If you want you can now add -DUSE_UTF8_SCRIPTS to your cflags

and the Perl will be built to do that by default (adding that
will break scripts having non-UTF-8 binary data, such as Latin-1.)

p4raw-id: //depot/perl@11656

diff --git a/op.c b/op.c
index f52ea29..677fe7a 100644 (file)
--- a/op.c
+++ b/op.c
@@ -115,7 +115,7 @@ Perl_pad_allocmy(pTHX_ char *name)
 
     if (!(PL_in_my == KEY_our ||
          isALPHA(name[1]) ||
-         (PL_hints & HINT_UTF8 && UTF8_IS_START(name[1])) ||
+         (USE_UTF8_IN_NAMES && UTF8_IS_START(name[1])) ||
          (name[1] == '_' && (int)strlen(name) > 2)))
     {
        if (!isPRINT(name[1]) || strchr("\t\n\r\f", name[1])) {

diff --git a/toke.c b/toke.c
index d535442..52706a3 100644 (file)
--- a/toke.c
+++ b/toke.c
@@ -36,11 +36,14 @@ static I32 utf16rev_textfilter(pTHXo_ int idx, SV *sv, int maxlen);
 #define XFAKEBRACK 128
 #define XENUMMASK 127
 
-#ifdef EBCDIC
-/* For now 'use utf8' does not affect tokenizer on EBCDIC */
-#define UTF (PL_linestr && DO_UTF8(PL_linestr))
+#ifdef USE_UTF8_SCRIPTS
+#   define UTF (!IN_BYTES)
 #else
-#define UTF ((PL_linestr && DO_UTF8(PL_linestr)) || (PL_hints & HINT_UTF8))
+#   ifdef EBCDIC /* For now 'use utf8' does not affect tokenizer on EBCDIC */
+#       define UTF (PL_linestr && DO_UTF8(PL_linestr))
+#   else
+#       define UTF ((PL_linestr && DO_UTF8(PL_linestr)) || (PL_hints & HINT_UTF8))
+#   endif
 #endif
 
 /* In variables named $^X, these are the legal values for X.

diff --git a/utf8.h b/utf8.h
index b793db6..c475d0f 100644 (file)
--- a/utf8.h
+++ b/utf8.h
@@ -7,6 +7,15 @@
  *
  */
 
+/* Use UTF-8 as the default script encoding?
+ * Turning this on will break scripts having non-UTF8 binary
+ * data (such as Latin-1) in string literals. */
+#ifdef USE_UTF8_SCRIPTS
+#    define USE_UTF8_IN_NAMES (!IN_BYTES)
+#else
+#    define USE_UTF8_IN_NAMES (PL_hints & HINT_UTF8)
+#endif
+
 #ifdef EBCDIC
 /* The equivalent of these macros but implementing UTF-EBCDIC
    are in the following header file: