Create a validator for plpgsql, so that some minimal syntax checking

is done at creation time for plpgsql functions.  Improve createlang and
droplang to support adding/dropping validators for PLs.  Initial steps
towards producing a syntax error position from plpgsql syntax errors
(this part is a work in progress, and will change depending on outcome
of current discussions).

src/pl/plpgsql/src/pl_comp.c

@@ -3,7 +3,7 @@
  *			  procedural language
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/pl/plpgsql/src/pl_comp.c,v 1.73 2004/01/07 18:56:30 neilc Exp $
+ *	  $PostgreSQL: pgsql/src/pl/plpgsql/src/pl_comp.c,v 1.74 2004/03/19 18:58:07 tgl Exp $
  *
  *	  This software is copyrighted by Jan Wieck - Hamburg.
  *
@@ -101,13 +101,15 @@ typedef struct plpgsql_hashent
  */
 static PLpgSQL_function *do_compile(FunctionCallInfo fcinfo,
 		   HeapTuple procTup,
-		   PLpgSQL_func_hashkey *hashkey);
+		   PLpgSQL_func_hashkey *hashkey,
+		   bool forValidator);
 static void plpgsql_compile_error_callback(void *arg);
 static char **fetchArgNames(HeapTuple procTup, int nargs);
 static PLpgSQL_type *build_datatype(HeapTuple typeTup, int32 typmod);
 static void compute_function_hashkey(FunctionCallInfo fcinfo,
 						 Form_pg_proc procStruct,
-						 PLpgSQL_func_hashkey *hashkey);
+						 PLpgSQL_func_hashkey *hashkey,
+						 bool forValidator);
 static PLpgSQL_function *plpgsql_HashTableLookup(PLpgSQL_func_hashkey *func_key);
 static void plpgsql_HashTableInsert(PLpgSQL_function *function,
 						PLpgSQL_func_hashkey *func_key);
@@ -134,12 +136,15 @@ perm_fmgr_info(Oid functionId, FmgrInfo *finfo)
 /* ----------
  * plpgsql_compile		Make an execution tree for a PL/pgSQL function.
  *
+ * If forValidator is true, we're only compiling for validation purposes,
+ * and so some checks are skipped.
+ *
  * Note: it's important for this to fall through quickly if the function
  * has already been compiled.
  * ----------
  */
 PLpgSQL_function *
-plpgsql_compile(FunctionCallInfo fcinfo)
+plpgsql_compile(FunctionCallInfo fcinfo, bool forValidator)
 {
 	Oid			funcOid = fcinfo->flinfo->fn_oid;
 	HeapTuple	procTup;
@@ -171,7 +176,7 @@ plpgsql_compile(FunctionCallInfo fcinfo)
 			plpgsql_HashTableInit();
 
 		/* Compute hashkey using function signature and actual arg types */
-		compute_function_hashkey(fcinfo, procStruct, &hashkey);
+		compute_function_hashkey(fcinfo, procStruct, &hashkey, forValidator);
 		hashkey_valid = true;
 
 		/* And do the lookup */
@@ -205,12 +210,13 @@ plpgsql_compile(FunctionCallInfo fcinfo)
 		 * the completed function.
 		 */
 		if (!hashkey_valid)
-			compute_function_hashkey(fcinfo, procStruct, &hashkey);
+			compute_function_hashkey(fcinfo, procStruct, &hashkey,
+									 forValidator);
 
 		/*
 		 * Do the hard part.
 		 */
-		function = do_compile(fcinfo, procTup, &hashkey);
+		function = do_compile(fcinfo, procTup, &hashkey, forValidator);
 	}
 
 	ReleaseSysCache(procTup);
@@ -232,7 +238,8 @@ plpgsql_compile(FunctionCallInfo fcinfo)
 static PLpgSQL_function *
 do_compile(FunctionCallInfo fcinfo,
 		   HeapTuple procTup,
-		   PLpgSQL_func_hashkey *hashkey)
+		   PLpgSQL_func_hashkey *hashkey,
+		   bool forValidator)
 {
 	Form_pg_proc procStruct = (Form_pg_proc) GETSTRUCT(procTup);
 	int			functype = CALLED_AS_TRIGGER(fcinfo) ? T_TRIGGER : T_FUNCTION;
@@ -308,7 +315,8 @@ do_compile(FunctionCallInfo fcinfo,
 			/*
 			 * Check for a polymorphic returntype. If found, use the
 			 * actual returntype type from the caller's FuncExpr node, if
-			 * we have one.
+			 * we have one.  (In validation mode we arbitrarily assume we
+			 * are dealing with integers.)
 			 *
 			 * Note: errcode is FEATURE_NOT_SUPPORTED because it should
 			 * always work; if it doesn't we're in some context that fails
@@ -317,7 +325,15 @@ do_compile(FunctionCallInfo fcinfo,
 			rettypeid = procStruct->prorettype;
 			if (rettypeid == ANYARRAYOID || rettypeid == ANYELEMENTOID)
 			{
-				rettypeid = get_fn_expr_rettype(fcinfo->flinfo);
+				if (forValidator)
+				{
+					if (rettypeid == ANYARRAYOID)
+						rettypeid = INT4ARRAYOID;
+					else
+						rettypeid = INT4OID;
+				}
+				else
+					rettypeid = get_fn_expr_rettype(fcinfo->flinfo);
 				if (!OidIsValid(rettypeid))
 					ereport(ERROR,
 							(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
@@ -1758,22 +1774,6 @@ plpgsql_add_initdatums(int **varnos)
 }
 
 
-/* ---------
- * plpgsql_yyerror			Handle parser error
- * ---------
- */
-
-void
-plpgsql_yyerror(const char *s)
-{
-	plpgsql_error_lineno = plpgsql_scanner_lineno();
-	ereport(ERROR,
-			(errcode(ERRCODE_SYNTAX_ERROR),
-	/* translator: first %s is a phrase like "syntax error" */
-			 errmsg("%s at or near \"%s\"", s, plpgsql_yytext)));
-}
-
-
 /*
  * Compute the hashkey for a given function invocation
  *
@@ -1782,7 +1782,8 @@ plpgsql_yyerror(const char *s)
 static void
 compute_function_hashkey(FunctionCallInfo fcinfo,
 						 Form_pg_proc procStruct,
-						 PLpgSQL_func_hashkey *hashkey)
+						 PLpgSQL_func_hashkey *hashkey,
+						 bool forValidator)
 {
 	int			i;
 
@@ -1792,8 +1793,12 @@ compute_function_hashkey(FunctionCallInfo fcinfo,
 	/* get function OID */
 	hashkey->funcOid = fcinfo->flinfo->fn_oid;
 
-	/* if trigger, get relation OID */
-	if (CALLED_AS_TRIGGER(fcinfo))
+	/*
+	 * if trigger, get relation OID.  In validation mode we do not know what
+	 * relation is intended to be used, so we leave trigrelOid zero; the
+	 * hash entry built in this case will never really be used.
+	 */
+	if (CALLED_AS_TRIGGER(fcinfo) && !forValidator)
 	{
 		TriggerData *trigdata = (TriggerData *) fcinfo->context;
 
@@ -1808,6 +1813,9 @@ compute_function_hashkey(FunctionCallInfo fcinfo,
 		/*
 		 * Check for polymorphic arguments. If found, use the actual
 		 * parameter type from the caller's FuncExpr node, if we have one.
+		 * (In validation mode we arbitrarily assume we are dealing with
+		 * integers.  This lets us build a valid, if possibly useless,
+		 * function hashtable entry.)
 		 *
 		 * We can support arguments of type ANY the same way as normal
 		 * polymorphic arguments.
@@ -1815,7 +1823,15 @@ compute_function_hashkey(FunctionCallInfo fcinfo,
 		if (argtypeid == ANYARRAYOID || argtypeid == ANYELEMENTOID ||
 			argtypeid == ANYOID)
 		{
-			argtypeid = get_fn_expr_argtype(fcinfo->flinfo, i);
+			if (forValidator)
+			{
+				if (argtypeid == ANYARRAYOID)
+					argtypeid = INT4ARRAYOID;
+				else
+					argtypeid = INT4OID;
+			}
+			else
+				argtypeid = get_fn_expr_argtype(fcinfo->flinfo, i);
 			if (!OidIsValid(argtypeid))
 				ereport(ERROR,
 						(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),

src/pl/plpgsql/src/pl_handler.c

@@ -3,7 +3,7 @@
  *			  procedural language
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/pl/plpgsql/src/pl_handler.c,v 1.19 2003/11/29 19:52:12 pgsql Exp $
+ *	  $PostgreSQL: pgsql/src/pl/plpgsql/src/pl_handler.c,v 1.20 2004/03/19 18:58:07 tgl Exp $
  *
  *	  This software is copyrighted by Jan Wieck - Hamburg.
  *
@@ -42,8 +42,11 @@
 #include "catalog/pg_proc.h"
 #include "catalog/pg_type.h"
 #include "utils/builtins.h"
+#include "utils/lsyscache.h"
 #include "utils/syscache.h"
 
+extern bool check_function_bodies;
+
 static int	plpgsql_firstcall = 1;
 
 static void plpgsql_init_all(void);
@@ -88,7 +91,6 @@ plpgsql_init_all(void)
 /* ----------
  * plpgsql_call_handler
  *
- * This is the only visible function of the PL interpreter.
  * The PostgreSQL function manager and trigger manager
  * call this function for execution of PL/pgSQL procedures.
  * ----------
@@ -111,7 +113,7 @@ plpgsql_call_handler(PG_FUNCTION_ARGS)
 		elog(ERROR, "SPI_connect failed");
 
 	/* Find or compile the function */
-	func = plpgsql_compile(fcinfo);
+	func = plpgsql_compile(fcinfo, false);
 
 	/*
 	 * Determine if called as function or trigger and call appropriate
@@ -131,3 +133,118 @@ plpgsql_call_handler(PG_FUNCTION_ARGS)
 
 	return retval;
 }
+
+/* ----------
+ * plpgsql_validator
+ *
+ * This function attempts to validate a PL/pgSQL function at
+ * CREATE FUNCTION time.
+ * ----------
+ */
+PG_FUNCTION_INFO_V1(plpgsql_validator);
+
+Datum
+plpgsql_validator(PG_FUNCTION_ARGS)
+{
+	Oid			funcoid = PG_GETARG_OID(0);
+	HeapTuple	tuple;
+	Form_pg_proc proc;
+	char		functyptype;
+	bool		istrigger = false;
+	bool		haspolyresult;
+	bool		haspolyarg;
+	int			i;
+
+	/* perform initialization */
+	plpgsql_init_all();
+
+	/* Get the new function's pg_proc entry */
+	tuple = SearchSysCache(PROCOID,
+						   ObjectIdGetDatum(funcoid),
+						   0, 0, 0);
+	if (!HeapTupleIsValid(tuple))
+		elog(ERROR, "cache lookup failed for function %u", funcoid);
+	proc = (Form_pg_proc) GETSTRUCT(tuple);
+
+	functyptype = get_typtype(proc->prorettype);
+
+	/* Disallow pseudotype result */
+	/* except for TRIGGER, RECORD, VOID, ANYARRAY, or ANYELEMENT */
+	if (functyptype == 'p')
+	{
+		/* we assume OPAQUE with no arguments means a trigger */
+		if (proc->prorettype == TRIGGEROID ||
+			(proc->prorettype == OPAQUEOID && proc->pronargs == 0))
+			istrigger = true;
+		else if (proc->prorettype == ANYARRAYOID ||
+				 proc->prorettype == ANYELEMENTOID)
+			haspolyresult = true;
+		else if (proc->prorettype != RECORDOID &&
+				 proc->prorettype != VOIDOID)
+			ereport(ERROR,
+					(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+					 errmsg("plpgsql functions cannot return type %s",
+							format_type_be(proc->prorettype))));
+	}
+
+	/* Disallow pseudotypes in arguments */
+	/* except for ANYARRAY or ANYELEMENT */
+	haspolyarg = false;
+	for (i = 0; i < proc->pronargs; i++)
+	{
+		if (get_typtype(proc->proargtypes[i]) == 'p')
+		{
+			if (proc->proargtypes[i] == ANYARRAYOID ||
+				proc->proargtypes[i] == ANYELEMENTOID)
+				haspolyarg = true;
+			else
+				ereport(ERROR,
+						(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+						 errmsg("plpgsql functions cannot take type %s",
+								format_type_be(proc->proargtypes[i]))));
+		}
+	}
+
+	/* Postpone body checks if !check_function_bodies */
+	if (check_function_bodies)
+	{
+		FunctionCallInfoData fake_fcinfo;
+		FmgrInfo	flinfo;
+		TriggerData trigdata;
+
+		/*
+		 * Connect to SPI manager (is this needed for compilation?)
+		 */
+		if (SPI_connect() != SPI_OK_CONNECT)
+			elog(ERROR, "SPI_connect failed");
+
+		/*
+		 * Set up a fake fcinfo with just enough info to satisfy
+		 * plpgsql_compile().
+		 */
+		MemSet(&fake_fcinfo, 0, sizeof(fake_fcinfo));
+		MemSet(&flinfo, 0, sizeof(flinfo));
+		fake_fcinfo.flinfo = &flinfo;
+		flinfo.fn_oid = funcoid;
+		flinfo.fn_mcxt = CurrentMemoryContext;
+		if (istrigger)
+		{
+			MemSet(&trigdata, 0, sizeof(trigdata));
+			trigdata.type = T_TriggerData;
+			fake_fcinfo.context = (Node *) &trigdata;
+		}
+
+		/* Test-compile the function */
+		plpgsql_compile(&fake_fcinfo, true);
+
+		/*
+		 * Disconnect from SPI manager
+		 */
+		if (SPI_finish() != SPI_OK_FINISH)
+			elog(ERROR, "SPI_finish failed");
+	}
+
+	ReleaseSysCache(tuple);
+
+	PG_RETURN_VOID();
+}

src/pl/plpgsql/src/plpgsql.h

@@ -3,7 +3,7 @@
  *			  procedural language
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/pl/plpgsql/src/plpgsql.h,v 1.44 2004/02/25 18:10:51 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/pl/plpgsql/src/plpgsql.h,v 1.45 2004/03/19 18:58:07 tgl Exp $
  *
  *	  This software is copyrighted by Jan Wieck - Hamburg.
  *
@@ -620,7 +620,8 @@ extern PLpgSQL_function *plpgsql_curr_compile;
  * Functions in pl_comp.c
  * ----------
  */
-extern PLpgSQL_function *plpgsql_compile(FunctionCallInfo fcinfo);
+extern PLpgSQL_function *plpgsql_compile(FunctionCallInfo fcinfo,
+										 bool forValidator);
 extern int	plpgsql_parse_word(char *word);
 extern int	plpgsql_parse_dblword(char *word);
 extern int	plpgsql_parse_tripword(char *word);
@@ -633,7 +634,6 @@ extern PLpgSQL_type *plpgsql_parse_datatype(char *string);
 extern PLpgSQL_row *plpgsql_build_rowtype(Oid classOid);
 extern void plpgsql_adddatum(PLpgSQL_datum * new);
 extern int	plpgsql_add_initdatums(int **varnos);
-extern void plpgsql_yyerror(const char *s);
 extern void plpgsql_HashTableInit(void);
 
 /* ----------
@@ -642,6 +642,7 @@ extern void plpgsql_HashTableInit(void);
  */
 extern void plpgsql_init(void);
 extern Datum plpgsql_call_handler(PG_FUNCTION_ARGS);
+extern Datum plpgsql_validator(PG_FUNCTION_ARGS);
 
 /* ----------
  * Functions in pl_exec.c
@@ -691,6 +692,7 @@ extern int	plpgsql_yyparse(void);
 extern int	plpgsql_base_yylex(void);
 extern int	plpgsql_yylex(void);
 extern void plpgsql_push_back_token(int token);
+extern void plpgsql_yyerror(const char *message);
 extern int	plpgsql_scanner_lineno(void);
 extern void plpgsql_scanner_init(const char *str, int functype);
 extern void plpgsql_scanner_finish(void);

src/pl/plpgsql/src/scan.l

@@ -4,7 +4,7 @@
  *			  procedural language
  *
  * IDENTIFICATION
- *    $PostgreSQL: pgsql/src/pl/plpgsql/src/scan.l,v 1.32 2004/02/25 18:10:51 tgl Exp $
+ *    $PostgreSQL: pgsql/src/pl/plpgsql/src/scan.l,v 1.33 2004/03/19 18:58:07 tgl Exp $
  *
  *    This software is copyrighted by Jan Wieck - Hamburg.
  *
@@ -38,6 +38,8 @@
 
 #include "plpgsql.h"
 
+#include "mb/pg_wchar.h"
+
 
 /* No reason to constrain amount of data slurped */
 #define YY_READ_BUF_SIZE 16777216
@@ -409,6 +411,38 @@ plpgsql_push_back_token(int token)
 	have_pushback_token = true;
 }
 
+/*
+ * Report a syntax error.
+ */
+void
+plpgsql_yyerror(const char *message)
+{
+	const char *loc = yytext;
+	int			cursorpos;
+
+	plpgsql_error_lineno = plpgsql_scanner_lineno();
+
+	/* in multibyte encodings, return index in characters not bytes */
+	cursorpos = pg_mbstrlen_with_len(scanbuf, loc - scanbuf) + 1;
+
+	if (*loc == YY_END_OF_BUFFER_CHAR)
+	{
+		ereport(ERROR,
+				(errcode(ERRCODE_SYNTAX_ERROR),
+				 /* translator: %s is typically "syntax error" */
+				 errmsg("%s at end of input", message),
+				 errposition(cursorpos)));
+	}
+	else
+	{
+		ereport(ERROR,
+				(errcode(ERRCODE_SYNTAX_ERROR),
+				 /* translator: first %s is typically "syntax error" */
+				 errmsg("%s at or near \"%s\"", message, loc),
+				 errposition(cursorpos)));
+	}
+}
+
 /*
  * Get the line number at which the current token ends.  This substitutes
  * for flex's very poorly implemented yylineno facility.
@@ -439,19 +473,6 @@ plpgsql_scanner_init(const char *str, int functype)
 {
 	Size	slen;
 
-	/*----------
-	 * Hack: skip any initial newline, so that in the common coding layout
-	 *		CREATE FUNCTION ... AS '
-	 *			code body
-	 *		' LANGUAGE 'plpgsql';
-	 * we will think "line 1" is what the programmer thinks of as line 1.
-	 *----------
-	 */
-    if (*str == '\r')
-        str++;
-    if (*str == '\n')
-        str++;
-
 	slen = strlen(str);
 
 	/*
@@ -478,6 +499,19 @@ plpgsql_scanner_init(const char *str, int functype)
 	cur_line_start = scanbuf;
 	cur_line_num = 1;
 
+	/*----------
+	 * Hack: skip any initial newline, so that in the common coding layout
+	 *		CREATE FUNCTION ... AS '
+	 *			code body
+	 *		' LANGUAGE 'plpgsql';
+	 * we will think "line 1" is what the programmer thinks of as line 1.
+	 *----------
+	 */
+    if (*cur_line_start == '\r')
+        cur_line_start++;
+    if (*cur_line_start == '\n')
+        cur_line_start++;
+
 	BEGIN(INITIAL);
 }