More EBCDIC fixes.

[perl5.git] / utfebcdic.h

diff --git a/utfebcdic.h b/utfebcdic.h
index efcd00d..ef67cb2 100644 (file)
--- a/utfebcdic.h
+++ b/utfebcdic.h
@@ -15,17 +15,18 @@ START_EXTERN_C
 #ifdef DOINIT
 /* Indexed by encoded byte this table gives the length of the sequence.
    Adapted from the shadow flags table in tr16.
-   The entries marked 9 are continuation bytes.
+   The entries marked 9 in tr6 are continuation bytes and are marked
+   as length 1 here so that we can recover.
 */
 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,
 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,
-1,9,9,9,9,9,9,9,9,9,9,1,1,1,1,1,
-1,9,9,9,9,9,9,9,9,9,1,1,1,1,1,1,
-1,1,9,9,9,9,9,9,9,9,9,1,1,1,1,1,
-9,9,9,9,2,2,2,2,2,1,1,1,1,1,1,1,
+1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+1,1,1,1,2,2,2,2,2,1,1,1,1,1,1,1,
 2,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,
 2,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,
 2,1,1,1,1,1,1,1,1,1,2,2,2,1,2,2,
@@ -221,7 +222,7 @@ END_EXTERN_C
 #define UTF_TO_NATIVE(ch)        PL_utf2e[(U8)(ch)]
 /* Transform in wide UV char space */
 #define NATIVE_TO_UNI(ch)        (((ch) > 255) ? (ch) : NATIVE_TO_ASCII(ch))
-#define UNI_TO_NATIVE(ch)        (((ch) > 255) ? (ch) : (UV) ASCII_TO_NATIVE(ch))
+#define UNI_TO_NATIVE(ch)        (((ch) > 255) ? (ch) : ASCII_TO_NATIVE(ch))
 /* Transform in invariant..byte space */
 #define NATIVE_TO_NEED(enc,ch)   ((enc) ? UTF_TO_NATIVE(NATIVE_TO_ASCII(ch)) : (ch))
 #define ASCII_TO_NEED(enc,ch)    ((enc) ? UTF_TO_NATIVE(ch) : ASCII_TO_NATIVE(ch))
@@ -233,10 +234,10 @@ END_EXTERN_C
  * unnecessarily.
  */
 
-#define isIDFIRST_lazy_if(p,c) ((IN_BYTE || (!c || UTF8_INVARIANT(*p)) \
+#define isIDFIRST_lazy_if(p,c) ((IN_BYTE || (!c || UTF8_IS_INVARIANT(*p))) \
                                ? isIDFIRST(*(p)) \
                                : isIDFIRST_utf8((U8*)p))
-#define isALNUM_lazy_if(p,c)   ((IN_BYTE || (!c || UTF8_INVARIANT(*p)) \
+#define isALNUM_lazy_if(p,c)   ((IN_BYTE || (!c || UTF8_IS_INVARIANT(*p))) \
                                ? isALNUM(*(p)) \
                                : isALNUM_utf8((U8*)p))
 
@@ -267,8 +268,11 @@ END_EXTERN_C
                      (uv) < 0x400000       ? 5 : \
                      (uv) < 0x4000000      ? 6 : 7 )
 
+
+#define UNI_IS_INVARIANT(c)            ((c) <  0xA0)
 /* UTF-EBCDIC sematic macros - transform back into UTF-8-Mod and then compare */
-#define UTF8_IS_INVARIANT(c)           (NATIVE_TO_UTF(c) <  0xA0)
+#define NATIVE_IS_INVARIANT(c)         UNI_IS_INVARIANT(NATIVE_TO_ASCII(c))
+#define UTF8_IS_INVARIANT(c)           UNI_IS_INVARIANT(NATIVE_TO_UTF(c))
 #define UTF8_IS_START(c)               (NATIVE_TO_UTF(c) >= 0xA0 && (NATIVE_TO_UTF(c) & 0xE0) != 0xA0)
 #define UTF8_IS_CONTINUATION(c)                (NATIVE_TO_UTF(c) >= 0xA0 && (NATIVE_TO_UTF(c) & 0xE0) == 0xA0)
 #define UTF8_IS_CONTINUED(c)           (NATIVE_TO_UTF(c) >= 0xA0)
@@ -280,7 +284,7 @@ END_EXTERN_C
 #define UTF_CONTINUATION_MASK          ((U8)0x1f)
 #define UTF_ACCUMULATION_SHIFT         5
 
-#define UTF8_ACCUMULATE(old, new)      (((old) << UTF_ACCUMULATION_SHIFT)|(NATIVE_TO_UTF(new) & UTFE_CONTINUATION_MASK))
+#define UTF8_ACCUMULATE(old, new)      (((old) << UTF_ACCUMULATION_SHIFT)|(NATIVE_TO_UTF(new) & UTF_CONTINUATION_MASK))
 
 /* UTF-EBCDIC encode a downgradeable value */
 #define UTF8_EIGHT_BIT_HI(c)   UTF_TO_NATIVE((((U8)(c))>>UTF_ACCUMULATION_SHIFT)|UTF_START_MARK(2))