Support XMLTABLE query expression

XMLTABLE is defined by the SQL/XML standard as a feature that allows
turning XML-formatted data into relational form, so that it can be used
as a <table primary> in the FROM clause of a query.

This new construct provides significant simplicity and performance
benefit for XML data processing; what in a client-side custom
implementation was reported to take 20 minutes can be executed in 400ms
using XMLTABLE.  (The same functionality was said to take 10 seconds
using nested PostgreSQL XPath function calls, and 5 seconds using
XMLReader under PL/Python).

The implemented syntax deviates slightly from what the standard
requires.  First, the standard indicates that the PASSING clause is
optional and that multiple XML input documents may be given to it; we
make it mandatory and accept a single document only.  Second, we don't
currently support a default namespace to be specified.

This implementation relies on a new executor node based on a hardcoded
method table.  (Because the grammar is fixed, there is no extensibility
in the current approach; further constructs can be implemented on top of
this such as JSON_TABLE, but they require changes to core code.)

Author: Pavel Stehule, Álvaro Herrera
Extensively reviewed by: Craig Ringer
Discussion: https://postgr.es/m/CAFj8pRAgfzMD-LoSmnMGybD0WsEznLHWap8DO79+-GTRAPR4qA@mail.gmail.com

1 files changed, 76 insertions, 3 deletions

diff --git a/src/backend/parser/parse_relation.c b/src/backend/parser/parse_relation.c
index cf69533b53d..2eea258d28d 100644
--- a/src/backend/parser/parse_relation.c
+++ b/src/backend/parser/parse_relation.c
@@ -1628,6 +1628,69 @@ addRangeTableEntryForFunction(ParseState *pstate,
 }
 
 /*
+ * Add an entry for a table function to the pstate's range table (p_rtable).
+ *
+ * This is much like addRangeTableEntry() except that it makes a tablefunc RTE.
+ */
+RangeTblEntry *
+addRangeTableEntryForTableFunc(ParseState *pstate,
+							   TableFunc *tf,
+							   Alias *alias,
+							   bool lateral,
+							   bool inFromCl)
+{
+	RangeTblEntry *rte = makeNode(RangeTblEntry);
+	char	   *refname = alias ? alias->aliasname : pstrdup("xmltable");
+	Alias	   *eref;
+	int			numaliases;
+
+	Assert(pstate != NULL);
+
+	rte->rtekind = RTE_TABLEFUNC;
+	rte->relid = InvalidOid;
+	rte->subquery = NULL;
+	rte->tablefunc = tf;
+	rte->coltypes = tf->coltypes;
+	rte->coltypmods = tf->coltypmods;
+	rte->colcollations = tf->colcollations;
+	rte->alias = alias;
+
+	eref = alias ? copyObject(alias) : makeAlias(refname, NIL);
+	numaliases = list_length(eref->colnames);
+
+	/* fill in any unspecified alias columns */
+	if (numaliases < list_length(tf->colnames))
+		eref->colnames = list_concat(eref->colnames,
+								   list_copy_tail(tf->colnames, numaliases));
+
+	rte->eref = eref;
+
+	/*
+	 * Set flags and access permissions.
+	 *
+	 * Tablefuncs are never checked for access rights (at least, not by the
+	 * RTE permissions mechanism).
+	 */
+	rte->lateral = lateral;
+	rte->inh = false;			/* never true for tablefunc RTEs */
+	rte->inFromCl = inFromCl;
+
+	rte->requiredPerms = 0;
+	rte->checkAsUser = InvalidOid;
+	rte->selectedCols = NULL;
+	rte->insertedCols = NULL;
+	rte->updatedCols = NULL;
+
+	/*
+	 * Add completed RTE to pstate's range table list, but not to join list
+	 * nor namespace --- caller must do that if appropriate.
+	 */
+	pstate->p_rtable = lappend(pstate->p_rtable, rte);
+
+	return rte;
+}
+
+/*
  * Add an entry for a VALUES list to the pstate's range table (p_rtable).
  *
  * This is much like addRangeTableEntry() except that it makes a values RTE.
@@ -2226,10 +2289,11 @@ expandRTE(RangeTblEntry *rte, int rtindex, int sublevels_up,
 				}
 			}
 			break;
+		case RTE_TABLEFUNC:
 		case RTE_VALUES:
 		case RTE_CTE:
 			{
-				/* Values or CTE RTE */
+				/* Tablefunc, Values or CTE RTE */
 				ListCell   *aliasp_item = list_head(rte->eref->colnames);
 				ListCell   *lct;
 				ListCell   *lcm;
@@ -2638,10 +2702,14 @@ get_rte_attribute_type(RangeTblEntry *rte, AttrNumber attnum,
 				*varcollid = exprCollation(aliasvar);
 			}
 			break;
+		case RTE_TABLEFUNC:
 		case RTE_VALUES:
 		case RTE_CTE:
 			{
-				/* VALUES or CTE RTE --- get type info from lists in the RTE */
+				/*
+				 * tablefunc, VALUES or CTE RTE --- get type info from lists
+				 * in the RTE
+				 */
 				Assert(attnum > 0 && attnum <= list_length(rte->coltypes));
 				*vartype = list_nth_oid(rte->coltypes, attnum - 1);
 				*vartypmod = list_nth_int(rte->coltypmods, attnum - 1);
@@ -2684,9 +2752,14 @@ get_rte_attribute_is_dropped(RangeTblEntry *rte, AttrNumber attnum)
 			}
 			break;
 		case RTE_SUBQUERY:
+		case RTE_TABLEFUNC:
 		case RTE_VALUES:
 		case RTE_CTE:
-			/* Subselect, Values, CTE RTEs never have dropped columns */
+
+			/*
+			 * Subselect, Table Functions, Values, CTE RTEs never have dropped
+			 * columns
+			 */
 			result = false;
 			break;
 		case RTE_JOIN: