test.com shebang handling (was Re: VMS@14369)

[p5sagit/p5-mst-13.2.git] / utf8.h

diff --git a/utf8.h b/utf8.h
index b793db6..c1c69d2 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:
@@ -126,7 +135,12 @@ END_EXTERN_C
 #define isIDFIRST_lazy(p)      isIDFIRST_lazy_if(p,1)
 #define isALNUM_lazy(p)                isALNUM_lazy_if(p,1)
 
-#define UTF8_MAXLEN 13 /* how wide can a single UTF8 encoded character become */
+/* how wide can a single UTF8 encoded character become */
+#define UTF8_MAXLEN 13
+/* how wide a character can become when upper/lowercased */
+#define UTF8_MAXLEN_UCLC (UTF8_MAXLEN*2)
+/* how wide a character can become when casefolded */
+#define UTF8_MAXLEN_FOLD (UTF8_MAXLEN*3)
 
 #define IN_BYTES (PL_curcop->op_private & HINT_BYTES)
 #define DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTES)
@@ -144,7 +158,7 @@ END_EXTERN_C
                                         UTF8_ALLOW_SURROGATE|UTF8_ALLOW_BOM|\
                                         UTF8_ALLOW_FFFF|UTF8_ALLOW_LONG)
 #define UTF8_ALLOW_ANY                 0x00ff
-#define UTF8_CHECK_ONLY                        0x0100
+#define UTF8_CHECK_ONLY                        0x0200
 
 #define UNICODE_SURROGATE_FIRST                0xd800
 #define UNICODE_SURROGATE_LAST         0xdfff
@@ -152,6 +166,17 @@ END_EXTERN_C
 #define UNICODE_BYTER_ORDER_MARK       0xfffe
 #define UNICODE_ILLEGAL                        0xffff
 
+/* Though our UTF-8 encoding can go beyond this,
+ * let's be conservative. */
+#define PERL_UNICODE_MAX       0x10FFFF
+
+#define UNICODE_ALLOW_SURROGATE 0x0001 /* Allow UTF-16 surrogates (EVIL) */
+#define UNICODE_ALLOW_FDD0     0x0002  /* Allow the U+FDD0...U+FDEF */
+#define UNICODE_ALLOW_FFFE     0x0004  /* Allow 0xFFFE, 0x1FFFE, ... */
+#define UNICODE_ALLOW_FFFF     0x0008  /* Allow 0xFFFE, 0x1FFFE, ... */
+#define UNICODE_ALLOW_SUPER    0x0010  /* Allow past 10xFFFF */
+#define UNICODE_ALLOW_ANY      0xFFFF
+
 #define UNICODE_IS_SURROGATE(c)                ((c) >= UNICODE_SURROGATE_FIRST && \
                                         (c) <= UNICODE_SURROGATE_LAST)
 #define UNICODE_IS_REPLACEMENT(c)      ((c) == UNICODE_REPLACEMENT)
@@ -163,3 +188,21 @@ END_EXTERN_C
 #endif
 
 #define UTF8_IS_ASCII(c) UTF8_IS_INVARIANT(c)
+
+#define UNICODE_LATIN_SMALL_LETTER_SHARP_S     0x00DF
+#define UNICODE_GREEK_CAPITAL_LETTER_SIGMA     0x03A3
+#define UNICODE_GREEK_SMALL_LETTER_FINAL_SIGMA 0x03C2
+#define UNICODE_GREEK_SMALL_LETTER_SIGMA       0x03C3
+
+#define UNI_DISPLAY_ISPRINT    0x0001
+#define UNI_DISPLAY_BACKSLASH  0x0002
+#define UNI_DISPLAY_QQ         (UNI_DISPLAY_ISPRINT|UNI_DISPLAY_BACKSLASH)
+#define UNI_DISPLAY_REGEX      (UNI_DISPLAY_ISPRINT|UNI_DISPLAY_BACKSLASH)
+
+#define ANYOF_UNICODE_FOLD_SHARP_S(node, input, end)   \
+       (ANYOF_BITMAP_TEST(node, UNICODE_LATIN_SMALL_LETTER_SHARP_S) && \
+        (ANYOF_FLAGS(node) & ANYOF_UNICODE) && \
+        (ANYOF_FLAGS(node) & ANYOF_FOLD) && \
+        ((end) > (input) + 1) && \
+        toLOWER((input)[0]) == 's' && \
+        toLOWER((input)[1]) == 's')