Initial SQL/XML support: xml data type and initial set of functions.

5 files changed, 323 insertions, 14 deletions

diff --git a/src/backend/parser/gram.y b/src/backend/parser/gram.y
index c90743a1017..cc400407363 100644
--- a/src/backend/parser/gram.y
+++ b/src/backend/parser/gram.y
@@ -11,7 +11,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/parser/gram.y,v 2.568 2006/11/05 22:42:09 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/parser/gram.y,v 2.569 2006/12/21 16:05:14 petere Exp $
  *
  * HISTORY
  *	  AUTHOR			DATE			MAJOR EVENT
@@ -106,6 +106,7 @@ static void insertSelectOptions(SelectStmt *stmt,
 static Node *makeSetOp(SetOperation op, bool all, Node *larg, Node *rarg);
 static Node *doNegate(Node *n, int location);
 static void doNegateFloat(Value *v);
+static Node *makeXmlExpr(XmlExprOp op, char *name, List *named_args, List *args);
 
 %}
 
@@ -345,6 +346,11 @@ static void doNegateFloat(Value *v);
 %type <str>		OptTableSpace OptConsTableSpace OptTableSpaceOwner
 %type <list>	opt_check_option
 
+%type <target>	xml_attribute_el
+%type <list>	xml_attribute_list xml_attributes
+%type <node>	xml_root_version
+%type <ival>	opt_xml_root_standalone document_or_content xml_whitespace_option
+
 
 /*
  * If you make any token changes, update the keyword table in
@@ -365,13 +371,13 @@ static void doNegateFloat(Value *v);
 	CHARACTER CHARACTERISTICS CHECK CHECKPOINT CLASS CLOSE
 	CLUSTER COALESCE COLLATE COLUMN COMMENT COMMIT
 	COMMITTED CONCURRENTLY CONNECTION CONSTRAINT CONSTRAINTS
-	CONVERSION_P CONVERT COPY CREATE CREATEDB
+	CONTENT CONVERSION_P CONVERT COPY CREATE CREATEDB
 	CREATEROLE CREATEUSER CROSS CSV CURRENT_DATE CURRENT_ROLE CURRENT_TIME
 	CURRENT_TIMESTAMP CURRENT_USER CURSOR CYCLE
 
 	DATABASE DAY_P DEALLOCATE DEC DECIMAL_P DECLARE DEFAULT DEFAULTS
 	DEFERRABLE DEFERRED DEFINER DELETE_P DELIMITER DELIMITERS
-	DESC DISABLE_P DISTINCT DO DOMAIN_P DOUBLE_P DROP
+	DESC DISABLE_P DISTINCT DO DOCUMENT DOMAIN_P DOUBLE_P DROP
 
 	EACH ELSE ENABLE_P ENCODING ENCRYPTED END_P ESCAPE EXCEPT EXCLUDING
 	EXCLUSIVE EXECUTE EXISTS EXPLAIN EXTERNAL EXTRACT
@@ -398,7 +404,7 @@ static void doNegateFloat(Value *v);
 
 	MATCH MAXVALUE MINUTE_P MINVALUE MODE MONTH_P MOVE
 
-	NAMES NATIONAL NATURAL NCHAR NEW NEXT NO NOCREATEDB
+	NAME NAMES NATIONAL NATURAL NCHAR NEW NEXT NO NOCREATEDB
 	NOCREATEROLE NOCREATEUSER NOINHERIT NOLOGIN_P NONE NOSUPERUSER
 	NOT NOTHING NOTIFY NOTNULL NOWAIT NULL_P NULLIF NUMERIC
 
@@ -417,8 +423,8 @@ static void doNegateFloat(Value *v);
 
 	SAVEPOINT SCHEMA SCROLL SECOND_P SECURITY SELECT SEQUENCE
 	SERIALIZABLE SESSION SESSION_USER SET SETOF SHARE
-	SHOW SIMILAR SIMPLE SMALLINT SOME STABLE START STATEMENT
-	STATISTICS STDIN STDOUT STORAGE STRICT_P SUBSTRING SUPERUSER_P SYMMETRIC
+	SHOW SIMILAR SIMPLE SMALLINT SOME STABLE STANDALONE START STATEMENT
+	STATISTICS STDIN STDOUT STORAGE STRICT_P STRIP SUBSTRING SUPERUSER_P SYMMETRIC
 	SYSID SYSTEM_P
 
 	TABLE TABLESPACE TEMP TEMPLATE TEMPORARY THEN TIME TIMESTAMP
@@ -428,12 +434,15 @@ static void doNegateFloat(Value *v);
 	UNCOMMITTED UNENCRYPTED UNION UNIQUE UNKNOWN UNLISTEN UNTIL
 	UPDATE USER USING
 
-	VACUUM VALID VALIDATOR VALUES VARCHAR VARYING
-	VERBOSE VIEW VOLATILE
+	VACUUM VALID VALIDATOR VALUE VALUES VARCHAR VARYING
+	VERBOSE VERSION VIEW VOLATILE
+
+	WHEN WHERE WHITESPACE WITH WITHOUT WORK WRITE
 
-	WHEN WHERE WITH WITHOUT WORK WRITE
+	XMLATTRIBUTES XMLCONCAT XMLELEMENT XMLFOREST XMLPARSE
+	XMLPI XMLROOT XMLSERIALIZE
 
-	YEAR_P
+	YEAR_P YES
 
 	ZONE
 
@@ -484,6 +493,7 @@ static void doNegateFloat(Value *v);
  * left-associativity among the JOIN rules themselves.
  */
 %left		JOIN CROSS LEFT FULL RIGHT INNER_P NATURAL
+%right		PRESERVE STRIP
 %%
 
 /*
@@ -7868,6 +7878,146 @@ func_expr:	func_name '(' ')'
 					v->op = IS_LEAST;
 					$$ = (Node *)v;
 				}
+			| XMLCONCAT '(' expr_list ')'
+				{		
+					$$ = makeXmlExpr(IS_XMLCONCAT, NULL, NULL, $3);
+				}
+			| XMLELEMENT '(' NAME ColLabel ')'
+				{
+					$$ = makeXmlExpr(IS_XMLELEMENT, $4, NULL, NULL);
+				}
+			| XMLELEMENT '(' NAME ColLabel ',' xml_attributes ')'
+				{
+					$$ = makeXmlExpr(IS_XMLELEMENT, $4, $6, NULL);
+				}
+			| XMLELEMENT '(' NAME ColLabel ',' expr_list ')'
+				{
+					$$ = makeXmlExpr(IS_XMLELEMENT, $4, NULL, $6);
+				}
+			| XMLELEMENT '(' NAME ColLabel ',' xml_attributes ',' expr_list ')'
+				{
+					$$ = makeXmlExpr(IS_XMLELEMENT, $4, $6, $8);
+				}
+			| XMLFOREST '(' xml_attribute_list ')'
+				{
+					$$ = makeXmlExpr(IS_XMLFOREST, NULL, $3, NULL);
+				}
+			| XMLPARSE '(' document_or_content a_expr xml_whitespace_option ')'
+				{
+					FuncCall *n = makeNode(FuncCall);
+					n->funcname = SystemFuncName("xmlparse");
+					n->args = list_make3(makeBoolAConst($3 == DOCUMENT), $4, makeBoolAConst($5 == PRESERVE));
+					n->agg_star = FALSE;
+					n->agg_distinct = FALSE;
+					n->location = @1;
+					$$ = (Node *)n;
+				}
+			| XMLPI '(' NAME ColLabel ')'
+				{
+					FuncCall *n = makeNode(FuncCall);
+					n->funcname = SystemFuncName("xmlpi");
+					n->args = list_make1(makeStringConst($4, NULL));
+					n->agg_star = FALSE;
+					n->agg_distinct = FALSE;
+					n->location = @1;
+					$$ = (Node *)n;
+				}
+			| XMLPI '(' NAME ColLabel ',' a_expr ')'
+				{
+					FuncCall *n = makeNode(FuncCall);
+					n->funcname = SystemFuncName("xmlpi");
+					n->args = list_make2(makeStringConst($4, NULL), $6);
+					n->agg_star = FALSE;
+					n->agg_distinct = FALSE;
+					n->location = @1;
+					$$ = (Node *)n;
+				}
+			| XMLROOT '(' a_expr ',' xml_root_version opt_xml_root_standalone ')'
+				{
+					FuncCall *n = makeNode(FuncCall);
+					Node *ver;
+					A_Const *sa;
+
+					if ($5)
+						ver = $5;
+					else
+					{
+						A_Const *val;
+
+						val = makeNode(A_Const);
+						val->val.type = T_Null;
+						ver = (Node *) val;
+					}
+
+					if ($6)
+						sa = makeBoolAConst($6 == 1);
+					else
+					{
+						sa = makeNode(A_Const);
+						sa->val.type = T_Null;
+					}
+
+					n->funcname = SystemFuncName("xmlroot");
+					n->args = list_make3($3, ver, sa);
+					n->agg_star = FALSE;
+					n->agg_distinct = FALSE;
+					n->location = @1;
+					$$ = (Node *)n;
+				}
+			| XMLSERIALIZE '(' document_or_content a_expr AS Typename ')'
+				{
+					/*
+					 * FIXME: This should be made distinguishable from
+					 * CAST (for reverse compilation at least).
+					 */
+					$$ = makeTypeCast($4, $6);
+				}
+		;
+
+/*
+ * SQL/XML support
+ */
+xml_root_version: VERSION a_expr	{ $$ = $2; }
+			| VERSION NO VALUE		{ $$ = NULL; }
+		;
+
+opt_xml_root_standalone: ',' STANDALONE YES	{ $$ = 1; }
+			| ',' STANDALONE NO				{ $$ = -1; }
+			| ',' STANDALONE NO VALUE		{ $$ = 0; }
+			| /*EMPTY*/						{ $$ = 0; }
+		;
+
+xml_attributes: XMLATTRIBUTES '(' xml_attribute_list ')' { $$ = $3; }
+		;
+
+xml_attribute_list:	xml_attribute_el					{ $$ = list_make1($1); }
+			| xml_attribute_list ',' xml_attribute_el	{ $$ = lappend($1, $3); }
+		;
+
+xml_attribute_el: a_expr AS ColLabel
+				{
+					$$ = makeNode(ResTarget);
+					$$->name = $3;
+					$$->indirection = NULL;
+					$$->val = (Node *) $1;
+
+				}
+			| a_expr
+				{
+					$$ = makeNode(ResTarget);
+					$$->name = NULL;
+					$$->indirection = NULL;
+					$$->val = (Node *) $1;				
+				}
+		;
+
+document_or_content: DOCUMENT { $$ = DOCUMENT; }
+			| CONTENT { $$ = CONTENT; }
+		;
+
+xml_whitespace_option: PRESERVE WHITESPACE { $$ = PRESERVE; }
+			| STRIP WHITESPACE { $$ = STRIP; }
+			| /*EMPTY*/ { $$ = STRIP; }
 		;
 
 /*
@@ -8562,6 +8712,7 @@ unreserved_keyword:
 			| CONCURRENTLY
 			| CONNECTION
 			| CONSTRAINTS
+			| CONTENT
 			| CONVERSION_P
 			| COPY
 			| CREATEDB
@@ -8581,6 +8732,7 @@ unreserved_keyword:
 			| DELIMITER
 			| DELIMITERS
 			| DISABLE_P
+			| DOCUMENT
 			| DOMAIN_P
 			| DOUBLE_P
 			| DROP
@@ -8640,6 +8792,7 @@ unreserved_keyword:
 			| MODE
 			| MONTH_P
 			| MOVE
+			| NAME
 			| NAMES
 			| NEXT
 			| NO
@@ -8700,12 +8853,14 @@ unreserved_keyword:
 			| SHOW
 			| SIMPLE
 			| STABLE
+			| STANDALONE
 			| START
 			| STATEMENT
 			| STATISTICS
 			| STDIN
 			| STDOUT
 			| STORAGE
+			| STRIP
 			| SUPERUSER_P
 			| SYSID
 			| SYSTEM_P
@@ -8729,13 +8884,17 @@ unreserved_keyword:
 			| VALID
 			| VALIDATOR
 			| VARYING
+			| VERSION
 			| VIEW
+			| VALUE
 			| VOLATILE
+			| WHITESPACE
 			| WITH
 			| WITHOUT
 			| WORK
 			| WRITE
 			| YEAR_P
+			| YES
 			| ZONE
 		;
 
