Make defn of UTF_IS_CONTINUATION_MASK common

This variable can be defined from the same base in both UTF-8 and
UTF-EBCDIC, and doing so eliminates an EBCDIC dependency.

diff --git a/utf8.h b/utf8.h
index 472527c..356c3e4 100644 (file)
--- a/utf8.h
+++ b/utf8.h
@@ -311,9 +311,6 @@ C<cp> is Unicode if above 255; otherwise is platform-native.
 #define UTF8_IS_START(c)      (__ASSERT_(FITS_IN_8_BITS(c))                 \
                                ((U8)((c) | 0)) >= 0xc2)
 
 #define UTF8_IS_START(c)      (__ASSERT_(FITS_IN_8_BITS(c))                 \
                                ((U8)((c) | 0)) >= 0xc2)
 

-/* For use in UTF8_IS_CONTINUATION() below */
-#define UTF_IS_CONTINUATION_MASK    0xC0
-

 /* Is the byte 'c' part of a multi-byte UTF8-8 encoded sequence, and not the
  * first byte thereof?  The |0 makes sure this isn't mistakenly called with a
  * ptr argument */
 /* Is the byte 'c' part of a multi-byte UTF8-8 encoded sequence, and not the
  * first byte thereof?  The |0 makes sure this isn't mistakenly called with a
  * ptr argument */


@@ -360,6 +357,10 @@ C<cp> is Unicode if above 255; otherwise is platform-native.
  * UTF-8, 0x1F in UTF-EBCDIC. */
 #define UTF_CONTINUATION_MASK  ((U8) ((1U << UTF_ACCUMULATION_SHIFT) - 1))
 
  * UTF-8, 0x1F in UTF-EBCDIC. */
 #define UTF_CONTINUATION_MASK  ((U8) ((1U << UTF_ACCUMULATION_SHIFT) - 1))
 

+/* For use in UTF8_IS_CONTINUATION().  This turns out to be 0xC0 in UTF-8,
+ * E0 in UTF-EBCDIC */
+#define UTF_IS_CONTINUATION_MASK    ((U8) (0xFF << UTF_ACCUMULATION_SHIFT))
+

 /* Internal macro to be used only in this file to aid in constructing other
  * publicly accessible macros.
  * The number of bytes required to express this uv in UTF-8, for just those
 /* Internal macro to be used only in this file to aid in constructing other
  * publicly accessible macros.
  * The number of bytes required to express this uv in UTF-8, for just those

diff --git a/utfebcdic.h b/utfebcdic.h
index d8278a1..f13f555 100644 (file)
--- a/utfebcdic.h
+++ b/utfebcdic.h
@@ -220,8 +220,6 @@ explicitly forbidden, and the shortest possible encoding should always be used
 
 #define UTF8_IS_START(c)               _generic_isCC(c, _CC_UTF8_IS_START)
 
 
 #define UTF8_IS_START(c)               _generic_isCC(c, _CC_UTF8_IS_START)
 

-#define UTF_IS_CONTINUATION_MASK    0xE0
-

 #define UTF8_IS_CONTINUATION(c)                _generic_isCC(c, _CC_UTF8_IS_CONTINUATION)
 
 /* The above instead could be written as this:
 #define UTF8_IS_CONTINUATION(c)                _generic_isCC(c, _CC_UTF8_IS_CONTINUATION)
 
 /* The above instead could be written as this: