Tighten overflow checks in tidin().

This code seems to have been written on the assumption that
"unsigned long" is 32 bits; or at any rate it ignored the
possibility of conversion overflow.  Rewrite, borrowing some
logic from oidin().

Discussion: https://postgr.es/m/3441768.1646343914@sss.pgh.pa.us

src/backend/utils/adt/tid.c

@@ -64,10 +64,10 @@ tidin(PG_FUNCTION_ARGS)
 	BlockNumber blockNumber;
 	OffsetNumber offsetNumber;
 	char	   *badp;
-	int			hold_offset;
+	unsigned long cvt;
 
 	for (i = 0, p = str; *p && i < NTIDARGS && *p != RDELIM; p++)
-		if (*p == DELIM || (*p == LDELIM && !i))
+		if (*p == DELIM || (*p == LDELIM && i == 0))
 			coord[i++] = p + 1;
 
 	if (i < NTIDARGS)
@@ -77,22 +77,36 @@ tidin(PG_FUNCTION_ARGS)
 						"tid", str)));
 
 	errno = 0;
-	blockNumber = strtoul(coord[0], &badp, 10);
+	cvt = strtoul(coord[0], &badp, 10);
 	if (errno || *badp != DELIM)
 		ereport(ERROR,
 				(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
 				 errmsg("invalid input syntax for type %s: \"%s\"",
 						"tid", str)));
+	blockNumber = (BlockNumber) cvt;
 
-	hold_offset = strtol(coord[1], &badp, 10);
-	if (errno || *badp != RDELIM ||
-		hold_offset > USHRT_MAX || hold_offset < 0)
+	/*
+	 * Cope with possibility that unsigned long is wider than BlockNumber, in
+	 * which case strtoul will not raise an error for some values that are out
+	 * of the range of BlockNumber.  (See similar code in oidin().)
+	 */
+#if SIZEOF_LONG > 4
+	if (cvt != (unsigned long) blockNumber &&
+		cvt != (unsigned long) ((int32) blockNumber))
 		ereport(ERROR,
 				(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
 				 errmsg("invalid input syntax for type %s: \"%s\"",
 						"tid", str)));
+#endif
 
-	offsetNumber = hold_offset;
+	cvt = strtoul(coord[1], &badp, 10);
+	if (errno || *badp != RDELIM ||
+		cvt > USHRT_MAX)
+		ereport(ERROR,
+				(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+				 errmsg("invalid input syntax for type %s: \"%s\"",
+						"tid", str)));
+	offsetNumber = (OffsetNumber) cvt;
 
 	result = (ItemPointer) palloc(sizeof(ItemPointerData));
 

src/test/regress/expected/tid.out

@@ -1,3 +1,22 @@
+-- basic tests for the TID data type
+SELECT
+  '(0,0)'::tid as tid00,
+  '(0,1)'::tid as tid01,
+  '(-1,0)'::tid as tidm10,
+  '(4294967295,65535)'::tid as tidmax;
+ tid00 | tid01 |     tidm10     |       tidmax       
+-------+-------+----------------+--------------------
+ (0,0) | (0,1) | (4294967295,0) | (4294967295,65535)
+(1 row)
+
+SELECT '(4294967296,1)'::tid;  -- error
+ERROR:  invalid input syntax for type tid: "(4294967296,1)"
+LINE 1: SELECT '(4294967296,1)'::tid;
+               ^
+SELECT '(1,65536)'::tid;  -- error
+ERROR:  invalid input syntax for type tid: "(1,65536)"
+LINE 1: SELECT '(1,65536)'::tid;
+               ^
 -- tests for functions related to TID handling
 CREATE TABLE tid_tab (a int);
 -- min() and max() for TIDs

src/test/regress/sql/tid.sql

@@ -1,3 +1,15 @@
+-- basic tests for the TID data type
+
+SELECT
+  '(0,0)'::tid as tid00,
+  '(0,1)'::tid as tid01,
+  '(-1,0)'::tid as tidm10,
+  '(4294967295,65535)'::tid as tidmax;
+
+SELECT '(4294967296,1)'::tid;  -- error
+SELECT '(1,65536)'::tid;  -- error
+
+
 -- tests for functions related to TID handling
 
 CREATE TABLE tid_tab (a int);