* Definitions for planner's internal data structures.
*
*
- * Portions Copyright (c) 1996-2007, PostgreSQL Global Development Group
+ * Portions Copyright (c) 1996-2010, PostgreSQL Global Development Group
* Portions Copyright (c) 1994, Regents of the University of California
*
- * $PostgreSQL: pgsql/src/include/nodes/relation.h,v 1.142 2007/05/22 01:40:33 tgl Exp $
+ * src/include/nodes/relation.h
*
*-------------------------------------------------------------------------
*/
List *subrtables; /* Rangetables for SubPlan nodes */
+ List *subrowmarks; /* PlanRowMarks for SubPlan nodes */
+
Bitmapset *rewindPlanIDs; /* indices of subplans that require REWIND */
List *finalrtable; /* "flat" rangetable for executor */
+
+ List *finalrowmarks; /* "flat" list of PlanRowMarks */
+
+ List *relationOids; /* OIDs of relations the plan depends on */
+
+ List *invalItems; /* other dependencies, as PlanInvalItems */
+
+ Index lastPHId; /* highest PlaceHolderVar ID assigned */
+
+ bool transientPlan; /* redo plan when TransactionXmin changes? */
} PlannerGlobal;
/* macro for fetching the Plan associated with a SubPlan node */
Index query_level; /* 1 at the outermost Query */
+ struct PlannerInfo *parent_root; /* NULL at outermost Query */
+
/*
* simple_rel_array holds pointers to "base rels" and "other rels" (see
* comments for RelOptInfo for more info). It is indexed by rangetable
* rt_fetch(), which can be a bit slow once large inheritance sets have
* been expanded.
*/
- RangeTblEntry **simple_rte_array; /* rangetable as an array */
+ RangeTblEntry **simple_rte_array; /* rangetable as an array */
/*
* join_rel_list is a list of all join-relation RelOptInfos we have
List *join_rel_list; /* list of join-relation RelOptInfos */
struct HTAB *join_rel_hash; /* optional hashtable for join relations */
+ /*
+ * When doing a dynamic-programming-style join search, join_rel_level[k]
+ * is a list of all join-relation RelOptInfos of level k, and
+ * join_cur_level is the current level. New join-relation RelOptInfos are
+ * automatically added to the join_rel_level[join_cur_level] list.
+ * join_rel_level is NULL if not in use.
+ */
+ List **join_rel_level; /* lists of join-relation RelOptInfos */
+ int join_cur_level; /* index of list being extended */
+
List *resultRelations; /* integer list of RT indexes, or NIL */
- List *returningLists; /* list of lists of TargetEntry, or NIL */
+ List *init_plans; /* init SubPlans for query */
- List *init_plans; /* init subplans for query */
+ List *cte_plan_ids; /* per-CTE-item list of subplan IDs */
- List *eq_classes; /* list of active EquivalenceClasses */
+ List *eq_classes; /* list of active EquivalenceClasses */
- List *canon_pathkeys; /* list of "canonical" PathKeys */
+ List *canon_pathkeys; /* list of "canonical" PathKeys */
List *left_join_clauses; /* list of RestrictInfos for
* mergejoinable outer join clauses
List *full_join_clauses; /* list of RestrictInfos for
* mergejoinable full join clauses */
- List *oj_info_list; /* list of OuterJoinInfos */
-
- List *in_info_list; /* list of InClauseInfos */
+ List *join_info_list; /* list of SpecialJoinInfos */
List *append_rel_list; /* list of AppendRelInfos */
+ List *rowMarks; /* list of PlanRowMarks */
+
+ List *placeholder_list; /* list of PlaceHolderInfos */
+
List *query_pathkeys; /* desired pathkeys for query_planner(), and
* actual pathkeys afterwards */
List *group_pathkeys; /* groupClause pathkeys, if any */
+ List *window_pathkeys; /* pathkeys of bottom window, if any */
+ List *distinct_pathkeys; /* distinctClause pathkeys, if any */
List *sort_pathkeys; /* sortClause pathkeys, if any */
+ List *minmax_aggs; /* List of MinMaxAggInfos */
+
+ List *initial_rels; /* RelOptInfos we are now trying to join */
+
MemoryContext planner_cxt; /* context holding PlannerInfo */
double total_table_pages; /* # of pages in all tables of query */
double tuple_fraction; /* tuple_fraction passed to query_planner */
+ double limit_tuples; /* limit_tuples passed to query_planner */
+ bool hasInheritedTarget; /* true if parse->resultRelation is an
+ * inheritance child rel */
bool hasJoinRTEs; /* true if any RTEs are RTE_JOIN kind */
- bool hasOuterJoins; /* true if any RTEs are outer joins */
bool hasHavingQual; /* true if havingQual was non-null */
bool hasPseudoConstantQuals; /* true if any RestrictInfo has
* pseudoconstant = true */
+ bool hasRecursion; /* true if planning a recursive WITH item */
+
+ /* These fields are used only when hasRecursion is true: */
+ int wt_param_id; /* PARAM_EXEC ID for the work table */
+ struct Plan *non_recursive_plan; /* plan for non-recursive term */
+
+ /* These fields are workspace for createplan.c */
+ Relids curOuterRels; /* outer rels above current node */
+ List *curOuterParams; /* not-yet-assigned NestLoopParams */
+
+ /* optional private data for join_search_hook, e.g., GEQO */
+ void *join_search_private;
} PlannerInfo;
*
* We also have "other rels", which are like base rels in that they refer to
* single RT indexes; but they are not part of the join tree, and are given
- * a different RelOptKind to identify them.
+ * a different RelOptKind to identify them. Lastly, there is a RelOptKind
+ * for "dead" relations, which are base rels that we have proven we don't
+ * need to join after all.
*
* Currently the only kind of otherrels are those made for member relations
* of an "append relation", that is an inheritance set or UNION ALL subquery.
* the entire append relation. The member RTEs are otherrels. The parent
* is present in the query join tree but the members are not. The member
* RTEs and otherrels are used to plan the scans of the individual tables or
- * subqueries of the append set; then the parent baserel is given an Append
- * plan comprising the best plans for the individual member rels. (See
- * comments for AppendRelInfo for more information.)
+ * subqueries of the append set; then the parent baserel is given Append
+ * and/or MergeAppend paths comprising the best paths for the individual
+ * member rels. (See comments for AppendRelInfo for more information.)
*
* At one time we also made otherrels to represent join RTEs, for use in
* handling join alias Vars. Currently this is not needed because all join
* clauses have been applied (ie, output rows of a plan for it)
* width - avg. number of bytes per tuple in the relation after the
* appropriate projections have been done (ie, output width)
- * reltargetlist - List of Var nodes for the attributes we need to
- * output from this relation (in no particular order)
- * NOTE: in a child relation, may contain RowExprs
+ * reltargetlist - List of Var and PlaceHolderVar nodes for the values
+ * we need to output from this relation.
+ * List is in no particular order, but all rels of an
+ * appendrel set must use corresponding orders.
+ * NOTE: in a child relation, may contain RowExpr or
+ * ConvertRowtypeExpr representing a whole-row Var.
* pathlist - List of Path nodes, one for each potentially useful
* method of generating the relation
* cheapest_startup_path - the pathlist member with lowest startup cost
* tuples - number of tuples in relation (not considering restrictions)
* subplan - plan for subquery (NULL if it's not a subquery)
* subrtable - rangetable for subquery (NIL if it's not a subquery)
+ * subrowmark - rowmarks for subquery (NIL if it's not a subquery)
*
* Note: for a subquery, tuples and subplan are not set immediately
* upon creation of the RelOptInfo object; they are filled in when
{
RELOPT_BASEREL,
RELOPT_JOINREL,
- RELOPT_OTHER_MEMBER_REL
+ RELOPT_OTHER_MEMBER_REL,
+ RELOPT_DEADREL
} RelOptKind;
typedef struct RelOptInfo
int width; /* estimated avg width of result tuples */
/* materialization information */
- List *reltargetlist; /* needed Vars */
+ List *reltargetlist; /* Vars to be output by scan of relation */
List *pathlist; /* Path structures */
struct Path *cheapest_startup_path;
struct Path *cheapest_total_path;
/* information about a base rel (not set for join rels!) */
Index relid;
+ Oid reltablespace; /* containing tablespace */
RTEKind rtekind; /* RELATION, SUBQUERY, or FUNCTION */
AttrNumber min_attr; /* smallest attrno of rel (often <0) */
AttrNumber max_attr; /* largest attrno of rel */
Relids *attr_needed; /* array indexed [min_attr .. max_attr] */
int32 *attr_widths; /* array indexed [min_attr .. max_attr] */
- List *indexlist;
+ List *indexlist; /* list of IndexOptInfo */
BlockNumber pages;
double tuples;
struct Plan *subplan; /* if subquery */
List *subrtable; /* if subquery */
+ List *subrowmark; /* if subquery */
/* used by various scans and joins: */
List *baserestrictinfo; /* RestrictInfo structures (if base
* and indexes, but that created confusion without actually doing anything
* useful. So now we have a separate IndexOptInfo struct for indexes.
*
- * opfamily[], indexkeys[], fwdsortop[], revsortop[], and nulls_first[]
- * each have ncolumns entries. Note: for historical reasons, the
- * opfamily array has an extra entry that is always zero. Some code
- * scans until it sees a zero entry, rather than looking at ncolumns.
+ * opfamily[], indexkeys[], opcintype[], fwdsortop[], revsortop[],
+ * and nulls_first[] each have ncolumns entries.
*
* Zeroes in the indexkeys[] array indicate index columns that are
* expressions; there is one element in indexprs for each such column.
*
- * For an unordered index, the sortop arrays contains zeroes. Note that
+ * For an unordered index, the sortop arrays contains zeroes. Note that
* fwdsortop[] and nulls_first[] describe the sort ordering of a forward
* indexscan; we can also consider a backward indexscan, which will
* generate sort order described by revsortop/!nulls_first.
NodeTag type;
Oid indexoid; /* OID of the index relation */
+ Oid reltablespace; /* tablespace of index (not table) */
RelOptInfo *rel; /* back-link to index's table */
/* statistics from pg_class */
int ncolumns; /* number of columns in index */
Oid *opfamily; /* OIDs of operator families for columns */
int *indexkeys; /* column numbers of index's keys, or 0 */
+ Oid *opcintype; /* OIDs of opclass declared input data types */
Oid *fwdsortop; /* OIDs of sort operators for each column */
Oid *revsortop; /* OIDs of sort operators for backward scan */
bool *nulls_first; /* do NULLs come first in the sort order? */
bool predOK; /* true if predicate matches query */
bool unique; /* true if a unique index */
+ bool amcanorderbyop; /* does AM support order by operator result? */
bool amoptionalkey; /* can query omit key for the first column? */
- bool amsearchnulls; /* can AM search for NULL index entries? */
+ bool amsearchnulls; /* can AM search for NULL/NOT NULL entries? */
+ bool amhasgettuple; /* does AM have amgettuple interface? */
+ bool amhasgetbitmap; /* does AM have amgetbitmap interface? */
} IndexOptInfo;
* us represent knowledge about different sort orderings being equivalent.
* Since every PathKey must reference an EquivalenceClass, we will end up
* with single-member EquivalenceClasses whenever a sort key expression has
- * not been equivalenced to anything else. It is also possible that such an
+ * not been equivalenced to anything else. It is also possible that such an
* EquivalenceClass will contain a volatile expression ("ORDER BY random()"),
* which is a case that can't arise otherwise since clauses containing
* volatile functions are never considered mergejoinable. We mark such
* EquivalenceClasses specially to prevent them from being merged with
- * ordinary EquivalenceClasses.
+ * ordinary EquivalenceClasses. Also, for volatile expressions we have
+ * to be careful to match the EquivalenceClass to the correct targetlist
+ * entry: consider SELECT random() AS a, random() AS b ... ORDER BY b,a.
+ * So we record the SortGroupRef of the originating sort clause.
*
* We allow equality clauses appearing below the nullable side of an outer join
* to form EquivalenceClasses, but these have a slightly different meaning:
* the included values might be all NULL rather than all the same non-null
- * values. See src/backend/optimizer/README for more on that point.
+ * values. See src/backend/optimizer/README for more on that point.
*
* NB: if ec_merged isn't NULL, this class has been merged into another, and
* should be ignored in favor of using the pointed-to class.
{
NodeTag type;
- List *ec_opfamilies; /* btree operator family OIDs */
- List *ec_members; /* list of EquivalenceMembers */
- List *ec_sources; /* list of generating RestrictInfos */
- List *ec_derives; /* list of derived RestrictInfos */
- Relids ec_relids; /* all relids appearing in ec_members */
- bool ec_has_const; /* any pseudoconstants in ec_members? */
+ List *ec_opfamilies; /* btree operator family OIDs */
+ List *ec_members; /* list of EquivalenceMembers */
+ List *ec_sources; /* list of generating RestrictInfos */
+ List *ec_derives; /* list of derived RestrictInfos */
+ Relids ec_relids; /* all relids appearing in ec_members */
+ bool ec_has_const; /* any pseudoconstants in ec_members? */
bool ec_has_volatile; /* the (sole) member is a volatile expr */
bool ec_below_outer_join; /* equivalence applies below an OJ */
- bool ec_broken; /* failed to generate needed clauses? */
- struct EquivalenceClass *ec_merged; /* set if merged into another EC */
+ bool ec_broken; /* failed to generate needed clauses? */
+ Index ec_sortref; /* originating sortclause label, or 0 */
+ struct EquivalenceClass *ec_merged; /* set if merged into another EC */
} EquivalenceClass;
+/*
+ * If an EC contains a const and isn't below-outer-join, any PathKey depending
+ * on it must be redundant, since there's only one possible value of the key.
+ */
+#define EC_MUST_BE_REDUNDANT(eclass) \
+ ((eclass)->ec_has_const && !(eclass)->ec_below_outer_join)
+
/*
* EquivalenceMember - one member expression of an EquivalenceClass
*
* em_is_child signifies that this element was built by transposing a member
* for an inheritance parent relation to represent the corresponding expression
- * on an inheritance child. The element should be ignored for all purposes
- * except constructing inner-indexscan paths for the child relation. (Other
- * types of join are driven from transposed joininfo-list entries.) Note
- * that the EC's ec_relids field does NOT include the child relation.
