⭐ 欢迎来到虫虫下载站! | 📦 资源下载 📁 资源专辑 ℹ️ 关于我们
⭐ 虫虫下载站

📄 joinrels.c

📁 postgresql8.3.4源码,开源数据库
💻 C
📖 第 1 页 / 共 2 页
字号:
/*------------------------------------------------------------------------- * * joinrels.c *	  Routines to determine which relations should be joined * * Portions Copyright (c) 1996-2008, PostgreSQL Global Development Group * Portions Copyright (c) 1994, Regents of the University of California * * * IDENTIFICATION *	  $PostgreSQL: pgsql/src/backend/optimizer/path/joinrels.c,v 1.91.2.1 2008/03/24 21:53:12 tgl Exp $ * *------------------------------------------------------------------------- */#include "postgres.h"#include "optimizer/joininfo.h"#include "optimizer/pathnode.h"#include "optimizer/paths.h"static List *make_rels_by_clause_joins(PlannerInfo *root,						  RelOptInfo *old_rel,						  ListCell *other_rels);static List *make_rels_by_clauseless_joins(PlannerInfo *root,							  RelOptInfo *old_rel,							  ListCell *other_rels);static bool has_join_restriction(PlannerInfo *root, RelOptInfo *rel);static bool has_legal_joinclause(PlannerInfo *root, RelOptInfo *rel);static bool is_dummy_rel(RelOptInfo *rel);static void mark_dummy_join(RelOptInfo *rel);/* * join_search_one_level *	  Consider ways to produce join relations containing exactly 'level' *	  jointree items.  (This is one step of the dynamic-programming method *	  embodied in standard_join_search.)  Join rel nodes for each feasible *	  combination of lower-level rels are created and returned in a list. *	  Implementation paths are created for each such joinrel, too. * * level: level of rels we want to make this time. * joinrels[j], 1 <= j < level, is a list of rels containing j items. */List *join_search_one_level(PlannerInfo *root, int level, List **joinrels){	List	   *result_rels = NIL;	List	   *new_rels;	ListCell   *r;	int			k;	/*	 * First, consider left-sided and right-sided plans, in which rels of	 * exactly level-1 member relations are joined against initial relations.	 * We prefer to join using join clauses, but if we find a rel of level-1	 * members that has no join clauses, we will generate Cartesian-product	 * joins against all initial rels not already contained in it.	 *	 * In the first pass (level == 2), we try to join each initial rel to each	 * initial rel that appears later in joinrels[1].  (The mirror-image joins	 * are handled automatically by make_join_rel.)  In later passes, we try	 * to join rels of size level-1 from joinrels[level-1] to each initial rel	 * in joinrels[1].	 */	foreach(r, joinrels[level - 1])	{		RelOptInfo *old_rel = (RelOptInfo *) lfirst(r);		ListCell   *other_rels;		if (level == 2)			other_rels = lnext(r);		/* only consider remaining initial										 * rels */		else			other_rels = list_head(joinrels[1]);		/* consider all initial														 * rels */		if (old_rel->joininfo != NIL || old_rel->has_eclass_joins ||			has_join_restriction(root, old_rel))		{			/*			 * Note that if all available join clauses for this rel require			 * more than one other rel, we will fail to make any joins against			 * it here.  In most cases that's OK; it'll be considered by			 * "bushy plan" join code in a higher-level pass where we have			 * those other rels collected into a join rel.			 *			 * See also the last-ditch case below.			 */			new_rels = make_rels_by_clause_joins(root,												 old_rel,												 other_rels);		}		else		{			/*			 * Oops, we have a relation that is not joined to any other			 * relation, either directly or by join-order restrictions.			 * Cartesian product time.			 */			new_rels = make_rels_by_clauseless_joins(root,													 old_rel,													 other_rels);		}		/*		 * At levels above 2 we will generate the same joined relation in		 * multiple ways --- for example (a join b) join c is the same		 * RelOptInfo as (b join c) join a, though the second case will add a		 * different set of Paths to it.  