@@ -8788,6 +8947,14 @@ col_name_keyword:
 			| TRIM
 			| VALUES
 			| VARCHAR
+			| XMLATTRIBUTES
+			| XMLELEMENT
+			| XMLCONCAT
+			| XMLFOREST
+			| XMLPARSE
+			| XMLPI
+			| XMLROOT
+			| XMLSERIALIZE
 		;
 
 /* Function identifier --- keywords that can be function names.
@@ -9322,6 +9489,17 @@ doNegateFloat(Value *v)
 	}
 }
 
+static Node *
+makeXmlExpr(XmlExprOp op, char *name, List *named_args, List *args)
+{
+	XmlExpr *x = makeNode(XmlExpr);
+	x->op = op;
+	x->name = name;
+	x->named_args = named_args;
+	x->args = args;
+	return (Node *) x;
+}
+
 /*
  * Must undefine base_yylex before including scan.c, since we want it
  * to create the function base_yylex not filtered_base_yylex.
diff --git a/src/backend/parser/keywords.c b/src/backend/parser/keywords.c
index 50fd3aac405..b5e49e955fc 100644
--- a/src/backend/parser/keywords.c
+++ b/src/backend/parser/keywords.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/parser/keywords.c,v 1.177 2006/10/07 21:51:02 petere Exp $
+ *	  $PostgreSQL: pgsql/src/backend/parser/keywords.c,v 1.178 2006/12/21 16:05:14 petere Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -89,6 +89,7 @@ static const ScanKeyword ScanKeywords[] = {
 	{"connection", CONNECTION},
 	{"constraint", CONSTRAINT},
 	{"constraints", CONSTRAINTS},
+	{"content", CONTENT},
 	{"conversion", CONVERSION_P},
 	{"convert", CONVERT},
 	{"copy", COPY},
@@ -123,6 +124,7 @@ static const ScanKeyword ScanKeywords[] = {
 	{"disable", DISABLE_P},
 	{"distinct", DISTINCT},
 	{"do", DO},
+	{"document", DOCUMENT},
 	{"domain", DOMAIN_P},
 	{"double", DOUBLE_P},
 	{"drop", DROP},
@@ -218,6 +220,7 @@ static const ScanKeyword ScanKeywords[] = {
 	{"mode", MODE},
 	{"month", MONTH_P},
 	{"move", MOVE},
+	{"name", NAME},
 	{"names", NAMES},
 	{"national", NATIONAL},
 	{"natural", NATURAL},
@@ -314,6 +317,7 @@ static const ScanKeyword ScanKeywords[] = {
 	{"smallint", SMALLINT},
 	{"some", SOME},
 	{"stable", STABLE},
+	{"standalone", STANDALONE},
 	{"start", START},
 	{"statement", STATEMENT},
 	{"statistics", STATISTICS},
@@ -321,6 +325,7 @@ static const ScanKeyword ScanKeywords[] = {
 	{"stdout", STDOUT},
 	{"storage", STORAGE},
 	{"strict", STRICT_P},
+	{"strip", STRIP},
 	{"substring", SUBSTRING},
 	{"superuser", SUPERUSER_P},
 	{"symmetric", SYMMETRIC},
@@ -357,19 +362,31 @@ static const ScanKeyword ScanKeywords[] = {
 	{"vacuum", VACUUM},
 	{"valid", VALID},
 	{"validator", VALIDATOR},
+	{"value", VALUE},
 	{"values", VALUES},
 	{"varchar", VARCHAR},
 	{"varying", VARYING},
 	{"verbose", VERBOSE},
+	{"version", VERSION},
 	{"view", VIEW},
 	{"volatile", VOLATILE},
 	{"when", WHEN},
 	{"where", WHERE},
+	{"whitespace", WHITESPACE},
 	{"with", WITH},
 	{"without", WITHOUT},
 	{"work", WORK},
 	{"write", WRITE},
+	{"xmlattributes", XMLATTRIBUTES},
+	{"xmlconcat", XMLCONCAT},
+	{"xmlelement", XMLELEMENT},
+	{"xmlforest", XMLFOREST},
+	{"xmlparse", XMLPARSE},
+	{"xmlpi", XMLPI},
+	{"xmlroot", XMLROOT},
+	{"xmlserialize", XMLSERIALIZE},
 	{"year", YEAR_P},
+	{"yes", YES},
 	{"zone", ZONE},
 };
 
diff --git a/src/backend/parser/parse_coerce.c b/src/backend/parser/parse_coerce.c
index 2a468c68271..5670ed4fe74 100644
--- a/src/backend/parser/parse_coerce.c
+++ b/src/backend/parser/parse_coerce.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/parser/parse_coerce.c,v 2.147 2006/12/10 22:13:26 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/parser/parse_coerce.c,v 2.148 2006/12/21 16:05:14 petere Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -919,6 +919,46 @@ coerce_to_bigint(ParseState *pstate, Node *node,
 	return node;
 }
 
+/*
+ * coerce_to_xml()
+ *		Coerce an argument of a construct that requires xml input.
+ *		Also check that input is not a set.
+ *
+ * Returns the possibly-transformed node tree.
+ *
+ * As with coerce_type, pstate may be NULL if no special unknown-Param
+ * processing is wanted.
+ */
+Node *
+coerce_to_xml(ParseState *pstate, Node *node,
+				 const char *constructName)
+{
+	Oid			inputTypeId = exprType(node);
+
+	if (inputTypeId != XMLOID)
+	{
+		node = coerce_to_target_type(pstate, node, inputTypeId,
+									 XMLOID, -1,
+									 COERCION_ASSIGNMENT,
+									 COERCE_IMPLICIT_CAST);
+		if (node == NULL)
+			ereport(ERROR,
+					(errcode(ERRCODE_DATATYPE_MISMATCH),
+			/* translator: first %s is name of a SQL construct, eg LIMIT */
+					 errmsg("argument of %s must be type xml, not type %s",
+							constructName, format_type_be(inputTypeId))));
+	}
+
+	if (expression_returns_set(node))
+		ereport(ERROR,
+				(errcode(ERRCODE_DATATYPE_MISMATCH),
+		/* translator: %s is name of a SQL construct, eg LIMIT */
+				 errmsg("argument of %s must not return a set",
+						constructName)));
+
+	return node;
+}
+
 
 /* select_common_type()
  *		Determine the common supertype of a list of input expression types.
diff --git a/src/backend/parser/parse_expr.c b/src/backend/parser/parse_expr.c
index b1b6ea81456..234a15b6afb 100644
--- a/src/backend/parser/parse_expr.c
+++ b/src/backend/parser/parse_expr.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/parser/parse_expr.c,v 1.199 2006/12/10 22:13:26 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/parser/parse_expr.c,v 1.200 2006/12/21 16:05:14 petere Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -33,6 +33,7 @@
 #include "parser/parse_type.h"
 #include "utils/builtins.h"
 #include "utils/lsyscache.h"
+#include "utils/xml.h"
 
 
 bool		Transform_null_equals = false;
@@ -55,6 +56,7 @@ static Node *transformArrayExpr(ParseState *pstate, ArrayExpr *a);
 static Node *transformRowExpr(ParseState *pstate, RowExpr *r);
 static Node *transformCoalesceExpr(ParseState *pstate, CoalesceExpr *c);
 static Node *transformMinMaxExpr(ParseState *pstate, MinMaxExpr *m);
+static Node *transformXmlExpr(ParseState *pstate, XmlExpr *x);
 static Node *transformBooleanTest(ParseState *pstate, BooleanTest *b);
 static Node *transformColumnRef(ParseState *pstate, ColumnRef *cref);
 static Node *transformWholeRowRef(ParseState *pstate, char *schemaname,
@@ -232,6 +234,10 @@ transformExpr(ParseState *pstate, Node *expr)
 			result = transformBooleanTest(pstate, (BooleanTest *) expr);
 			break;
 
