Revise collation derivation method and expression-tree representation.

All expression nodes now have an explicit output-collation field, unless
they are known to only return a noncollatable data type (such as boolean
or record).  Also, nodes that can invoke collation-aware functions store
a separate field that is the collation value to pass to the function.
This avoids confusion that arises when a function has collatable inputs
and noncollatable output type, or vice versa.

Also, replace the parser's on-the-fly collation assignment method with
a post-pass over the completed expression tree.  This allows us to use
a more complex (and hopefully more nearly spec-compliant) assignment
rule without paying for it in extra storage in every expression node.

Fix assorted bugs in the planner's handling of collations by making
collation one of the defining properties of an EquivalenceClass and
by converting CollateExprs into discardable RelabelType nodes during
expression preprocessing.

7 files changed, 61 insertions, 70 deletions

diff --git a/src/backend/executor/execQual.c b/src/backend/executor/execQual.c
index 0faf52dfd79..e410818900d 100644
--- a/src/backend/executor/execQual.c
+++ b/src/backend/executor/execQual.c
@@ -82,7 +82,7 @@ static Datum ExecEvalParamExec(ExprState *exprstate, ExprContext *econtext,
 			  bool *isNull, ExprDoneCond *isDone);
 static Datum ExecEvalParamExtern(ExprState *exprstate, ExprContext *econtext,
 			  bool *isNull, ExprDoneCond *isDone);
-static void init_fcache(Oid foid, FuncExprState *fcache,
+static void init_fcache(Oid foid, Oid input_collation, FuncExprState *fcache,
 			MemoryContext fcacheCxt, bool needDescForSets);
 static void ShutdownFuncExpr(Datum arg);
 static TupleDesc get_cached_rowtype(Oid type_id, int32 typmod,
@@ -120,9 +120,6 @@ static Datum ExecEvalAnd(BoolExprState *andExpr, ExprContext *econtext,
 static Datum ExecEvalConvertRowtype(ConvertRowtypeExprState *cstate,
 					   ExprContext *econtext,
 					   bool *isNull, ExprDoneCond *isDone);
-static Datum ExecEvalCollateExpr(GenericExprState *exprstate,
-					ExprContext *econtext,
-					bool *isNull, ExprDoneCond *isDone);
 static Datum ExecEvalCase(CaseExprState *caseExpr, ExprContext *econtext,
 			 bool *isNull, ExprDoneCond *isDone);
 static Datum ExecEvalCaseTestExpr(ExprState *exprstate,
@@ -1179,7 +1176,7 @@ GetAttributeByName(HeapTupleHeader tuple, const char *attname, bool *isNull)
  * init_fcache - initialize a FuncExprState node during first use
  */
 static void
-init_fcache(Oid foid, FuncExprState *fcache,
+init_fcache(Oid foid, Oid input_collation, FuncExprState *fcache,
 			MemoryContext fcacheCxt, bool needDescForSets)
 {
 	AclResult	aclresult;
@@ -1205,7 +1202,8 @@ init_fcache(Oid foid, FuncExprState *fcache,
 
 	/* Set up the primary fmgr lookup information */
 	fmgr_info_cxt(foid, &(fcache->func), fcacheCxt);
-	fmgr_info_expr((Node *) fcache->xprstate.expr, &(fcache->func));
+	fmgr_info_set_collation(input_collation, &(fcache->func));
+	fmgr_info_set_expr((Node *) fcache->xprstate.expr, &(fcache->func));
 
 	/* Initialize the function call parameter struct as well */
 	InitFunctionCallInfoData(fcache->fcinfo_data, &(fcache->func),
@@ -1976,7 +1974,7 @@ ExecMakeTableFunctionResult(ExprState *funcexpr,
 		{
 			FuncExpr   *func = (FuncExpr *) fcache->xprstate.expr;
 
-			init_fcache(func->funcid, fcache,
+			init_fcache(func->funcid, func->inputcollid, fcache,
 						econtext->ecxt_per_query_memory, false);
 		}
 		returnsSet = fcache->func.fn_retset;
@@ -2255,7 +2253,8 @@ ExecEvalFunc(FuncExprState *fcache,
 	FuncExpr   *func = (FuncExpr *) fcache->xprstate.expr;
 
 	/* Initialize function lookup info */
-	init_fcache(func->funcid, fcache, econtext->ecxt_per_query_memory, true);
+	init_fcache(func->funcid, func->inputcollid, fcache,
+				econtext->ecxt_per_query_memory, true);
 
 	/* Go directly to ExecMakeFunctionResult on subsequent uses */
 	fcache->xprstate.evalfunc = (ExprStateEvalFunc) ExecMakeFunctionResult;
@@ -2277,7 +2276,8 @@ ExecEvalOper(FuncExprState *fcache,
 	OpExpr	   *op = (OpExpr *) fcache->xprstate.expr;
 
 	/* Initialize function lookup info */
-	init_fcache(op->opfuncid, fcache, econtext->ecxt_per_query_memory, true);
+	init_fcache(op->opfuncid, op->inputcollid, fcache,
+				econtext->ecxt_per_query_memory, true);
 
 	/* Go directly to ExecMakeFunctionResult on subsequent uses */
 	fcache->xprstate.evalfunc = (ExprStateEvalFunc) ExecMakeFunctionResult;
@@ -2318,7 +2318,7 @@ ExecEvalDistinct(FuncExprState *fcache,
 	{
 		DistinctExpr *op = (DistinctExpr *) fcache->xprstate.expr;
 
-		init_fcache(op->opfuncid, fcache,
+		init_fcache(op->opfuncid, op->inputcollid, fcache,
 					econtext->ecxt_per_query_memory, true);
 		Assert(!fcache->func.fn_retset);
 	}
@@ -2395,7 +2395,7 @@ ExecEvalScalarArrayOp(ScalarArrayOpExprState *sstate,
 	 */
 	if (sstate->fxprstate.func.fn_oid == InvalidOid)
 	{
-		init_fcache(opexpr->opfuncid, &sstate->fxprstate,
+		init_fcache(opexpr->opfuncid, opexpr->inputcollid, &sstate->fxprstate,
 					econtext->ecxt_per_query_memory, true);
 		Assert(!sstate->fxprstate.func.fn_retset);
 	}
@@ -2754,20 +2754,6 @@ ExecEvalConvertRowtype(ConvertRowtypeExprState *cstate,
 }
 
 /* ----------------------------------------------------------------
- *		ExecEvalCollateExpr
- *
- *		Evaluate a CollateExpr node.
- * ----------------------------------------------------------------
- */
-static Datum
-ExecEvalCollateExpr(GenericExprState *exprstate,
-					ExprContext *econtext,
-					bool *isNull, ExprDoneCond *isDone)
-{
-	return ExecEvalExpr(exprstate->arg, econtext, isNull, isDone);
-}
-
-/* ----------------------------------------------------------------
  *		ExecEvalCase
  *
  *		Evaluate a CASE clause. Will have boolean expressions
@@ -3542,7 +3528,7 @@ ExecEvalNullIf(FuncExprState *nullIfExpr,
 	{
 		NullIfExpr *op = (NullIfExpr *) nullIfExpr->xprstate.expr;
 
-		init_fcache(op->opfuncid, nullIfExpr,
+		init_fcache(op->opfuncid, op->inputcollid, nullIfExpr,
 					econtext->ecxt_per_query_memory, true);
 		Assert(!nullIfExpr->func.fn_retset);
 	}
@@ -4129,9 +4115,8 @@ ExecEvalArrayCoerceExpr(ArrayCoerceExprState *astate,
 		/* Set up the primary fmgr lookup information */
 		fmgr_info_cxt(acoerce->elemfuncid, &(astate->elemfunc),
 					  econtext->ecxt_per_query_memory);
-
-		/* Initialize additional info */
-		fmgr_info_expr((Node *) acoerce, &(astate->elemfunc));
+		/* Note: coercion functions are assumed to not use collation */
+		fmgr_info_set_expr((Node *) acoerce, &(astate->elemfunc));
 	}
 
 	/*
@@ -4400,6 +4385,18 @@ ExecInitExpr(Expr *node, PlanState *parent)
 				state = (ExprState *) fstate;
 			}
 			break;
+		case T_NullIfExpr:
+			{
+				NullIfExpr *nullifexpr = (NullIfExpr *) node;
+				FuncExprState *fstate = makeNode(FuncExprState);
+
+				fstate->xprstate.evalfunc = (ExprStateEvalFunc) ExecEvalNullIf;
+				fstate->args = (List *)
+					ExecInitExpr((Expr *) nullifexpr->args, parent);
+				fstate->func.fn_oid = InvalidOid;		/* not initialized */
+				state = (ExprState *) fstate;
+			}
+			break;
 		case T_ScalarArrayOpExpr:
 			{
 				ScalarArrayOpExpr *opexpr = (ScalarArrayOpExpr *) node;
@@ -4551,16 +4548,6 @@ ExecInitExpr(Expr *node, PlanState *parent)
 				state = (ExprState *) cstate;
 			}
 			break;
-		case T_CollateExpr:
-			{
-				CollateExpr *collate = (CollateExpr *) node;
-				GenericExprState *gstate = makeNode(GenericExprState);
-
-				gstate->xprstate.evalfunc = (ExprStateEvalFunc) ExecEvalCollateExpr;
-				gstate->arg = ExecInitExpr(collate->arg, parent);
-				state = (ExprState *) gstate;
-			}
-			break;
 		case T_CaseExpr:
 			{
 				CaseExpr   *caseexpr = (CaseExpr *) node;
@@ -4713,11 +4700,11 @@ ExecInitExpr(Expr *node, PlanState *parent)
 				Assert(list_length(rcexpr->opfamilies) == nopers);
 				rstate->funcs = (FmgrInfo *) palloc(nopers * sizeof(FmgrInfo));
 				i = 0;
-				forthree(l, rcexpr->opnos, l2, rcexpr->opfamilies, l3, rcexpr->collids)
+				forthree(l, rcexpr->opnos, l2, rcexpr->opfamilies, l3, rcexpr->inputcollids)
 				{
 					Oid			opno = lfirst_oid(l);
 					Oid			opfamily = lfirst_oid(l2);
-					Oid			collid = lfirst_oid(l3);
+					Oid			inputcollid = lfirst_oid(l3);
 					int			strategy;
 					Oid			lefttype;
 					Oid			righttype;
@@ -4739,7 +4726,7 @@ ExecInitExpr(Expr *node, PlanState *parent)
 					 * does this code.
 					 */
 					fmgr_info(proc, &(rstate->funcs[i]));
-					fmgr_info_collation(collid, &(rstate->funcs[i]));
+					fmgr_info_set_collation(inputcollid, &(rstate->funcs[i]));
 					i++;
 				}
 				state = (ExprState *) rstate;
@@ -4799,7 +4786,8 @@ ExecInitExpr(Expr *node, PlanState *parent)
 				 * code.
 				 */
 				fmgr_info(typentry->cmp_proc, &(mstate->cfunc));
-				fmgr_info_collation(minmaxexpr->collid, &(mstate->cfunc));
+				fmgr_info_set_collation(minmaxexpr->inputcollid,
+										&(mstate->cfunc));
 				state = (ExprState *) mstate;
 			}
 			break;
@@ -4836,18 +4824,6 @@ ExecInitExpr(Expr *node, PlanState *parent)
 				state = (ExprState *) xstate;
 			}
 			break;
-		case T_NullIfExpr:
-			{
-				NullIfExpr *nullifexpr = (NullIfExpr *) node;
-				FuncExprState *fstate = makeNode(FuncExprState);
-
-				fstate->xprstate.evalfunc = (ExprStateEvalFunc) ExecEvalNullIf;
-				fstate->args = (List *)
-					ExecInitExpr((Expr *) nullifexpr->args, parent);
-				fstate->func.fn_oid = InvalidOid;		/* not initialized */
-				state = (ExprState *) fstate;
-			}
-			break;
 		case T_NullTest:
 			{
 				NullTest   *ntest = (NullTest *) node;
diff --git a/src/backend/executor/functions.c b/src/backend/executor/functions.c
index a2f4fac74cb..0421be57a41 100644
--- a/src/backend/executor/functions.c
+++ b/src/backend/executor/functions.c
@@ -1290,6 +1290,7 @@ check_sql_fn_retval(Oid func_id, Oid rettype, List *queryTreeList,
 			tle->expr = (Expr *) makeRelabelType(tle->expr,
 												 rettype,
 												 -1,
+												 get_typcollation(rettype),
 												 COERCE_DONTCARE);
 			/* Relabel is dangerous if TLE is a sort/group or setop column */
 			if (tle->ressortgroupref != 0 || parse->setOperations)
@@ -1335,6 +1336,7 @@ check_sql_fn_retval(Oid func_id, Oid rettype, List *queryTreeList,
 					tle->expr = (Expr *) makeRelabelType(tle->expr,
 														 rettype,
 														 -1,
+														 get_typcollation(rettype),
 														 COERCE_DONTCARE);
 					/* Relabel is dangerous if sort/group or setop column */
 					if (tle->ressortgroupref != 0 || parse->setOperations)
@@ -1437,6 +1439,7 @@ check_sql_fn_retval(Oid func_id, Oid rettype, List *queryTreeList,
 					tle->expr = (Expr *) makeRelabelType(tle->expr,
 														 atttype,
 														 -1,
+														 get_typcollation(atttype),
 														 COERCE_DONTCARE);
 					/* Relabel is dangerous if sort/group or setop column */
 					if (tle->ressortgroupref != 0 || parse->setOperations)
diff --git a/src/backend/executor/nodeAgg.c b/src/backend/executor/nodeAgg.c
index d9bed220e4a..51b1228c26f 100644
--- a/src/backend/executor/nodeAgg.c
+++ b/src/backend/executor/nodeAgg.c
@@ -1669,19 +1669,21 @@ ExecInitAgg(Agg *node, EState *estate, int eflags)
 								numArguments,
 								aggtranstype,
 								aggref->aggtype,
+								aggref->inputcollid,
 								transfn_oid,
 								finalfn_oid,
-								aggref->collid,
 								&transfnexpr,
 								&finalfnexpr);
 
