Branch data Line data Source code
1 : : /*-------------------------------------------------------------------------
2 : : *
3 : : * equivclass.c
4 : : * Routines for managing EquivalenceClasses
5 : : *
6 : : * See src/backend/optimizer/README for discussion of EquivalenceClasses.
7 : : *
8 : : *
9 : : * Portions Copyright (c) 1996-2026, PostgreSQL Global Development Group
10 : : * Portions Copyright (c) 1994, Regents of the University of California
11 : : *
12 : : * IDENTIFICATION
13 : : * src/backend/optimizer/path/equivclass.c
14 : : *
15 : : *-------------------------------------------------------------------------
16 : : */
17 : : #include "postgres.h"
18 : :
19 : : #include <limits.h>
20 : :
21 : : #include "access/stratnum.h"
22 : : #include "catalog/pg_type.h"
23 : : #include "common/hashfn.h"
24 : : #include "nodes/makefuncs.h"
25 : : #include "nodes/nodeFuncs.h"
26 : : #include "optimizer/appendinfo.h"
27 : : #include "optimizer/clauses.h"
28 : : #include "optimizer/optimizer.h"
29 : : #include "optimizer/pathnode.h"
30 : : #include "optimizer/paths.h"
31 : : #include "optimizer/planmain.h"
32 : : #include "optimizer/restrictinfo.h"
33 : : #include "rewrite/rewriteManip.h"
34 : : #include "utils/lsyscache.h"
35 : :
36 : :
37 : : static EquivalenceMember *make_eq_member(EquivalenceClass *ec,
38 : : Expr *expr, Relids relids,
39 : : JoinDomain *jdomain,
40 : : EquivalenceMember *parent,
41 : : Oid datatype);
42 : : static EquivalenceMember *add_eq_member(EquivalenceClass *ec,
43 : : Expr *expr, Relids relids,
44 : : JoinDomain *jdomain,
45 : : Oid datatype);
46 : : static EquivalenceMember *add_child_eq_member(PlannerInfo *root,
47 : : EquivalenceClass *ec,
48 : : int ec_index, Expr *expr,
49 : : Relids relids,
50 : : JoinDomain *jdomain,
51 : : EquivalenceMember *parent_em,
52 : : Oid datatype,
53 : : Index child_relid);
54 : : static void generate_base_implied_equalities_const(PlannerInfo *root,
55 : : EquivalenceClass *ec);
56 : : static void generate_base_implied_equalities_no_const(PlannerInfo *root,
57 : : EquivalenceClass *ec);
58 : : static void generate_base_implied_equalities_broken(PlannerInfo *root,
59 : : EquivalenceClass *ec);
60 : : static List *generate_join_implied_equalities_normal(PlannerInfo *root,
61 : : EquivalenceClass *ec,
62 : : Relids join_relids,
63 : : Relids outer_relids,
64 : : Relids inner_relids);
65 : : static List *generate_join_implied_equalities_broken(PlannerInfo *root,
66 : : EquivalenceClass *ec,
67 : : Relids nominal_join_relids,
68 : : Relids outer_relids,
69 : : Relids nominal_inner_relids,
70 : : RelOptInfo *inner_rel);
71 : : static Oid select_equality_operator(EquivalenceClass *ec,
72 : : Oid lefttype, Oid righttype);
73 : : static RestrictInfo *create_join_clause(PlannerInfo *root,
74 : : EquivalenceClass *ec, Oid opno,
75 : : EquivalenceMember *leftem,
76 : : EquivalenceMember *rightem,
77 : : EquivalenceClass *parent_ec);
78 : : static bool reconsider_outer_join_clause(PlannerInfo *root,
79 : : OuterJoinClauseInfo *ojcinfo,
80 : : bool outer_on_left);
81 : : static bool reconsider_full_join_clause(PlannerInfo *root,
82 : : OuterJoinClauseInfo *ojcinfo);
83 : : static JoinDomain *find_join_domain(PlannerInfo *root, Relids relids);
84 : : static Bitmapset *get_eclass_indexes_for_relids(PlannerInfo *root,
85 : : Relids relids);
86 : : static Bitmapset *get_common_eclass_indexes(PlannerInfo *root, Relids relids1,
87 : : Relids relids2);
88 : : static void ec_build_derives_hash(PlannerInfo *root, EquivalenceClass *ec);
89 : : static void ec_add_derived_clauses(EquivalenceClass *ec, List *clauses);
90 : : static void ec_add_derived_clause(EquivalenceClass *ec, RestrictInfo *clause);
91 : : static void ec_add_clause_to_derives_hash(EquivalenceClass *ec, RestrictInfo *rinfo);
92 : : static void ec_clear_derived_clauses(EquivalenceClass *ec);
93 : : static RestrictInfo *ec_search_clause_for_ems(PlannerInfo *root, EquivalenceClass *ec,
94 : : EquivalenceMember *leftem,
95 : : EquivalenceMember *rightem,
96 : : EquivalenceClass *parent_ec);
97 : : static RestrictInfo *ec_search_derived_clause_for_ems(PlannerInfo *root,
98 : : EquivalenceClass *ec,
99 : : EquivalenceMember *leftem,
100 : : EquivalenceMember *rightem,
101 : : EquivalenceClass *parent_ec);
102 : :
103 : : /*
104 : : * Hash key identifying a derived clause.
105 : : *
106 : : * This structure should not be filled manually. Use fill_ec_derives_key() to
107 : : * set it up in canonical form.
108 : : */
109 : : typedef struct
110 : : {
111 : : EquivalenceMember *em1;
112 : : EquivalenceMember *em2;
113 : : EquivalenceClass *parent_ec;
114 : : } ECDerivesKey;
115 : :
116 : : /* Hash table entry in ec_derives_hash. */
117 : : typedef struct
118 : : {
119 : : uint32 status;
120 : : ECDerivesKey key;
121 : : RestrictInfo *rinfo;
122 : : } ECDerivesEntry;
123 : :
124 : : /* Threshold for switching from list to hash table */
125 : : #define EC_DERIVES_HASH_THRESHOLD 32
126 : :
127 : : #define SH_PREFIX derives
128 : : #define SH_ELEMENT_TYPE ECDerivesEntry
129 : : #define SH_KEY_TYPE ECDerivesKey
130 : : #define SH_KEY key
131 : : #define SH_HASH_KEY(tb, key) \
132 : : hash_bytes((const unsigned char *) &(key), sizeof(ECDerivesKey))
133 : : #define SH_EQUAL(tb, a, b) \
134 : : ((a).em1 == (b).em1 && (a).em2 == (b).em2 && (a).parent_ec == (b).parent_ec)
135 : : #define SH_SCOPE static inline
136 : : #define SH_DECLARE
137 : : #define SH_DEFINE
138 : : #include "lib/simplehash.h"
139 : :
140 : : /*
141 : : * process_equivalence
142 : : * The given clause has a mergejoinable operator and is not an outer-join
143 : : * qualification, so its two sides can be considered equal
144 : : * anywhere they are both computable; moreover that equality can be
145 : : * extended transitively. Record this knowledge in the EquivalenceClass
146 : : * data structure, if applicable. Returns true if successful, false if not
147 : : * (in which case caller should treat the clause as ordinary, not an
148 : : * equivalence).
149 : : *
150 : : * In some cases, although we cannot convert a clause into EquivalenceClass
151 : : * knowledge, we can still modify it to a more useful form than the original.
152 : : * Then, *p_restrictinfo will be replaced by a new RestrictInfo, which is what
153 : : * the caller should use for further processing.
154 : : *
155 : : * jdomain is the join domain within which the given clause was found.
156 : : * This limits the applicability of deductions from the EquivalenceClass,
157 : : * as described in optimizer/README.
158 : : *
159 : : * We reject proposed equivalence clauses if they contain leaky functions
160 : : * and have security_level above zero. The EC evaluation rules require us to
161 : : * apply certain tests at certain joining levels, and we can't tolerate
162 : : * delaying any test on security_level grounds. By rejecting candidate clauses
163 : : * that might require security delays, we ensure it's safe to apply an EC
164 : : * clause as soon as it's supposed to be applied.
165 : : *
166 : : * On success return, we have also initialized the clause's left_ec/right_ec
167 : : * fields to point to the EquivalenceClass representing it. This saves lookup
168 : : * effort later.
169 : : *
170 : : * Note: constructing merged EquivalenceClasses is a standard UNION-FIND
171 : : * problem, for which there exist better data structures than simple lists.
172 : : * If this code ever proves to be a bottleneck then it could be sped up ---
173 : : * but for now, simple is beautiful.
174 : : *
175 : : * Note: this is only called during planner startup, not during GEQO
176 : : * exploration, so we need not worry about whether we're in the right
177 : : * memory context.
178 : : */
179 : : bool
180 : 240493 : process_equivalence(PlannerInfo *root,
181 : : RestrictInfo **p_restrictinfo,
182 : : JoinDomain *jdomain)
183 : : {
184 : 240493 : RestrictInfo *restrictinfo = *p_restrictinfo;
185 : 240493 : Expr *clause = restrictinfo->clause;
186 : : Oid opno,
187 : : collation,
188 : : item1_type,
189 : : item2_type;
190 : : Expr *item1;
191 : : Expr *item2;
192 : : Relids item1_relids,
193 : : item2_relids;
194 : : List *opfamilies;
195 : : EquivalenceClass *ec1,
196 : : *ec2;
197 : : EquivalenceMember *em1,
198 : : *em2;
199 : : ListCell *lc1;
200 : : int ec2_idx;
201 : :
202 : : /* Should not already be marked as having generated an eclass */
203 : : Assert(restrictinfo->left_ec == NULL);
204 : : Assert(restrictinfo->right_ec == NULL);
205 : :
206 : : /* Reject if it is potentially postponable by security considerations */
207 [ + + + + ]: 240493 : if (restrictinfo->security_level > 0 && !restrictinfo->leakproof)
208 : 172 : return false;
209 : :
210 : : /* Extract info from given clause */
211 : : Assert(is_opclause(clause));
212 : 240321 : opno = ((OpExpr *) clause)->opno;
213 : 240321 : collation = ((OpExpr *) clause)->inputcollid;
214 : 240321 : item1 = (Expr *) get_leftop(clause);
215 : 240321 : item2 = (Expr *) get_rightop(clause);
216 : 240321 : item1_relids = restrictinfo->left_relids;
217 : 240321 : item2_relids = restrictinfo->right_relids;
218 : :
219 : : /*
220 : : * Ensure both input expressions expose the desired collation (their types
221 : : * should be OK already); see comments for canonicalize_ec_expression.
222 : : */
223 : 240321 : item1 = canonicalize_ec_expression(item1,
224 : : exprType((Node *) item1),
225 : : collation);
226 : 240321 : item2 = canonicalize_ec_expression(item2,
227 : : exprType((Node *) item2),
228 : : collation);
229 : :
230 : : /*
231 : : * Clauses of the form X=X cannot be translated into EquivalenceClasses.
232 : : * We'd either end up with a single-entry EC, losing the knowledge that
233 : : * the clause was present at all, or else make an EC with duplicate
234 : : * entries, causing other issues.
235 : : */
236 [ + + ]: 240321 : if (equal(item1, item2))
237 : : {
238 : : /*
239 : : * If the operator is strict, then the clause can be treated as just
240 : : * "X IS NOT NULL". (Since we know we are considering a top-level
241 : : * qual, we can ignore the difference between FALSE and NULL results.)
242 : : * It's worth making the conversion because we'll typically get a much
243 : : * better selectivity estimate than we would for X=X.
244 : : *
245 : : * If the operator is not strict, we can't be sure what it will do
246 : : * with NULLs, so don't attempt to optimize it.
247 : : */
248 : 45 : set_opfuncid((OpExpr *) clause);
249 [ + - ]: 45 : if (func_strict(((OpExpr *) clause)->opfuncid))
250 : : {
251 : 45 : NullTest *ntest = makeNode(NullTest);
252 : :
253 : 45 : ntest->arg = item1;
254 : 45 : ntest->nulltesttype = IS_NOT_NULL;
255 : 45 : ntest->argisrow = false; /* correct even if composite arg */
256 : 45 : ntest->location = -1;
257 : :
258 : 45 : *p_restrictinfo =
259 : 45 : make_restrictinfo(root,
260 : : (Expr *) ntest,
261 : 45 : restrictinfo->is_pushed_down,
262 : 45 : restrictinfo->has_clone,
263 : 45 : restrictinfo->is_clone,
264 : 45 : restrictinfo->pseudoconstant,
265 : : restrictinfo->security_level,
266 : : NULL,
267 : : restrictinfo->incompatible_relids,
268 : : restrictinfo->outer_relids);
269 : : }
270 : 45 : return false;
271 : : }
272 : :
273 : : /*
274 : : * We use the declared input types of the operator, not exprType() of the
275 : : * inputs, as the nominal datatypes for opfamily lookup. This presumes
276 : : * that btree operators are always registered with amoplefttype and
277 : : * amoprighttype equal to their declared input types. We will need this
278 : : * info anyway to build EquivalenceMember nodes, and by extracting it now
279 : : * we can use type comparisons to short-circuit some equal() tests.
280 : : */
281 : 240276 : op_input_types(opno, &item1_type, &item2_type);
282 : :
283 : 240276 : opfamilies = restrictinfo->mergeopfamilies;
284 : :
285 : : /*
286 : : * Sweep through the existing EquivalenceClasses looking for matches to
287 : : * item1 and item2. These are the possible outcomes:
288 : : *
289 : : * 1. We find both in the same EC. The equivalence is already known, so
290 : : * there's nothing to do.
291 : : *
292 : : * 2. We find both in different ECs. Merge the two ECs together.
293 : : *
294 : : * 3. We find just one. Add the other to its EC.
295 : : *
296 : : * 4. We find neither. Make a new, two-entry EC.
297 : : *
298 : : * Note: since all ECs are built through this process or the similar
299 : : * search in get_eclass_for_sort_expr(), it's impossible that we'd match
300 : : * an item in more than one existing nonvolatile EC. So it's okay to stop
301 : : * at the first match.
302 : : */
303 : 240276 : ec1 = ec2 = NULL;
304 : 240276 : em1 = em2 = NULL;
305 : 240276 : ec2_idx = -1;
306 [ + + + + : 412706 : foreach(lc1, root->eq_classes)
+ + ]
307 : : {
308 : 172508 : EquivalenceClass *cur_ec = (EquivalenceClass *) lfirst(lc1);
309 : : ListCell *lc2;
310 : :
311 : : /* Never match to a volatile EC */
312 [ - + ]: 172508 : if (cur_ec->ec_has_volatile)
313 : 0 : continue;
314 : :
315 : : /*
316 : : * The collation has to match; check this first since it's cheaper
317 : : * than the opfamily comparison.
318 : : */
319 [ + + ]: 172508 : if (collation != cur_ec->ec_collation)
320 : 17250 : continue;
321 : :
322 : : /*
323 : : * A "match" requires matching sets of btree opfamilies. Use of
324 : : * equal() for this test has implications discussed in the comments
325 : : * for get_mergejoin_opfamilies().
326 : : */
327 [ + + ]: 155258 : if (!equal(opfamilies, cur_ec->ec_opfamilies))
328 : 42053 : continue;
329 : :
330 : : /* We don't expect any children yet */
331 : : Assert(cur_ec->ec_childmembers == NULL);
332 : :
333 [ + - + + : 336804 : foreach(lc2, cur_ec->ec_members)
+ + ]
334 : : {
335 : 223677 : EquivalenceMember *cur_em = (EquivalenceMember *) lfirst(lc2);
336 : :
337 : : /* Child members should not exist in ec_members */
338 : : Assert(!cur_em->em_is_child);
339 : :
340 : : /*
341 : : * Match constants only within the same JoinDomain (see
342 : : * optimizer/README).
343 : : */
344 [ + + + + ]: 223677 : if (cur_em->em_is_const && cur_em->em_jdomain != jdomain)
345 : 3019 : continue;
346 : :
347 [ + + ]: 220658 : if (!ec1 &&
348 [ + + + + ]: 424787 : item1_type == cur_em->em_datatype &&
349 : 212241 : equal(item1, cur_em->em_expr))
350 : : {
351 : 13000 : ec1 = cur_ec;
352 : 13000 : em1 = cur_em;
353 [ + + ]: 13000 : if (ec2)
354 : 60 : break;
355 : : }
356 : :
357 [ + + ]: 220598 : if (!ec2 &&
358 [ + + + + ]: 436696 : item2_type == cur_em->em_datatype &&
359 : 218042 : equal(item2, cur_em->em_expr))
360 : : {
361 : 2961 : ec2 = cur_ec;
362 : 2961 : ec2_idx = foreach_current_index(lc1);
363 : 2961 : em2 = cur_em;
364 [ + + ]: 2961 : if (ec1)
365 : 18 : break;
366 : : }
367 : : }
368 : :
369 [ + + + + ]: 113205 : if (ec1 && ec2)
370 : 78 : break;
371 : : }
372 : :
373 : : /* Sweep finished, what did we find? */
374 : :
375 [ + + + + ]: 240276 : if (ec1 && ec2)
376 : : {
377 : : /* If case 1, nothing to do, except add to sources */
378 [ + + ]: 78 : if (ec1 == ec2)
379 : : {
380 : 45 : ec1->ec_sources = lappend(ec1->ec_sources, restrictinfo);
381 : 45 : ec1->ec_min_security = Min(ec1->ec_min_security,
382 : : restrictinfo->security_level);
383 : 45 : ec1->ec_max_security = Max(ec1->ec_max_security,
384 : : restrictinfo->security_level);
385 : : /* mark the RI as associated with this eclass */
386 : 45 : restrictinfo->left_ec = ec1;
387 : 45 : restrictinfo->right_ec = ec1;
388 : : /* mark the RI as usable with this pair of EMs */
389 : 45 : restrictinfo->left_em = em1;
390 : 45 : restrictinfo->right_em = em2;
391 : 45 : return true;
392 : : }
393 : :
394 : : /*
395 : : * Case 2: need to merge ec1 and ec2. This should never happen after
396 : : * the ECs have reached canonical state; otherwise, pathkeys could be
397 : : * rendered non-canonical by the merge, and relation eclass indexes
398 : : * would get broken by removal of an eq_classes list entry.
399 : : */
400 [ - + ]: 33 : if (root->ec_merging_done)
401 [ # # ]: 0 : elog(ERROR, "too late to merge equivalence classes");
402 : :
403 : : /*
404 : : * We add ec2's items to ec1, then set ec2's ec_merged link to point
405 : : * to ec1 and remove ec2 from the eq_classes list. We cannot simply
406 : : * delete ec2 because that could leave dangling pointers in existing
407 : : * PathKeys. We leave it behind with a link so that the merged EC can
408 : : * be found.
409 : : */
410 : 33 : ec1->ec_members = list_concat(ec1->ec_members, ec2->ec_members);
411 : 33 : ec1->ec_sources = list_concat(ec1->ec_sources, ec2->ec_sources);
412 : :
413 : : /*
414 : : * Appends ec2's derived clauses to ec1->ec_derives_list and adds them
415 : : * to ec1->ec_derives_hash if present.
416 : : */
417 : 33 : ec_add_derived_clauses(ec1, ec2->ec_derives_list);
418 : 33 : ec1->ec_relids = bms_join(ec1->ec_relids, ec2->ec_relids);
419 : 33 : ec1->ec_has_const |= ec2->ec_has_const;
420 : : /* can't need to set has_volatile */
421 : 33 : ec1->ec_min_security = Min(ec1->ec_min_security,
422 : : ec2->ec_min_security);
423 : 33 : ec1->ec_max_security = Max(ec1->ec_max_security,
424 : : ec2->ec_max_security);
425 : 33 : ec2->ec_merged = ec1;
426 : 33 : root->eq_classes = list_delete_nth_cell(root->eq_classes, ec2_idx);
427 : : /* just to avoid debugging confusion w/ dangling pointers: */
428 : 33 : ec2->ec_members = NIL;
429 : 33 : ec2->ec_sources = NIL;
430 : 33 : ec_clear_derived_clauses(ec2);
431 : 33 : ec2->ec_relids = NULL;
432 : 33 : ec1->ec_sources = lappend(ec1->ec_sources, restrictinfo);
433 : 33 : ec1->ec_min_security = Min(ec1->ec_min_security,
434 : : restrictinfo->security_level);
435 : 33 : ec1->ec_max_security = Max(ec1->ec_max_security,
436 : : restrictinfo->security_level);
437 : : /* mark the RI as associated with this eclass */
438 : 33 : restrictinfo->left_ec = ec1;
439 : 33 : restrictinfo->right_ec = ec1;
440 : : /* mark the RI as usable with this pair of EMs */
441 : 33 : restrictinfo->left_em = em1;
442 : 33 : restrictinfo->right_em = em2;
443 : : }
444 [ + + ]: 240198 : else if (ec1)
445 : : {
446 : : /* Case 3: add item2 to ec1 */
447 : 12922 : em2 = add_eq_member(ec1, item2, item2_relids,
448 : : jdomain, item2_type);
449 : 12922 : ec1->ec_sources = lappend(ec1->ec_sources, restrictinfo);
450 : 12922 : ec1->ec_min_security = Min(ec1->ec_min_security,
451 : : restrictinfo->security_level);
452 : 12922 : ec1->ec_max_security = Max(ec1->ec_max_security,
453 : : restrictinfo->security_level);
454 : : /* mark the RI as associated with this eclass */
455 : 12922 : restrictinfo->left_ec = ec1;
456 : 12922 : restrictinfo->right_ec = ec1;
457 : : /* mark the RI as usable with this pair of EMs */
458 : 12922 : restrictinfo->left_em = em1;
459 : 12922 : restrictinfo->right_em = em2;
460 : : }
461 [ + + ]: 227276 : else if (ec2)
462 : : {
463 : : /* Case 3: add item1 to ec2 */
464 : 2883 : em1 = add_eq_member(ec2, item1, item1_relids,
465 : : jdomain, item1_type);
466 : 2883 : ec2->ec_sources = lappend(ec2->ec_sources, restrictinfo);
467 : 2883 : ec2->ec_min_security = Min(ec2->ec_min_security,
468 : : restrictinfo->security_level);
469 : 2883 : ec2->ec_max_security = Max(ec2->ec_max_security,
470 : : restrictinfo->security_level);
471 : : /* mark the RI as associated with this eclass */
472 : 2883 : restrictinfo->left_ec = ec2;
473 : 2883 : restrictinfo->right_ec = ec2;
474 : : /* mark the RI as usable with this pair of EMs */
475 : 2883 : restrictinfo->left_em = em1;
476 : 2883 : restrictinfo->right_em = em2;
477 : : }
478 : : else
479 : : {
480 : : /* Case 4: make a new, two-entry EC */
481 : 224393 : EquivalenceClass *ec = makeNode(EquivalenceClass);
482 : :
483 : 224393 : ec->ec_opfamilies = opfamilies;
484 : 224393 : ec->ec_collation = collation;
485 : 224393 : ec->ec_childmembers_size = 0;
486 : 224393 : ec->ec_members = NIL;
487 : 224393 : ec->ec_childmembers = NULL;
488 : 224393 : ec->ec_sources = list_make1(restrictinfo);
489 : 224393 : ec->ec_derives_list = NIL;
490 : 224393 : ec->ec_derives_hash = NULL;
491 : 224393 : ec->ec_relids = NULL;
492 : 224393 : ec->ec_has_const = false;
493 : 224393 : ec->ec_has_volatile = false;
494 : 224393 : ec->ec_broken = false;
495 : 224393 : ec->ec_sortref = 0;
496 : 224393 : ec->ec_min_security = restrictinfo->security_level;
497 : 224393 : ec->ec_max_security = restrictinfo->security_level;
498 : 224393 : ec->ec_merged = NULL;
499 : 224393 : em1 = add_eq_member(ec, item1, item1_relids,
500 : : jdomain, item1_type);
501 : 224393 : em2 = add_eq_member(ec, item2, item2_relids,
502 : : jdomain, item2_type);
503 : :
504 : 224393 : root->eq_classes = lappend(root->eq_classes, ec);
505 : :
506 : : /* mark the RI as associated with this eclass */
507 : 224393 : restrictinfo->left_ec = ec;
508 : 224393 : restrictinfo->right_ec = ec;
509 : : /* mark the RI as usable with this pair of EMs */
510 : 224393 : restrictinfo->left_em = em1;
511 : 224393 : restrictinfo->right_em = em2;
512 : : }
513 : :
514 : 240231 : return true;
515 : : }
516 : :
517 : : /*
518 : : * canonicalize_ec_expression
519 : : *
520 : : * This function ensures that the expression exposes the expected type and
521 : : * collation, so that it will be equal() to other equivalence-class expressions
522 : : * that it ought to be equal() to.