+ * on an inheritance child. These elements are used for constructing
+ * inner-indexscan paths for the child relation (other types of join are
+ * driven from transposed joininfo-list entries) and for constructing
+ * MergeAppend paths for the whole inheritance tree. Note that the EC's
+ * ec_relids field does NOT include the child relation.
*
* em_datatype is usually the same as exprType(em_expr), but can be
* different when dealing with a binary-compatible opfamily; in particular
- * anyarray_ops would never work without this. Use em_datatype when
+ * anyarray_ops would never work without this. Use em_datatype when
* looking up a specific btree operator to work with this expression.
*/
typedef struct EquivalenceMember
* etc. The value being sorted is represented by linking to an
* EquivalenceClass containing that value and including pk_opfamily among its
* ec_opfamilies. This is a convenient method because it makes it trivial
- * to detect equivalent and closely-related orderings. (See optimizer/README
+ * to detect equivalent and closely-related orderings. (See optimizer/README
* for more information.)
*
* Note: pk_strategy is either BTLessStrategyNumber (for ASC) or
- * BTGreaterStrategyNumber (for DESC). We assume that all ordering-capable
+ * BTGreaterStrategyNumber (for DESC). We assume that all ordering-capable
* index types will use btree-compatible strategy numbers.
*/
NodeTag type;
EquivalenceClass *pk_eclass; /* the value that is ordered */
- Oid pk_opfamily; /* btree opfamily defining the ordering */
- int pk_strategy; /* sort direction (ASC or DESC) */
- bool pk_nulls_first; /* do NULLs come before normal values? */
+ Oid pk_opfamily; /* btree opfamily defining the ordering */
+ int pk_strategy; /* sort direction (ASC or DESC) */
+ bool pk_nulls_first; /* do NULLs come before normal values? */
} PathKey;
/*
- * Type "Path" is used as-is for sequential-scan paths. For other
- * path types it is the first component of a larger struct.
+ * Type "Path" is used as-is for sequential-scan paths, as well as some other
+ * simple plan types that we don't need any extra information in the path for.
+ * For other path types it is the first component of a larger struct.
*
* Note: "pathtype" is the NodeTag of the Plan node we could build from this
* Path. It is partially redundant with the Path's NodeTag, but allows us
*
* Note: it is possible for "subpaths" to contain only one, or even no,
* elements. These cases are optimized during create_append_plan.
+ * In particular, an AppendPath with no subpaths is a "dummy" path that
+ * is created to represent the case that a relation is provably empty.
*/
typedef struct AppendPath
{
List *subpaths; /* list of component Paths */
} AppendPath;
+#define IS_DUMMY_PATH(p) \
+ (IsA((p), AppendPath) && ((AppendPath *) (p))->subpaths == NIL)
+
+/*
+ * MergeAppendPath represents a MergeAppend plan, ie, the merging of sorted
+ * results from several member plans to produce similarly-sorted output.
+ */
+typedef struct MergeAppendPath
+{
+ Path path;
+ List *subpaths; /* list of component Paths */
+ double limit_tuples; /* hard limit on output tuples, or -1 */
+} MergeAppendPath;
+
/*
* ResultPath represents use of a Result plan node to compute a variable-free
* targetlist with no underlying tables (a "SELECT expressions" query).
Path path;
Path *subpath;
UniquePathMethod umethod;
+ List *in_operators; /* equality operators of the IN clause */
+ List *uniq_exprs; /* expressions to be made unique */
double rows; /* estimated number of result tuples */
} UniquePath;
/*
* A mergejoin path has these fields.
*
+ * Unlike other path types, a MergePath node doesn't represent just a single
+ * run-time plan node: it can represent up to four. Aside from the MergeJoin
+ * node itself, there can be a Sort node for the outer input, a Sort node
+ * for the inner input, and/or a Material node for the inner input. We could
+ * represent these nodes by separate path nodes, but considering how many
+ * different merge paths are investigated during a complex join problem,
+ * it seems better to avoid unnecessary palloc overhead.
+ *
* path_mergeclauses lists the clauses (in the form of RestrictInfos)
* that will be used in the merge.
*
* outersortkeys (resp. innersortkeys) is NIL if the outer path
* (resp. inner path) is already ordered appropriately for the
* mergejoin. If it is not NIL then it is a PathKeys list describing
- * the ordering that must be created by an explicit sort step.
+ * the ordering that must be created by an explicit Sort node.
+ *
+ * materialize_inner is TRUE if a Material node should be placed atop the
+ * inner input. This may appear with or without an inner Sort step.
