--- gforth/utf-8.fs 2005/01/12 21:21:53 1.15 +++ gforth/utf-8.fs 2005/12/31 15:46:10 1.18 @@ -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 @@ -141,7 +143,11 @@ $80 Value max-single-byte ; : utf-8-cold ( -- ) - s" LANG" getenv s" .UTF-8" search nip nip + s" LC_ALL" getenv 2dup d0= IF 2drop + s" LC_CTYPE" getenv 2dup d0= IF 2drop + s" LANG" getenv 2dup d0= IF 2drop + s" C" THEN THEN THEN + s" UTF-8" search nip nip IF set-encoding-utf-8 ELSE set-encoding-fixed-width THEN ; ' utf-8-cold INIT8 chained