523 : : *
524 : : * The rule for datatypes is that the exposed type should match what it would
525 : : * be for an input to an operator of the EC's opfamilies; which is usually
526 : : * the declared input type of the operator, but in the case of polymorphic
527 : : * operators no relabeling is wanted (compare the behavior of parse_coerce.c).
528 : : * Expressions coming in from quals will generally have the right type
529 : : * already, but expressions coming from indexkeys may not (because they are
530 : : * represented without any explicit relabel in pg_index), and the same problem
531 : : * occurs for sort expressions (because the parser is likewise cavalier about
532 : : * putting relabels on them). Such cases will be binary-compatible with the
533 : : * real operators, so adding a RelabelType is sufficient.
534 : : *
535 : : * Also, the expression's exposed collation must match the EC's collation.
536 : : * This is important because in comparisons like "foo < bar COLLATE baz",
537 : : * only one of the expressions has the correct exposed collation as we receive
538 : : * it from the parser. Forcing both of them to have it ensures that all
539 : : * variant spellings of such a construct behave the same. Again, we can
540 : : * stick on a RelabelType to force the right exposed collation. (It might
541 : : * work to not label the collation at all in EC members, but this is risky
542 : : * since some parts of the system expect exprCollation() to deliver the
543 : : * right answer for a sort key.)
544 : : */
545 : : Expr *
546 : 2349391 : canonicalize_ec_expression(Expr *expr, Oid req_type, Oid req_collation)
547 : : {
548 : 2349391 : Oid expr_type = exprType((Node *) expr);
549 : :
550 : : /*
551 : : * For a polymorphic-input-type opclass, just keep the same exposed type.
552 : : * RECORD opclasses work like polymorphic-type ones for this purpose.
553 : : */
554 [ + - + + : 2349391 : if (IsPolymorphicType(req_type) || req_type == RECORDOID)
+ - + + +
+ + + + -
+ - + - +
- + - +
+ ]
555 : 7893 : req_type = expr_type;
556 : :
557 : : /*
558 : : * No work if the expression exposes the right type/collation already.
559 : : */
560 [ + + + + ]: 4636593 : if (expr_type != req_type ||
561 : 2287202 : exprCollation((Node *) expr) != req_collation)
562 : : {
563 : : /*
564 : : * If we have to change the type of the expression, set typmod to -1,
565 : : * since the new type may not have the same typmod interpretation.
566 : : * When we only have to change collation, preserve the exposed typmod.
567 : : */
568 : : int32 req_typmod;
569 : :
570 [ + + ]: 64128 : if (expr_type != req_type)
571 : 62189 : req_typmod = -1;
572 : : else
573 : 1939 : req_typmod = exprTypmod((Node *) expr);
574 : :
575 : : /*
576 : : * Use applyRelabelType so that we preserve const-flatness. This is
577 : : * important since eval_const_expressions has already been applied.
578 : : */
579 : 64128 : expr = (Expr *) applyRelabelType((Node *) expr,
580 : : req_type, req_typmod, req_collation,
581 : : COERCE_IMPLICIT_CAST, -1, false);
582 : : }
583 : :
584 : 2349391 : return expr;
585 : : }
586 : :
587 : : /*
588 : : * make_eq_member
589 : : * Build a new EquivalenceMember without adding it to an EC. If 'parent'
590 : : * is NULL, the result will be a parent member, otherwise a child member.
591 : : */
592 : : static EquivalenceMember *
593 : 768012 : make_eq_member(EquivalenceClass *ec, Expr *expr, Relids relids,
594 : : JoinDomain *jdomain, EquivalenceMember *parent, Oid datatype)
595 : : {
596 : 768012 : EquivalenceMember *em = makeNode(EquivalenceMember);
597 : :
598 : 768012 : em->em_expr = expr;
599 : 768012 : em->em_relids = relids;
600 : 768012 : em->em_is_const = false;
601 : 768012 : em->em_is_child = (parent != NULL);
602 : 768012 : em->em_datatype = datatype;
603 : 768012 : em->em_jdomain = jdomain;
604 : 768012 : em->em_parent = parent;
605 : :
606 [ + + ]: 768012 : if (bms_is_empty(relids))
607 : : {
608 : : /*
609 : : * No Vars, assume it's a pseudoconstant. This is correct for entries
610 : : * generated from process_equivalence(), because a WHERE clause can't
611 : : * contain aggregates or SRFs, and non-volatility was checked before
612 : : * process_equivalence() ever got called. But
613 : : * get_eclass_for_sort_expr() has to work harder. We put the tests
614 : : * there not here to save cycles in the equivalence case.
615 : : */
616 : : Assert(!parent);
617 : 171964 : em->em_is_const = true;
618 : 171964 : ec->ec_has_const = true;
619 : : /* it can't affect ec_relids */
620 : : }
621 : :
622 : 768012 : return em;
623 : : }
624 : :
625 : : /*
626 : : * add_eq_member - build a new non-child EquivalenceMember and add it to 'ec'.
627 : : */
628 : : static EquivalenceMember *
629 : 685632 : add_eq_member(EquivalenceClass *ec, Expr *expr, Relids relids,
630 : : JoinDomain *jdomain, Oid datatype)
631 : : {
632 : 685632 : EquivalenceMember *em = make_eq_member(ec, expr, relids, jdomain,
633 : : NULL, datatype);
634 : :
635 : : /* add to the members list */
636 : 685632 : ec->ec_members = lappend(ec->ec_members, em);
637 : :
638 : : /* record the relids for parent members */
639 : 685632 : ec->ec_relids = bms_add_members(ec->ec_relids, relids);
640 : :
641 : 685632 : return em;
642 : : }
643 : :
644 : : /*
645 : : * add_child_eq_member
646 : : * Create an em_is_child=true EquivalenceMember and add it to 'ec'.
647 : : *
648 : : * 'root' is the PlannerInfo that 'ec' belongs to.
649 : : * 'ec' is the EquivalenceClass to add the child member to.
650 : : * 'ec_index' the index of 'ec' within root->eq_classes, or -1 if maintaining
651 : : * the RelOptInfo.eclass_indexes isn't needed.
652 : : * 'expr' is the em_expr for the new member.
653 : : * 'relids' is the 'em_relids' for the new member.
654 : : * 'jdomain' is the 'em_jdomain' for the new member.
655 : : * 'parent_em' is the parent member of the child to create.
656 : : * 'datatype' is the em_datatype of the new member.
657 : : * 'child_relid' defines which element of ec_childmembers to add this member
658 : : * to. This is generally a RELOPT_OTHER_MEMBER_REL, but for set operations
659 : : * can be a RELOPT_BASEREL representing the set-op children.
660 : : */
661 : : static EquivalenceMember *
662 : 82380 : add_child_eq_member(PlannerInfo *root, EquivalenceClass *ec, int ec_index,
663 : : Expr *expr, Relids relids, JoinDomain *jdomain,
664 : : EquivalenceMember *parent_em, Oid datatype,
665 : : Index child_relid)
666 : : {
667 : : EquivalenceMember *em;
668 : :
669 : : Assert(parent_em != NULL);
670 : :
671 : : /*
672 : : * Allocate the array to store child members; an array of Lists indexed by
673 : : * relid, or expand the existing one, if necessary.
674 : : */
675 [ + + ]: 82380 : if (unlikely(ec->ec_childmembers_size < root->simple_rel_array_size))
676 : : {
677 [ + - ]: 23123 : if (ec->ec_childmembers == NULL)
678 : 23123 : ec->ec_childmembers = palloc0_array(List *, root->simple_rel_array_size);
679 : : else
680 : 0 : ec->ec_childmembers = repalloc0_array(ec->ec_childmembers, List *,
681 : : ec->ec_childmembers_size,
682 : : root->simple_rel_array_size);
683 : :
684 : 23123 : ec->ec_childmembers_size = root->simple_rel_array_size;
685 : : }
686 : :
687 : 82380 : em = make_eq_member(ec, expr, relids, jdomain, parent_em, datatype);
688 : :
689 : : /* add member to the ec_childmembers List for the given child_relid */
690 : 82380 : ec->ec_childmembers[child_relid] = lappend(ec->ec_childmembers[child_relid], em);
691 : :
692 : : /* Record this EC index for the child rel */
693 [ + + ]: 82380 : if (ec_index >= 0)
694 : : {
695 : 49442 : RelOptInfo *child_rel = root->simple_rel_array[child_relid];
696 : :
697 : 49442 : child_rel->eclass_indexes =
698 : 49442 : bms_add_member(child_rel->eclass_indexes, ec_index);
699 : : }
700 : :
701 : 82380 : return em;
702 : : }
703 : :
704 : :
705 : : /*
706 : : * get_eclass_for_sort_expr
707 : : * Given an expression and opfamily/collation info, find an existing
708 : : * equivalence class it is a member of; if none, optionally build a new
709 : : * single-member EquivalenceClass for it.
710 : : *
711 : : * sortref is the SortGroupRef of the originating SortGroupClause, if any,
712 : : * or zero if not. (It should never be zero if the expression is volatile!)
713 : : *
714 : : * If rel is not NULL, it identifies a specific relation we're considering
715 : : * a path for, and indicates that child EC members for that relation can be
716 : : * considered. Otherwise child members are ignored. (Note: since child EC
717 : : * members aren't guaranteed unique, a non-NULL value means that there could
718 : : * be more than one EC that matches the expression; if so it's order-dependent
719 : : * which one you get. This is annoying but it only happens in corner cases,
720 : : * so for now we live with just reporting the first match. See also
721 : : * generate_implied_equalities_for_column and match_pathkeys_to_index.)
722 : : *
723 : : * If create_it is true, we'll build a new EquivalenceClass when there is no
724 : : * match. If create_it is false, we just return NULL when no match.
725 : : *
726 : : * This can be used safely both before and after EquivalenceClass merging;
727 : : * since it never causes merging it does not invalidate any existing ECs
728 : : * or PathKeys. However, ECs added after path generation has begun are
729 : : * of limited usefulness, so usually it's best to create them beforehand.
730 : : *
731 : : * Note: opfamilies must be chosen consistently with the way
732 : : * process_equivalence() would do; that is, generated from a mergejoinable
733 : : * equality operator. Else we might fail to detect valid equivalences,
734 : : * generating poor (but not incorrect) plans.
735 : : */
736 : : EquivalenceClass *
737 : 1736694 : get_eclass_for_sort_expr(PlannerInfo *root,
738 : : Expr *expr,
739 : : List *opfamilies,
740 : : Oid opcintype,
741 : : Oid collation,
742 : : Index sortref,
743 : : Relids relids,
744 : : bool create_it)
745 : : {
746 : : JoinDomain *jdomain;
747 : : Relids expr_relids;
748 : : EquivalenceClass *newec;
749 : : EquivalenceMember *newem;
750 : : ListCell *lc1;
751 : : MemoryContext oldcontext;
752 : :
753 : : /*
754 : : * Ensure the expression exposes the correct type and collation.
755 : : */
756 : 1736694 : expr = canonicalize_ec_expression(expr, opcintype, collation);
757 : :
758 : : /*
759 : : * Since SortGroupClause nodes are top-level expressions (GROUP BY, ORDER
760 : : * BY, etc), they can be presumed to belong to the top JoinDomain.
761 : : */
762 : 1736694 : jdomain = linitial_node(JoinDomain, root->join_domains);
763 : :
764 : : /*
765 : : * Scan through the existing EquivalenceClasses for a match
766 : : */
767 [ + + + + : 5685944 : foreach(lc1, root->eq_classes)
+ + ]
768 : : {
769 : 4954969 : EquivalenceClass *cur_ec = (EquivalenceClass *) lfirst(lc1);
770 : : EquivalenceMemberIterator it;
771 : : EquivalenceMember *cur_em;
772 : :
773 : : /*
774 : : * Never match to a volatile EC, except when we are looking at another
775 : : * reference to the same volatile SortGroupClause.
776 : : */
777 [ + + + + ]: 4954969 : if (cur_ec->ec_has_volatile &&
778 [ + + ]: 28 : (sortref == 0 || sortref != cur_ec->ec_sortref))
779 : 2101292 : continue;
780 : :
781 [ + + ]: 4954459 : if (collation != cur_ec->ec_collation)
782 : 1388732 : continue;
783 [ + + ]: 3565727 : if (!equal(opfamilies, cur_ec->ec_opfamilies))
784 : 712050 : continue;
785 : :
786 : 2853677 : setup_eclass_member_iterator(&it, cur_ec, relids);
787 [ + + ]: 6339680 : while ((cur_em = eclass_member_iterator_next(&it)) != NULL)
788 : : {
789 : : /*
790 : : * Ignore child members unless they match the request.
791 : : */
792 [ + + ]: 4491722 : if (cur_em->em_is_child &&
793 [ - + ]: 85706 : !bms_equal(cur_em->em_relids, relids))
794 : 0 : continue;
795 : :
796 : : /*
797 : : * Match constants only within the same JoinDomain (see
798 : : * optimizer/README).
799 : : */
800 [ + + + + ]: 4491722 : if (cur_em->em_is_const && cur_em->em_jdomain != jdomain)
801 : 57312 : continue;
802 : :
803 [ + + + + ]: 8835090 : if (opcintype == cur_em->em_datatype &&
804 : 4400680 : equal(expr, cur_em->em_expr))
805 : 1005719 : return cur_ec; /* Match! */
806 : : }
807 : : }
808 : :
809 : : /* No match; does caller want a NULL result? */
810 [ + + ]: 730975 : if (!create_it)
811 : 509934 : return NULL;
812 : :
813 : : /*
814 : : * OK, build a new single-member EC
815 : : *
816 : : * Here, we must be sure that we construct the EC in the right context.
817 : : */
818 : 221041 : oldcontext = MemoryContextSwitchTo(root->planner_cxt);
819 : :
820 : 221041 : newec = makeNode(EquivalenceClass);
821 : 221041 : newec->ec_opfamilies = list_copy(opfamilies);
822 : 221041 : newec->ec_collation = collation;
823 : 221041 : newec->ec_childmembers_size = 0;
824 : 221041 : newec->ec_members = NIL;
825 : 221041 : newec->ec_childmembers = NULL;
826 : 221041 : newec->ec_sources = NIL;
827 : 221041 : newec->ec_derives_list = NIL;
828 : 221041 : newec->ec_derives_hash = NULL;
829 : 221041 : newec->ec_relids = NULL;
830 : 221041 : newec->ec_has_const = false;
831 : 221041 : newec->ec_has_volatile = contain_volatile_functions((Node *) expr);
832 : 221041 : newec->ec_broken = false;
833 : 221041 : newec->ec_sortref = sortref;
834 : 221041 : newec->ec_min_security = UINT_MAX;
835 : 221041 : newec->ec_max_security = 0;
836 : 221041 : newec->ec_merged = NULL;
837 : :
838 [ + + - + ]: 221041 : if (newec->ec_has_volatile && sortref == 0) /* should not happen */
839 [ # # ]: 0 : elog(ERROR, "volatile EquivalenceClass has no sortref");
840 : :
841 : : /*
842 : : * Get the precise set of relids appearing in the expression.
843 : : */
844 : 221041 : expr_relids = pull_varnos(root, (Node *) expr);
845 : :
846 : 221041 : newem = add_eq_member(newec, copyObject(expr), expr_relids,
847 : : jdomain, opcintype);
848 : :
849 : : /*
850 : : * add_eq_member doesn't check for volatile functions, set-returning
851 : : * functions, aggregates, or window functions, but such could appear in
852 : : * sort expressions; so we have to check whether its const-marking was
853 : : * correct.
854 : : */
855 [ + + ]: 221041 : if (newec->ec_has_const)
856 : : {
857 [ + + + + ]: 15312 : if (newec->ec_has_volatile ||
858 [ + + ]: 15084 : expression_returns_set((Node *) expr) ||
859 [ + + ]: 14813 : contain_agg_clause((Node *) expr) ||
860 : 7331 : contain_window_function((Node *) expr))
861 : : {
862 : 384 : newec->ec_has_const = false;
863 : 384 : newem->em_is_const = false;
864 : : }
865 : : }
866 : :
867 : 221041 : root->eq_classes = lappend(root->eq_classes, newec);
868 : :
869 : : /*
870 : : * If EC merging is already complete, we have to mop up by adding the new
871 : : * EC to the eclass_indexes of the relation(s) mentioned in it.
872 : : */
873 [ + + ]: 221041 : if (root->ec_merging_done)
874 : : {
875 : 127735 : int ec_index = list_length(root->eq_classes) - 1;
876 : 127735 : int i = -1;
877 : :
878 [ + + ]: 246348 : while ((i = bms_next_member(newec->ec_relids, i)) > 0)
879 : : {
880 : 118613 : RelOptInfo *rel = root->simple_rel_array[i];
881 : :
882 : : /* ignore the RTE_GROUP RTE */
883 [ + + ]: 118613 : if (i == root->group_rtindex)
884 : 584 : continue;
885 : :
886 [ + + ]: 118029 : if (rel == NULL) /* must be an outer join */
887 : : {
888 : : Assert(bms_is_member(i, root->outer_join_rels));
889 : 5385 : continue;
890 : : }
891 : :
892 : : Assert(rel->reloptkind == RELOPT_BASEREL);
893 : :
894 : 112644 : rel->eclass_indexes = bms_add_member(rel->eclass_indexes,
895 : : ec_index);
896 : : }
897 : : }
898 : :
899 : 221041 : MemoryContextSwitchTo(oldcontext);
900 : :
901 : 221041 : return newec;
902 : : }
903 : :
904 : : /*
905 : : * find_ec_member_matching_expr
906 : : * Locate an EquivalenceClass member matching the given expr, if any;
907 : : * return NULL if no match.
908 : : *
909 : : * "Matching" is defined as "equal after stripping RelabelTypes".
910 : : * This is used for identifying sort expressions, and we need to allow
911 : : * binary-compatible relabeling for some cases involving binary-compatible
912 : : * sort operators.
913 : : *
914 : : * Child EC members are ignored unless they belong to given 'relids'.
915 : : */
916 : : EquivalenceMember *
917 : 263120 : find_ec_member_matching_expr(EquivalenceClass *ec,
918 : : Expr *expr,
919 : : Relids relids)
920 : : {
921 : : EquivalenceMemberIterator it;
922 : : EquivalenceMember *em;
923 : :
924 : : /* We ignore binary-compatible relabeling on both ends */
925 [ + - + + ]: 284685 : while (expr && IsA(expr, RelabelType))
926 : 21565 : expr = ((RelabelType *) expr)->arg;
927 : :
928 : 263120 : setup_eclass_member_iterator(&it, ec, relids);
929 [ + + ]: 440492 : while ((em = eclass_member_iterator_next(&it)) != NULL)
930 : : {
931 : : Expr *emexpr;
932 : :
933 : : /*
934 : : * We shouldn't be trying to sort by an equivalence class that
935 : : * contains a constant, so no need to consider such cases any further.
936 : : */
937 [ - + ]: 292150 : if (em->em_is_const)
938 : 0 : continue;
939 : :
940 : : /*
941 : : * Ignore child members unless they belong to the requested rel.
942 : : */
943 [ + + ]: 292150 : if (em->em_is_child &&
944 [ + + ]: 10055 : !bms_is_subset(em->em_relids, relids))
945 : 3396 : continue;
946 : :
947 : : /*
948 : : * Match if same expression (after stripping relabel).
949 : : */
950 : 288754 : emexpr = em->em_expr;
951 [ + - + + ]: 293717 : while (emexpr && IsA(emexpr, RelabelType))
952 : 4963 : emexpr = ((RelabelType *) emexpr)->arg;
953 : :
954 [ + + ]: 288754 : if (equal(emexpr, expr))
955 : 114778 : return em;
956 : : }
957 : :
958 : 148342 : return NULL;
959 : : }
960 : :
961 : : /*
962 : : * find_computable_ec_member
963 : : * Locate an EquivalenceClass member that can be computed from the
964 : : * expressions appearing in "exprs"; return NULL if no match.
965 : : *
966 : : * "exprs" can be either a list of bare expression trees, or a list of
967 : : * TargetEntry nodes. Typically it will contain Vars and possibly Aggrefs
968 : : * and WindowFuncs; however, when considering an appendrel member the list
969 : : * could contain arbitrary expressions. We consider an EC member to be
970 : : * computable if all the Vars, PlaceHolderVars, Aggrefs, and WindowFuncs
971 : : * it needs are present in "exprs".
972 : : *
973 : : * There is some subtlety in that definition: for example, if an EC member is
974 : : * Var_A + 1 while what is in "exprs" is Var_A + 2, it's still computable.
975 : : * This works because in the final plan tree, the EC member's expression will
976 : : * be computed as part of the same plan node targetlist that is currently
977 : : * represented by "exprs". So if we have Var_A available for the existing
978 : : * tlist member, it must be OK to use it in the EC expression too.
979 : : *
980 : : * Unlike find_ec_member_matching_expr, there's no special provision here
981 : : * for binary-compatible relabeling. This is intentional: if we have to
982 : : * compute an expression in this way, setrefs.c is going to insist on exact
983 : : * matches of Vars to the source tlist.
984 : : *
985 : : * Child EC members are ignored unless they belong to given 'relids'.
986 : : * Also, non-parallel-safe expressions are ignored if 'require_parallel_safe'.
