Add required functions and hook to support OPENXML with WITH clause (#587)

To add support for Openxml with WITH clause, this commit exposes wrapper hooks to facilitate support for T-SQL
OPENXML functionalities by allowing Babelfish to customised XML table processing during query parsing.

Added two new hooks:
1. pre_transform_openxml_columns_hook , to allow extensions to pre-process OPENXML column definitions
before standard XMLTABLE transformation.
2. openxml_set_namespaces_hook, which fetches and register namespaces in xpath context.

Extensions PR : babelfish-for-postgresql/babelfish_extensions#3820

Task: BABEL-3635, BABEL 6045
Signed-off-by: Harsh Dubey <[email protected]>

Author: harshdubey166

src/backend/parser/parse_clause.c

@@ -56,6 +56,8 @@ sortby_nulls_hook_type  sortby_nulls_hook = NULL;
 
 optimize_explicit_cast_hook_type optimize_explicit_cast_hook = NULL;
 
+pre_transform_openxml_columns_hook_type pre_transform_openxml_columns_hook = NULL;
+
 static int	extractRemainingColumns(ParseState *pstate,
 									ParseNamespaceColumn *src_nscolumns,
 									List *src_colnames,
@@ -711,6 +713,13 @@ transformRangeTableFunc(ParseState *pstate, RangeTableFunc *rtf)
 	constructName = "XMLTABLE";
 	docType = XMLOID;
 
+	/*
+	 * Hook to allow extensions to pre-process OPENXML column definitions
+	 * before standard XMLTABLE transformation.
+	 */
+	if (pre_transform_openxml_columns_hook && tf->functype == TFT_XMLTABLE)
+		pre_transform_openxml_columns_hook(pstate, rtf);
+
 	/*
 	 * We make lateral_only names of this level visible, whether or not the
 	 * RangeTableFunc is explicitly marked LATERAL.  This is needed for SQL

src/backend/utils/adt/xml.c

@@ -232,6 +232,10 @@ const TableFuncRoutine XmlTableRoutine =
 	.DestroyOpaque = XmlTableDestroyOpaque
 };
 
+#ifdef USE_LIBXML
+openxml_set_namespaces_hook_type openxml_set_namespaces_hook = NULL;
+#endif
+
 #define NO_XML_SUPPORT() \
 	ereport(ERROR, \
 			(errcode(ERRCODE_FEATURE_NOT_SUPPORTED), \
@@ -4795,6 +4799,10 @@ XmlTableSetNamespace(TableFuncScanState *state, const char *name, const char *ur
 				 errmsg("DEFAULT namespace is not supported")));
 	xtCxt = GetXmlTableBuilderPrivateData(state, "XmlTableSetNamespace");
 
+	/* For TSQL OPENXML, following hook will fetch and register namespaces in Xpath context. */
+	if (openxml_set_namespaces_hook)
+        return openxml_set_namespaces_hook(xtCxt->xpathcxt, xtCxt->xmlerrcxt, uri);
+
 	if (xmlXPathRegisterNs(xtCxt->xpathcxt,
 						   pg_xmlCharStrndup(name, strlen(name)),
 						   pg_xmlCharStrndup(uri, strlen(uri))))

src/include/parser/parse_clause.h

@@ -62,4 +62,8 @@ extern PGDLLEXPORT optimize_explicit_cast_hook_type optimize_explicit_cast_hook;
 /* functions in parse_jsontable.c */
 extern ParseNamespaceItem *transformJsonTable(ParseState *pstate, JsonTable *jt);
 
+/* Hook for preprocessing OPENXML column definitions before XMLTABLE transformation */
+typedef void (*pre_transform_openxml_columns_hook_type) (ParseState *pstate, RangeTableFunc *rtf);
+extern PGDLLIMPORT pre_transform_openxml_columns_hook_type pre_transform_openxml_columns_hook;
+
 #endif							/* PARSE_CLAUSE_H */

src/include/utils/xml.h

@@ -21,6 +21,7 @@
 #include "nodes/primnodes.h"
 #ifdef USE_LIBXML
 #include <libxml/tree.h>
+#include <libxml/xpath.h>
 #endif
 
 typedef struct varlena xmltype;
@@ -96,6 +97,10 @@ extern xmlDocPtr xml_parse_wrapper(text *data, XmlOptionType xmloption_arg,
 extern xmlChar *pg_xmlCharStrndup_wrapper(const char *str, size_t len);
 extern int parse_xml_decl_wrapper(const xmlChar *str, size_t *lenp,
 						   xmlChar **version, xmlChar **encoding, int *standalone);
+						   
+/* Hook function type for TSQL OPENXML namespace handling */
+typedef void (*openxml_set_namespaces_hook_type) (xmlXPathContext *xpathctx, PgXmlErrorContext *xmlerrcxt, char *doc_id_str);
+extern PGDLLIMPORT openxml_set_namespaces_hook_type openxml_set_namespaces_hook;
 #endif							/* USE_LIBXML */
 
 extern PGDLLIMPORT int xmlbinary;	/* XmlBinaryType, but int for guc enum */