, FunctionCall2Coll(), i, PartitionBoundInfoData::indexes, PartitionDispatchData::key, sort-test::key, PartitionBoundInfoData::kind, PartitionDescData::last_found_count, PartitionDescData::last_found_datum_index, PartitionDescData::last_found_part_index, PartitionBoundInfoData::ndatums, PartitionBoundInfoData::nindexes, PartitionBoundInfoData::null_index, PartitionDispatchData::partdesc, partition_bound_accepts_nulls, PARTITION_CACHED_FIND_THRESHOLD, partition_list_bsearch(), partition_range_datum_bsearch(), partition_rbound_datum_cmp(), PARTITION_STRATEGY_HASH, PARTITION_STRATEGY_LIST, PARTITION_STRATEGY_RANGE, and values.

Referenced by ExecFindPartition().

InitExecPartitionPruneContexts()

static void InitExecPartitionPruneContexts ( PartitionPruneState *  prunestate, PlanState *  parent_plan, Bitmapset *  initially_valid_subplans, int  n_total_subplans  )

static

Definition at line 2474 of file execPartition.c.

{
    EState     *estate;
    int        *new_subplan_indexes = NULL;
    Bitmapset  *new_other_subplans;
    int         i;
    int         newidx;
    bool        fix_subplan_map = false;
 
    Assert(prunestate->do_exec_prune);
    Assert(parent_plan != NULL);
    estate = parent_plan->state;
 
    /*
     * No need to fix subplans maps if initial pruning didn't eliminate any
     * subplans.
     */
    if (bms_num_members(initially_valid_subplans) < n_total_subplans)
    {
        fix_subplan_map = true;
 
        /*
         * First we must build a temporary array which maps old subplan
         * indexes to new ones.  For convenience of initialization, we use
         * 1-based indexes in this array and leave pruned items as 0.
         */
        new_subplan_indexes = (int *) palloc0(sizeof(int) * n_total_subplans);
        newidx = 1;
        i = -1;
        while ((i = bms_next_member(initially_valid_subplans, i)) >= 0)
        {
            Assert(i < n_total_subplans);
            new_subplan_indexes[i] = newidx++;
        }
    }
 
    /*
     * Now we can update each PartitionedRelPruneInfo's subplan_map with new
     * subplan indexes.  We must also recompute its present_parts bitmap.
     */
    for (i = 0; i < prunestate->num_partprunedata; i++)
    {
        PartitionPruningData *prunedata = prunestate->partprunedata[i];
        int         j;
 
        /*
         * Within each hierarchy, we perform this loop in back-to-front order
         * so that we determine present_parts for the lowest-level partitioned
         * tables first.  This way we can tell whether a sub-partitioned
         * table's partitions were entirely pruned so we can exclude it from
         * the current level's present_parts.
         */
        for (j = prunedata->num_partrelprunedata - 1; j >= 0; j--)
        {
            PartitionedRelPruningData *pprune = &prunedata->partrelprunedata[j];
            int         nparts = pprune->nparts;
            int         k;
 
            /* Initialize PartitionPruneContext for exec pruning, if needed. */
            if (pprune->exec_pruning_steps != NIL)
            {
                PartitionKey partkey;
                PartitionDesc partdesc;
 
                /*
                 * See the comment in CreatePartitionPruneState() regarding
                 * the usage of partdesc and partkey.
                 */
                partkey = RelationGetPartitionKey(pprune->partrel);
                partdesc = PartitionDirectoryLookup(estate->es_partition_directory,
                                                    pprune->partrel);
 
                InitPartitionPruneContext(&pprune->exec_context,
                                          pprune->exec_pruning_steps,
                                          partdesc, partkey, parent_plan,
                                          prunestate->econtext);
            }
 
            if (!fix_subplan_map)
                continue;
 
            /* We just rebuild present_parts from scratch */
            bms_free(pprune->present_parts);
            pprune->present_parts = NULL;
 
            for (k = 0; k < nparts; k++)
            {
                int         oldidx = pprune->subplan_map[k];
                int         subidx;
 
                /*
                 * If this partition existed as a subplan then change the old
                 * subplan index to the new subplan index.  The new index may
                 * become -1 if the partition was pruned above, or it may just
                 * come earlier in the subplan list due to some subplans being
                 * removed earlier in the list.  If it's a subpartition, add
                 * it to present_parts unless it's entirely pruned.
                 */
                if (oldidx >= 0)
                {
                    Assert(oldidx < n_total_subplans);
                    pprune->subplan_map[k] = new_subplan_indexes[oldidx] - 1;
 
                    if (new_subplan_indexes[oldidx] > 0)
                        pprune->present_parts =
                            bms_add_member(pprune->present_parts, k);
                }
                else if ((subidx = pprune->subpart_map[k]) >= 0)
                {
                    PartitionedRelPruningData *subprune;
 
                    subprune = &prunedata->partrelprunedata[subidx];
 
                    if (!bms_is_empty(subprune->present_parts))
                        pprune->present_parts =
                            bms_add_member(pprune->present_parts, k);
                }
            }
        }
    }
 
    /*
     * If we fixed subplan maps, we must also recompute the other_subplans
     * set, since indexes in it may change.
     */
    if (fix_subplan_map)
    {
        new_other_subplans = NULL;
        i = -1;
        while ((i = bms_next_member(prunestate->other_subplans, i)) >= 0)
            new_other_subplans = bms_add_member(new_other_subplans,
                                                new_subplan_indexes[i] - 1);
 
        bms_free(prunestate->other_subplans);
        prunestate->other_subplans = new_other_subplans;
 
        pfree(new_subplan_indexes);
    }
}

References Assert(), bms_add_member(), bms_free(), bms_is_empty, bms_next_member(), bms_num_members(), PartitionPruneState::do_exec_prune, PartitionPruneState::econtext, EState::es_partition_directory, PartitionedRelPruningData::exec_context, PartitionedRelPruningData::exec_pruning_steps, i, InitPartitionPruneContext(), j, NIL, PartitionedRelPruningData::nparts, PartitionPruneState::num_partprunedata, PartitionPruningData::num_partrelprunedata, PartitionPruneState::other_subplans, palloc0(), PartitionDirectoryLookup(), PartitionPruneState::partprunedata, PartitionedRelPruningData::partrel, PartitionPruningData::partrelprunedata, pfree(), PartitionedRelPruningData::present_parts, RelationGetPartitionKey(), PlanState::state, PartitionedRelPruningData::subpart_map, and PartitionedRelPruningData::subplan_map.

Referenced by ExecInitPartitionExecPruning().

InitPartitionPruneContext()

static void InitPartitionPruneContext ( PartitionPruneContext *  context, List *  pruning_steps, PartitionDesc  partdesc, PartitionKey  partkey, PlanState *  planstate, ExprContext *  econtext  )

static

Definition at line 2372 of file execPartition.c.

{
    int         n_steps;
    int         partnatts;
    ListCell   *lc;
 
    n_steps = list_length(pruning_steps);
 
    context->strategy = partkey->strategy;
    context->partnatts = partnatts = partkey->partnatts;
    context->nparts = partdesc->nparts;
    context->boundinfo = partdesc->boundinfo;
    context->partcollation = partkey->partcollation;
    context->partsupfunc = partkey->partsupfunc;
 
    /* We'll look up type-specific support functions as needed */
    context->stepcmpfuncs = (FmgrInfo *)
        palloc0(sizeof(FmgrInfo) * n_steps * partnatts);
 
    context->ppccontext = CurrentMemoryContext;
    context->planstate = planstate;
    context->exprcontext = econtext;
 
    /* Initialize expression state for each expression we need */
    context->exprstates = (ExprState **)
        palloc0(sizeof(ExprState *) * n_steps * partnatts);
    foreach(lc, pruning_steps)
    {
        PartitionPruneStepOp *step = (PartitionPruneStepOp *) lfirst(lc);
        ListCell   *lc2 = list_head(step->exprs);
        int         keyno;
 
        /* not needed for other step kinds */
        if (!IsA(step, PartitionPruneStepOp))
            continue;
 
        Assert(list_length(step->exprs) <= partnatts);
 
        for (keyno = 0; keyno < partnatts; keyno++)
        {
            if (bms_is_member(keyno, step->nullkeys))
                continue;
 
            if (lc2 != NULL)
            {
                Expr       *expr = lfirst(lc2);
 
                /* not needed for Consts */
                if (!IsA(expr, Const))
                {
                    int         stateidx = PruneCxtStateIdx(partnatts,
                                                            step->step.step_id,
                                                            keyno);
 