987 : : *
988 : : * Note: some callers pass root == NULL for notational reasons. This is OK
989 : : * when require_parallel_safe is false.
990 : : */
991 : : EquivalenceMember *
992 : 6927 : find_computable_ec_member(PlannerInfo *root,
993 : : EquivalenceClass *ec,
994 : : List *exprs,
995 : : Relids relids,
996 : : bool require_parallel_safe)
997 : : {
998 : : List *exprvars;
999 : : EquivalenceMemberIterator it;
1000 : : EquivalenceMember *em;
1001 : :
1002 : : /*
1003 : : * Pull out the Vars and quasi-Vars present in "exprs". In the typical
1004 : : * non-appendrel case, this is just another representation of the same
1005 : : * list. However, it does remove the distinction between the case of a
1006 : : * list of plain expressions and a list of TargetEntrys.
1007 : : */
1008 : 6927 : exprvars = pull_var_clause((Node *) exprs,
1009 : : PVC_INCLUDE_AGGREGATES |
1010 : : PVC_INCLUDE_WINDOWFUNCS |
1011 : : PVC_INCLUDE_PLACEHOLDERS);
1012 : :
1013 : 6927 : setup_eclass_member_iterator(&it, ec, relids);
1014 [ + + ]: 13952 : while ((em = eclass_member_iterator_next(&it)) != NULL)
1015 : : {
1016 : : List *emvars;
1017 : : ListCell *lc2;
1018 : :
1019 : : /*
1020 : : * We shouldn't be trying to sort by an equivalence class that
1021 : : * contains a constant, so no need to consider such cases any further.
1022 : : */
1023 [ - + ]: 7415 : if (em->em_is_const)
1024 : 0 : continue;
1025 : :
1026 : : /*
1027 : : * Ignore child members unless they belong to the requested rel.
1028 : : */
1029 [ + + ]: 7415 : if (em->em_is_child &&
1030 [ + + ]: 270 : !bms_is_subset(em->em_relids, relids))
1031 : 110 : continue;
1032 : :
1033 : : /*
1034 : : * Match if all Vars and quasi-Vars are present in "exprs".
1035 : : */
1036 : 7305 : emvars = pull_var_clause((Node *) em->em_expr,
1037 : : PVC_INCLUDE_AGGREGATES |
1038 : : PVC_INCLUDE_WINDOWFUNCS |
1039 : : PVC_INCLUDE_PLACEHOLDERS);
1040 [ + + + + : 8071 : foreach(lc2, emvars)
+ + ]
1041 : : {
1042 [ + + ]: 7641 : if (!list_member(exprvars, lfirst(lc2)))
1043 : 6875 : break;
1044 : : }
1045 : 7305 : list_free(emvars);
1046 [ + + ]: 7305 : if (lc2)
1047 : 6875 : continue; /* we hit a non-available Var */
1048 : :
1049 : : /*
1050 : : * If requested, reject expressions that are not parallel-safe. We
1051 : : * check this last because it's a rather expensive test.
1052 : : */
1053 [ + + ]: 430 : if (require_parallel_safe &&
1054 [ + + ]: 121 : !is_parallel_safe(root, (Node *) em->em_expr))
1055 : 40 : continue;
1056 : :
1057 : 390 : return em; /* found usable expression */
1058 : : }
1059 : :
1060 : 6537 : return NULL;
1061 : : }
1062 : :
1063 : : /*
1064 : : * relation_can_be_sorted_early
1065 : : * Can this relation be sorted on this EC before the final output step?
1066 : : *
1067 : : * To succeed, we must find an EC member that prepare_sort_from_pathkeys knows
1068 : : * how to sort on, given the rel's reltarget as input. There are also a few
1069 : : * additional constraints based on the fact that the desired sort will be done
1070 : : * "early", within the scan/join part of the plan. Also, non-parallel-safe
1071 : : * expressions are ignored if 'require_parallel_safe'.
1072 : : *
1073 : : * At some point we might want to return the identified EquivalenceMember,
1074 : : * but for now, callers only want to know if there is one.
1075 : : */
1076 : : bool
1077 : 15750 : relation_can_be_sorted_early(PlannerInfo *root, RelOptInfo *rel,
1078 : : EquivalenceClass *ec, bool require_parallel_safe)
1079 : : {
1080 : 15750 : PathTarget *target = rel->reltarget;
1081 : : EquivalenceMember *em;
1082 : : ListCell *lc;
1083 : :
1084 : : /*
1085 : : * Reject volatile ECs immediately; such sorts must always be postponed.
1086 : : */
1087 [ + + ]: 15750 : if (ec->ec_has_volatile)
1088 : 60 : return false;
1089 : :
1090 : : /*
1091 : : * Try to find an EM directly matching some reltarget member.
1092 : : */
1093 [ + + + + : 34623 : foreach(lc, target->exprs)
+ + ]
1094 : : {
1095 : 28005 : Expr *targetexpr = (Expr *) lfirst(lc);
1096 : :
1097 : 28005 : em = find_ec_member_matching_expr(ec, targetexpr, rel->relids);
1098 [ + + ]: 28005 : if (!em)
1099 : 18933 : continue;
1100 : :
1101 : : /*
1102 : : * Reject expressions involving set-returning functions, as those
1103 : : * can't be computed early either. (Note: this test and the following
1104 : : * one are effectively checking properties of targetexpr, so there's
1105 : : * no point in asking whether some other EC member would be better.)
1106 : : */
1107 [ - + ]: 9072 : if (expression_returns_set((Node *) em->em_expr))
1108 : 0 : continue;
1109 : :
1110 : : /*
1111 : : * If requested, reject expressions that are not parallel-safe. We
1112 : : * check this last because it's a rather expensive test.
1113 : : */
1114 [ + - ]: 9072 : if (require_parallel_safe &&
1115 [ - + ]: 9072 : !is_parallel_safe(root, (Node *) em->em_expr))
1116 : 0 : continue;
1117 : :
1118 : 9072 : return true;
1119 : : }
1120 : :
1121 : : /*
1122 : : * Try to find an expression computable from the reltarget.
1123 : : */
1124 : 6618 : em = find_computable_ec_member(root, ec, target->exprs, rel->relids,
1125 : : require_parallel_safe);
1126 [ + + ]: 6618 : if (!em)
1127 : 6537 : return false;
1128 : :
1129 : : /*
1130 : : * Reject expressions involving set-returning functions, as those can't be
1131 : : * computed early either. (There's no point in looking for another EC
1132 : : * member in this case; since SRFs can't appear in WHERE, they cannot
1133 : : * belong to multi-member ECs.)
1134 : : */
1135 [ + + ]: 81 : if (expression_returns_set((Node *) em->em_expr))
1136 : 10 : return false;
1137 : :
1138 : 71 : return true;
1139 : : }
1140 : :
1141 : : /*
1142 : : * generate_base_implied_equalities
1143 : : * Generate any restriction clauses that we can deduce from equivalence
1144 : : * classes.
1145 : : *
1146 : : * When an EC contains pseudoconstants, our strategy is to generate
1147 : : * "member = const1" clauses where const1 is the first constant member, for
1148 : : * every other member (including other constants). If we are able to do this
1149 : : * then we don't need any "var = var" comparisons because we've successfully
1150 : : * constrained all the vars at their points of creation. If we fail to
1151 : : * generate any of these clauses due to lack of cross-type operators, we fall
1152 : : * back to the "ec_broken" strategy described below. (XXX if there are
1153 : : * multiple constants of different types, it's possible that we might succeed
1154 : : * in forming all the required clauses if we started from a different const
1155 : : * member; but this seems a sufficiently hokey corner case to not be worth
1156 : : * spending lots of cycles on.)
1157 : : *
1158 : : * For ECs that contain no pseudoconstants, we generate derived clauses
1159 : : * "member1 = member2" for each pair of members belonging to the same base
1160 : : * relation (actually, if there are more than two for the same base relation,
1161 : : * we only need enough clauses to link each to each other). This provides
1162 : : * the base case for the recursion: each row emitted by a base relation scan
1163 : : * will constrain all computable members of the EC to be equal. As each
1164 : : * join path is formed, we'll add additional derived clauses on-the-fly
1165 : : * to maintain this invariant (see generate_join_implied_equalities).
1166 : : *
1167 : : * If the opfamilies used by the EC do not provide complete sets of cross-type
1168 : : * equality operators, it is possible that we will fail to generate a clause
1169 : : * that must be generated to maintain the invariant. (An example: given
1170 : : * "WHERE a.x = b.y AND b.y = a.z", the scheme breaks down if we cannot
1171 : : * generate "a.x = a.z" as a restriction clause for A.) In this case we mark
1172 : : * the EC "ec_broken" and fall back to regurgitating its original source
1173 : : * RestrictInfos at appropriate times. We do not try to retract any derived
1174 : : * clauses already generated from the broken EC, so the resulting plan could
1175 : : * be poor due to bad selectivity estimates caused by redundant clauses. But
1176 : : * the correct solution to that is to fix the opfamilies ...
1177 : : *
1178 : : * Equality clauses derived by this function are passed off to
1179 : : * process_implied_equality (in plan/initsplan.c) to be inserted into the
1180 : : * restrictinfo datastructures. Note that this must be called after initial
1181 : : * scanning of the quals and before Path construction begins.
1182 : : *
1183 : : * We make no attempt to avoid generating duplicate RestrictInfos here: we
1184 : : * don't search existing source or derived clauses in the EC for matches. It
1185 : : * doesn't really seem worth the trouble to do so.
1186 : : */
1187 : : void
1188 : 254054 : generate_base_implied_equalities(PlannerInfo *root)
1189 : : {
1190 : : int ec_index;
1191 : : ListCell *lc;
1192 : :
1193 : : /*
1194 : : * At this point, we're done absorbing knowledge of equivalences in the
1195 : : * query, so no further EC merging should happen, and ECs remaining in the
1196 : : * eq_classes list can be considered canonical. (But note that it's still
1197 : : * possible for new single-member ECs to be added through
1198 : : * get_eclass_for_sort_expr().)
1199 : : */
1200 : 254054 : root->ec_merging_done = true;
1201 : :
1202 : 254054 : ec_index = 0;
1203 [ + + + + : 571720 : foreach(lc, root->eq_classes)
+ + ]
1204 : : {
1205 : 317666 : EquivalenceClass *ec = (EquivalenceClass *) lfirst(lc);
1206 : 317666 : bool can_generate_joinclause = false;
1207 : : int i;
1208 : :
1209 : : Assert(ec->ec_merged == NULL); /* else shouldn't be in list */
1210 : : Assert(!ec->ec_broken); /* not yet anyway... */
1211 : :
1212 : : /*
1213 : : * Generate implied equalities that are restriction clauses.
1214 : : * Single-member ECs won't generate any deductions, either here or at
1215 : : * the join level.
1216 : : */
1217 [ + + ]: 317666 : if (list_length(ec->ec_members) > 1)
1218 : : {
1219 [ + + ]: 226033 : if (ec->ec_has_const)
1220 : 164114 : generate_base_implied_equalities_const(root, ec);
1221 : : else
1222 : 61919 : generate_base_implied_equalities_no_const(root, ec);
1223 : :
1224 : : /* Recover if we failed to generate required derived clauses */
1225 [ + + ]: 226033 : if (ec->ec_broken)
1226 : 25 : generate_base_implied_equalities_broken(root, ec);
1227 : :
1228 : : /* Detect whether this EC might generate join clauses */
1229 : 226033 : can_generate_joinclause =
1230 : 226033 : (bms_membership(ec->ec_relids) == BMS_MULTIPLE);
1231 : : }
1232 : :
1233 : : /*
1234 : : * Mark the base rels cited in each eclass (which should all exist by
1235 : : * now) with the eq_classes indexes of all eclasses mentioning them.
1236 : : * This will let us avoid searching in subsequent lookups. While
1237 : : * we're at it, we can mark base rels that have pending eclass joins;
1238 : : * this is a cheap version of has_relevant_eclass_joinclause().
1239 : : */
1240 : 317666 : i = -1;
1241 [ + + ]: 712721 : while ((i = bms_next_member(ec->ec_relids, i)) > 0)
1242 : : {
1243 : 395055 : RelOptInfo *rel = root->simple_rel_array[i];
1244 : :
1245 : : /* ignore the RTE_GROUP RTE */
1246 [ - + ]: 395055 : if (i == root->group_rtindex)
1247 : 0 : continue;
1248 : :
1249 [ + + ]: 395055 : if (rel == NULL) /* must be an outer join */
1250 : : {
1251 : : Assert(bms_is_member(i, root->outer_join_rels));
1252 : 3455 : continue;
1253 : : }
1254 : :
1255 : : Assert(rel->reloptkind == RELOPT_BASEREL);
1256 : :
1257 : 391600 : rel->eclass_indexes = bms_add_member(rel->eclass_indexes,
1258 : : ec_index);
1259 : :
1260 [ + + ]: 391600 : if (can_generate_joinclause)
1261 : 147198 : rel->has_eclass_joins = true;
1262 : : }
1263 : :
1264 : 317666 : ec_index++;
1265 : : }
1266 : 254054 : }
1267 : :
1268 : : /*
1269 : : * generate_base_implied_equalities when EC contains pseudoconstant(s)
1270 : : */
1271 : : static void
1272 : 164114 : generate_base_implied_equalities_const(PlannerInfo *root,
1273 : : EquivalenceClass *ec)
1274 : : {
1275 : 164114 : EquivalenceMember *const_em = NULL;
1276 : : ListCell *lc;
1277 : :
1278 : : /*
1279 : : * In the trivial case where we just had one "var = const" clause, push
1280 : : * the original clause back into the main planner machinery. There is
1281 : : * nothing to be gained by doing it differently, and we save the effort to
1282 : : * re-build and re-analyze an equality clause that will be exactly
1283 : : * equivalent to the old one.
1284 : : */
1285 [ + + + + ]: 315999 : if (list_length(ec->ec_members) == 2 &&
1286 : 151885 : list_length(ec->ec_sources) == 1)
1287 : : {
1288 : 151850 : RestrictInfo *restrictinfo = (RestrictInfo *) linitial(ec->ec_sources);
1289 : :
1290 : 151850 : distribute_restrictinfo_to_rels(root, restrictinfo);
1291 : 151850 : return;
1292 : : }
1293 : :
1294 : : /* We don't expect any children yet */
1295 : : Assert(ec->ec_childmembers == NULL);
1296 : :
1297 : : /*
1298 : : * Find the constant member to use. We prefer an actual constant to
1299 : : * pseudo-constants (such as Params), because the constraint exclusion
1300 : : * machinery might be able to exclude relations on the basis of generated
1301 : : * "var = const" equalities, but "var = param" won't work for that.
1302 : : */
1303 [ + - + + : 29525 : foreach(lc, ec->ec_members)
+ + ]
1304 : : {
1305 : 29441 : EquivalenceMember *cur_em = (EquivalenceMember *) lfirst(lc);
1306 : :
1307 [ + + ]: 29441 : if (cur_em->em_is_const)
1308 : : {
1309 : 12269 : const_em = cur_em;
1310 [ + + ]: 12269 : if (IsA(cur_em->em_expr, Const))
1311 : 12180 : break;
1312 : : }
1313 : : }
1314 : : Assert(const_em != NULL);
1315 : :
1316 : : /* Generate a derived equality against each other member */
1317 [ + - + + : 49137 : foreach(lc, ec->ec_members)
+ + ]
1318 : : {
1319 : 36898 : EquivalenceMember *cur_em = (EquivalenceMember *) lfirst(lc);
1320 : : Oid eq_op;
1321 : : RestrictInfo *rinfo;
1322 : :
1323 : : /* Child members should not exist in ec_members */
1324 : : Assert(!cur_em->em_is_child);
1325 [ + + ]: 36898 : if (cur_em == const_em)
1326 : 12244 : continue;
1327 : 24654 : eq_op = select_equality_operator(ec,
1328 : : cur_em->em_datatype,
1329 : : const_em->em_datatype);
1330 [ + + ]: 24654 : if (!OidIsValid(eq_op))
1331 : : {
1332 : : /* failed... */
1333 : 25 : ec->ec_broken = true;
1334 : 25 : break;
1335 : : }
1336 : :
1337 : : /*
1338 : : * We use the constant's em_jdomain as qualscope, so that if the
1339 : : * generated clause is variable-free (i.e, both EMs are consts) it
1340 : : * will be enforced at the join domain level.
1341 : : */
1342 : 24629 : rinfo = process_implied_equality(root, eq_op, ec->ec_collation,
1343 : : cur_em->em_expr, const_em->em_expr,
1344 : 24629 : const_em->em_jdomain->jd_relids,
1345 : : ec->ec_min_security,
1346 : 24629 : cur_em->em_is_const);
1347 : :
1348 : : /*
1349 : : * If the clause didn't degenerate to a constant, fill in the correct
1350 : : * markings for a mergejoinable clause, and save it as a derived
1351 : : * clause. (We will not re-use such clauses directly, but selectivity
1352 : : * estimation may consult those later. Note that this use of derived
1353 : : * clauses does not overlap with its use for join clauses, since we
1354 : : * never generate join clauses from an ec_has_const eclass.)
1355 : : */
1356 [ + - + + ]: 24629 : if (rinfo && rinfo->mergeopfamilies)
1357 : : {
1358 : : /* it's not redundant, so don't set parent_ec */
1359 : 24494 : rinfo->left_ec = rinfo->right_ec = ec;
1360 : 24494 : rinfo->left_em = cur_em;
1361 : 24494 : rinfo->right_em = const_em;
1362 : 24494 : ec_add_derived_clause(ec, rinfo);
1363 : : }
1364 : : }
1365 : : }
1366 : :
1367 : : /*
1368 : : * generate_base_implied_equalities when EC contains no pseudoconstants
1369 : : */
1370 : : static void
1371 : 61919 : generate_base_implied_equalities_no_const(PlannerInfo *root,
1372 : : EquivalenceClass *ec)
1373 : : {
1374 : : EquivalenceMember **prev_ems;
1375 : : ListCell *lc;
1376 : :
1377 : : /*
1378 : : * We scan the EC members once and track the last-seen member for each
1379 : : * base relation. When we see another member of the same base relation,
1380 : : * we generate "prev_em = cur_em". This results in the minimum number of
1381 : : * derived clauses, but it's possible that it will fail when a different
1382 : : * ordering would succeed. XXX FIXME: use a UNION-FIND algorithm similar
1383 : : * to the way we build merged ECs. (Use a list-of-lists for each rel.)
1384 : : */
1385 : 61919 : prev_ems = palloc0_array(EquivalenceMember *, root->simple_rel_array_size);
1386 : :
1387 : : /* We don't expect any children yet */
1388 : : Assert(ec->ec_childmembers == NULL);
1389 : :
1390 [ + - + + : 187550 : foreach(lc, ec->ec_members)
+ + ]
1391 : : {
1392 : 125631 : EquivalenceMember *cur_em = (EquivalenceMember *) lfirst(lc);
1393 : : int relid;
1394 : :
1395 : : /* Child members should not exist in ec_members */
1396 : : Assert(!cur_em->em_is_child);
1397 : :
1398 [ + + ]: 125631 : if (!bms_get_singleton_member(cur_em->em_relids, &relid))
1399 : 205 : continue;
1400 : : Assert(relid < root->simple_rel_array_size);
1401 : :
1402 [ + + ]: 125426 : if (prev_ems[relid] != NULL)
1403 : : {
1404 : 394 : EquivalenceMember *prev_em = prev_ems[relid];
1405 : : Oid eq_op;
1406 : : RestrictInfo *rinfo;
1407 : :
1408 : 394 : eq_op = select_equality_operator(ec,
1409 : : prev_em->em_datatype,
1410 : : cur_em->em_datatype);
1411 [ - + ]: 394 : if (!OidIsValid(eq_op))
1412 : : {
1413 : : /* failed... */
1414 : 0 : ec->ec_broken = true;
1415 : 0 : break;
1416 : : }
1417 : :
1418 : : /*
1419 : : * The expressions aren't constants, so the passed qualscope will
1420 : : * never be used to place the generated clause. We just need to
1421 : : * be sure it covers both expressions, which em_relids should do.
1422 : : */
1423 : 394 : rinfo = process_implied_equality(root, eq_op, ec->ec_collation,
1424 : : prev_em->em_expr, cur_em->em_expr,
1425 : : cur_em->em_relids,
1426 : : ec->ec_min_security,
1427 : : false);
1428 : :
1429 : : /*
1430 : : * If the clause didn't degenerate to a constant, fill in the
1431 : : * correct markings for a mergejoinable clause. We don't record
1432 : : * it as a derived clause, since we don't currently need to
1433 : : * re-find such clauses, and don't want to clutter the
1434 : : * derived-clause set with non-join clauses.
1435 : : */
1436 [ + - + - ]: 394 : if (rinfo && rinfo->mergeopfamilies)
1437 : : {
1438 : : /* it's not redundant, so don't set parent_ec */
1439 : 394 : rinfo->left_ec = rinfo->right_ec = ec;
1440 : 394 : rinfo->left_em = prev_em;
1441 : 394 : rinfo->right_em = cur_em;
1442 : : }
1443 : : }
1444 : 125426 : prev_ems[relid] = cur_em;
1445 : : }
1446 : :
1447 : 61919 : pfree(prev_ems);
1448 : :
1449 : : /*
1450 : : * We also have to make sure that all the Vars used in the member clauses
1451 : : * will be available at any join node we might try to reference them at.
1452 : : * For the moment we force all the Vars to be available at all join nodes
1453 : : * for this eclass. Perhaps this could be improved by doing some
1454 : : * pre-analysis of which members we prefer to join, but it's no worse than
1455 : : * what happened in the pre-8.3 code.
1456 : : */
1457 [ + - + + : 187550 : foreach(lc, ec->ec_members)
+ + ]
1458 : : {
1459 : 125631 : EquivalenceMember *cur_em = (EquivalenceMember *) lfirst(lc);
1460 : 125631 : List *vars = pull_var_clause((Node *) cur_em->em_expr,
1461 : : PVC_RECURSE_AGGREGATES |
1462 : : PVC_RECURSE_WINDOWFUNCS |
1463 : : PVC_INCLUDE_PLACEHOLDERS);
1464 : :
1465 : 125631 : add_vars_to_targetlist(root, vars, ec->ec_relids);
1466 : 125631 : list_free(vars);
1467 : : }
1468 : 61919 : }
1469 : :
1470 : : /*
1471 : : * generate_base_implied_equalities cleanup after failure
1472 : : *
1473 : : * What we must do here is push any zero- or one-relation source RestrictInfos
1474 : : * of the EC back into the main restrictinfo datastructures. Multi-relation
1475 : : * clauses will be regurgitated later by generate_join_implied_equalities().
1476 : : * (We do it this way to maintain continuity with the case that ec_broken
1477 : : * becomes set only after we've gone up a join level or two.) However, for
1478 : : * an EC that contains constants, we can adopt a simpler strategy and just
1479 : : * throw back all the source RestrictInfos immediately; that works because
1480 : : * we know that such an EC can't become broken later. (This rule justifies
1481 : : * ignoring ec_has_const ECs in generate_join_implied_equalities, even when
1482 : : * they are broken.)
1483 : : */
1484 : : static void
1485 : 25 : generate_base_implied_equalities_broken(PlannerInfo *root,
1486 : : EquivalenceClass *ec)
1487 : : {
1488 : : ListCell *lc;
1489 : :
1490 [ + - + + : 80 : foreach(lc, ec->ec_sources)
+ + ]
1491 : : {
1492 : 55 : RestrictInfo *restrictinfo = (RestrictInfo *) lfirst(lc);
1493 : :
1494 [ - + - - ]: 55 : if (ec->ec_has_const ||
1495 : 0 : bms_membership(restrictinfo->required_relids) != BMS_MULTIPLE)
1496 : 55 : distribute_restrictinfo_to_rels(root, restrictinfo);
1497 : : }
1498 : 25 : }
1499 : :
1500 : :
1501 : : /*
1502 : : * generate_join_implied_equalities
1503 : : * Generate any join clauses that we can deduce from equivalence classes.
1504 : : *
1505 : : * At a join node, we must enforce restriction clauses sufficient to ensure
1506 : : * that all equivalence-class members computable at that node are equal.
1507 : : * Since the set of clauses to enforce can vary depending on which subset
1508 : : * relations are the inputs, we have to compute this afresh for each join
1509 : : * relation pair. Hence a fresh List of RestrictInfo nodes is built and
1510 : : * passed back on each call.
1511 : : *
1512 : : * In addition to its use at join nodes, this can be applied to generate
1513 : : * eclass-based join clauses for use in a parameterized scan of a base rel.
1514 : : * The reason for the asymmetry of specifying the inner rel as a RelOptInfo
1515 : : * and the outer rel by Relids is that this usage occurs before we have
1516 : : * built any join RelOptInfos.
1517 : : *
1518 : : * An annoying special case for parameterized scans is that the inner rel can
1519 : : * be an appendrel child (an "other rel"). In this case we must generate
1520 : : * appropriate clauses using child EC members. add_child_rel_equivalences
1521 : : * must already have been done for the child rel.
1522 : : *
1523 : : * The results are sufficient for use in merge, hash, and plain nestloop join
1524 : : * methods. We do not worry here about selecting clauses that are optimal
1525 : : * for use in a parameterized indexscan. indxpath.c makes its own selections
1526 : : * of clauses to use, and if the ones we pick here are redundant with those,
1527 : : * the extras will be eliminated at createplan time, using the parent_ec
1528 : : * markers that we provide (see is_redundant_derived_clause()).
1529 : : *
1530 : : * Because the same join clauses are likely to be needed multiple times as
1531 : : * we consider different join paths, we avoid generating multiple copies:
1532 : : * whenever we select a particular pair of EquivalenceMembers to join,
1533 : : * we check to see if the pair matches any original clause (in ec_sources)
1534 : : * or previously-built derived clause. This saves memory and allows
1535 : : * re-use of information cached in RestrictInfos. We also avoid generating
1536 : : * commutative duplicates, i.e. if the algorithm selects "a.x = b.y" but
1537 : : * we already have "b.y = a.x", we return the existing clause.
1538 : : *
1539 : : * If we are considering an outer join, sjinfo is the associated OJ info,
1540 : : * otherwise it can be NULL.
1541 : : *
1542 : : * join_relids should always equal bms_union(outer_relids, inner_rel->relids)
1543 : : * plus whatever add_outer_joins_to_relids() would add. We could simplify
1544 : : * this function's API by computing it internally, but most callers have the
1545 : : * value at hand anyway.
1546 : : */
1547 : : List *
1548 : 430347 : generate_join_implied_equalities(PlannerInfo *root,
1549 : : Relids join_relids,
1550 : : Relids outer_relids,
1551 : : RelOptInfo *inner_rel,
1552 : : SpecialJoinInfo *sjinfo)
1553 : : {
1554 : 430347 : List *result = NIL;
1555 : 430347 : Relids inner_relids = inner_rel->relids;
1556 : : Relids nominal_inner_relids;
1557 : : Relids nominal_join_relids;
1558 : : Bitmapset *matching_ecs;
1559 : : int i;
1560 : :
1561 : : /* If inner rel is a child, extra setup work is needed */
1562 [ + + + + : 430347 : if (IS_OTHER_REL(inner_rel))
- + ]
1563 : : {
1564 : : Assert(!bms_is_empty(inner_rel->top_parent_relids));
1565 : :
1566 : : /* Fetch relid set for the topmost parent rel */
1567 : 6609 : nominal_inner_relids = inner_rel->top_parent_relids;
1568 : : /* ECs will be marked with the parent's relid, not the child's */
1569 : 6609 : nominal_join_relids = bms_union(outer_relids, nominal_inner_relids);
1570 : 6609 : nominal_join_relids = add_outer_joins_to_relids(root,
1571 : : nominal_join_relids,
1572 : : sjinfo,
1573 : : NULL);
1574 : : }
1575 : : else
1576 : : {
1577 : 423738 : nominal_inner_relids = inner_relids;
1578 : 423738 : nominal_join_relids = join_relids;
1579 : : }
1580 : :
1581 : : /*
1582 : : * Examine all potentially-relevant eclasses.
1583 : : *
1584 : : * If we are considering an outer join, we must include "join" clauses
1585 : : * that mention either input rel plus the outer join's relid; these
1586 : : * represent post-join filter clauses that have to be applied at this
1587 : : * join. We don't have infrastructure that would let us identify such
1588 : : * eclasses cheaply, so just fall back to considering all eclasses
1589 : : * mentioning anything in nominal_join_relids.
1590 : : *
1591 : : * At inner joins, we can be smarter: only consider eclasses mentioning
1592 : : * both input rels.
1593 : : */
1594 [ + + + + ]: 430347 : if (sjinfo && sjinfo->ojrelid != 0)
1595 : 62869 : matching_ecs = get_eclass_indexes_for_relids(root, nominal_join_relids);
1596 : : else
1597 : 367478 : matching_ecs = get_common_eclass_indexes(root, nominal_inner_relids,
1598 : : outer_relids);
1599 : :
1600 : 430347 : i = -1;
1601 [ + + ]: 1144295 : while ((i = bms_next_member(matching_ecs, i)) >= 0)
1602 : : {
1603 : 713948 : EquivalenceClass *ec = (EquivalenceClass *) list_nth(root->eq_classes, i);
1604 : 713948 : List *sublist = NIL;
1605 : :
1606 : : /* ECs containing consts do not need any further enforcement */
1607 [ + + ]: 713948 : if (ec->ec_has_const)
1608 : 87446 : continue;
1609 : :
1610 : : /* Single-member ECs won't generate any deductions */
1611 [ + + ]: 626502 : if (list_length(ec->ec_members) <= 1)
1612 : 301420 : continue;
1613 : :
1614 : : /* Sanity check that this eclass overlaps the join */
1615 : : Assert(bms_overlap(ec->ec_relids, nominal_join_relids));
1616 : :
1617 [ + + ]: 325082 : if (!ec->ec_broken)
1618 : 324812 : sublist = generate_join_implied_equalities_normal(root,
1619 : : ec,
1620 : : join_relids,
1621 : : outer_relids,
1622 : : inner_relids);
1623 : :
1624 : : /* Recover if we failed to generate required derived clauses */
1625 [ + + ]: 325082 : if (ec->ec_broken)
1626 : 300 : sublist = generate_join_implied_equalities_broken(root,
1627 : : ec,
1628 : : nominal_join_relids,
1629 : : outer_relids,
1630 : : nominal_inner_relids,
1631 : : inner_rel);
1632 : :
1633 : 325082 : result = list_concat(result, sublist);
1634 : : }
1635 : :
1636 : 430347 : return result;
1637 : : }
1638 : :
1639 : : /*
1640 : : * generate_join_implied_equalities_for_ecs
1641 : : * As above, but consider only the listed ECs.
1642 : : *
1643 : : * For the sole current caller, we can assume sjinfo == NULL, that is we are
1644 : : * not interested in outer-join filter clauses. This might need to change
1645 : : * in future.
1646 : : */
1647 : : List *
1648 : 4499 : generate_join_implied_equalities_for_ecs(PlannerInfo *root,
1649 : : List *eclasses,
1650 : : Relids join_relids,
1651 : : Relids outer_relids,
1652 : : RelOptInfo *inner_rel)
1653 : : {
1654 : 4499 : List *result = NIL;
1655 : 4499 : Relids inner_relids = inner_rel->relids;
1656 : : Relids nominal_inner_relids;
1657 : : Relids nominal_join_relids;
1658 : : ListCell *lc;
1659 : :
1660 : : /* If inner rel is a child, extra setup work is needed */
1661 [ + + + - : 4499 : if (IS_OTHER_REL(inner_rel))
- + ]
1662 : : {
1663 : : Assert(!bms_is_empty(inner_rel->top_parent_relids));
1664 : :
1665 : : /* Fetch relid set for the topmost parent rel */
1666 : 80 : nominal_inner_relids = inner_rel->top_parent_relids;
1667 : : /* ECs will be marked with the parent's relid, not the child's */
1668 : 80 : nominal_join_relids = bms_union(outer_relids, nominal_inner_relids);
1669 : : }
1670 : : else
1671 : : {
1672 : 4419 : nominal_inner_relids = inner_relids;
1673 : 4419 : nominal_join_relids = join_relids;
1674 : : }
1675 : :
1676 [ + - + + : 9411 : foreach(lc, eclasses)
+ + ]
1677 : : {
1678 : 4912 : EquivalenceClass *ec = (EquivalenceClass *) lfirst(lc);
1679 : 4912 : List *sublist = NIL;
1680 : :
1681 : : /* ECs containing consts do not need any further enforcement */
1682 [ - + ]: 4912 : if (ec->ec_has_const)
1683 : 0 : continue;
1684 : :
1685 : : /* Single-member ECs won't generate any deductions */
1686 [ - + ]: 4912 : if (list_length(ec->ec_members) <= 1)
1687 : 0 : continue;
1688 : :
1689 : : /* We can quickly ignore any that don't overlap the join, too */
1690 [ - + ]: 4912 : if (!bms_overlap(ec->ec_relids, nominal_join_relids))
1691 : 0 : continue;
1692 : :
1693 [ + - ]: 4912 : if (!ec->ec_broken)
1694 : 4912 : sublist = generate_join_implied_equalities_normal(root,
1695 : : ec,
1696 : : join_relids,
1697 : : outer_relids,
1698 : : inner_relids);
1699 : :
1700 : : /* Recover if we failed to generate required derived clauses */
1701 [ - + ]: 4912 : if (ec->ec_broken)
1702 : 0 : sublist = generate_join_implied_equalities_broken(root,
1703 : : ec,
1704 : : nominal_join_relids,
1705 : : outer_relids,
1706 : : nominal_inner_relids,
1707 : : inner_rel);
1708 : :
1709 : 4912 : result = list_concat(result, sublist);
1710 : : }
1711 : :
1712 : 4499 : return result;
1713 : : }
1714 : :
1715 : : /*
1716 : : * generate_join_implied_equalities for a still-valid EC
1717 : : */
1718 : : static List *
1719 : 329724 : generate_join_implied_equalities_normal(PlannerInfo *root,
1720 : : EquivalenceClass *ec,
1721 : : Relids join_relids,
1722 : : Relids outer_relids,
1723 : : Relids inner_relids)
1724 : : {
1725 : 329724 : List *result = NIL;
1726 : 329724 : List *new_members = NIL;
1727 : 329724 : List *outer_members = NIL;
1728 : 329724 : List *inner_members = NIL;
1729 : : EquivalenceMemberIterator it;
1730 : : EquivalenceMember *cur_em;
1731 : :
1732 : : /*
1733 : : * First, scan the EC to identify member values that are computable at the
1734 : : * outer rel, at the inner rel, or at this relation but not in either
1735 : : * input rel. The outer-rel members should already be enforced equal,
1736 : : * likewise for the inner-rel members. We'll need to create clauses to
1737 : : * enforce that any newly computable members are all equal to each other
1738 : : * as well as to at least one input member, plus enforce at least one
1739 : : * outer-rel member equal to at least one inner-rel member.
1740 : : */
1741 : 329724 : setup_eclass_member_iterator(&it, ec, join_relids);
1742 [ + + ]: 1047955 : while ((cur_em = eclass_member_iterator_next(&it)) != NULL)
1743 : : {
1744 : : /*
1745 : : * We don't need to check explicitly for child EC members. This test
1746 : : * against join_relids will cause them to be ignored except when
1747 : : * considering a child inner rel, which is what we want.
1748 : : */
1749 [ + + ]: 718231 : if (!bms_is_subset(cur_em->em_relids, join_relids))
1750 : 59663 : continue; /* not computable yet, or wrong child */
1751 : :
1752 [ + + ]: 658568 : if (bms_is_subset(cur_em->em_relids, outer_relids))
1753 : 369094 : outer_members = lappend(outer_members, cur_em);
1754 [ + + ]: 289474 : else if (bms_is_subset(cur_em->em_relids, inner_relids))
1755 : 287404 : inner_members = lappend(inner_members, cur_em);
1756 : : else
1757 : 2070 : new_members = lappend(new_members, cur_em);
1758 : : }
1759 : :
1760 : : /*
1761 : : * First, select the joinclause if needed. We can equate any one outer
1762 : : * member to any one inner member, but we have to find a datatype
1763 : : * combination for which an opfamily member operator exists. If we have
1764 : : * choices, we prefer simple Var members (possibly with RelabelType) since
1765 : : * these are (a) cheapest to compute at runtime and (b) most likely to
1766 : : * have useful statistics. Also, prefer operators that are also
1767 : : * hashjoinable.
1768 : : */
1769 [ + + + + ]: 329724 : if (outer_members && inner_members)
1770 : : {
1771 : 276578 : EquivalenceMember *best_outer_em = NULL;
1772 : 276578 : EquivalenceMember *best_inner_em = NULL;
1773 : 276578 : Oid best_eq_op = InvalidOid;
1774 : 276578 : int best_score = -1;
1775 : : RestrictInfo *rinfo;
1776 : : ListCell *lc1;
1777 : :
1778 [ + - + + : 289344 : foreach(lc1, outer_members)
+ + ]
1779 : : {
1780 : 276643 : EquivalenceMember *outer_em = (EquivalenceMember *) lfirst(lc1);
1781 : : ListCell *lc2;
1782 : :
1783 [ + - + + : 289429 : foreach(lc2, inner_members)
+ + ]
1784 : : {
1785 : 276663 : EquivalenceMember *inner_em = (EquivalenceMember *) lfirst(lc2);
1786 : : Oid eq_op;
1787 : : int score;
1788 : :
1789 : 276663 : eq_op = select_equality_operator(ec,
1790 : : outer_em->em_datatype,
1791 : : inner_em->em_datatype);
1792 [ + + ]: 276663 : if (!OidIsValid(eq_op))
1793 : 30 : continue;
1794 : 276633 : score = 0;
1795 [ + + ]: 276633 : if (IsA(outer_em->em_expr, Var) ||
1796 [ + + ]: 14268 : (IsA(outer_em->em_expr, RelabelType) &&
1797 [ + + ]: 3662 : IsA(((RelabelType *) outer_em->em_expr)->arg, Var)))
1798 : 265816 : score++;
1799 [ + + ]: 276633 : if (IsA(inner_em->em_expr, Var) ||
1800 [ + + ]: 9728 : (IsA(inner_em->em_expr, RelabelType) &&
1801 [ + + ]: 7572 : IsA(((RelabelType *) inner_em->em_expr)->arg, Var)))
1802 : 274351 : score++;
1803 [ + + ]: 276633 : if (op_hashjoinable(eq_op,
1804 : 276633 : exprType((Node *) outer_em->em_expr)))
1805 : 276574 : score++;
1806 [ + + ]: 276633 : if (score > best_score)
1807 : : {
1808 : 276548 : best_outer_em = outer_em;
1809 : 276548 : best_inner_em = inner_em;
1810 : 276548 : best_eq_op = eq_op;
1811 : 276548 : best_score = score;
1812 [ + + ]: 276548 : if (best_score == 3)
1813 : 263877 : break; /* no need to look further */
1814 : : }
1815 : : }
1816 [ + + ]: 276643 : if (best_score == 3)
1817 : 263877 : break; /* no need to look further */
1818 : : }
1819 [ + + ]: 276578 : if (best_score < 0)
1820 : : {
1821 : : /* failed... */
1822 : 30 : ec->ec_broken = true;
1823 : 30 : return NIL;
1824 : : }
1825 : :
1826 : : /*
1827 : : * Create clause, setting parent_ec to mark it as redundant with other
1828 : : * joinclauses
1829 : : */
1830 : 276548 : rinfo = create_join_clause(root, ec, best_eq_op,
1831 : : best_outer_em, best_inner_em,
1832 : : ec);
1833 : :
1834 : 276548 : result = lappend(result, rinfo);
1835 : : }
1836 : :
1837 : : /*
1838 : : * Now deal with building restrictions for any expressions that involve
1839 : : * Vars from both sides of the join. We have to equate all of these to
1840 : : * each other as well as to at least one old member (if any).
1841 : : *
1842 : : * XXX as in generate_base_implied_equalities_no_const, we could be a lot
1843 : : * smarter here to avoid unnecessary failures in cross-type situations.
1844 : : * For now, use the same left-to-right method used there.
1845 : : */
1846 [ + + ]: 329694 : if (new_members)
1847 : : {
1848 : 2040 : List *old_members = list_concat(outer_members, inner_members);
1849 : 2040 : EquivalenceMember *prev_em = NULL;
1850 : : RestrictInfo *rinfo;
1851 : : ListCell *lc1;
1852 : :
1853 : : /* For now, arbitrarily take the first old_member as the one to use */
1854 [ + + ]: 2040 : if (old_members)
1855 : 1683 : new_members = lappend(new_members, linitial(old_members));
1856 : :
1857 [ + - + + : 5793 : foreach(lc1, new_members)
+ + ]
1858 : : {
1859 : 3753 : cur_em = (EquivalenceMember *) lfirst(lc1);
1860 : :
1861 [ + + ]: 3753 : if (prev_em != NULL)
1862 : : {
1863 : : Oid eq_op;
1864 : :
1865 : 1713 : eq_op = select_equality_operator(ec,
1866 : : prev_em->em_datatype,
1867 : : cur_em->em_datatype);
1868 [ - + ]: 1713 : if (!OidIsValid(eq_op))
1869 : : {
1870 : : /* failed... */
1871 : 0 : ec->ec_broken = true;
1872 : 0 : return NIL;
1873 : : }
1874 : : /* do NOT set parent_ec, this qual is not redundant! */
1875 : 1713 : rinfo = create_join_clause(root, ec, eq_op,
1876 : : prev_em, cur_em,
1877 : : NULL);
1878 : :
1879 : 1713 : result = lappend(result, rinfo);
1880 : : }
1881 : 3753 : prev_em = cur_em;
1882 : : }
1883 : : }
1884 : :
1885 : 329694 : return result;
1886 : : }
1887 : :
1888 : : /*
1889 : : * generate_join_implied_equalities cleanup after failure
1890 : : *
1891 : : * Return any original RestrictInfos that are enforceable at this join.
1892 : : *
1893 : : * In the case of a child inner relation, we have to translate the
1894 : : * original RestrictInfos from parent to child Vars.
1895 : : */
1896 : : static List *
1897 : 300 : generate_join_implied_equalities_broken(PlannerInfo *root,
1898 : : EquivalenceClass *ec,
1899 : : Relids nominal_join_relids,
1900 : : Relids outer_relids,
1901 : : Relids nominal_inner_relids,
1902 : : RelOptInfo *inner_rel)
1903 : : {
1904 : 300 : List *result = NIL;
1905 : : ListCell *lc;
1906 : :
1907 [ + - + + : 820 : foreach(lc, ec->ec_sources)
+ + ]
1908 : : {
1909 : 520 : RestrictInfo *restrictinfo = (RestrictInfo *) lfirst(lc);
1910 : 520 : Relids clause_relids = restrictinfo->required_relids;
1911 : :
1912 [ + + ]: 520 : if (bms_is_subset(clause_relids, nominal_join_relids) &&
1913 [ + + ]: 280 : !bms_is_subset(clause_relids, outer_relids) &&
1914 [ + - ]: 260 : !bms_is_subset(clause_relids, nominal_inner_relids))
1915 : 260 : result = lappend(result, restrictinfo);
1916 : : }
1917 : :
1918 : : /*
1919 : : * If we have to translate, just brute-force apply adjust_appendrel_attrs
1920 : : * to all the RestrictInfos at once. This will result in returning
1921 : : * RestrictInfos that are not included in EC's derived clauses, but there
1922 : : * shouldn't be any duplication, and it's a sufficiently narrow corner
1923 : : * case that we shouldn't sweat too much over it anyway.
1924 : : *
1925 : : * Since inner_rel might be an indirect descendant of the baserel
1926 : : * mentioned in the ec_sources clauses, we have to be prepared to apply
1927 : : * multiple levels of Var translation.
1928 : : */
1929 [ + + + - : 300 : if (IS_OTHER_REL(inner_rel) && result != NIL)
- + + + ]
1930 : 135 : result = (List *) adjust_appendrel_attrs_multilevel(root,
1931 : : (Node *) result,
1932 : : inner_rel,
1933 : 135 : inner_rel->top_parent);
1934 : :
1935 : 300 : return result;
1936 : : }
1937 : :
1938 : :
1939 : : /*
1940 : : * select_equality_operator
1941 : : * Select a suitable equality operator for comparing two EC members
1942 : : *
1943 : : * Returns InvalidOid if no operator can be found for this datatype combination
1944 : : */
1945 : : static Oid
1946 : 404829 : select_equality_operator(EquivalenceClass *ec, Oid lefttype, Oid righttype)
1947 : : {
1948 : : ListCell *lc;
1949 : :
1950 [ + - + + : 404884 : foreach(lc, ec->ec_opfamilies)
+ + ]
1951 : : {
1952 : 404829 : Oid opfamily = lfirst_oid(lc);
1953 : : Oid opno;
1954 : :
1955 : 404829 : opno = get_opfamily_member_for_cmptype(opfamily, lefttype, righttype, COMPARE_EQ);
1956 [ + + ]: 404829 : if (!OidIsValid(opno))
1957 : 55 : continue;
1958 : : /* If no barrier quals in query, don't worry about leaky operators */
1959 [ + + ]: 404774 : if (ec->ec_max_security == 0)
1960 : 404774 : return opno;
1961 : : /* Otherwise, insist that selected operators be leakproof */
1962 [ + - ]: 479 : if (get_func_leakproof(get_opcode(opno)))
1963 : 479 : return opno;
1964 : : }
1965 : 55 : return InvalidOid;
1966 : : }
1967 : :
1968 : :
1969 : : /*
1970 : : * create_join_clause
1971 : : * Find or make a RestrictInfo comparing the two given EC members
1972 : : * with the given operator (or, possibly, its commutator, because
1973 : : * the ordering of the operands in the result is not guaranteed).
