[Openmcl-devel] %ioblock-read-u16-encoded-char

[Takehiko Abe]

Thanks for the explanation. It is very informative.

Gary Byers wrote:

> - I'm tempted to prefer the short-term-fix of changing the limits
>    for the UTF-16 encodings to #xd800 over the short-term fix of
>    removing the test in %IOBLOCK-READ-U16-ENCODED-CHAR, since I
>    think that the test would probably want to be put back once
>    the issues of separate encode/decode limits are unscrambled.

I agree.

;;;

I found a typo in stream-decode-function of utf-8 character
encoding.

The change is

    (< s2 #xf0)  --> (< 1st-unit #xf0)

"(< s2 #xf0)" appears only once in the file


My mailer probably wraps lines of the diff output.


Index: level-1/l1-unicode.lisp
===================================================================
RCS file: /usr/local/tmpcvs/openmcl-dev/ccl/level-1/l1-unicode.lisp,v
retrieving revision 1.18
diff -u -r1.18 l1-unicode.lisp
--- level-1/l1-unicode.lisp	8 Apr 2007 17:38:51 -0000	1.18
+++ level-1/l1-unicode.lisp	25 Apr 2007 13:05:10 -0000
@@ -2699,7 +2699,7 @@
                        s2
                        (locally
                            (declare (type (unsigned-byte 8) s2))
-                         (if (< s2 #xf0)
+                         (if (< 1st-unit #xf0)
                            (if (and (< (the fixnum (logxor s1 #x80)) #x40)
                                     (< (the fixnum (logxor s2 #x80)) #x40)
                                     (or (>= 1st-unit #xe1)