                    /*
                     * When planstate is NULL, pruning_steps is known not to
                     * contain any expressions that depend on the parent plan.
                     * Information of any available EXTERN parameters must be
                     * passed explicitly in that case, which the caller must
                     * have made available via econtext.
                     */
                    if (planstate == NULL)
                        context->exprstates[stateidx] =
                            ExecInitExprWithParams(expr,
                                                   econtext->ecxt_param_list_info);
                    else
                        context->exprstates[stateidx] =
                            ExecInitExpr(expr, context->planstate);
                }
                lc2 = lnext(step->exprs, lc2);
            }
        }
    }
}

References Assert(), bms_is_member(), PartitionDescData::boundinfo, PartitionPruneContext::boundinfo, CurrentMemoryContext, ExprContext::ecxt_param_list_info, ExecInitExpr(), ExecInitExprWithParams(), PartitionPruneContext::exprcontext, PartitionPruneStepOp::exprs, PartitionPruneContext::exprstates, IsA, lfirst, list_head(), list_length(), lnext(), PartitionDescData::nparts, PartitionPruneContext::nparts, PartitionPruneStepOp::nullkeys, palloc0(), PartitionPruneContext::partcollation, PartitionKeyData::partcollation, PartitionPruneContext::partnatts, PartitionKeyData::partnatts, PartitionPruneContext::partsupfunc, PartitionKeyData::partsupfunc, PartitionPruneContext::planstate, PartitionPruneContext::ppccontext, PruneCxtStateIdx, PartitionPruneStepOp::step, PartitionPruneStep::step_id, PartitionPruneContext::stepcmpfuncs, PartitionPruneContext::strategy, and PartitionKeyData::strategy.

Referenced by CreatePartitionPruneState(), and InitExecPartitionPruneContexts().

IsIndexCompatibleAsArbiter()

static bool IsIndexCompatibleAsArbiter ( Relation  arbiterIndexRelation, IndexInfo *  arbiterIndexInfo, Relation  indexRelation, IndexInfo *  indexInfo  )

static

Definition at line 502 of file execPartition.c.

{
    Assert(arbiterIndexRelation->rd_index->indrelid == indexRelation->rd_index->indrelid);
 
    /* must match whether they're unique */
    if (arbiterIndexInfo->ii_Unique != indexInfo->ii_Unique)
        return false;
 
    /* No support currently for comparing exclusion indexes. */
    if (arbiterIndexInfo->ii_ExclusionOps != NULL ||
        indexInfo->ii_ExclusionOps != NULL)
        return false;
 
    /* the "nulls not distinct" criterion must match */
    if (arbiterIndexInfo->ii_NullsNotDistinct !=
        indexInfo->ii_NullsNotDistinct)
        return false;
 
    /* number of key attributes must match */
    if (arbiterIndexInfo->ii_NumIndexKeyAttrs !=
        indexInfo->ii_NumIndexKeyAttrs)
        return false;
 
    for (int i = 0; i < arbiterIndexInfo->ii_NumIndexKeyAttrs; i++)
    {
        if (arbiterIndexRelation->rd_indcollation[i] !=
            indexRelation->rd_indcollation[i])
            return false;
 
        if (arbiterIndexRelation->rd_opfamily[i] !=
            indexRelation->rd_opfamily[i])
            return false;
 
        if (arbiterIndexRelation->rd_index->indkey.values[i] !=
            indexRelation->rd_index->indkey.values[i])
            return false;
    }
 
    if (list_difference(RelationGetIndexExpressions(arbiterIndexRelation),
                        RelationGetIndexExpressions(indexRelation)) != NIL)
        return false;
 
    if (list_difference(RelationGetIndexPredicate(arbiterIndexRelation),
                        RelationGetIndexPredicate(indexRelation)) != NIL)
        return false;
    return true;
}

References Assert(), i, IndexInfo::ii_ExclusionOps, IndexInfo::ii_NullsNotDistinct, IndexInfo::ii_NumIndexKeyAttrs, IndexInfo::ii_Unique, list_difference(), NIL, RelationData::rd_indcollation, RelationData::rd_index, RelationData::rd_opfamily, RelationGetIndexExpressions(), and RelationGetIndexPredicate().

Referenced by ExecInitPartitionInfo().