1974 : : *
1975 : : * parent_ec is either equal to ec (if the clause is a potentially-redundant
1976 : : * join clause) or NULL (if not). We have to treat this as part of the
1977 : : * match requirements --- it's possible that a clause comparing the same two
1978 : : * EMs is a join clause in one join path and a restriction clause in another.
1979 : : */
1980 : : static RestrictInfo *
1981 : 382007 : create_join_clause(PlannerInfo *root,
1982 : : EquivalenceClass *ec, Oid opno,
1983 : : EquivalenceMember *leftem,
1984 : : EquivalenceMember *rightem,
1985 : : EquivalenceClass *parent_ec)
1986 : : {
1987 : : RestrictInfo *rinfo;
1988 : 382007 : RestrictInfo *parent_rinfo = NULL;
1989 : : MemoryContext oldcontext;
1990 : :
1991 : 382007 : rinfo = ec_search_clause_for_ems(root, ec, leftem, rightem, parent_ec);
1992 [ + + ]: 382007 : if (rinfo)
1993 : 316302 : return rinfo;
1994 : :
1995 : : /*
1996 : : * Not there, so build it, in planner context so we can re-use it. (Not
1997 : : * important in normal planning, but definitely so in GEQO.)
1998 : : */
1999 : 65705 : oldcontext = MemoryContextSwitchTo(root->planner_cxt);
2000 : :
2001 : : /*
2002 : : * If either EM is a child, recursively create the corresponding
2003 : : * parent-to-parent clause, so that we can duplicate its rinfo_serial.
2004 : : */
2005 [ + + + + ]: 65705 : if (leftem->em_is_child || rightem->em_is_child)
2006 : : {
2007 [ + + ]: 4044 : EquivalenceMember *leftp = leftem->em_parent ? leftem->em_parent : leftem;
2008 [ + + ]: 4044 : EquivalenceMember *rightp = rightem->em_parent ? rightem->em_parent : rightem;
2009 : :
2010 : 4044 : parent_rinfo = create_join_clause(root, ec, opno,
2011 : : leftp, rightp,
2012 : : parent_ec);
2013 : : }
2014 : :
2015 : 65705 : rinfo = build_implied_join_equality(root,
2016 : : opno,
2017 : : ec->ec_collation,
2018 : : leftem->em_expr,
2019 : : rightem->em_expr,
2020 : 65705 : bms_union(leftem->em_relids,
2021 : 65705 : rightem->em_relids),
2022 : : ec->ec_min_security);
2023 : :
2024 : : /*
2025 : : * If either EM is a child, force the clause's clause_relids to include
2026 : : * the relid(s) of the child rel. In normal cases it would already, but
2027 : : * not if we are considering appendrel child relations with pseudoconstant
2028 : : * translated variables (i.e., UNION ALL sub-selects with constant output
2029 : : * items). We must do this so that join_clause_is_movable_into() will
2030 : : * think that the clause should be evaluated at the correct place.
2031 : : */
2032 [ + + ]: 65705 : if (leftem->em_is_child)
2033 : 3513 : rinfo->clause_relids = bms_add_members(rinfo->clause_relids,
2034 : 3513 : leftem->em_relids);
2035 [ + + ]: 65705 : if (rightem->em_is_child)
2036 : 591 : rinfo->clause_relids = bms_add_members(rinfo->clause_relids,
2037 : 591 : rightem->em_relids);
2038 : :
2039 : : /* If it's a child clause, copy the parent's rinfo_serial */
2040 [ + + ]: 65705 : if (parent_rinfo)
2041 : 4044 : rinfo->rinfo_serial = parent_rinfo->rinfo_serial;
2042 : : else
2043 : : {
2044 : : RestrictInfo *counterpart;
2045 : :
2046 : : /*
2047 : : * If a clause comparing the same two EMs already exists with the
2048 : : * opposite parent_ec marking, adopt its rinfo_serial: the two clauses
2049 : : * enforce the same condition, and they must share a serial number
2050 : : * lest we enforce that condition more than once in a plan.
2051 : : */
2052 [ + + ]: 61661 : counterpart = ec_search_clause_for_ems(root, ec, leftem, rightem,
2053 : : parent_ec ? NULL : ec);
2054 [ + + ]: 61661 : if (counterpart)
2055 : 59930 : rinfo->rinfo_serial = counterpart->rinfo_serial;
2056 : : }
2057 : :
2058 : : /* Mark the clause as redundant, or not */
2059 : 65705 : rinfo->parent_ec = parent_ec;
2060 : :
2061 : : /*
2062 : : * We know the correct values for left_ec/right_ec, ie this particular EC,
2063 : : * so we can just set them directly instead of forcing another lookup.
2064 : : */
2065 : 65705 : rinfo->left_ec = ec;
2066 : 65705 : rinfo->right_ec = ec;
2067 : :
2068 : : /* Mark it as usable with these EMs */
2069 : 65705 : rinfo->left_em = leftem;
2070 : 65705 : rinfo->right_em = rightem;
2071 : : /* and save it for possible re-use */
2072 : 65705 : ec_add_derived_clause(ec, rinfo);
2073 : :
2074 : 65705 : MemoryContextSwitchTo(oldcontext);
2075 : :
2076 : 65705 : return rinfo;
2077 : : }
2078 : :
2079 : :
2080 : : /*
2081 : : * reconsider_outer_join_clauses
2082 : : * Re-examine any outer-join clauses that were set aside by
2083 : : * distribute_qual_to_rels(), and see if we can derive any
2084 : : * EquivalenceClasses from them. Then, if they were not made
2085 : : * redundant, push them out into the regular join-clause lists.
2086 : : *
2087 : : * When we have mergejoinable clauses A = B that are outer-join clauses,
2088 : : * we can't blindly combine them with other clauses A = C to deduce B = C,
2089 : : * since in fact the "equality" A = B won't necessarily hold above the
2090 : : * outer join (one of the variables might be NULL instead). Nonetheless
2091 : : * there are cases where we can add qual clauses using transitivity.
2092 : : *
2093 : : * One case that we look for here is an outer-join clause OUTERVAR = INNERVAR
2094 : : * for which there is also an equivalence clause OUTERVAR = CONSTANT.
2095 : : * It is safe and useful to push a clause INNERVAR = CONSTANT into the
2096 : : * evaluation of the inner (nullable) relation, because any inner rows not
2097 : : * meeting this condition will not contribute to the outer-join result anyway.
2098 : : * (Any outer rows they could join to will be eliminated by the pushed-down
2099 : : * equivalence clause.)
2100 : : *
2101 : : * Note that the above rule does not work for full outer joins; nor is it
2102 : : * very interesting to consider cases where the generated equivalence clause
2103 : : * would involve relations outside the outer join, since such clauses couldn't
2104 : : * be pushed into the inner side's scan anyway. So the restriction to
2105 : : * outervar = pseudoconstant is not really giving up anything.
2106 : : *
2107 : : * For full-join cases, we can only do something useful if it's a FULL JOIN
2108 : : * USING and a merged column has an equivalence MERGEDVAR = CONSTANT.
2109 : : * By the time it gets here, the merged column will look like
2110 : : * COALESCE(LEFTVAR, RIGHTVAR)
2111 : : * and we will have a full-join clause LEFTVAR = RIGHTVAR that we can match
2112 : : * the COALESCE expression to. In this situation we can push LEFTVAR = CONSTANT
2113 : : * and RIGHTVAR = CONSTANT into the input relations, since any rows not
2114 : : * meeting these conditions cannot contribute to the join result.
2115 : : *
2116 : : * Again, there isn't any traction to be gained by trying to deal with
2117 : : * clauses comparing a mergedvar to a non-pseudoconstant. So we can make
2118 : : * use of the EquivalenceClasses to search for matching variables that were
2119 : : * equivalenced to constants. The interesting outer-join clauses were
2120 : : * accumulated for us by distribute_qual_to_rels.
2121 : : *
2122 : : * When we find one of these cases, we implement the changes we want by
2123 : : * generating a new equivalence clause INNERVAR = CONSTANT (or LEFTVAR, etc)
2124 : : * and pushing it into the EquivalenceClass structures. This is because we
2125 : : * may already know that INNERVAR is equivalenced to some other var(s), and
2126 : : * we'd like the constant to propagate to them too. Note that it would be
2127 : : * unsafe to merge any existing EC for INNERVAR with the OUTERVAR's EC ---
2128 : : * that could result in propagating constant restrictions from
2129 : : * INNERVAR to OUTERVAR, which would be very wrong.
2130 : : *
2131 : : * It's possible that the INNERVAR is also an OUTERVAR for some other
2132 : : * outer-join clause, in which case the process can be repeated. So we repeat
2133 : : * looping over the lists of clauses until no further deductions can be made.
2134 : : * Whenever we do make a deduction, we remove the generating clause from the
2135 : : * lists, since we don't want to make the same deduction twice.
2136 : : *
2137 : : * If we don't find any match for a set-aside outer join clause, we must
2138 : : * throw it back into the regular joinclause processing by passing it to
2139 : : * distribute_restrictinfo_to_rels(). If we do generate a derived clause,
2140 : : * however, the outer-join clause is redundant. We must still put some
2141 : : * clause into the regular processing, because otherwise the join will be
2142 : : * seen as a clauseless join and avoided during join order searching.
2143 : : * We handle this by generating a constant-TRUE clause that is marked with
2144 : : * the same required_relids etc as the removed outer-join clause, thus
2145 : : * making it a join clause between the correct relations.
2146 : : */
2147 : : void
2148 : 254054 : reconsider_outer_join_clauses(PlannerInfo *root)
2149 : : {
2150 : : bool found;
2151 : : ListCell *cell;
2152 : :
2153 : : /* Outer loop repeats until we find no more deductions */
2154 : : do
2155 : : {
2156 : 255707 : found = false;
2157 : :
2158 : : /* Process the LEFT JOIN clauses */
2159 [ + + + + : 278297 : foreach(cell, root->left_join_clauses)
+ + ]
2160 : : {
2161 : 22590 : OuterJoinClauseInfo *ojcinfo = (OuterJoinClauseInfo *) lfirst(cell);
2162 : :
2163 [ + + ]: 22590 : if (reconsider_outer_join_clause(root, ojcinfo, true))
2164 : : {
2165 : 518 : RestrictInfo *rinfo = ojcinfo->rinfo;
2166 : :
2167 : 518 : found = true;
2168 : : /* remove it from the list */
2169 : 518 : root->left_join_clauses =
2170 : 518 : foreach_delete_current(root->left_join_clauses, cell);
2171 : : /* throw back a dummy replacement clause (see notes above) */
2172 : 518 : rinfo = make_restrictinfo(root,
2173 : 518 : (Expr *) makeBoolConst(true, false),
2174 : 518 : rinfo->is_pushed_down,
2175 : 518 : rinfo->has_clone,
2176 : 518 : rinfo->is_clone,
2177 : : false, /* pseudoconstant */
2178 : : 0, /* security_level */
2179 : : rinfo->required_relids,
2180 : : rinfo->incompatible_relids,
2181 : : rinfo->outer_relids);
2182 : 518 : distribute_restrictinfo_to_rels(root, rinfo);
2183 : : }
2184 : : }
2185 : :
2186 : : /* Process the RIGHT JOIN clauses */
2187 [ + + + + : 284727 : foreach(cell, root->right_join_clauses)
+ + ]
2188 : : {
2189 : 29020 : OuterJoinClauseInfo *ojcinfo = (OuterJoinClauseInfo *) lfirst(cell);
2190 : :
2191 [ + + ]: 29020 : if (reconsider_outer_join_clause(root, ojcinfo, false))
2192 : : {
2193 : 1140 : RestrictInfo *rinfo = ojcinfo->rinfo;
2194 : :
2195 : 1140 : found = true;
2196 : : /* remove it from the list */
2197 : 1140 : root->right_join_clauses =
2198 : 1140 : foreach_delete_current(root->right_join_clauses, cell);
2199 : : /* throw back a dummy replacement clause (see notes above) */
2200 : 1140 : rinfo = make_restrictinfo(root,
2201 : 1140 : (Expr *) makeBoolConst(true, false),
2202 : 1140 : rinfo->is_pushed_down,
2203 : 1140 : rinfo->has_clone,
2204 : 1140 : rinfo->is_clone,
2205 : : false, /* pseudoconstant */
2206 : : 0, /* security_level */
2207 : : rinfo->required_relids,
2208 : : rinfo->incompatible_relids,
2209 : : rinfo->outer_relids);
2210 : 1140 : distribute_restrictinfo_to_rels(root, rinfo);
2211 : : }
2212 : : }
2213 : :
2214 : : /* Process the FULL JOIN clauses */
2215 [ + + + + : 256762 : foreach(cell, root->full_join_clauses)
+ + ]
2216 : : {
2217 : 1055 : OuterJoinClauseInfo *ojcinfo = (OuterJoinClauseInfo *) lfirst(cell);
2218 : :
2219 [ + + ]: 1055 : if (reconsider_full_join_clause(root, ojcinfo))
2220 : : {
2221 : 5 : RestrictInfo *rinfo = ojcinfo->rinfo;
2222 : :
2223 : 5 : found = true;
2224 : : /* remove it from the list */
2225 : 5 : root->full_join_clauses =
2226 : 5 : foreach_delete_current(root->full_join_clauses, cell);
2227 : : /* throw back a dummy replacement clause (see notes above) */
2228 : 5 : rinfo = make_restrictinfo(root,
2229 : 5 : (Expr *) makeBoolConst(true, false),
2230 : 5 : rinfo->is_pushed_down,
2231 : 5 : rinfo->has_clone,
2232 : 5 : rinfo->is_clone,
2233 : : false, /* pseudoconstant */
2234 : : 0, /* security_level */
2235 : : rinfo->required_relids,
2236 : : rinfo->incompatible_relids,
2237 : : rinfo->outer_relids);
2238 : 5 : distribute_restrictinfo_to_rels(root, rinfo);
2239 : : }
2240 : : }
2241 [ + + ]: 255707 : } while (found);
2242 : :
2243 : : /* Now, any remaining clauses have to be thrown back */
2244 [ + + + + : 275810 : foreach(cell, root->left_join_clauses)
+ + ]
2245 : : {
2246 : 21756 : OuterJoinClauseInfo *ojcinfo = (OuterJoinClauseInfo *) lfirst(cell);
2247 : :
2248 : 21756 : distribute_restrictinfo_to_rels(root, ojcinfo->rinfo);
2249 : : }
2250 [ + + + + : 280749 : foreach(cell, root->right_join_clauses)
+ + ]
2251 : : {
2252 : 26695 : OuterJoinClauseInfo *ojcinfo = (OuterJoinClauseInfo *) lfirst(cell);
2253 : :
2254 : 26695 : distribute_restrictinfo_to_rels(root, ojcinfo->rinfo);
2255 : : }
2256 [ + + + + : 255104 : foreach(cell, root->full_join_clauses)
+ + ]
2257 : : {
2258 : 1050 : OuterJoinClauseInfo *ojcinfo = (OuterJoinClauseInfo *) lfirst(cell);
2259 : :
2260 : 1050 : distribute_restrictinfo_to_rels(root, ojcinfo->rinfo);
2261 : : }
2262 : 254054 : }
2263 : :
2264 : : /*
2265 : : * reconsider_outer_join_clauses for a single LEFT/RIGHT JOIN clause
2266 : : *
2267 : : * Returns true if we were able to propagate a constant through the clause.
2268 : : */
2269 : : static bool
2270 : 51610 : reconsider_outer_join_clause(PlannerInfo *root, OuterJoinClauseInfo *ojcinfo,
2271 : : bool outer_on_left)
2272 : : {
2273 : 51610 : RestrictInfo *rinfo = ojcinfo->rinfo;
2274 : 51610 : SpecialJoinInfo *sjinfo = ojcinfo->sjinfo;
2275 : : Expr *outervar,
2276 : : *innervar;
2277 : : Oid opno,
2278 : : collation,
2279 : : left_type,
2280 : : right_type,
2281 : : inner_datatype;
2282 : : Relids inner_relids;
2283 : : ListCell *lc1;
2284 : :
2285 : : Assert(is_opclause(rinfo->clause));
2286 : 51610 : opno = ((OpExpr *) rinfo->clause)->opno;
2287 : 51610 : collation = ((OpExpr *) rinfo->clause)->inputcollid;
2288 : :
2289 : : /* Extract needed info from the clause */
2290 : 51610 : op_input_types(opno, &left_type, &right_type);
2291 [ + + ]: 51610 : if (outer_on_left)
2292 : : {
2293 : 22590 : outervar = (Expr *) get_leftop(rinfo->clause);
2294 : 22590 : innervar = (Expr *) get_rightop(rinfo->clause);
2295 : 22590 : inner_datatype = right_type;
2296 : 22590 : inner_relids = rinfo->right_relids;
2297 : : }
2298 : : else
2299 : : {
2300 : 29020 : outervar = (Expr *) get_rightop(rinfo->clause);
2301 : 29020 : innervar = (Expr *) get_leftop(rinfo->clause);
2302 : 29020 : inner_datatype = left_type;
2303 : 29020 : inner_relids = rinfo->left_relids;
2304 : : }
2305 : :
2306 : : /* Scan EquivalenceClasses for a match to outervar */
2307 [ + - + + : 315043 : foreach(lc1, root->eq_classes)
+ + ]
2308 : : {
2309 : 265091 : EquivalenceClass *cur_ec = (EquivalenceClass *) lfirst(lc1);
2310 : : bool match;
2311 : : ListCell *lc2;
2312 : :
2313 : : /* We don't expect any children yet */
2314 : : Assert(cur_ec->ec_childmembers == NULL);
2315 : :
2316 : : /* Ignore EC unless it contains pseudoconstants */
2317 [ + + ]: 265091 : if (!cur_ec->ec_has_const)
2318 : 206024 : continue;
2319 : : /* Never match to a volatile EC */
2320 [ - + ]: 59067 : if (cur_ec->ec_has_volatile)
2321 : 0 : continue;
2322 : : /* It has to match the outer-join clause as to semantics, too */
2323 [ + + ]: 59067 : if (collation != cur_ec->ec_collation)
2324 : 3922 : continue;
2325 [ + + ]: 55145 : if (!equal(rinfo->mergeopfamilies, cur_ec->ec_opfamilies))
2326 : 10130 : continue;
2327 : : /* Does it contain a match to outervar? */
2328 : 45015 : match = false;
2329 [ + - + + : 143302 : foreach(lc2, cur_ec->ec_members)
+ + ]
2330 : : {
2331 : 99945 : EquivalenceMember *cur_em = (EquivalenceMember *) lfirst(lc2);
2332 : :
2333 : : /* Child members should not exist in ec_members */
2334 : : Assert(!cur_em->em_is_child);
2335 [ + + ]: 99945 : if (equal(outervar, cur_em->em_expr))
2336 : : {
2337 : 1658 : match = true;
2338 : 1658 : break;
2339 : : }
2340 : : }
2341 [ + + ]: 45015 : if (!match)
2342 : 43357 : continue; /* no match, so ignore this EC */
2343 : :
2344 : : /*
2345 : : * Yes it does! Try to generate a clause INNERVAR = CONSTANT for each
2346 : : * CONSTANT in the EC. Note that we must succeed with at least one
2347 : : * constant before we can decide to throw away the outer-join clause.
2348 : : */
2349 : 1658 : match = false;
2350 [ + - + + : 5867 : foreach(lc2, cur_ec->ec_members)
+ + ]
2351 : : {
2352 : 4209 : EquivalenceMember *cur_em = (EquivalenceMember *) lfirst(lc2);
2353 : : Oid eq_op;
2354 : : RestrictInfo *newrinfo;
2355 : : JoinDomain *jdomain;
2356 : :
2357 [ + + ]: 4209 : if (!cur_em->em_is_const)
2358 : 2516 : continue; /* ignore non-const members */
2359 : 1693 : eq_op = select_equality_operator(cur_ec,
2360 : : inner_datatype,
2361 : : cur_em->em_datatype);
2362 [ - + ]: 1693 : if (!OidIsValid(eq_op))
2363 : 0 : continue; /* can't generate equality */
2364 : 1693 : newrinfo = build_implied_join_equality(root,
2365 : : eq_op,
2366 : : cur_ec->ec_collation,
2367 : : innervar,
2368 : : cur_em->em_expr,
2369 : : bms_copy(inner_relids),
2370 : : cur_ec->ec_min_security);
2371 : : /* This equality holds within the OJ's child JoinDomain */
2372 : 1693 : jdomain = find_join_domain(root, sjinfo->syn_righthand);
2373 [ + - ]: 1693 : if (process_equivalence(root, &newrinfo, jdomain))
2374 : 1693 : match = true;
2375 : : }
2376 : :
2377 : : /*
2378 : : * If we were able to equate INNERVAR to any constant, report success.
2379 : : * Otherwise, fall out of the search loop, since we know the OUTERVAR
2380 : : * appears in at most one EC.
2381 : : */
2382 [ + - ]: 1658 : if (match)
2383 : 1658 : return true;
2384 : : else
2385 : 0 : break;
2386 : : }
2387 : :
2388 : 49952 : return false; /* failed to make any deduction */
2389 : : }
2390 : :
2391 : : /*
2392 : : * reconsider_outer_join_clauses for a single FULL JOIN clause
2393 : : *
2394 : : * Returns true if we were able to propagate a constant through the clause.
2395 : : */
2396 : : static bool
2397 : 1055 : reconsider_full_join_clause(PlannerInfo *root, OuterJoinClauseInfo *ojcinfo)
2398 : : {
2399 : 1055 : RestrictInfo *rinfo = ojcinfo->rinfo;
2400 : 1055 : SpecialJoinInfo *sjinfo = ojcinfo->sjinfo;
2401 : 1055 : Relids fjrelids = bms_make_singleton(sjinfo->ojrelid);
2402 : : Expr *leftvar;
2403 : : Expr *rightvar;
2404 : : Oid opno,
2405 : : collation,
2406 : : left_type,
2407 : : right_type;
2408 : : Relids left_relids,
2409 : : right_relids;
2410 : : ListCell *lc1;
2411 : :
2412 : : /* Extract needed info from the clause */
2413 : : Assert(is_opclause(rinfo->clause));
2414 : 1055 : opno = ((OpExpr *) rinfo->clause)->opno;
2415 : 1055 : collation = ((OpExpr *) rinfo->clause)->inputcollid;
2416 : 1055 : op_input_types(opno, &left_type, &right_type);
2417 : 1055 : leftvar = (Expr *) get_leftop(rinfo->clause);
2418 : 1055 : rightvar = (Expr *) get_rightop(rinfo->clause);
2419 : 1055 : left_relids = rinfo->left_relids;
2420 : 1055 : right_relids = rinfo->right_relids;
2421 : :
2422 [ + - + + : 5335 : foreach(lc1, root->eq_classes)
+ + ]
2423 : : {
2424 : 4285 : EquivalenceClass *cur_ec = (EquivalenceClass *) lfirst(lc1);
2425 : 4285 : EquivalenceMember *coal_em = NULL;
2426 : : bool match;
2427 : : bool matchleft;
2428 : : bool matchright;
2429 : : ListCell *lc2;
2430 : 4285 : int coal_idx = -1;
2431 : :
2432 : : /* We don't expect any children yet */
2433 : : Assert(cur_ec->ec_childmembers == NULL);
2434 : :
2435 : : /* Ignore EC unless it contains pseudoconstants */
2436 [ + + ]: 4285 : if (!cur_ec->ec_has_const)
2437 : 4035 : continue;
2438 : : /* Never match to a volatile EC */
2439 [ - + ]: 250 : if (cur_ec->ec_has_volatile)
2440 : 0 : continue;
2441 : : /* It has to match the outer-join clause as to semantics, too */
2442 [ + + ]: 250 : if (collation != cur_ec->ec_collation)
2443 : 30 : continue;
2444 [ - + ]: 220 : if (!equal(rinfo->mergeopfamilies, cur_ec->ec_opfamilies))
2445 : 0 : continue;
2446 : :
2447 : : /*
2448 : : * Does it contain a COALESCE(leftvar, rightvar) construct?
2449 : : *
2450 : : * We can assume the COALESCE() inputs are in the same order as the
2451 : : * join clause, since both were automatically generated in the cases
2452 : : * we care about.
2453 : : *
2454 : : * XXX currently this may fail to match in cross-type cases because
2455 : : * the COALESCE will contain typecast operations while the join clause
2456 : : * may not (if there is a cross-type mergejoin operator available for
2457 : : * the two column types). Is it OK to strip implicit coercions from
2458 : : * the COALESCE arguments?
2459 : : */
2460 : 220 : match = false;
2461 [ + - + + : 645 : foreach(lc2, cur_ec->ec_members)
+ + ]
2462 : : {
2463 : 430 : coal_em = (EquivalenceMember *) lfirst(lc2);
2464 : :
2465 : : /* Child members should not exist in ec_members */
2466 : : Assert(!coal_em->em_is_child);
2467 [ + + ]: 430 : if (IsA(coal_em->em_expr, CoalesceExpr))
2468 : : {
2469 : 15 : CoalesceExpr *cexpr = (CoalesceExpr *) coal_em->em_expr;
2470 : : Node *cfirst;
2471 : : Node *csecond;
2472 : :
2473 [ - + ]: 15 : if (list_length(cexpr->args) != 2)
2474 : 0 : continue;
2475 : 15 : cfirst = (Node *) linitial(cexpr->args);
2476 : 15 : csecond = (Node *) lsecond(cexpr->args);
2477 : :
2478 : : /*
2479 : : * The COALESCE arguments will be marked as possibly nulled by
2480 : : * the full join, while we wish to generate clauses that apply
2481 : : * to the join's inputs. So we must strip the join from the
2482 : : * nullingrels fields of cfirst/csecond before comparing them
2483 : : * to leftvar/rightvar. (Perhaps with a less hokey
2484 : : * representation for FULL JOIN USING output columns, this
2485 : : * wouldn't be needed?)
2486 : : */
2487 : 15 : cfirst = remove_nulling_relids(cfirst, fjrelids, NULL);
2488 : 15 : csecond = remove_nulling_relids(csecond, fjrelids, NULL);
2489 : :
2490 [ + + + - ]: 15 : if (equal(leftvar, cfirst) && equal(rightvar, csecond))
2491 : : {
2492 : 5 : coal_idx = foreach_current_index(lc2);
2493 : 5 : match = true;
2494 : 5 : break;
2495 : : }
2496 : : }
2497 : : }
2498 [ + + ]: 220 : if (!match)
2499 : 215 : continue; /* no match, so ignore this EC */
2500 : :
2501 : : /*
2502 : : * Yes it does! Try to generate clauses LEFTVAR = CONSTANT and
2503 : : * RIGHTVAR = CONSTANT for each CONSTANT in the EC. Note that we must
2504 : : * succeed with at least one constant for each var before we can
2505 : : * decide to throw away the outer-join clause.
2506 : : */
2507 : 5 : matchleft = matchright = false;
2508 [ + - + + : 15 : foreach(lc2, cur_ec->ec_members)
+ + ]
2509 : : {
2510 : 10 : EquivalenceMember *cur_em = (EquivalenceMember *) lfirst(lc2);
2511 : : Oid eq_op;
2512 : : RestrictInfo *newrinfo;
2513 : : JoinDomain *jdomain;
2514 : :
2515 [ + + ]: 10 : if (!cur_em->em_is_const)
2516 : 5 : continue; /* ignore non-const members */
2517 : 5 : eq_op = select_equality_operator(cur_ec,
2518 : : left_type,
2519 : : cur_em->em_datatype);
2520 [ + - ]: 5 : if (OidIsValid(eq_op))
2521 : : {
2522 : 5 : newrinfo = build_implied_join_equality(root,
2523 : : eq_op,
2524 : : cur_ec->ec_collation,
2525 : : leftvar,
2526 : : cur_em->em_expr,
2527 : : bms_copy(left_relids),
2528 : : cur_ec->ec_min_security);
2529 : : /* This equality holds within the lefthand child JoinDomain */
2530 : 5 : jdomain = find_join_domain(root, sjinfo->syn_lefthand);
2531 [ + - ]: 5 : if (process_equivalence(root, &newrinfo, jdomain))
2532 : 5 : matchleft = true;
2533 : : }
2534 : 5 : eq_op = select_equality_operator(cur_ec,
2535 : : right_type,
2536 : : cur_em->em_datatype);
2537 [ + - ]: 5 : if (OidIsValid(eq_op))
2538 : : {
2539 : 5 : newrinfo = build_implied_join_equality(root,
2540 : : eq_op,
2541 : : cur_ec->ec_collation,
2542 : : rightvar,
2543 : : cur_em->em_expr,
2544 : : bms_copy(right_relids),
2545 : : cur_ec->ec_min_security);
2546 : : /* This equality holds within the righthand child JoinDomain */
2547 : 5 : jdomain = find_join_domain(root, sjinfo->syn_righthand);
2548 [ + - ]: 5 : if (process_equivalence(root, &newrinfo, jdomain))
2549 : 5 : matchright = true;
2550 : : }
2551 : : }
2552 : :
2553 : : /*
2554 : : * If we were able to equate both vars to constants, we're done, and
2555 : : * we can throw away the full-join clause as redundant. Moreover, we
2556 : : * can remove the COALESCE entry from the EC, since the added
2557 : : * restrictions ensure it will always have the expected value. (We
2558 : : * don't bother trying to update ec_relids or ec_sources.)
2559 : : */
2560 [ + - + - ]: 5 : if (matchleft && matchright)
2561 : : {
2562 : 5 : cur_ec->ec_members = list_delete_nth_cell(cur_ec->ec_members, coal_idx);
2563 : 5 : return true;
2564 : : }
2565 : :
2566 : : /*
2567 : : * Otherwise, fall out of the search loop, since we know the COALESCE
2568 : : * appears in at most one EC (XXX might stop being true if we allow
2569 : : * stripping of coercions above?)
2570 : : */
2571 : 0 : break;
2572 : : }
2573 : :
2574 : 1050 : return false; /* failed to make any deduction */
2575 : : }
2576 : :
2577 : : /*
2578 : : * find_join_domain
2579 : : * Find the highest JoinDomain enclosed within the given relid set.
2580 : : *
2581 : : * (We could avoid this search at the cost of complicating APIs elsewhere,
2582 : : * which doesn't seem worth it.)
2583 : : */
2584 : : static JoinDomain *
2585 : 1703 : find_join_domain(PlannerInfo *root, Relids relids)
2586 : : {
2587 : : ListCell *lc;
2588 : :
2589 [ + - + - : 3491 : foreach(lc, root->join_domains)
+ - ]
2590 : : {
2591 : 3491 : JoinDomain *jdomain = (JoinDomain *) lfirst(lc);
2592 : :
2593 [ + + ]: 3491 : if (bms_is_subset(jdomain->jd_relids, relids))
2594 : 1703 : return jdomain;
2595 : : }
2596 [ # # ]: 0 : elog(ERROR, "failed to find appropriate JoinDomain");
2597 : : return NULL; /* keep compiler quiet */
2598 : : }
2599 : :
2600 : :
2601 : : /*
2602 : : * exprs_known_equal
2603 : : * Detect whether two expressions are known equal due to equivalence
2604 : : * relationships.
2605 : : *
2606 : : * If opfamily is given, the expressions must be known equal per the semantics
2607 : : * of that opfamily (note it has to be a btree opfamily, since those are the
2608 : : * only opfamilies equivclass.c deals with). If opfamily is InvalidOid, we'll
2609 : : * return true if they're equal according to any opfamily, which is fuzzy but
2610 : : * OK for estimation purposes.
2611 : : *
2612 : : * Note: does not bother to check for "equal(item1, item2)"; caller must
2613 : : * check that case if it's possible to pass identical items.
2614 : : */
2615 : : bool
2616 : 25318 : exprs_known_equal(PlannerInfo *root, Node *item1, Node *item2, Oid opfamily)
2617 : : {
2618 : : ListCell *lc1;
2619 : :
2620 [ + + + + : 193411 : foreach(lc1, root->eq_classes)
+ + ]
2621 : : {
2622 : 170919 : EquivalenceClass *ec = (EquivalenceClass *) lfirst(lc1);
2623 : 170919 : bool item1member = false;
2624 : 170919 : bool item2member = false;
2625 : : ListCell *lc2;
2626 : :
2627 : : /* Never match to a volatile EC */
2628 [ - + ]: 170919 : if (ec->ec_has_volatile)
2629 : 0 : continue;
2630 : :
2631 : : /*
2632 : : * It's okay to consider ec_broken ECs here. Brokenness just means we
2633 : : * couldn't derive all the implied clauses we'd have liked to; it does
2634 : : * not invalidate our knowledge that the members are equal.
2635 : : */
2636 : :
2637 : : /* Ignore if this EC doesn't use specified opfamily */
2638 [ + + ]: 170919 : if (OidIsValid(opfamily) &&
2639 [ + + ]: 550 : !list_member_oid(ec->ec_opfamilies, opfamily))
2640 : 190 : continue;
2641 : :
2642 : : /* Ignore children here */
2643 [ + - + + : 408525 : foreach(lc2, ec->ec_members)
+ + ]
2644 : : {
2645 : 240622 : EquivalenceMember *em = (EquivalenceMember *) lfirst(lc2);
2646 : :
2647 : : /* Child members should not exist in ec_members */
2648 : : Assert(!em->em_is_child);
2649 [ + + ]: 240622 : if (equal(item1, em->em_expr))
2650 : 12723 : item1member = true;
2651 [ + + ]: 227899 : else if (equal(item2, em->em_expr))
2652 : 20820 : item2member = true;
2653 : : /* Exit as soon as equality is proven */
2654 [ + + + + ]: 240622 : if (item1member && item2member)
2655 : 2826 : return true;
2656 : : }
2657 : : }
2658 : 22492 : return false;
2659 : : }
2660 : :
2661 : :
2662 : : /*
2663 : : * match_eclasses_to_foreign_key_col
2664 : : * See whether a foreign key column match is proven by any eclass.
2665 : : *
2666 : : * If the referenced and referencing Vars of the fkey's colno'th column are
2667 : : * known equal due to any eclass, return that eclass; otherwise return NULL.
2668 : : * (In principle there might be more than one matching eclass if multiple
2669 : : * collations are involved, but since collation doesn't matter for equality,
2670 : : * we ignore that fine point here.) This is much like exprs_known_equal,
2671 : : * except for the format of the input.
2672 : : *
2673 : : * On success, we also set fkinfo->eclass[colno] to the matching eclass,
2674 : : * and set fkinfo->fk_eclass_member[colno] to the eclass member for the
2675 : : * referencing Var.
2676 : : */
2677 : : EquivalenceClass *
2678 : 1944 : match_eclasses_to_foreign_key_col(PlannerInfo *root,
2679 : : ForeignKeyOptInfo *fkinfo,
2680 : : int colno)
2681 : : {
2682 : 1944 : Index var1varno = fkinfo->con_relid;
2683 : 1944 : AttrNumber var1attno = fkinfo->conkey[colno];
2684 : 1944 : Index var2varno = fkinfo->ref_relid;
2685 : 1944 : AttrNumber var2attno = fkinfo->confkey[colno];
2686 : 1944 : Oid eqop = fkinfo->conpfeqop[colno];
2687 : 1944 : RelOptInfo *rel1 = root->simple_rel_array[var1varno];
2688 : 1944 : RelOptInfo *rel2 = root->simple_rel_array[var2varno];
2689 : 1944 : List *opfamilies = NIL; /* compute only if needed */
2690 : : Bitmapset *matching_ecs;
2691 : : int i;
2692 : :
2693 : : /* Consider only eclasses mentioning both relations */
2694 : : Assert(root->ec_merging_done);
2695 : : Assert(IS_SIMPLE_REL(rel1));
2696 : : Assert(IS_SIMPLE_REL(rel2));
2697 : 1944 : matching_ecs = bms_intersect(rel1->eclass_indexes,
2698 : 1944 : rel2->eclass_indexes);
2699 : :
2700 : 1944 : i = -1;
2701 [ + + ]: 2024 : while ((i = bms_next_member(matching_ecs, i)) >= 0)
2702 : : {
2703 : 421 : EquivalenceClass *ec = (EquivalenceClass *) list_nth(root->eq_classes,
2704 : : i);
2705 : 421 : EquivalenceMember *item1_em = NULL;
2706 : 421 : EquivalenceMember *item2_em = NULL;
2707 : : ListCell *lc2;
2708 : :
2709 : : /* Never match to a volatile EC */
2710 [ - + ]: 421 : if (ec->ec_has_volatile)
2711 : 0 : continue;
2712 : :
2713 : : /*
2714 : : * It's okay to consider "broken" ECs here, see exprs_known_equal.
2715 : : * Ignore children here.
2716 : : */
2717 [ + - + + : 1007 : foreach(lc2, ec->ec_members)
+ + ]
2718 : : {
2719 : 927 : EquivalenceMember *em = (EquivalenceMember *) lfirst(lc2);
2720 : : Var *var;
2721 : :
2722 : : /* Child members should not exist in ec_members */
2723 : : Assert(!em->em_is_child);
2724 : :
2725 : : /* EM must be a Var, possibly with RelabelType */
2726 : 927 : var = (Var *) em->em_expr;
2727 [ + - - + ]: 927 : while (var && IsA(var, RelabelType))
2728 : 0 : var = (Var *) ((RelabelType *) var)->arg;
2729 [ + - + + ]: 927 : if (!(var && IsA(var, Var)))
2730 : 5 : continue;
2731 : :
2732 : : /* Match? */
2733 [ + + + + ]: 922 : if (var->varno == var1varno && var->varattno == var1attno)
2734 : 341 : item1_em = em;
2735 [ + + + + ]: 581 : else if (var->varno == var2varno && var->varattno == var2attno)
2736 : 341 : item2_em = em;
2737 : :
2738 : : /* Have we found both PK and FK column in this EC? */
2739 [ + + + + ]: 922 : if (item1_em && item2_em)
2740 : : {
2741 : : /*
2742 : : * Succeed if eqop matches EC's opfamilies. We could test
2743 : : * this before scanning the members, but it's probably cheaper
2744 : : * to test for member matches first.
2745 : : */
2746 [ + - ]: 341 : if (opfamilies == NIL) /* compute if we didn't already */
2747 : 341 : opfamilies = get_mergejoin_opfamilies(eqop);
2748 [ + - ]: 341 : if (equal(opfamilies, ec->ec_opfamilies))
2749 : : {
2750 : 341 : fkinfo->eclass[colno] = ec;
2751 : 341 : fkinfo->fk_eclass_member[colno] = item2_em;
2752 : 341 : return ec;
2753 : : }
2754 : : /* Otherwise, done with this EC, move on to the next */
2755 : 0 : break;
2756 : : }
2757 : : }
2758 : : }
2759 : 1603 : return NULL;
2760 : : }
2761 : :
2762 : : /*
2763 : : * find_derived_clause_for_ec_member
2764 : : * Search for a previously-derived clause mentioning the given EM.
2765 : : *
2766 : : * The eclass should be an ec_has_const EC, of which the EM is a non-const
2767 : : * member. This should ensure there is just one derived clause mentioning
2768 : : * the EM (and equating it to a constant).
2769 : : * Returns NULL if no such clause can be found.
2770 : : */
2771 : : RestrictInfo *
2772 : 5 : find_derived_clause_for_ec_member(PlannerInfo *root,
2773 : : EquivalenceClass *ec,
2774 : : EquivalenceMember *em)
2775 : : {
2776 : : Assert(ec->ec_has_const);
2777 : : Assert(!em->em_is_const);
2778 : :
2779 : 5 : return ec_search_derived_clause_for_ems(root, ec, em, NULL, NULL);
2780 : : }
2781 : :
2782 : :
2783 : : /*
2784 : : * add_child_rel_equivalences
2785 : : * Search for EC members that reference the root parent of child_rel, and
2786 : : * add transformed members referencing the child_rel.
2787 : : *
2788 : : * Note that this function won't be called at all unless we have at least some
2789 : : * reason to believe that the EC members it generates will be useful.
2790 : : *
2791 : : * parent_rel and child_rel could be derived from appinfo, but since the
2792 : : * caller has already computed them, we might as well just pass them in.
2793 : : *
2794 : : * The passed-in AppendRelInfo is not used when the parent_rel is not a
2795 : : * top-level baserel, since it shows the mapping from the parent_rel but
2796 : : * we need to translate EC expressions that refer to the top-level parent.
2797 : : * Using it is faster than using adjust_appendrel_attrs_multilevel(), though,
2798 : : * so we prefer it when we can.
2799 : : */
2800 : : void
2801 : 27987 : add_child_rel_equivalences(PlannerInfo *root,
2802 : : AppendRelInfo *appinfo,
2803 : : RelOptInfo *parent_rel,
2804 : : RelOptInfo *child_rel)
2805 : : {
2806 : 27987 : Relids top_parent_relids = child_rel->top_parent_relids;
2807 : 27987 : Relids child_relids = child_rel->relids;
2808 : : int i;
2809 : :
2810 : : /*
2811 : : * EC merging should be complete already, so we can use the parent rel's
2812 : : * eclass_indexes to avoid searching all of root->eq_classes.
2813 : : */
2814 : : Assert(root->ec_merging_done);
2815 : : Assert(IS_SIMPLE_REL(parent_rel));
2816 : :
2817 : 27987 : i = -1;
2818 [ + + ]: 79398 : while ((i = bms_next_member(parent_rel->eclass_indexes, i)) >= 0)
2819 : : {
2820 : 51411 : EquivalenceClass *cur_ec = (EquivalenceClass *) list_nth(root->eq_classes, i);
2821 : :
2822 : : /*
2823 : : * If this EC contains a volatile expression, then generating child
2824 : : * EMs would be downright dangerous, so skip it. We rely on a
2825 : : * volatile EC having only one EM.
2826 : : */
2827 [ - + ]: 51411 : if (cur_ec->ec_has_volatile)
2828 : 0 : continue;
2829 : :
2830 : : /* Sanity check eclass_indexes only contain ECs for parent_rel */
2831 : : Assert(bms_is_subset(top_parent_relids, cur_ec->ec_relids));
2832 : :
2833 [ + - + + : 176719 : foreach_node(EquivalenceMember, cur_em, cur_ec->ec_members)
+ + ]
2834 : : {
2835 [ + + ]: 73897 : if (cur_em->em_is_const)
2836 : 2812 : continue; /* ignore consts here */
2837 : :
2838 : : /* Child members should not exist in ec_members */
2839 : : Assert(!cur_em->em_is_child);
2840 : :
2841 : : /*
2842 : : * Consider only members that reference and can be computed at
2843 : : * child's topmost parent rel. In particular we want to exclude
2844 : : * parent-rel Vars that have nonempty varnullingrels. Translating
2845 : : * those might fail, if the transformed expression wouldn't be a
2846 : : * simple Var; and in any case it wouldn't produce a member that
2847 : : * has any use in creating plans for the child rel.
2848 : : */
2849 [ + + ]: 71085 : if (bms_is_subset(cur_em->em_relids, top_parent_relids) &&
2850 [ + - ]: 49442 : !bms_is_empty(cur_em->em_relids))
2851 : : {
2852 : : /* OK, generate transformed child version */
2853 : : Expr *child_expr;
2854 : : Relids new_relids;
2855 : :
2856 [ + + ]: 49442 : if (parent_rel->reloptkind == RELOPT_BASEREL)
2857 : : {
2858 : : /* Simple single-level transformation */
2859 : : child_expr = (Expr *)
2860 : 42245 : adjust_appendrel_attrs(root,
2861 : 42245 : (Node *) cur_em->em_expr,
2862 : : 1, &appinfo);
2863 : : }
2864 : : else
2865 : : {
2866 : : /* Must do multi-level transformation */
2867 : : child_expr = (Expr *)
2868 : 7197 : adjust_appendrel_attrs_multilevel(root,
2869 : 7197 : (Node *) cur_em->em_expr,
2870 : : child_rel,
2871 : 7197 : child_rel->top_parent);
2872 : : }
2873 : :
2874 : : /*
2875 : : * Transform em_relids to match. Note we do *not* do
2876 : : * pull_varnos(child_expr) here, as for example the
2877 : : * transformation might have substituted a constant, but we
2878 : : * don't want the child member to be marked as constant.
2879 : : */
2880 : 49442 : new_relids = bms_difference(cur_em->em_relids,
2881 : : top_parent_relids);
2882 : 49442 : new_relids = bms_add_members(new_relids, child_relids);
2883 : :
2884 : 49442 : add_child_eq_member(root,
2885 : : cur_ec,
2886 : : i,
2887 : : child_expr,
2888 : : new_relids,
2889 : : cur_em->em_jdomain,
2890 : : cur_em,
2891 : : cur_em->em_datatype,
2892 : : child_rel->relid);
2893 : : }
2894 : : }
2895 : : }
2896 : 27987 : }
2897 : :
2898 : : /*
2899 : : * add_child_join_rel_equivalences
2900 : : * Like add_child_rel_equivalences(), but for joinrels
2901 : : *
2902 : : * Here we find the ECs relevant to the top parent joinrel and add transformed
2903 : : * member expressions that refer to this child joinrel.
2904 : : *
2905 : : * Note that this function won't be called at all unless we have at least some
2906 : : * reason to believe that the EC members it generates will be useful.
2907 : : */
2908 : : void
2909 : 15035 : add_child_join_rel_equivalences(PlannerInfo *root,
2910 : : int nappinfos, AppendRelInfo **appinfos,
2911 : : RelOptInfo *parent_joinrel,
2912 : : RelOptInfo *child_joinrel)
2913 : : {
2914 : 15035 : Relids top_parent_relids = child_joinrel->top_parent_relids;
2915 : 15035 : Relids child_relids = child_joinrel->relids;
2916 : : Bitmapset *matching_ecs;
2917 : : MemoryContext oldcontext;
2918 : : int i;
2919 : :
2920 : : Assert(IS_JOIN_REL(child_joinrel) && IS_JOIN_REL(parent_joinrel));
2921 : :
2922 : : /* We need consider only ECs that mention the parent joinrel */
2923 : 15035 : matching_ecs = get_eclass_indexes_for_relids(root, top_parent_relids);
2924 : :
2925 : : /*
2926 : : * If we're being called during GEQO join planning, we still have to
2927 : : * create any new EC members in the main planner context, to avoid having
2928 : : * a corrupt EC data structure after the GEQO context is reset. This is
2929 : : * problematic since we'll leak memory across repeated GEQO cycles. For
2930 : : * now, though, bloat is better than crash. If it becomes a real issue
2931 : : * we'll have to do something to avoid generating duplicate EC members.
2932 : : */
2933 : 15035 : oldcontext = MemoryContextSwitchTo(root->planner_cxt);
2934 : :
2935 : 15035 : i = -1;
2936 [ + + ]: 40371 : while ((i = bms_next_member(matching_ecs, i)) >= 0)
2937 : : {
2938 : 25336 : EquivalenceClass *cur_ec = (EquivalenceClass *) list_nth(root->eq_classes, i);
2939 : :
2940 : : /*
2941 : : * If this EC contains a volatile expression, then generating child
2942 : : * EMs would be downright dangerous, so skip it. We rely on a
2943 : : * volatile EC having only one EM.
2944 : : */
2945 [ - + ]: 25336 : if (cur_ec->ec_has_volatile)
2946 : 0 : continue;
2947 : :
2948 : : /* Sanity check on get_eclass_indexes_for_relids result */
2949 : : Assert(bms_overlap(top_parent_relids, cur_ec->ec_relids));
2950 : :
2951 [ + - + + : 94422 : foreach_node(EquivalenceMember, cur_em, cur_ec->ec_members)
+ + ]
2952 : : {
2953 [ + + ]: 43750 : if (cur_em->em_is_const)
2954 : 2018 : continue; /* ignore consts here */
2955 : :
2956 : : /* Child members should not exist in ec_members */
2957 : : Assert(!cur_em->em_is_child);
2958 : :
2959 : : /*
2960 : : * We may ignore expressions that reference a single baserel,
2961 : : * because add_child_rel_equivalences should have handled them.
2962 : : */
2963 [ + + ]: 41732 : if (bms_membership(cur_em->em_relids) != BMS_MULTIPLE)
2964 : 39435 : continue;
2965 : :
2966 : : /* Does this member reference child's topmost parent rel? */
2967 [ + - ]: 2297 : if (bms_overlap(cur_em->em_relids, top_parent_relids))
2968 : : {
2969 : : /* Yes, generate transformed child version */
2970 : : Expr *child_expr;
2971 : : Relids new_relids;
2972 : :
2973 [ + + ]: 2297 : if (parent_joinrel->reloptkind == RELOPT_JOINREL)
2974 : : {
2975 : : /* Simple single-level transformation */
2976 : : child_expr = (Expr *)
2977 : 2217 : adjust_appendrel_attrs(root,
2978 : 2217 : (Node *) cur_em->em_expr,
2979 : : nappinfos, appinfos);
2980 : : }
2981 : : else
2982 : : {
2983 : : /* Must do multi-level transformation */
2984 : : Assert(parent_joinrel->reloptkind == RELOPT_OTHER_JOINREL);
2985 : : child_expr = (Expr *)
2986 : 80 : adjust_appendrel_attrs_multilevel(root,
2987 : 80 : (Node *) cur_em->em_expr,
2988 : : child_joinrel,
2989 : 80 : child_joinrel->top_parent);
2990 : : }
2991 : :
2992 : : /*
2993 : : * Transform em_relids to match. Note we do *not* do
2994 : : * pull_varnos(child_expr) here, as for example the
2995 : : * transformation might have substituted a constant, but we
2996 : : * don't want the child member to be marked as constant.
2997 : : */
2998 : 2297 : new_relids = bms_difference(cur_em->em_relids,
2999 : : top_parent_relids);
3000 : 2297 : new_relids = bms_add_members(new_relids, child_relids);
3001 : :
3002 : : /*
3003 : : * Add new child member to the EquivalenceClass. Because this
3004 : : * is a RELOPT_OTHER_JOINREL which has multiple component
3005 : : * relids, there is no ideal place to store these members in
3006 : : * the class. Ordinarily, child members are stored in the
3007 : : * ec_childmembers[] array element corresponding to their
3008 : : * relid, however, here we have multiple component relids, so
3009 : : * there's no single ec_childmembers[] array element to store
3010 : : * this member. So that we still correctly find this member
3011 : : * in loops iterating over an EquivalenceMemberIterator, we
3012 : : * opt to store the member in the ec_childmembers array in
3013 : : * only the first component relid slot of the array. This
3014 : : * allows the member to be found, providing callers of
3015 : : * setup_eclass_member_iterator() specify all the component
3016 : : * relids for the RELOPT_OTHER_JOINREL, which they do. If we
3017 : : * opted to store the member in each ec_childmembers[] element
3018 : : * for all the component relids, then that would just result
3019 : : * in eclass_member_iterator_next() finding the member
3020 : : * multiple times, which is a waste of effort.
3021 : : */
3022 : 2297 : add_child_eq_member(root,
3023 : : cur_ec,
3024 : : -1,
3025 : : child_expr,
3026 : : new_relids,
3027 : : cur_em->em_jdomain,
3028 : : cur_em,
3029 : : cur_em->em_datatype,
3030 : 2297 : bms_next_member(child_joinrel->relids, -1));
3031 : : }
3032 : : }
3033 : : }
3034 : :
3035 : 15035 : MemoryContextSwitchTo(oldcontext);
3036 : 15035 : }
3037 : :
3038 : : /*
3039 : : * add_setop_child_rel_equivalences
3040 : : * Add equivalence members for each non-resjunk target in 'child_tlist'
3041 : : * to the EquivalenceClass in the corresponding setop_pathkey's pk_eclass.
3042 : : *
3043 : : * 'root' is the PlannerInfo belonging to the top-level set operation.
3044 : : * 'child_rel' is the RelOptInfo of the child relation we're adding
3045 : : * EquivalenceMembers for.
3046 : : * 'child_tlist' is the target list for the setop child relation. The target
3047 : : * list expressions are what we add as EquivalenceMembers.
3048 : : * 'setop_pathkeys' is a list of PathKeys which must contain an entry for each
3049 : : * non-resjunk target in 'child_tlist'.
3050 : : */
3051 : : void
3052 : 10285 : add_setop_child_rel_equivalences(PlannerInfo *root, RelOptInfo *child_rel,
3053 : : List *child_tlist, List *setop_pathkeys)
3054 : : {
3055 : : ListCell *lc;
3056 : 10285 : ListCell *lc2 = list_head(setop_pathkeys);
3057 : :
3058 [ + - + + : 40926 : foreach(lc, child_tlist)
+ + ]
3059 : : {
3060 : 30641 : TargetEntry *tle = lfirst_node(TargetEntry, lc);
3061 : : EquivalenceMember *parent_em;
3062 : : PathKey *pk;
3063 : :
3064 [ - + ]: 30641 : if (tle->resjunk)
3065 : 0 : continue;
3066 : :
3067 [ - + ]: 30641 : if (lc2 == NULL)
3068 [ # # ]: 0 : elog(ERROR, "too few pathkeys for set operation");
3069 : :
3070 : 30641 : pk = lfirst_node(PathKey, lc2);
3071 : 30641 : parent_em = linitial(pk->pk_eclass->ec_members);
3072 : :
3073 : : /*
3074 : : * We can safely pass the parent member as the first member in the
3075 : : * ec_members list as this is added first in generate_union_paths,
3076 : : * likewise, the JoinDomain can be that of the initial member of the
3077 : : * Pathkey's EquivalenceClass. We pass -1 for ec_index since we
3078 : : * maintain the eclass_indexes for the child_rel after the loop.
3079 : : */
3080 : 30641 : add_child_eq_member(root,
3081 : : pk->pk_eclass,
3082 : : -1,
3083 : : tle->expr,
3084 : : child_rel->relids,
3085 : : parent_em->em_jdomain,
3086 : : parent_em,
3087 : 30641 : exprType((Node *) tle->expr),
3088 : : child_rel->relid);
3089 : :
3090 : 30641 : lc2 = lnext(setop_pathkeys, lc2);
3091 : : }
3092 : :
3093 : : /*
3094 : : * transformSetOperationStmt() ensures that the targetlist never contains
3095 : : * any resjunk columns, so all eclasses that exist in 'root' must have
3096 : : * received a new member in the loop above. Add them to the child_rel's
3097 : : * eclass_indexes.
3098 : : */
3099 : 10285 : child_rel->eclass_indexes = bms_add_range(child_rel->eclass_indexes, 0,
3100 : 10285 : list_length(root->eq_classes) - 1);
3101 : 10285 : }
3102 : :
3103 : : /*
3104 : : * setup_eclass_member_iterator
3105 : : * Setup an EquivalenceMemberIterator 'it' to iterate over all parent
3106 : : * EquivalenceMembers and child members belonging to the given 'ec'.
3107 : : *
3108 : : * This iterator returns:
3109 : : * - All parent members stored directly in ec_members for 'ec', and;
3110 : : * - Any child member added to the given ec by add_child_eq_member() where
3111 : : * the child_relid specified in the add_child_eq_member() call is a member
3112 : : * of the 'child_relids' parameter.
3113 : : *
3114 : : * Note:
3115 : : * The given 'child_relids' must remain allocated and not be changed for the
3116 : : * lifetime of the iterator.
3117 : : *
3118 : : * Parameters:
3119 : : * 'it' is a pointer to the iterator to set up. Normally stack allocated.
3120 : : * 'ec' is the EquivalenceClass from which to iterate members for.
3121 : : * 'child_relids' is the relids to return child members for.
3122 : : */
3123 : : void
3124 : 3910181 : setup_eclass_member_iterator(EquivalenceMemberIterator *it,
3125 : : EquivalenceClass *ec, Relids child_relids)
3126 : : {
3127 : 3910181 : it->ec = ec;
3128 : : /* no need to set this if the class has no child members array set */
3129 [ + + ]: 3910181 : it->child_relids = ec->ec_childmembers != NULL ? child_relids : NULL;
3130 : 3910181 : it->current_relid = -1;
3131 : 3910181 : it->current_list = ec->ec_members;
3132 : 3910181 : it->current_cell = list_head(it->current_list);
3133 : 3910181 : }
3134 : :
3135 : : /*
3136 : : * eclass_member_iterator_next
3137 : : * Get the next EquivalenceMember from the EquivalenceMemberIterator 'it',
3138 : : * as setup by setup_eclass_member_iterator(). NULL is returned if there
3139 : : * are no members left, after which callers must not call
3140 : : * eclass_member_iterator_next() again for the given iterator.
3141 : : */
3142 : : EquivalenceMember *
3143 : 9098756 : eclass_member_iterator_next(EquivalenceMemberIterator *it)
3144 : : {
3145 [ + - ]: 9098756 : while (it->current_list != NULL)
3146 : : {
3147 [ + + ]: 9098756 : while (it->current_cell != NULL)
3148 : : {
3149 : : EquivalenceMember *em;
3150 : :
3151 : 6296171 : nextcell:
3152 : 6406190 : em = lfirst_node(EquivalenceMember, it->current_cell);
3153 : 6406190 : it->current_cell = lnext(it->current_list, it->current_cell);
3154 : 6406190 : return em;
3155 : : }
3156 : :
3157 : : /* Search for the next list to return members from */
3158 [ + + ]: 2908637 : while ((it->current_relid = bms_next_member(it->child_relids, it->current_relid)) > 0)
3159 : : {
3160 : : /*
3161 : : * Be paranoid in case we're given relids above what we've sized
3162 : : * the ec_childmembers array to.
3163 : : */
3164 [ - + ]: 216071 : if (it->current_relid >= it->ec->ec_childmembers_size)
3165 : 0 : return NULL;
3166 : :
3167 : 216071 : it->current_list = it->ec->ec_childmembers[it->current_relid];
3168 : :
3169 : : /* If there are members in this list, use it. */
3170 [ + + ]: 216071 : if (it->current_list != NIL)
3171 : : {
3172 : : /* point current_cell to the head of this list */
3173 : 110019 : it->current_cell = list_head(it->current_list);
3174 : 110019 : goto nextcell;
3175 : : }
3176 : : }
3177 : 2692566 : return NULL;
3178 : : }
3179 : :
3180 : 0 : return NULL;
3181 : : }
3182 : :
3183 : : /*
3184 : : * generate_implied_equalities_for_column
3185 : : * Create EC-derived joinclauses usable with a specific column.
3186 : : *
3187 : : * This is used by indxpath.c to extract potentially indexable joinclauses
3188 : : * from ECs, and can be used by foreign data wrappers for similar purposes.
3189 : : * We assume that only expressions in Vars of a single table are of interest,
3190 : : * but the caller provides a callback function to identify exactly which
3191 : : * such expressions it would like to know about.
3192 : : *
3193 : : * We assume that any given table/index column could appear in only one EC.
3194 : : * (This should be true in all but the most pathological cases, and if it
3195 : : * isn't, we stop on the first match anyway.) Therefore, what we return
3196 : : * is a redundant list of clauses equating the table/index column to each of
3197 : : * the other-relation values it is known to be equal to. Any one of
3198 : : * these clauses can be used to create a parameterized path, and there
3199 : : * is no value in using more than one. (But it *is* worthwhile to create
3200 : : * a separate parameterized path for each one, since that leads to different
3201 : : * join orders.)
3202 : : *
3203 : : * The caller can pass a Relids set of rels we aren't interested in joining
3204 : : * to, so as to save the work of creating useless clauses.
3205 : : */
3206 : : List *
3207 : 462352 : generate_implied_equalities_for_column(PlannerInfo *root,
3208 : : RelOptInfo *rel,
3209 : : ec_matches_callback_type callback,
3210 : : void *callback_arg,
3211 : : Relids prohibited_rels)
3212 : : {
3213 : 462352 : List *result = NIL;
3214 : 462352 : bool is_child_rel = (rel->reloptkind == RELOPT_OTHER_MEMBER_REL);
3215 : : Relids parent_relids;
3216 : : int i;
3217 : :
3218 : : /* Should be OK to rely on eclass_indexes */
3219 : : Assert(root->ec_merging_done);
3220 : :
3221 : : /* Indexes are available only on base or "other" member relations. */
3222 : : Assert(IS_SIMPLE_REL(rel));
3223 : :
3224 : : /* If it's a child rel, we'll need to know what its parent(s) are */
3225 [ + + ]: 462352 : if (is_child_rel)
3226 : 10088 : parent_relids = find_childrel_parents(root, rel);
3227 : : else
3228 : 452264 : parent_relids = NULL; /* not used, but keep compiler quiet */
3229 : :
3230 : 462352 : i = -1;
3231 [ + + ]: 1331039 : while ((i = bms_next_member(rel->eclass_indexes, i)) >= 0)
3232 : : {
3233 : 963302 : EquivalenceClass *cur_ec = (EquivalenceClass *) list_nth(root->eq_classes, i);
3234 : : EquivalenceMemberIterator it;
3235 : : EquivalenceMember *cur_em;
3236 : : ListCell *lc2;
3237 : :
3238 : : /* Sanity check eclass_indexes only contain ECs for rel */
3239 : : Assert(is_child_rel || bms_is_subset(rel->relids, cur_ec->ec_relids));
3240 : :
3241 : : /*
3242 : : * Won't generate joinclauses if const or single-member (the latter
3243 : : * test covers the volatile case too)
3244 : : */
3245 [ + + + + ]: 963302 : if (cur_ec->ec_has_const || list_length(cur_ec->ec_members) <= 1)
3246 : 868527 : continue;
3247 : :
3248 : : /*
3249 : : * Scan members, looking for a match to the target column. Note that
3250 : : * child EC members are considered, but only when they belong to the
3251 : : * target relation. (Unlike regular members, the same expression
3252 : : * could be a child member of more than one EC. Therefore, it's
3253 : : * potentially order-dependent which EC a child relation's target
3254 : : * column gets matched to. This is annoying but it only happens in
3255 : : * corner cases, so for now we live with just reporting the first
3256 : : * match. See also get_eclass_for_sort_expr.)
3257 : : */
3258 : 454046 : setup_eclass_member_iterator(&it, cur_ec, rel->relids);
3259 [ + + ]: 1706326 : while ((cur_em = eclass_member_iterator_next(&it)) != NULL)
3260 : : {
3261 [ + + + + ]: 1347225 : if (bms_equal(cur_em->em_relids, rel->relids) &&
3262 : 454216 : callback(root, rel, cur_ec, cur_em, callback_arg))
3263 : 94775 : break;
3264 : : }
3265 : :
3266 [ + + ]: 454046 : if (!cur_em)
3267 : 359271 : continue;
3268 : :
3269 : : /*
3270 : : * Found our match. Scan the other EC members and attempt to generate
3271 : : * joinclauses. Ignore children here.
3272 : : */
3273 [ + - + + : 289537 : foreach(lc2, cur_ec->ec_members)
+ + ]
3274 : : {
3275 : 194762 : EquivalenceMember *other_em = (EquivalenceMember *) lfirst(lc2);
3276 : : Oid eq_op;
3277 : : RestrictInfo *rinfo;
3278 : :
3279 : : /* Child members should not exist in ec_members */
3280 : : Assert(!other_em->em_is_child);
3281 : :
3282 : : /* Make sure it'll be a join to a different rel */
3283 [ + + + + ]: 297631 : if (other_em == cur_em ||
3284 : 102869 : bms_overlap(other_em->em_relids, rel->relids))
3285 : 91978 : continue;
3286 : :
3287 : : /* Forget it if caller doesn't want joins to this rel */
3288 [ + + ]: 102784 : if (bms_overlap(other_em->em_relids, prohibited_rels))
3289 : 130 : continue;
3290 : :
3291 : : /*
3292 : : * Also, if this is a child rel, avoid generating a useless join
3293 : : * to its parent rel(s).
3294 : : */
3295 [ + + + + ]: 109125 : if (is_child_rel &&
3296 : 6471 : bms_overlap(parent_relids, other_em->em_relids))
3297 : 2952 : continue;
3298 : :
3299 : 99702 : eq_op = select_equality_operator(cur_ec,
3300 : : cur_em->em_datatype,
3301 : : other_em->em_datatype);
3302 [ - + ]: 99702 : if (!OidIsValid(eq_op))
3303 : 0 : continue;
3304 : :
3305 : : /* set parent_ec to mark as redundant with other joinclauses */
3306 : 99702 : rinfo = create_join_clause(root, cur_ec, eq_op,
3307 : : cur_em, other_em,
3308 : : cur_ec);
3309 : :
3310 : 99702 : result = lappend(result, rinfo);
3311 : : }
3312 : :
3313 : : /*
3314 : : * If somehow we failed to create any join clauses, we might as well
3315 : : * keep scanning the ECs for another match. But if we did make any,
3316 : : * we're done, because we don't want to return non-redundant clauses.
3317 : : */
3318 [ + + ]: 94775 : if (result)
3319 : 94615 : break;
3320 : : }
3321 : :
3322 : 462352 : return result;
3323 : : }
3324 : :
3325 : : /*
3326 : : * have_relevant_eclass_joinclause
3327 : : * Detect whether there is an EquivalenceClass that could produce
3328 : : * a joinclause involving the two given relations.
3329 : : *
3330 : : * This is essentially a very cut-down version of
3331 : : * generate_join_implied_equalities(). Note it's OK to occasionally say "yes"
3332 : : * incorrectly. Hence we don't bother with details like whether the lack of a
3333 : : * cross-type operator might prevent the clause from actually being generated.
3334 : : * False negatives are not always fatal either: they will discourage, but not
3335 : : * completely prevent, investigation of particular join pathways.
3336 : : */
3337 : : bool
3338 : 149990 : have_relevant_eclass_joinclause(PlannerInfo *root,
3339 : : RelOptInfo *rel1, RelOptInfo *rel2)
3340 : : {
3341 : : Bitmapset *matching_ecs;
3342 : : int i;
3343 : :
3344 : : /*
3345 : : * Examine only eclasses mentioning both rel1 and rel2.
3346 : : *
3347 : : * Note that we do not consider the possibility of an eclass generating
3348 : : * "join" clauses that mention just one of the rels plus an outer join
3349 : : * that could be formed from them. Although such clauses must be
3350 : : * correctly enforced when we form the outer join, they don't seem like
3351 : : * sufficient reason to prioritize this join over other ones. The join
3352 : : * ordering rules will force the join to be made when necessary.
3353 : : */
3354 : 149990 : matching_ecs = get_common_eclass_indexes(root, rel1->relids,
3355 : : rel2->relids);
3356 : :
3357 : 149990 : i = -1;
3358 [ + + ]: 150050 : while ((i = bms_next_member(matching_ecs, i)) >= 0)
3359 : : {
3360 : 128325 : EquivalenceClass *ec = (EquivalenceClass *) list_nth(root->eq_classes,
3361 : : i);
3362 : :
3363 : : /*
3364 : : * Sanity check that get_common_eclass_indexes gave only ECs
3365 : : * containing both rels.
3366 : : */
3367 : : Assert(bms_overlap(rel1->relids, ec->ec_relids));
3368 : : Assert(bms_overlap(rel2->relids, ec->ec_relids));
3369 : :
3370 : : /*
3371 : : * Won't generate joinclauses if single-member (this test covers the
3372 : : * volatile case too)
3373 : : */
3374 [ + + ]: 128325 : if (list_length(ec->ec_members) <= 1)
3375 : 60 : continue;
3376 : :
3377 : : /*
3378 : : * We do not need to examine the individual members of the EC, because
3379 : : * all that we care about is whether each rel overlaps the relids of
3380 : : * at least one member, and get_common_eclass_indexes() and the single
3381 : : * member check above are sufficient to prove that. (As with
3382 : : * have_relevant_joinclause(), it is not necessary that the EC be able
3383 : : * to form a joinclause relating exactly the two given rels, only that
3384 : : * it be able to form a joinclause mentioning both, and this will
3385 : : * surely be true if both of them overlap ec_relids.)
3386 : : *
3387 : : * Note we don't test ec_broken; if we did, we'd need a separate code
3388 : : * path to look through ec_sources. Checking the membership anyway is
3389 : : * OK as a possibly-overoptimistic heuristic.
3390 : : *
3391 : : * We don't test ec_has_const either, even though a const eclass won't
3392 : : * generate real join clauses. This is because if we had "WHERE a.x =
3393 : : * b.y and a.x = 42", it is worth considering a join between a and b,
3394 : : * since the join result is likely to be small even though it'll end
3395 : : * up being an unqualified nestloop.
3396 : : */
3397 : :
3398 : 128265 : return true;
3399 : : }
3400 : :
3401 : 21725 : return false;
3402 : : }
3403 : :
3404 : :
3405 : : /*
3406 : : * has_relevant_eclass_joinclause
3407 : : * Detect whether there is an EquivalenceClass that could produce
3408 : : * a joinclause involving the given relation and anything else.
3409 : : *
3410 : : * This is the same as have_relevant_eclass_joinclause with the other rel
3411 : : * implicitly defined as "everything else in the query".
3412 : : */
3413 : : bool
3414 : 171908 : has_relevant_eclass_joinclause(PlannerInfo *root, RelOptInfo *rel1)
3415 : : {
3416 : : Bitmapset *matched_ecs;
3417 : : int i;
3418 : :
3419 : : /* Examine only eclasses mentioning rel1 */
3420 : 171908 : matched_ecs = get_eclass_indexes_for_relids(root, rel1->relids);
3421 : :
3422 : 171908 : i = -1;
3423 [ + + ]: 592488 : while ((i = bms_next_member(matched_ecs, i)) >= 0)
3424 : : {
3425 : 485646 : EquivalenceClass *ec = (EquivalenceClass *) list_nth(root->eq_classes,
3426 : : i);
3427 : :
3428 : : /*
3429 : : * Won't generate joinclauses if single-member (this test covers the
3430 : : * volatile case too)
3431 : : */
3432 [ + + ]: 485646 : if (list_length(ec->ec_members) <= 1)
3433 : 217422 : continue;
3434 : :
3435 : : /*
3436 : : * Per the comment in have_relevant_eclass_joinclause, it's sufficient
3437 : : * to find an EC that mentions both this rel and some other rel.
3438 : : */
3439 [ + + ]: 268224 : if (!bms_is_subset(ec->ec_relids, rel1->relids))
3440 : 65066 : return true;
3441 : : }
3442 : :
3443 : 106842 : return false;
3444 : : }
3445 : :
3446 : :
3447 : : /*
3448 : : * eclass_useful_for_merging
3449 : : * Detect whether the EC could produce any mergejoinable join clauses
3450 : : * against the specified relation.
3451 : : *
3452 : : * This is just a heuristic test and doesn't have to be exact; it's better
3453 : : * to say "yes" incorrectly than "no". Hence we don't bother with details
3454 : : * like whether the lack of a cross-type operator might prevent the clause
3455 : : * from actually being generated.
3456 : : */
3457 : : bool
3458 : 563343 : eclass_useful_for_merging(PlannerInfo *root,
3459 : : EquivalenceClass *eclass,
3460 : : RelOptInfo *rel)
3461 : : {
3462 : : Relids relids;
3463 : : ListCell *lc;
3464 : :
3465 : : Assert(!eclass->ec_merged);
3466 : :
3467 : : /*
3468 : : * Won't generate joinclauses if const or single-member (the latter test
3469 : : * covers the volatile case too)
3470 : : */
3471 [ + + + + ]: 563343 : if (eclass->ec_has_const || list_length(eclass->ec_members) <= 1)
3472 : 30530 : return false;
3473 : :
3474 : : /*
3475 : : * Note we don't test ec_broken; if we did, we'd need a separate code path
3476 : : * to look through ec_sources. Checking the members anyway is OK as a
3477 : : * possibly-overoptimistic heuristic.
3478 : : */
3479 : :
3480 : : /* If specified rel is a child, we must consider the topmost parent rel */
3481 [ + + + + : 532813 : if (IS_OTHER_REL(rel))
- + ]
3482 : : {
3483 : : Assert(!bms_is_empty(rel->top_parent_relids));
3484 : 5127 : relids = rel->top_parent_relids;
3485 : : }
3486 : : else
3487 : 527686 : relids = rel->relids;
3488 : :
3489 : : /* If rel already includes all members of eclass, no point in searching */
3490 [ + + ]: 532813 : if (bms_is_subset(eclass->ec_relids, relids))
3491 : 193018 : return false;
3492 : :
3493 : : /*
3494 : : * To join, we need a member not in the given rel. Ignore children here.
3495 : : */
3496 [ + - + + : 528319 : foreach(lc, eclass->ec_members)
+ + ]
3497 : : {
3498 : 527524 : EquivalenceMember *cur_em = (EquivalenceMember *) lfirst(lc);
3499 : :
3500 : : /* Child members should not exist in ec_members */
3501 : : Assert(!cur_em->em_is_child);
3502 : :
3503 [ + + ]: 527524 : if (!bms_overlap(cur_em->em_relids, relids))
3504 : 339000 : return true;
3505 : : }
3506 : :
3507 : 795 : return false;
3508 : : }
3509 : :
3510 : :
3511 : : /*
3512 : : * is_redundant_derived_clause
3513 : : * Test whether rinfo is derived from same EC as any clause in clauselist;
3514 : : * if so, it can be presumed to represent a condition that's redundant
3515 : : * with that member of the list.
3516 : : */
3517 : : bool
3518 : 70 : is_redundant_derived_clause(RestrictInfo *rinfo, List *clauselist)
3519 : : {
3520 : 70 : EquivalenceClass *parent_ec = rinfo->parent_ec;
3521 : : ListCell *lc;
3522 : :
3523 : : /* Fail if it's not a potentially-redundant clause from some EC */
3524 [ + - ]: 70 : if (parent_ec == NULL)
3525 : 70 : return false;
3526 : :
3527 [ # # # # : 0 : foreach(lc, clauselist)
# # ]
3528 : : {
3529 : 0 : RestrictInfo *otherrinfo = (RestrictInfo *) lfirst(lc);
3530 : :
3531 [ # # ]: 0 : if (otherrinfo->parent_ec == parent_ec)
3532 : 0 : return true;
3533 : : }
3534 : :
3535 : 0 : return false;
3536 : : }
3537 : :
3538 : : /*
3539 : : * is_redundant_with_indexclauses
3540 : : * Test whether rinfo is redundant with any clause in the IndexClause
3541 : : * list. Here, for convenience, we test both simple identity and
3542 : : * whether it is derived from the same EC as any member of the list.
3543 : : */
3544 : : bool
3545 : 1141234 : is_redundant_with_indexclauses(RestrictInfo *rinfo, List *indexclauses)
3546 : : {
3547 : 1141234 : EquivalenceClass *parent_ec = rinfo->parent_ec;
3548 : : ListCell *lc;
3549 : :
3550 [ + + + + : 1589277 : foreach(lc, indexclauses)
+ + ]
3551 : : {
3552 : 1155489 : IndexClause *iclause = lfirst_node(IndexClause, lc);
3553 : 1155489 : RestrictInfo *otherrinfo = iclause->rinfo;
3554 : :
3555 : : /* If indexclause is lossy, it won't enforce the condition exactly */
3556 [ + + ]: 1155489 : if (iclause->lossy)
3557 : 32584 : continue;
3558 : :
3559 : : /* Match if it's same clause (pointer equality should be enough) */
3560 [ + + ]: 1122905 : if (rinfo == otherrinfo)
3561 : 707446 : return true;
3562 : : /* Match if derived from same EC */
3563 [ + + + + ]: 415903 : if (parent_ec && otherrinfo->parent_ec == parent_ec)
3564 : 444 : return true;
3565 : :
3566 : : /*
3567 : : * No need to look at the derived clauses in iclause->indexquals; they
3568 : : * couldn't match if the parent clause didn't.
3569 : : */
3570 : : }
3571 : :
3572 : 433788 : return false;
3573 : : }
3574 : :
3575 : : /*
3576 : : * get_eclass_indexes_for_relids
3577 : : * Build and return a Bitmapset containing the indexes into root's
3578 : : * eq_classes list for all eclasses that mention any of these relids
3579 : : */
3580 : : static Bitmapset *
3581 : 885857 : get_eclass_indexes_for_relids(PlannerInfo *root, Relids relids)
3582 : : {
3583 : 885857 : Bitmapset *ec_indexes = NULL;
3584 : 885857 : int i = -1;
3585 : :
3586 : : /* Should be OK to rely on eclass_indexes */
3587 : : Assert(root->ec_merging_done);
3588 : :
3589 [ + + ]: 2733284 : while ((i = bms_next_member(relids, i)) > 0)
3590 : : {
3591 : 1847427 : RelOptInfo *rel = root->simple_rel_array[i];
3592 : :
3593 : : /* ignore the RTE_GROUP RTE */
3594 [ - + ]: 1847427 : if (i == root->group_rtindex)
3595 : 0 : continue;
3596 : :
3597 [ + + ]: 1847427 : if (rel == NULL) /* must be an outer join */
3598 : : {
3599 : : Assert(bms_is_member(i, root->outer_join_rels));
3600 : 214916 : continue;
3601 : : }
3602 : :
3603 : 1632511 : ec_indexes = bms_add_members(ec_indexes, rel->eclass_indexes);
3604 : : }
3605 : 885857 : return ec_indexes;
3606 : : }
3607 : :
3608 : : /*
3609 : : * get_common_eclass_indexes
3610 : : * Build and return a Bitmapset containing the indexes into root's
3611 : : * eq_classes list for all eclasses that mention rels in both
3612 : : * relids1 and relids2.
3613 : : */
3614 : : static Bitmapset *
3615 : 517468 : get_common_eclass_indexes(PlannerInfo *root, Relids relids1, Relids relids2)
3616 : : {
3617 : : Bitmapset *rel1ecs;
3618 : : Bitmapset *rel2ecs;
3619 : : int relid;
3620 : :
3621 : 517468 : rel1ecs = get_eclass_indexes_for_relids(root, relids1);
3622 : :
3623 : : /*
3624 : : * We can get away with just using the relation's eclass_indexes directly
3625 : : * when relids2 is a singleton set.
3626 : : */
3627 [ + + ]: 517468 : if (bms_get_singleton_member(relids2, &relid))
3628 : 398891 : rel2ecs = root->simple_rel_array[relid]->eclass_indexes;
3629 : : else
3630 : 118577 : rel2ecs = get_eclass_indexes_for_relids(root, relids2);
3631 : :
3632 : : /* Calculate and return the common EC indexes, recycling the left input. */
3633 : 517468 : return bms_int_members(rel1ecs, rel2ecs);
3634 : : }
3635 : :
3636 : : /*
3637 : : * ec_build_derives_hash
3638 : : * Construct the auxiliary hash table for derived clause lookups.
3639 : : */
3640 : : static void
3641 : 0 : ec_build_derives_hash(PlannerInfo *root, EquivalenceClass *ec)
3642 : : {
3643 : : Assert(!ec->ec_derives_hash);
3644 : :
3645 : : /*
3646 : : * Create the hash table.
3647 : : *
3648 : : * We pass list_length(ec->ec_derives_list) as the initial size.
3649 : : * Simplehash will divide this by the fillfactor (typically 0.9) and round
3650 : : * up to the next power of two, so this will usually give us at least 64
3651 : : * buckets around the threshold. That avoids immediate resizing without
3652 : : * hardcoding a specific size.
3653 : : */
3654 : 0 : ec->ec_derives_hash = derives_create(root->planner_cxt,
3655 : 0 : list_length(ec->ec_derives_list),
3656 : : NULL);
3657 : :
3658 [ # # # # : 0 : foreach_node(RestrictInfo, rinfo, ec->ec_derives_list)
# # ]
3659 : 0 : ec_add_clause_to_derives_hash(ec, rinfo);
3660 : 0 : }
3661 : :
3662 : : /*
3663 : : * ec_add_derived_clause
3664 : : * Add a clause to the set of derived clauses for the given
3665 : : * EquivalenceClass. Always appends to ec_derives_list; also adds
3666 : : * to ec_derives_hash if it exists.
3667 : : *
3668 : : * Also asserts expected invariants of derived clauses.
3669 : : */
3670 : : static void
3671 : 90199 : ec_add_derived_clause(EquivalenceClass *ec, RestrictInfo *clause)
3672 : : {
3673 : : /*
3674 : : * Constant, if present, is always placed on the RHS; see
3675 : : * generate_base_implied_equalities_const(). LHS is never a constant.
3676 : : */
3677 : : Assert(!clause->left_em->em_is_const);
3678 : :
3679 : : /*
3680 : : * Clauses containing a constant are never considered redundant, so
3681 : : * parent_ec is not set.
3682 : : */
3683 : : Assert(!clause->parent_ec || !clause->right_em->em_is_const);
3684 : :
3685 : 90199 : ec->ec_derives_list = lappend(ec->ec_derives_list, clause);
3686 [ - + ]: 90199 : if (ec->ec_derives_hash)
3687 : 0 : ec_add_clause_to_derives_hash(ec, clause);
3688 : 90199 : }
3689 : :
3690 : : /*
3691 : : * ec_add_derived_clauses
3692 : : * Add a list of clauses to the set of clauses derived from the given
3693 : : * EquivalenceClass; adding to the list and hash table if needed.
3694 : : *
3695 : : * This function is similar to ec_add_derived_clause() but optimized for adding
3696 : : * multiple clauses at a time to the ec_derives_list. The assertions from
3697 : : * ec_add_derived_clause() are not repeated here, as the input clauses are
3698 : : * assumed to have already been validated.
3699 : : */
3700 : : static void
3701 : 33 : ec_add_derived_clauses(EquivalenceClass *ec, List *clauses)
3702 : : {
3703 : 33 : ec->ec_derives_list = list_concat(ec->ec_derives_list, clauses);
3704 [ - + ]: 33 : if (ec->ec_derives_hash)
3705 [ # # # # : 0 : foreach_node(RestrictInfo, rinfo, clauses)
# # ]
3706 : 0 : ec_add_clause_to_derives_hash(ec, rinfo);
3707 : 33 : }
3708 : :
3709 : : /*
3710 : : * fill_ec_derives_key
3711 : : * Compute a canonical key for ec_derives_hash lookup or insertion.
3712 : : *
3713 : : * Derived clauses are looked up using a pair of EquivalenceMembers and a
3714 : : * parent EquivalenceClass. To avoid storing or searching for both EM orderings,
3715 : : * we canonicalize the key:
3716 : : *
3717 : : * - For clauses involving two non-constant EMs, em1 is set to the EM with lower
3718 : : * memory address and em2 is set to the other one.
3719 : : * - For clauses involving a constant EM, the caller must pass the non-constant
3720 : : * EM as leftem and NULL as rightem; we then set em1 = NULL and em2 = leftem.
3721 : : */
3722 : : static inline void
3723 : 0 : fill_ec_derives_key(ECDerivesKey *key,
3724 : : EquivalenceMember *leftem,
3725 : : EquivalenceMember *rightem,
3726 : : EquivalenceClass *parent_ec)
3727 : : {
3728 : : Assert(leftem); /* Always required for lookup or insertion */
3729 : :
3730 [ # # ]: 0 : if (rightem == NULL)
3731 : : {
3732 : 0 : key->em1 = NULL;
3733 : 0 : key->em2 = leftem;
3734 : : }
3735 [ # # ]: 0 : else if (leftem < rightem)
3736 : : {
3737 : 0 : key->em1 = leftem;
3738 : 0 : key->em2 = rightem;
3739 : : }
3740 : : else
3741 : : {
3742 : 0 : key->em1 = rightem;
3743 : 0 : key->em2 = leftem;
3744 : : }
3745 : 0 : key->parent_ec = parent_ec;
3746 : 0 : }
3747 : :
3748 : : /*
3749 : : * ec_add_clause_to_derives_hash
3750 : : * Add a derived clause to ec_derives_hash in the given EquivalenceClass.
3751 : : *
3752 : : * Each clause is associated with a canonicalized key. For constant-containing
3753 : : * clauses, only the non-constant EM is used for lookup; see comments in
3754 : : * fill_ec_derives_key().
3755 : : */
3756 : : static void
3757 : 0 : ec_add_clause_to_derives_hash(EquivalenceClass *ec, RestrictInfo *rinfo)
3758 : : {
3759 : : ECDerivesKey key;
3760 : : ECDerivesEntry *entry;
3761 : : bool found;
3762 : :
3763 : : /*
3764 : : * Constants are always placed on the RHS; see
3765 : : * generate_base_implied_equalities_const().
3766 : : */
3767 : : Assert(!rinfo->left_em->em_is_const);
3768 : :
3769 : : /*
3770 : : * Clauses containing a constant are never considered redundant, so
3771 : : * parent_ec is not set.
3772 : : */
3773 : : Assert(!rinfo->parent_ec || !rinfo->right_em->em_is_const);
3774 : :
3775 : : /*
3776 : : * See fill_ec_derives_key() for details: we use a canonicalized key to
3777 : : * avoid storing both EM orderings. For constant EMs, only the
3778 : : * non-constant EM is included in the key.
3779 : : */
3780 : 0 : fill_ec_derives_key(&key,
3781 : : rinfo->left_em,
3782 [ # # ]: 0 : rinfo->right_em->em_is_const ? NULL : rinfo->right_em,
3783 : : rinfo->parent_ec);
3784 : 0 : entry = derives_insert(ec->ec_derives_hash, key, &found);
3785 : : Assert(!found);
3786 : 0 : entry->rinfo = rinfo;
3787 : 0 : }
3788 : :
3789 : : /*
3790 : : * ec_clear_derived_clauses
3791 : : * Reset ec_derives_list and ec_derives_hash.
3792 : : *
3793 : : * We destroy the hash table explicitly, since it may consume significant
3794 : : * space. The list holds the same set of entries and can become equally large
3795 : : * when thousands of partitions are involved, so we free it as well -- even
3796 : : * though we do not typically free lists.
3797 : : */
3798 : : static void
3799 : 33 : ec_clear_derived_clauses(EquivalenceClass *ec)
3800 : : {
3801 : 33 : list_free(ec->ec_derives_list);
3802 : 33 : ec->ec_derives_list = NIL;
3803 : :
3804 [ - + ]: 33 : if (ec->ec_derives_hash)
3805 : : {
3806 : 0 : derives_destroy(ec->ec_derives_hash);
3807 : 0 : ec->ec_derives_hash = NULL;
3808 : : }
3809 : 33 : }
3810 : :
3811 : : /*
3812 : : * ec_search_clause_for_ems
3813 : : * Search for an existing RestrictInfo that equates the given pair
3814 : : * of EquivalenceMembers, either from ec_sources or ec_derives.
3815 : : *
3816 : : * Returns a clause with matching operands in either given order or commuted
3817 : : * order. We used to require matching operator OIDs, but dropped that since any
3818 : : * semantically different operator here would indicate a broken operator family.
3819 : : *
3820 : : * Returns NULL if no matching clause is found.
3821 : : */
3822 : : static RestrictInfo *
3823 : 443668 : ec_search_clause_for_ems(PlannerInfo *root, EquivalenceClass *ec,
3824 : : EquivalenceMember *leftem, EquivalenceMember *rightem,
3825 : : EquivalenceClass *parent_ec)
3826 : : {
3827 : : /* Check original source clauses */
3828 [ + - + + : 1270725 : foreach_node(RestrictInfo, rinfo, ec->ec_sources)
+ + ]
3829 : : {
3830 [ + + ]: 506147 : if (rinfo->left_em == leftem &&
3831 [ + + ]: 254694 : rinfo->right_em == rightem &&
3832 [ + + ]: 231903 : rinfo->parent_ec == parent_ec)
3833 : 61379 : return rinfo;
3834 [ + + ]: 463377 : if (rinfo->left_em == rightem &&
3835 [ + + ]: 207123 : rinfo->right_em == leftem &&
3836 [ + + ]: 189194 : rinfo->parent_ec == parent_ec)
3837 : 18609 : return rinfo;
3838 : : }
3839 : :
3840 : : /* Not found in ec_sources; search derived clauses */
3841 : 382289 : return ec_search_derived_clause_for_ems(root, ec, leftem, rightem,
3842 : : parent_ec);
3843 : : }
3844 : :
3845 : : /*
3846 : : * ec_search_derived_clause_for_ems
3847 : : * Search for an existing derived clause between two EquivalenceMembers.
3848 : : *
3849 : : * If the number of derived clauses exceeds a threshold, switch to hash table
3850 : : * lookup; otherwise, scan ec_derives_list linearly.
3851 : : *
3852 : : * Clauses involving constants are looked up by passing the non-constant EM
3853 : : * as leftem and setting rightem to NULL. In that case, we expect to find a
3854 : : * clause with a constant on the RHS.
3855 : : *
3856 : : * While searching the list, we compare each given EM with both sides of each
3857 : : * clause. But for hash table lookups, we construct a canonicalized key and
3858 : : * perform a single lookup.
3859 : : */
3860 : : static RestrictInfo *
3861 : 382294 : ec_search_derived_clause_for_ems(PlannerInfo *root, EquivalenceClass *ec,
3862 : : EquivalenceMember *leftem,
3863 : : EquivalenceMember *rightem,
3864 : : EquivalenceClass *parent_ec)
3865 : : {
3866 : : /* Switch to using hash lookup when list grows "too long". */
3867 [ + - - + ]: 764588 : if (!ec->ec_derives_hash &&
3868 : 382294 : list_length(ec->ec_derives_list) >= EC_DERIVES_HASH_THRESHOLD)
3869 : 0 : ec_build_derives_hash(root, ec);
3870 : :
3871 : : /* Perform hash table lookup if available */
3872 [ - + ]: 382294 : if (ec->ec_derives_hash)
3873 : : {
3874 : : ECDerivesKey key;
3875 : : RestrictInfo *rinfo;
3876 : : ECDerivesEntry *entry;
3877 : :
3878 : 0 : fill_ec_derives_key(&key, leftem, rightem, parent_ec);
3879 : 0 : entry = derives_lookup(ec->ec_derives_hash, key);
3880 [ # # ]: 0 : if (entry)
3881 : : {
3882 : 0 : rinfo = entry->rinfo;
3883 : : Assert(rinfo);
3884 : : Assert(rightem || rinfo->right_em->em_is_const);
3885 : 0 : return rinfo;
3886 : : }
3887 : : }
3888 : : else
3889 : : {
3890 : : /* Fallback to linear search over ec_derives_list */
3891 [ + + + + : 556912 : foreach_node(RestrictInfo, rinfo, ec->ec_derives_list)
+ + ]
3892 : : {
3893 : : /* Handle special case: lookup by non-const EM alone */
3894 [ + + ]: 422040 : if (!rightem &&
3895 [ + - ]: 5 : rinfo->left_em == leftem)
3896 : : {
3897 : : Assert(rinfo->right_em->em_is_const);
3898 : 314858 : return rinfo;
3899 : : }
3900 [ + + ]: 422035 : if (rinfo->left_em == leftem &&
3901 [ + + ]: 169528 : rinfo->right_em == rightem &&
3902 [ + + ]: 153365 : rinfo->parent_ec == parent_ec)
3903 : 153355 : return rinfo;
3904 [ + + ]: 268680 : if (rinfo->left_em == rightem &&
3905 [ + + ]: 170779 : rinfo->right_em == leftem &&
3906 [ + - ]: 161498 : rinfo->parent_ec == parent_ec)
3907 : 161498 : return rinfo;
3908 : : }
3909 : : }
3910 : :
3911 : 67436 : return NULL;
3912 : : }
|
