--- gforth/utf-8.fs	2005/01/13 21:26:46	1.16
+++ gforth/utf-8.fs	2006/02/19 22:35:40	1.21
@@ -1,6 +1,6 @@
 \ UTF-8 handling                                       12dec04py
 
-\ Copyright (C) 2004 Free Software Foundation, Inc.
+\ Copyright (C) 2004,2005 Free Software Foundation, Inc.
 
 \ This file is part of Gforth.
 
@@ -27,11 +27,12 @@ $80 Value max-single-byte
 : u8len ( u8 -- n )
     dup      max-single-byte u< IF  drop 1  EXIT  THEN \ special case ASCII
     $800  2 >r
-    BEGIN  2dup u>=  WHILE  5 lshift r> 1+ >r  REPEAT
+    BEGIN  2dup u>=  WHILE  5 lshift r> 1+ >r  dup 0= UNTIL  THEN
     2drop r> ;
 
 : u8@+ ( u8addr -- u8addr' u )
     count  dup max-single-byte u< ?EXIT  \ special case ASCII
+    dup $C2 u< IF  UTF-8-err throw  THEN  \ malformed character
     $7F and  $40 >r
     BEGIN  dup r@ and  WHILE  r@ xor
 	    6 lshift r> 5 lshift >r >r count
@@ -72,6 +73,7 @@ $80 Value max-single-byte
 
 : u8key ( -- u )
     defers key dup max-single-byte u< ?EXIT  \ special case ASCII
+    dup $C2 u< IF  UTF-8-err throw  THEN  \ malformed character
     $7F and  $40 >r
     BEGIN  dup r@ and  WHILE  r@ xor
 	    6 lshift r> 5 lshift >r >r defers key
@@ -126,6 +128,13 @@ $80 Value max-single-byte
 	nip nip over -
     then ;
 
+[IFDEF] wcwidth
+: u8width ( xcaddr u -- n )
+    0 rot rot over + swap ?DO
+	I xc@+ swap >r wcwidth +
+    r> I - +LOOP ;
+[THEN]
+
 : set-encoding-utf-8 ( -- )
     ['] u8emit is xemit
     ['] u8key is xkey
@@ -137,6 +146,9 @@ $80 Value max-single-byte
     ['] u8!+? is xc!+?
     ['] u8@+ is xc@+
     ['] u8len is xc-size
+[ [IFDEF] x-width ]
+    ['] u8width is x-width
+[ [THEN] ]
     ['] -u8trailing-garbage is -trailing-garbage
 ;