To avoid making extra work for		 * subsequent passes, do not enter the same RelOptInfo into our output		 * list multiple times.		 */		result_rels = list_concat_unique_ptr(result_rels, new_rels);	}	/*	 * Now, consider "bushy plans" in which relations of k initial rels are	 * joined to relations of level-k initial rels, for 2 <= k <= level-2.	 *	 * We only consider bushy-plan joins for pairs of rels where there is a	 * suitable join clause (or join order restriction), in order to avoid	 * unreasonable growth of planning time.	 */	for (k = 2;; k++)	{		int			other_level = level - k;		/*		 * Since make_join_rel(x, y) handles both x,y and y,x cases, we only		 * need to go as far as the halfway point.		 */		if (k > other_level)			break;		foreach(r, joinrels[k])		{			RelOptInfo *old_rel = (RelOptInfo *) lfirst(r);			ListCell   *other_rels;			ListCell   *r2;			/*			 * We can ignore clauseless joins here, *except* when they			 * participate in join-order restrictions --- then we might have			 * to force a bushy join plan.			 */			if (old_rel->joininfo == NIL && !old_rel->has_eclass_joins &&				!has_join_restriction(root, old_rel))				continue;			if (k == other_level)				other_rels = lnext(r);	/* only consider remaining rels */			else				other_rels = list_head(joinrels[other_level]);			for_each_cell(r2, other_rels)			{				RelOptInfo *new_rel = (RelOptInfo *) lfirst(r2);				if (!bms_overlap(old_rel->relids, new_rel->relids))				{					/*					 * OK, we can build a rel of the right level from this					 * pair of rels.  Do so if there is at least one usable					 * join clause or a relevant join restriction.					 */					if (have_relevant_joinclause(root, old_rel, new_rel) ||						have_join_order_restriction(root, old_rel, new_rel))					{						RelOptInfo *jrel;						jrel = make_join_rel(root, old_rel, new_rel);						/* Avoid making duplicate entries ... */						if (jrel)							result_rels = list_append_unique_ptr(result_rels,																 jrel);					}				}			}		}	}	/*	 * Last-ditch effort: if we failed to find any usable joins so far, force	 * a set of cartesian-product joins to be generated.  This handles the	 * special case where all the available rels have join clauses but we	 * cannot use any of the joins yet.  An example is	 *	 * SELECT * FROM a,b,c WHERE (a.f1 + b.f2 + c.f3) = 0;	 *	 * The join clause will be usable at level 3, but at level 2 we have no	 * choice but to make cartesian joins.	We consider only left-sided and	 * right-sided cartesian joins in this case (no bushy).	 */	if (result_rels == NIL)	{		/*		 * This loop is just like the first one, except we always call		 * make_rels_by_clauseless_joins().		 */		foreach(r, joinrels[level - 1])		{			RelOptInfo *old_rel = (RelOptInfo *) lfirst(r);			ListCell   *other_rels;			if (level == 2)				other_rels = lnext(r);	/* only consider remaining initial										 * rels */			else				other_rels = list_head(joinrels[1]);	/* consider all initial														 * rels */			new_rels = make_rels_by_clauseless_joins(root,													 old_rel,													 other_rels);			result_rels = list_concat_unique_ptr(result_rels, new_rels);		}		/*----------		 * When OJs or IN clauses are involved, there may be no legal way		 * to make an N-way join for some values of N.	For example consider		 *		 * SELECT ... FROM t1 WHERE		 *	 x IN (SELECT ... FROM t2,t3 WHERE ...) AND		 *	 y IN (SELECT ... FROM t4,t5 WHERE ...)		 *		 * We will flatten this query to a 5-way join problem, but there are		 * no 4-way joins that join_is_legal() will consider legal.  We have		 * to accept failure at level 4 and go on to discover a workable		 * bushy plan at level 5.		 *		 * However, if there are no such clauses then join_is_legal() should		 * never fail, and so the following sanity check is useful.		 *----------		 */		if (result_rels == NIL &&			root->oj_info_list == NIL && root->in_info_list == NIL)			elog(ERROR, "failed to build any %d-way joins", level);	}	return result_rels;}/* * make_rels_by_clause_joins *	  Build joins between the given relation 'old_rel' and other relations *	  that participate in join clauses that 'old_rel' also participates in *	  (or participate in join-order restrictions with it). *	  The join rel nodes are returned in a list. * * 'old_rel' is the relation entry for the relation to be joined * 'other_rels': the first cell in a linked list containing the other * rels to be considered for joining * * Currently, this is only used with initial rels in other_rels, but it * will work for joining to joinrels too. */static List *make_rels_by_clause_joins(PlannerInfo *root,						  RelOptInfo *old_rel,						  ListCell *other_rels){	List	   *result = NIL;	ListCell   *l;	for_each_cell(l, other_rels)	{		RelOptInfo *other_rel = (RelOptInfo *) lfirst(l);		if (!bms_overlap(old_rel->relids, other_rel->relids) &&			(have_relevant_joinclause(root, old_rel, other_rel) ||			 have_join_order_restriction(root, old_rel, other_rel)))		{			RelOptInfo *jrel;			jrel = make_join_rel(root, old_rel, other_rel);			if (jrel)				result = lcons(jrel, result);		}	}	return result;}/* * make_rels_by_clauseless_joins *	  Given a relation 'old_rel' and a list of other relations *	  'other_rels', create a join relation between 'old_rel' and each *	  member of 'other_rels' that isn't already included in 'old_rel'. *	  The join rel nodes are returned in a list. * * 'old_rel' is the relation entry for the relation to be joined * 'other_rels': the first cell of a linked list containing the * other rels to be considered for joining * * Currently, this is only used with initial rels in other_rels, but it would * work for joining to joinrels too. */static List *make_rels_by_clauseless_joins(PlannerInfo *root,							  RelOptInfo *old_rel,							  ListCell *other_rels){	List	   *result = NIL;	ListCell   *i;	for_each_cell(i, other_rels)	{		RelOptInfo *other_rel = (RelOptInfo *) lfirst(i);		if (!bms_overlap(other_rel->relids, old_rel->relids))		{			RelOptInfo *jrel;			jrel = make_join_rel(root, old_rel, other_rel);			/*			 * As long as given other_rels are distinct, don't need to test to			 * see if jrel is already part of output list.			 */			if (jrel)				result = lcons(jrel, result);		}	}	return result;}/* * join_is_legal *	   Determine whether a proposed join is legal given the query's *	   join order constraints; and if it is, determine the join type. * * Caller must supply not only the two rels, but the union of their relids. * (We could simplify the API by computing joinrelids locally, but this * would be redundant work in the normal path through make_join_rel.) * * On success, *jointype_p is set to the required join type. */static booljoin_is_legal(PlannerInfo *root, RelOptInfo *rel1, RelOptInfo *rel2,			  Relids joinrelids, JoinType *jointype_p){	JoinType	jointype;	bool		is_valid_inner;	ListCell   *l;	/*	 * Ensure *jointype_p is set on failure return.  This is just to suppress	 * uninitialized-variable warnings from overly anal compilers.	 */	*jointype_p = JOIN_INNER;	/*	 * If we have any outer joins, the proposed join might be illegal; and in	 * any case we have to determine its join type.  Scan the OJ list for	 * conflicts.	 */	jointype = JOIN_INNER;		/* default if no match to an OJ */	is_valid_inner = true;	foreach(l, root->oj_info_list)	{		OuterJoinInfo *ojinfo = (OuterJoinInfo *) lfirst(l);		/*		 * This OJ is not relevant unless its RHS overlaps the proposed join.		 * (Check this first as a fast path for dismissing most irrelevant OJs		 * quickly.)		 */		if (!bms_overlap(ojinfo->min_righthand, joinrelids))			continue;		/*		 * Also, not relevant if proposed join is fully contained within RHS		 * (ie, we're still building up the RHS).		 */		if (bms_is_subset(joinrelids, ojinfo->min_righthand))			continue;		/*		 * Also, not relevant if OJ is already done within either input.		 */		if (bms_is_subset(ojinfo->min_lefthand, rel1->relids) &&			bms_is_subset(ojinfo->min_righthand, rel1->relids))			continue;		if (bms_is_subset(ojinfo->min_lefthand, rel2->relids) &&			bms_is_subset(ojinfo->min_righthand, rel2->relids))			continue;		/*		 * If one input contains min_lefthand and the other contains		 * min_righthand, then we can perform the OJ at this join.		 *		 * Barf if we get matches to more than one OJ (is that possible?)		 */		if (bms_is_subset(ojinfo->min_lefthand, rel1->relids) &&			bms_is_subset(ojinfo->min_righthand, rel2->relids))		{			if (jointype != JOIN_INNER)				return false;	/* invalid join path */			jointype = ojinfo->is_full_join ? JOIN_FULL : JOIN_LEFT;		}		else if (bms_is_subset(ojinfo->min_lefthand, rel2->relids) &&				 bms_is_subset(ojinfo->min_righthand, rel1->relids))		{			if (jointype != JOIN_INNER)				return false;	/* invalid join path */			jointype = ojinfo->is_full_join ? JOIN_FULL : JOIN_RIGHT;		}		else		{			/*----------			 * Otherwise, the proposed join overlaps the RHS but isn't			 * a valid implementation of this OJ.  It might still be			 * a legal join, however.  If both inputs overlap the RHS,			 * assume that it's OK.  Since the inputs presumably got past			 * this function's checks previously, they can't overlap the			 * LHS and their violations of the RHS boundary must represent			 * OJs that have been determined to commute with this one.			 * We have to allow this to work correctly in cases like			 *		(a LEFT JOIN (b JOIN (c LEFT JOIN d)))			 * when the c/d join has been determined to commute with the join			 * to a, and hence d is not part of min_righthand for the upper			 * join.  It should be legal to join b to c/d but this will appear			 * as a violation of the upper join's RHS.			 * Furthermore, if one input overlaps the RHS and the other does			 * not, we should still allow the join if it is a valid			 * implementation of some other OJ.  We have to allow this to			 * support the associative identity			 *		(a LJ b on Pab) LJ c ON Pbc = a LJ (b LJ c ON Pbc) on Pab			 * since joining B directly to C violates the lower OJ's RHS.			 * We assume that make_outerjoininfo() set things up correctly			 * so that we'll only match to some OJ if the join is valid.			 * Set flag here to check at bottom of loop.			 *----------			 */			if (bms_overlap(rel1->relids, ojinfo->min_righthand) &&				bms_overlap(rel2->relids, ojinfo->min_righthand))			{				/* seems OK */				Assert(!bms_overlap(joinrelids, ojinfo->min_lefthand));			}			else				is_valid_inner = false;		}	}	/* Fail if violated some OJ's RHS and didn't match to another OJ */	if (jointype == JOIN_INNER && !is_valid_inner)		return false;			/* invalid join path */	/*	 * Similarly, if we are implementing IN clauses as joins, check for	 * illegal join path and detect whether we need a non-default join type.	 */	foreach(l, root->in_info_list)	{		InClauseInfo *ininfo = (InClauseInfo *) lfirst(l);		/*		 * This IN clause is not relevant unless its RHS overlaps the proposed		 * join.  (Check this first as a fast path for dismissing most		 * irrelevant INs quickly.)		 */		if (!bms_overlap(ininfo->righthand, joinrelids))			continue;		/*		 * If we are still building the IN clause's RHS, then this IN clause		 * isn't relevant yet.		 */		if (bms_is_subset(joinrelids, ininfo->righthand))			continue;		/*		 * Cannot join if proposed join contains rels not in the RHS *and*		 * contains only part of the RHS.  We must build the complete RHS		 * (subselect's join) before it can be joined to rels outside the		 * subselect.		 */		if (!bms_is_subset(ininfo->righthand, joinrelids))			return false;		/*		 * At this point we are considering a join of the IN's RHS to some		 * other rel(s).		 *

⌨️ 快捷键说明

复制代码 Ctrl + C
搜索代码 Ctrl + F
全屏模式 F11
切换主题 Ctrl + Shift + D
显示快捷键 ?
增大字号 Ctrl + =
减小字号 Ctrl + -