Fix for "[ID 20010213.005] utf8 + localized hash elems + 64 bits?"

The hash key got wrongly UTF8fied.

p4raw-id: //depot/perl@8835

diff --git a/pod/perlapi.pod b/pod/perlapi.pod
index 40d40fe..ef3a260 100644 (file)
--- a/pod/perlapi.pod
+++ b/pod/perlapi.pod
@@ -1035,9 +1035,10 @@ Found in file handy.h
 
 =item is_utf8_char
 
-Tests if some arbitrary number of bytes begins in a valid UTF-8 character.
-The actual number of bytes in the UTF-8 character will be returned if it
-is valid, otherwise 0. 
+Tests if some arbitrary number of bytes begins in a valid UTF-8
+character.  Note that an ASCII character is a valid UTF-8 character.
+The actual number of bytes in the UTF-8 character will be returned if
+it is valid, otherwise 0.
  
        STRLEN  is_utf8_char(U8 *p)
 
@@ -1046,8 +1047,10 @@ Found in file utf8.c
 
 =item is_utf8_string
 
-Returns true if first C<len> bytes of the given string form valid a UTF8
-string, false otherwise.
+Returns true if first C<len> bytes of the given string form a valid UTF8
+string, false otherwise.  Note that 'a valid UTF8 string' does not mean
+'a string that contains UTF8' because a valid ASCII string is a valid
+UTF8 string.
 
        bool    is_utf8_string(U8 *s, STRLEN len)
 

diff --git a/toke.c b/toke.c
index 5243fea..f8d7145 100644 (file)
--- a/toke.c
+++ b/toke.c
@@ -3185,9 +3185,6 @@ Perl_yylex(pTHX)
                if (*d == '}') {
                    char minus = (PL_tokenbuf[0] == '-');
                    s = force_word(s + minus, WORD, FALSE, TRUE, FALSE);
-                   if (UTF && !IN_BYTE && is_utf8_string((U8*)PL_tokenbuf, 0) &&
-                       PL_nextval[PL_nexttoke-1].opval)
-                     SvUTF8_on(((SVOP*)PL_nextval[PL_nexttoke-1].opval)->op_sv);
                    if (minus)
                        force_next('-');
                }

diff --git a/utf8.c b/utf8.c
index 918b669..13b953a 100644 (file)
--- a/utf8.c
+++ b/utf8.c
@@ -121,12 +121,12 @@ Perl_uv_to_utf8(pTHX_ U8 *d, UV uv)
 /*
 =for apidoc A|STRLEN|is_utf8_char|U8 *s
 
-Tests if some arbitrary number of bytes begins in a valid UTF-8 character.
-The actual number of bytes in the UTF-8 character will be returned if it
-is valid, otherwise 0. 
+Tests if some arbitrary number of bytes begins in a valid UTF-8
+character.  Note that an ASCII character is a valid UTF-8 character.
+The actual number of bytes in the UTF-8 character will be returned if
+it is valid, otherwise 0.
  
-=cut
-*/
+=cut */
 STRLEN
 Perl_is_utf8_char(pTHX_ U8 *s)
 {
@@ -168,8 +168,10 @@ Perl_is_utf8_char(pTHX_ U8 *s)
 /*
 =for apidoc A|bool|is_utf8_string|U8 *s|STRLEN len
 
-Returns true if first C<len> bytes of the given string form valid a UTF8
-string, false otherwise.
+Returns true if first C<len> bytes of the given string form a valid UTF8
+string, false otherwise.  Note that 'a valid UTF8 string' does not mean
+'a string that contains UTF8' because a valid ASCII string is a valid
+UTF8 string.
 
 =cut
 */