 		fmgr_info(transfn_oid, &peraggstate->transfn);
-		fmgr_info_expr((Node *) transfnexpr, &peraggstate->transfn);
+		fmgr_info_set_collation(aggref->inputcollid, &peraggstate->transfn);
+		fmgr_info_set_expr((Node *) transfnexpr, &peraggstate->transfn);
 
 		if (OidIsValid(finalfn_oid))
 		{
 			fmgr_info(finalfn_oid, &peraggstate->finalfn);
-			fmgr_info_expr((Node *) finalfnexpr, &peraggstate->finalfn);
+			fmgr_info_set_collation(aggref->inputcollid, &peraggstate->finalfn);
+			fmgr_info_set_expr((Node *) finalfnexpr, &peraggstate->finalfn);
 		}
 
 		get_typlenbyval(aggref->aggtype,
@@ -1831,6 +1833,8 @@ ExecInitAgg(Agg *node, EState *estate, int eflags)
 				SortGroupClause *sortcl = (SortGroupClause *) lfirst(lc);
 
 				fmgr_info(get_opcode(sortcl->eqop), &peraggstate->equalfns[i]);
+				fmgr_info_set_collation(aggref->inputcollid,
+										&peraggstate->equalfns[i]);
 				i++;
 			}
 			Assert(i == numDistinctCols);
diff --git a/src/backend/executor/nodeIndexscan.c b/src/backend/executor/nodeIndexscan.c
index 55fce94b321..e60db7813a4 100644
--- a/src/backend/executor/nodeIndexscan.c
+++ b/src/backend/executor/nodeIndexscan.c
@@ -732,7 +732,6 @@ ExecIndexBuildScanKeys(PlanState *planstate, Relation index, Index scanrelid,
 		int			op_strategy;	/* operator's strategy number */
 		Oid			op_lefttype;	/* operator's declared input types */
 		Oid			op_righttype;
-		Oid			collation;
 		Expr	   *leftop;		/* expr on lhs of operator */
 		Expr	   *rightop;	/* expr on rhs ... */
 		AttrNumber	varattno;	/* att number used in scan */
@@ -833,7 +832,7 @@ ExecIndexBuildScanKeys(PlanState *planstate, Relation index, Index scanrelid,
 								   opfuncid,	/* reg proc to use */
 								   scanvalue);	/* constant */
 			ScanKeyEntryInitializeCollation(this_scan_key,
-											((OpExpr *) clause)->collid);
+											((OpExpr *) clause)->inputcollid);
 		}
 		else if (IsA(clause, RowCompareExpr))
 		{
@@ -842,7 +841,7 @@ ExecIndexBuildScanKeys(PlanState *planstate, Relation index, Index scanrelid,
 			ListCell   *largs_cell = list_head(rc->largs);
 			ListCell   *rargs_cell = list_head(rc->rargs);
 			ListCell   *opnos_cell = list_head(rc->opnos);
-			ListCell   *collids_cell = list_head(rc->collids);
+			ListCell   *collids_cell = list_head(rc->inputcollids);
 			ScanKey		first_sub_key;
 			int			n_sub_key;
 
@@ -858,6 +857,7 @@ ExecIndexBuildScanKeys(PlanState *planstate, Relation index, Index scanrelid,
 				ScanKey		this_sub_key = &first_sub_key[n_sub_key];
 				int			flags = SK_ROW_MEMBER;
 				Datum		scanvalue;
+				Oid			inputcollation;
 
 				/*
 				 * leftop should be the index key Var, possibly relabeled
@@ -901,7 +901,7 @@ ExecIndexBuildScanKeys(PlanState *planstate, Relation index, Index scanrelid,
 											 op_righttype,
 											 BTORDER_PROC);
 
-				collation = lfirst_oid(collids_cell);
+				inputcollation = lfirst_oid(collids_cell);
 				collids_cell = lnext(collids_cell);
 
 				/*
@@ -960,7 +960,7 @@ ExecIndexBuildScanKeys(PlanState *planstate, Relation index, Index scanrelid,
 									   opfuncid,		/* reg proc to use */
 									   scanvalue);		/* constant */
 				ScanKeyEntryInitializeCollation(this_sub_key,
-												collation);
+												inputcollation);
 				n_sub_key++;
 			}
 
@@ -1045,7 +1045,7 @@ ExecIndexBuildScanKeys(PlanState *planstate, Relation index, Index scanrelid,
 								   opfuncid,	/* reg proc to use */
 								   (Datum) 0);	/* constant */
 			ScanKeyEntryInitializeCollation(this_scan_key,
-											saop->collid);
+											saop->inputcollid);
 		}
 		else if (IsA(clause, NullTest))
 		{
diff --git a/src/backend/executor/nodeMergejoin.c b/src/backend/executor/nodeMergejoin.c
index c0b9f230855..75c3a645359 100644
--- a/src/backend/executor/nodeMergejoin.c
+++ b/src/backend/executor/nodeMergejoin.c
@@ -242,7 +242,7 @@ MJExamineQuals(List *mergeclauses,
 
 		/* Set up the fmgr lookup information */
 		fmgr_info(cmpproc, &(clause->cmpfinfo));
-		fmgr_info_collation(collation, &(clause->cmpfinfo));
+		fmgr_info_set_collation(collation, &(clause->cmpfinfo));
 