+		case T_XmlExpr:
+			result = transformXmlExpr(pstate, (XmlExpr *) expr);
+			break;
+
 			/*********************************************
 			 * Quietly accept node types that may be presented when we are
 			 * called on an already-transformed tree.
@@ -1409,6 +1415,56 @@ transformBooleanTest(ParseState *pstate, BooleanTest *b)
 	return (Node *) b;
 }
 
+static Node *
+transformXmlExpr(ParseState *pstate, XmlExpr *x)
+{
+	ListCell	*lc;
+	XmlExpr *newx = makeNode(XmlExpr);
+
+	newx->op = x->op;
+	if (x->name)
+		newx->name = map_sql_identifier_to_xml_name(x->name, false);
+	else
+		newx->name = NULL;
+
+	foreach(lc, x->named_args)
+	{
+		ResTarget 	*r = (ResTarget *) lfirst(lc);
+		Node 		*expr = transformExpr(pstate, r->val);
+		char 		*argname = NULL;
+
+		if (r->name)
+			argname = map_sql_identifier_to_xml_name(r->name, false);
+		else if (IsA(r->val, ColumnRef))
+			argname = map_sql_identifier_to_xml_name(FigureColname(r->val), true);
+		else
+			ereport(ERROR,
+					(errcode(ERRCODE_SYNTAX_ERROR),
+					 x->op == IS_XMLELEMENT
+					 ? errmsg("unnamed attribute value must be a column reference")
+					 : errmsg("unnamed element value must be a column reference")));
+
+		newx->named_args = lappend(newx->named_args, 
+								   makeTargetEntry((Expr *) expr, 0, argname, false)); 
+	}
+
+	foreach(lc, x->args)
+	{
+		Node	   *e = (Node *) lfirst(lc);
+		Node	   *newe;
+
+		newe = coerce_to_xml(pstate, transformExpr(pstate, e),
+							 (x->op == IS_XMLCONCAT 
+							  ? "XMLCONCAT"
+							  : (x->op == IS_XMLELEMENT
+								 ? "XMLELEMENT"
+								 : "XMLFOREST")));
+		newx->args = lappend(newx->args, newe);
+	}
+		
+	return (Node *) newx;
+}
+
 /*
  * Construct a whole-row reference to represent the notation "relation.*".
  *
@@ -1668,6 +1724,9 @@ exprType(Node *expr)
 		case T_BooleanTest:
 			type = BOOLOID;
 			break;
+		case T_XmlExpr:
+			type = XMLOID;
+			break;
 		case T_CoerceToDomain:
 			type = ((CoerceToDomain *) expr)->resulttype;
 			break;
diff --git a/src/backend/parser/parse_target.c b/src/backend/parser/parse_target.c
index bb4b065eebb..906d96e45c6 100644
--- a/src/backend/parser/parse_target.c
+++ b/src/backend/parser/parse_target.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/parser/parse_target.c,v 1.149 2006/10/04 00:29:56 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/parser/parse_target.c,v 1.150 2006/12/21 16:05:14 petere Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -1315,6 +1315,21 @@ FigureColnameInternal(Node *node, char **name)
 					return 2;
 			}
 			break;
+		case T_XmlExpr:
+			/* make SQL/XML functions act like a regular function */
+			switch (((XmlExpr*) node)->op)
+			{		
+				case IS_XMLCONCAT:
+					*name = "xmlconcat";
+					return 2;
+				case IS_XMLELEMENT:
+					*name = "xmlelement";
+					return 2;
+				case IS_XMLFOREST:
+					*name = "xmlforest";
+					return 2;
+			} 
+			break;
 		default:
 			break;
 	}