Make the UTF-8 decoding stricter and more verbose when

[perl5.git] / utf8.h

diff --git a/utf8.h b/utf8.h
index 6f86f72..bb494ab 100644 (file)
--- a/utf8.h
+++ b/utf8.h
@@ -1,12 +1,14 @@
 /*    utf8.h
  *
- *    Copyright (c) 1998, Larry Wall
+ *    Copyright (c) 1998-2000, Larry Wall
  *
  *    You may distribute under the terms of either the GNU General Public
  *    License or the Artistic License, as specified in the README file.
  *
  */
 
+START_EXTERN_C
+
 #ifdef DOINIT
 EXTCONST unsigned char PL_utf8skip[] = {
 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
@@ -16,12 +18,59 @@ EXTCONST unsigned char PL_utf8skip[] = {
 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus */
 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus */
 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* scripts */
-3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,6,6,7,8, /* cjk etc. */
+3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,6,6,    /* cjk etc. */
+7,13, /* Perl extended (not UTF-8).  Up to 72bit allowed (64-bit + reserved). */
 };
 #else
 EXTCONST unsigned char PL_utf8skip[];
 #endif
 
-#define IN_UTF8 (curcop->op_private & HINT_UTF8)
+END_EXTERN_C
+
+#define UTF8_MAXLEN 13 /* how wide can a single UTF8 encoded character become */
+
+#define IN_UTF8 (PL_curcop->op_private & HINT_UTF8)
+#define IN_BYTE (PL_curcop->op_private & HINT_BYTE)
+#define DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTE)
 
 #define UTF8SKIP(s) PL_utf8skip[*(U8*)s]
+
+#ifdef HAS_QUAD
+#define UTF8LEN(uv) ( (uv) < 0x80           ? 1 : \
+                     (uv) < 0x800          ? 2 : \
+                     (uv) < 0x10000        ? 3 : \
+                     (uv) < 0x200000       ? 4 : \
+                     (uv) < 0x4000000      ? 5 : \
+                     (uv) < 0x80000000     ? 6 : \
+                      (uv) < 0x1000000000LL ? 7 : 13 ) 
+#else
+/* No, I'm not even going to *TRY* putting #ifdef inside a #define */
+#define UTF8LEN(uv) ( (uv) < 0x80           ? 1 : \
+                     (uv) < 0x800          ? 2 : \
+                     (uv) < 0x10000        ? 3 : \
+                     (uv) < 0x200000       ? 4 : \
+                     (uv) < 0x4000000      ? 5 : \
+                     (uv) < 0x80000000     ? 6 : 7 )
+#endif
+
+#define UNICODE_REPLACEMENT_CHARACTER  0xfffd
+
+/*
+ * Note: we try to be careful never to call the isXXX_utf8() functions
+ * unless we're pretty sure we've seen the beginning of a UTF-8 character
+ * (that is, the two high bits are set).  Otherwise we risk loading in the
+ * heavy-duty SWASHINIT and SWASHGET routines unnecessarily.
+ */
+#ifdef EBCDIC
+#define isIDFIRST_lazy_if(p,c) isIDFIRST(*(p))
+#define isALNUM_lazy_if(p,c)   isALNUM(*(p))
+#else
+#define isIDFIRST_lazy_if(p,c) ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \
+                               ? isIDFIRST(*(p)) \
+                               : isIDFIRST_utf8((U8*)p))
+#define isALNUM_lazy_if(p,c)   ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \
+                               ? isALNUM(*(p)) \
+                               : isALNUM_utf8((U8*)p))
+#endif
+#define isIDFIRST_lazy(p)      isIDFIRST_lazy_if(p,1)
+#define isALNUM_lazy(p)                isALNUM_lazy_if(p,1)