 		/* Fill the additional comparison-strategy flags */
 		if (opstrategy == BTLessStrategyNumber)
diff --git a/src/backend/executor/nodeSubplan.c b/src/backend/executor/nodeSubplan.c
index e9b3d76df1c..08a3017e614 100644
--- a/src/backend/executor/nodeSubplan.c
+++ b/src/backend/executor/nodeSubplan.c
@@ -831,7 +831,9 @@ ExecInitSubPlan(SubPlan *subplan, PlanState *parent)
 
 			/* Lookup the equality function (potentially cross-type) */
 			fmgr_info(opexpr->opfuncid, &sstate->cur_eq_funcs[i - 1]);
-			fmgr_info_expr((Node *) opexpr, &sstate->cur_eq_funcs[i - 1]);
+			fmgr_info_set_collation(opexpr->inputcollid,
+									&sstate->cur_eq_funcs[i - 1]);
+			fmgr_info_set_expr((Node *) opexpr, &sstate->cur_eq_funcs[i - 1]);
 
 			/* Look up the equality function for the RHS type */
 			if (!get_compatible_hash_operators(opexpr->opno,
@@ -839,6 +841,8 @@ ExecInitSubPlan(SubPlan *subplan, PlanState *parent)
 				elog(ERROR, "could not find compatible hash operator for operator %u",
 					 opexpr->opno);
 			fmgr_info(get_opcode(rhs_eq_oper), &sstate->tab_eq_funcs[i - 1]);
+			fmgr_info_set_collation(opexpr->inputcollid,
+									&sstate->tab_eq_funcs[i - 1]);
 
 			/* Lookup the associated hash functions */
 			if (!get_op_hash_functions(opexpr->opno,
diff --git a/src/backend/executor/nodeWindowAgg.c b/src/backend/executor/nodeWindowAgg.c
index 372262ad7f6..5680efeb69e 100644
--- a/src/backend/executor/nodeWindowAgg.c
+++ b/src/backend/executor/nodeWindowAgg.c
@@ -1561,7 +1561,9 @@ ExecInitWindowAgg(WindowAgg *node, EState *estate, int eflags)
 
 		fmgr_info_cxt(wfunc->winfnoid, &perfuncstate->flinfo,
 					  econtext->ecxt_per_query_memory);
-		fmgr_info_expr((Node *) wfunc, &perfuncstate->flinfo);
+		fmgr_info_set_collation(wfunc->inputcollid, &perfuncstate->flinfo);
+		fmgr_info_set_expr((Node *) wfunc, &perfuncstate->flinfo);
+
 		get_typlenbyval(wfunc->wintype,
 						&perfuncstate->resulttypeLen,
 						&perfuncstate->resulttypeByVal);
@@ -1792,19 +1794,21 @@ initialize_peragg(WindowAggState *winstate, WindowFunc *wfunc,
 							numArguments,
 							aggtranstype,
 							wfunc->wintype,
+							wfunc->inputcollid,
 							transfn_oid,
 							finalfn_oid,
-							wfunc->collid,
 							&transfnexpr,
 							&finalfnexpr);
 
 	fmgr_info(transfn_oid, &peraggstate->transfn);
-	fmgr_info_expr((Node *) transfnexpr, &peraggstate->transfn);
+	fmgr_info_set_collation(wfunc->inputcollid, &peraggstate->transfn);
+	fmgr_info_set_expr((Node *) transfnexpr, &peraggstate->transfn);
 
 	if (OidIsValid(finalfn_oid))
 	{
 		fmgr_info(finalfn_oid, &peraggstate->finalfn);
-		fmgr_info_expr((Node *) finalfnexpr, &peraggstate->finalfn);
+		fmgr_info_set_collation(wfunc->inputcollid, &peraggstate->finalfn);
+		fmgr_info_set_expr((Node *) finalfnexpr, &peraggstate->finalfn);
 	}
 
 	get_typlenbyval(wfunc->wintype,