*/
typedef struct MergePath
List *path_mergeclauses; /* join clauses to be used for merge */
List *outersortkeys; /* keys for explicit sort, if any */
List *innersortkeys; /* keys for explicit sort, if any */
+ bool materialize_inner; /* add Materialize to inner? */
} MergePath;
/*
{
JoinPath jpath;
List *path_hashclauses; /* join clauses used for hashing */
+ int num_batches; /* number of batches expected */
} HashPath;
/*
*
* When dealing with outer joins we have to be very careful about pushing qual
* clauses up and down the tree. An outer join's own JOIN/ON conditions must
- * be evaluated exactly at that join node, and any quals appearing in WHERE or
- * in a JOIN above the outer join cannot be pushed down below the outer join.
- * Otherwise the outer join will produce wrong results because it will see the
- * wrong sets of input rows. All quals are stored as RestrictInfo nodes
- * during planning, but there's a flag to indicate whether a qual has been
+ * be evaluated exactly at that join node, unless they are "degenerate"
+ * conditions that reference only Vars from the nullable side of the join.
+ * Quals appearing in WHERE or in a JOIN above the outer join cannot be pushed
+ * down below the outer join, if they reference any nullable Vars.
+ * RestrictInfo nodes contain a flag to indicate whether a qual has been
* pushed down to a lower level than its original syntactic placement in the
* join tree would suggest. If an outer join prevents us from pushing a qual
* down to its "natural" semantic level (the level associated with just the
* base rels used in the qual) then we mark the qual with a "required_relids"
* value including more than just the base rels it actually uses. By
- * pretending that the qual references all the rels appearing in the outer
+ * pretending that the qual references all the rels required to form the outer
* join, we prevent it from being evaluated below the outer join's joinrel.
* When we do form the outer join's joinrel, we still need to distinguish
* those quals that are actually in that join's JOIN/ON condition from those
* that appeared elsewhere in the tree and were pushed down to the join rel
* because they used no other rels. That's what the is_pushed_down flag is
* for; it tells us that a qual is not an OUTER JOIN qual for the set of base
- * rels listed in required_relids. A clause that originally came from WHERE
+ * rels listed in required_relids. A clause that originally came from WHERE
* or an INNER JOIN condition will *always* have its is_pushed_down flag set.
* It's possible for an OUTER JOIN clause to be marked is_pushed_down too,
* if we decide that it can be pushed down into the nullable side of the join.
* In that case it acts as a plain filter qual for wherever it gets evaluated.
- *
- * When application of a qual must be delayed by outer join, we also mark it
- * with outerjoin_delayed = true. This isn't redundant with required_relids
- * because that might equal clause_relids whether or not it's an outer-join
- * clause.
+ * (In short, is_pushed_down is only false for non-degenerate outer join
+ * conditions. Possibly we should rename it to reflect that meaning?)
+ *
+ * RestrictInfo nodes also contain an outerjoin_delayed flag, which is true
+ * if the clause's applicability must be delayed due to any outer joins
+ * appearing below it (ie, it has to be postponed to some join level higher
+ * than the set of relations it actually references). There is also a
+ * nullable_relids field, which is the set of rels it references that can be
+ * forced null by some outer join below the clause. outerjoin_delayed = true
+ * is subtly different from nullable_relids != NULL: a clause might reference
+ * some nullable rels and yet not be outerjoin_delayed because it also
+ * references all the other rels of the outer join(s). A clause that is not
+ * outerjoin_delayed can be enforced anywhere it is computable.
*
* In general, the referenced clause might be arbitrarily complex. The
* kinds of clauses we can handle as indexscan quals, mergejoin clauses,
*
* When join clauses are generated from EquivalenceClasses, there may be
* several equally valid ways to enforce join equivalence, of which we need
- * apply only one. We mark clauses of this kind by setting parent_ec to
+ * apply only one. We mark clauses of this kind by setting parent_ec to
* point to the generating EquivalenceClass. Multiple clauses with the same
* parent_ec in the same join are redundant.
*/
bool is_pushed_down; /* TRUE if clause was pushed down in level */
- bool outerjoin_delayed; /* TRUE if delayed by outer join */
+ bool outerjoin_delayed; /* TRUE if delayed by lower outer join */
bool can_join; /* see comment above */
/* The set of relids required to evaluate the clause: */
Relids required_relids;
+ /* The relids used in the clause that are nullable by lower outer joins: */
+ Relids nullable_relids;
+
/* These fields are set for any binary opclause: */
Relids left_relids; /* relids in left side of clause */
Relids right_relids; /* relids in right side of clause */
/* cache space for cost and selectivity */
QualCost eval_cost; /* eval cost of clause; -1 if not yet set */
- Selectivity this_selec; /* selectivity; -1 if not yet set */
+ Selectivity norm_selec; /* selectivity for "normal" (JOIN_INNER)
+ * semantics; -1 if not yet set; >1 means a
+ * redundant clause */
+ Selectivity outer_selec; /* selectivity for outer join semantics; -1 if
+ * not yet set */
/* valid if clause is mergejoinable, else NIL */
List *mergeopfamilies; /* opfamilies containing clause operator */
/* cache space for mergeclause processing; NULL if not yet set */
EquivalenceClass *left_ec; /* EquivalenceClass containing lefthand */
- EquivalenceClass *right_ec; /* EquivalenceClass containing righthand */
- EquivalenceMember *left_em; /* EquivalenceMember for lefthand */
+ EquivalenceClass *right_ec; /* EquivalenceClass containing righthand */
+ EquivalenceMember *left_em; /* EquivalenceMember for lefthand */
EquivalenceMember *right_em; /* EquivalenceMember for righthand */
List *scansel_cache; /* list of MergeScanSelCache structs */
int strategy; /* sort direction (ASC or DESC) */
bool nulls_first; /* do NULLs come before normal values? */
/* Results */
- Selectivity leftscansel; /* scan fraction for clause left side */
- Selectivity rightscansel; /* scan fraction for clause right side */
+ Selectivity leftstartsel; /* first-join fraction for clause left side */
+ Selectivity leftendsel; /* last-join fraction for clause left side */
+ Selectivity rightstartsel; /* first-join fraction for clause right side */
+ Selectivity rightendsel; /* last-join fraction for clause right side */
} MergeScanSelCache;
/*
Relids other_relids; /* a set of relevant other relids */
bool isouterjoin; /* true if join is outer */
/* Best paths for this lookup key (NULL if no available indexscans): */
- Path *cheapest_startup_innerpath; /* cheapest startup cost */
- Path *cheapest_total_innerpath; /* cheapest total cost */
+ Path *cheapest_startup_innerpath; /* cheapest startup cost */
+ Path *cheapest_total_innerpath; /* cheapest total cost */
} InnerIndexscanInfo;
/*
- * Outer join info.
+ * Placeholder node for an expression to be evaluated below the top level
+ * of a plan tree. This is used during planning to represent the contained
+ * expression. At the end of the planning process it is replaced by either
+ * the contained expression or a Var referring to a lower-level evaluation of
+ * the contained expression. Typically the evaluation occurs below an outer
+ * join, and Var references above the outer join might thereby yield NULL
+ * instead of the expression value.
+ *
+ * Although the planner treats this as an expression node type, it is not
+ * recognized by the parser or executor, so we declare it here rather than
+ * in primnodes.h.
+ */
+
+typedef struct PlaceHolderVar
+{
+ Expr xpr;
+ Expr *phexpr; /* the represented expression */
+ Relids phrels; /* base relids syntactically within expr src */
+ Index phid; /* ID for PHV (unique within planner run) */
+ Index phlevelsup; /* > 0 if PHV belongs to outer query */
+} PlaceHolderVar;
+
+/*
+ * "Special join" info.
*
* One-sided outer joins constrain the order of joining partially but not
* completely. We flatten such joins into the planner's top-level list of
- * relations to join, but record information about each outer join in an
- * OuterJoinInfo struct. These structs are kept in the PlannerInfo node's
- * oj_info_list.
+ * relations to join, but record information about each outer join in a
+ * SpecialJoinInfo struct. These structs are kept in the PlannerInfo node's
+ * join_info_list.
+ *
+ * Similarly, semijoins and antijoins created by flattening IN (subselect)
+ * and EXISTS(subselect) clauses create partial constraints on join order.
+ * These are likewise recorded in SpecialJoinInfo structs.
+ *
+ * We make SpecialJoinInfos for FULL JOINs even though there is no flexibility
+ * of planning for them, because this simplifies make_join_rel()'s API.
*
* min_lefthand and min_righthand are the sets of base relids that must be
- * available on each side when performing the outer join. lhs_strict is
- * true if the outer join's condition cannot succeed when the LHS variables
- * are all NULL (this means that the outer join can commute with upper-level
+ * available on each side when performing the special join. lhs_strict is
+ * true if the special join's condition cannot succeed when the LHS variables
+ * are all NULL (this means that an outer join can commute with upper-level
* outer joins even if it appears in their RHS). We don't bother to set
* lhs_strict for FULL JOINs, however.
*
* It is not valid for either min_lefthand or min_righthand to be empty sets;
* if they were, this would break the logic that enforces join order.
*
- * Note: OuterJoinInfo directly represents only LEFT JOIN and FULL JOIN;
- * RIGHT JOIN is handled by switching the inputs to make it a LEFT JOIN.
- * We make an OuterJoinInfo for FULL JOINs even though there is no flexibility
- * of planning for them, because this simplifies make_join_rel()'s API.
+ * syn_lefthand and syn_righthand are the sets of base relids that are
+ * syntactically below this special join. (These are needed to help compute
+ * min_lefthand and min_righthand for higher joins.)
+ *
+ * delay_upper_joins is set TRUE if we detect a pushed-down clause that has
+ * to be evaluated after this join is formed (because it references the RHS).
+ * Any outer joins that have such a clause and this join in their RHS cannot
+ * commute with this join, because that would leave noplace to check the
+ * pushed-down clause. (We don't track this for FULL JOINs, either.)
+ *
+ * join_quals is an implicit-AND list of the quals syntactically associated
+ * with the join (they may or may not end up being applied at the join level).
+ * This is just a side list and does not drive actual application of quals.
+ * For JOIN_SEMI joins, this is cleared to NIL in create_unique_path() if
+ * the join is found not to be suitable for a uniqueify-the-RHS plan.
+ *
+ * jointype is never JOIN_RIGHT; a RIGHT JOIN is handled by switching
+ * the inputs to make it a LEFT JOIN. So the allowed values of jointype
+ * in a join_info_list member are only LEFT, FULL, SEMI, or ANTI.
+ *
+ * For purposes of join selectivity estimation, we create transient
+ * SpecialJoinInfo structures for regular inner joins; so it is possible
+ * to have jointype == JOIN_INNER in such a structure, even though this is
+ * not allowed within join_info_list. We also create transient
+ * SpecialJoinInfos with jointype == JOIN_INNER for outer joins, since for
+ * cost estimation purposes it is sometimes useful to know the join size under
+ * plain innerjoin semantics. Note that lhs_strict, delay_upper_joins, and
+ * join_quals are not set meaningfully within such structs.
*/
-typedef struct OuterJoinInfo
+typedef struct SpecialJoinInfo
{
NodeTag type;
Relids min_lefthand; /* base relids in minimum LHS for join */
Relids min_righthand; /* base relids in minimum RHS for join */
- bool is_full_join; /* it's a FULL OUTER JOIN */
+ Relids syn_lefthand; /* base relids syntactically within LHS */
+ Relids syn_righthand; /* base relids syntactically within RHS */
+ JoinType jointype; /* always INNER, LEFT, FULL, SEMI, or ANTI */
bool lhs_strict; /* joinclause is strict for some LHS rel */
-} OuterJoinInfo;
-
-/*
- * IN clause info.
- *
- * When we convert top-level IN quals into join operations, we must restrict
- * the order of joining and use special join methods at some join points.
- * We record information about each such IN clause in an InClauseInfo struct.
- * These structs are kept in the PlannerInfo node's in_info_list.
- *
- * Note: sub_targetlist is just a list of Vars or expressions; it does not
- * contain TargetEntry nodes.
- */
-
-typedef struct InClauseInfo
-{
- NodeTag type;
- Relids lefthand; /* base relids in lefthand expressions */
- Relids righthand; /* base relids coming from the subselect */
- List *sub_targetlist; /* targetlist of original RHS subquery */
- List *in_operators; /* OIDs of the IN's equality operator(s) */
-} InClauseInfo;
+ bool delay_upper_joins; /* can't commute with upper RHS */
+ List *join_quals; /* join quals, in implicit-AND list format */
+} SpecialJoinInfo;
/*
* Append-relation info.
Oid parent_reltype; /* OID of parent's composite type */
Oid child_reltype; /* OID of child's composite type */
- /*
- * The N'th element of this list is the integer column number of the child
- * column corresponding to the N'th column of the parent. A list element
- * is zero if it corresponds to a dropped column of the parent (this is
- * only possible for inheritance cases, not UNION ALL).
- */
- List *col_mappings; /* list of child attribute numbers */
-
/*
* The N'th element of this list is a Var or expression representing the
* child column corresponding to the N'th column of the parent. This is
* used to translate Vars referencing the parent rel into references to
* the child. A list element is NULL if it corresponds to a dropped
* column of the parent (this is only possible for inheritance cases, not
- * UNION ALL).
- *
- * This might seem redundant with the col_mappings data, but it is handy
- * because flattening of sub-SELECTs that are members of a UNION ALL will
- * cause changes in the expressions that need to be substituted for a
- * parent Var. Adjusting this data structure lets us track what really
- * needs to be substituted.
+ * UNION ALL). The list elements are always simple Vars for inheritance
+ * cases, but can be arbitrary expressions in UNION ALL cases.
*
* Notice we only store entries for user columns (attno > 0). Whole-row
* Vars are special-cased, and system columns (attno < 0) need no special
Oid parent_reloid; /* OID of parent relation */
} AppendRelInfo;
+/*
+ * For each distinct placeholder expression generated during planning, we
+ * store a PlaceHolderInfo node in the PlannerInfo node's placeholder_list.
+ * This stores info that is needed centrally rather than in each copy of the
+ * PlaceHolderVar. The phid fields identify which PlaceHolderInfo goes with
+ * each PlaceHolderVar. Note that phid is unique throughout a planner run,
+ * not just within a query level --- this is so that we need not reassign ID's
+ * when pulling a subquery into its parent.
+ *
+ * The idea is to evaluate the expression at (only) the ph_eval_at join level,
+ * then allow it to bubble up like a Var until the ph_needed join level.
+ * ph_needed has the same definition as attr_needed for a regular Var.
+ *
+ * ph_may_need is an initial estimate of ph_needed, formed using the
+ * syntactic locations of references to the PHV. We need this in order to
+ * determine whether the PHV reference forces a join ordering constraint:
+ * if the PHV has to be evaluated below the nullable side of an outer join,
+ * and then used above that outer join, we must constrain join order to ensure
+ * there's a valid place to evaluate the PHV below the join. The final
+ * actual ph_needed level might be lower than ph_may_need, but we can't
+ * determine that until later on. Fortunately this doesn't matter for what
+ * we need ph_may_need for: if there's a PHV reference syntactically
+ * above the outer join, it's not going to be allowed to drop below the outer
+ * join, so we would come to the same conclusions about join order even if
+ * we had the final ph_needed value to compare to.
+ *
+ * We create a PlaceHolderInfo only after determining that the PlaceHolderVar
+ * is actually referenced in the plan tree, so that unreferenced placeholders
+ * don't result in unnecessary constraints on join order.
+ */
+
+typedef struct PlaceHolderInfo
+{
+ NodeTag type;
+
+ Index phid; /* ID for PH (unique within planner run) */
+ PlaceHolderVar *ph_var; /* copy of PlaceHolderVar tree */
+ Relids ph_eval_at; /* lowest level we can evaluate value at */
+ Relids ph_needed; /* highest level the value is needed at */
+ Relids ph_may_need; /* highest level it might be needed at */
+ int32 ph_width; /* estimated attribute width */
+} PlaceHolderInfo;
+
+/*
+ * For each potentially index-optimizable MIN/MAX aggregate function,
+ * root->minmax_aggs stores a MinMaxAggInfo describing it.
+ *
+ * Note: a MIN/MAX agg doesn't really care about the nulls_first property,
+ * so the pathkey's nulls_first flag should be ignored.
+ */
+typedef struct MinMaxAggInfo
+{
+ NodeTag type;
+
+ Oid aggfnoid; /* pg_proc Oid of the aggregate */
+ Oid aggsortop; /* Oid of its sort operator */
+ Expr *target; /* expression we are aggregating on */
+ List *pathkeys; /* pathkeys representing needed sort order */
+} MinMaxAggInfo;
+
/*
* glob->paramlist keeps track of the PARAM_EXEC slots that we have decided
* we need for the query. At runtime these slots are used to pass values
- * either down into subqueries (for outer references in subqueries) or up out
- * of subqueries (for the results of a subplan). The n'th entry in the list
- * (n counts from 0) corresponds to Param->paramid = n.
+ * around from one plan node to another. They can be used to pass values
+ * down into subqueries (for outer references in subqueries), or up out of
+ * subqueries (for the results of a subplan), or from a NestLoop plan node
+ * into its inner relation (when the inner scan is parameterized with values
+ * from the outer relation). The n'th entry in the list (n counts from 0)
+ * corresponds to Param->paramid = n.
*
* Each paramlist item shows the absolute query level it is associated with,
* where the outermost query is level 1 and nested subqueries have higher
* numbers. The item the parameter slot represents can be one of three kinds:
*
* A Var: the slot represents a variable of that level that must be passed
- * down because subqueries have outer references to it. The varlevelsup
+ * down because subqueries have outer references to it, or must be passed
+ * from a NestLoop node of that level to its inner scan. The varlevelsup
* value in the Var will always be zero.
*
* An Aggref (with an expression tree representing its argument): the slot
* to the parent query of the subplan.
*
* Note: we detect duplicate Var parameters and coalesce them into one slot,
- * but we do not do this for Aggref or Param slots.
+ * but we do not bother to do this for Aggrefs, and it would be incorrect
+ * to do so for Param slots. Duplicate detection is actually *necessary*
+ * in the case of NestLoop parameters since it serves to match up the usage
+ * of a Param (in the inner scan) with the assignment of the value (in the
+ * NestLoop node). This might result in the same PARAM_EXEC slot being used
+ * by multiple NestLoop nodes or SubPlan nodes, but no harm is done since
+ * the same value would be assigned anyway.
*/
typedef struct PlannerParamItem
{
projects / postgresql / blobdiff