utf8.c: Rmv an EBCDIC dependency

This is now generated by regcharclass.pl

3 files changed, 5 insertions, 10 deletions

diff --git a/regcharclass.h b/regcharclass.h
index 4d2b3eaf3c..fa72120616 100644
--- a/regcharclass.h
+++ b/regcharclass.h
@@ -3765,6 +3765,6 @@
  * d0fe91c50f5e6ca796f9a7dbe273c95e44ddd50d8e840364e925baf255f9bb00 lib/unicore/mktables
  * 50b85a67451145545a65cea370dab8d3444fbfe07e9c34cef560c5b7da9d3eef lib/unicore/version
  * 0a6b5ab33bb1026531f816efe81aea1a8ffcd34a27cbea37dd6a70a63d73c844 regen/charset_translations.pl
- * 3fb6bafb4c830dd501868e34f550cdad3bf8d2c9eed44756488f36c484969417 regen/regcharclass.pl
+ * 1aa94679c695efd507b7e4491629dba1021b74c21a5324dfd3a582a5d654bd32 regen/regcharclass.pl
  * b2f896452d2b30da3e04800f478c60c1fd0b03d6b668689b020f1e3cf1f1cdd9 regen/regcharclass_multi_char_folds.pl
  * ex: set ro: */
diff --git a/regen/regcharclass.pl b/regen/regcharclass.pl
index 890dd23208..32460b72b9 100755
--- a/regen/regcharclass.pl
+++ b/regen/regcharclass.pl
@@ -1805,6 +1805,8 @@ LARGER_NON_CHARS:   # 5 bytes
 0xFFFFE - 0xFFFFF
 0x10FFFE - 0x10FFFF
 
+# Note that code in utf8.c is counting on the 'fast' version to look at no
+# more than two bytes
 SURROGATE: Surrogate code points
 => UTF8 :safe fast
 \p{_Perl_Surrogate}
diff --git a/utf8.c b/utf8.c
index 12568591a4..6df64810d5 100644
--- a/utf8.c
+++ b/utf8.c
@@ -863,13 +863,9 @@ Perl_is_utf8_char_helper(const U8 * const s, const U8 * e, const U32 flags)
 #  define FIRST_START_BYTE_THAT_IS_DEFINITELY_SUPER  0xFA
 #  define IS_UTF8_2_BYTE_SUPER(s0, s1)           ((s0) == 0xF9 && (s1) >= 0xA2)
 
-#  define IS_UTF8_2_BYTE_SURROGATE(s0, s1)       ((s0) == 0xF1              \
-                                                       /* B6 and B7 */      \
-                                              && ((s1) & 0xFE ) == 0xB6)
 #else
 #  define FIRST_START_BYTE_THAT_IS_DEFINITELY_SUPER  0xF5
 #  define IS_UTF8_2_BYTE_SUPER(s0, s1)           ((s0) == 0xF4 && (s1) >= 0x90)
-#  define IS_UTF8_2_BYTE_SURROGATE(s0, s1)       ((s0) == 0xED && (s1) >= 0xA0)
 #endif
 
         if (  (flags & UTF8_DISALLOW_SUPER)
@@ -894,7 +890,7 @@ Perl_is_utf8_char_helper(const U8 * const s, const U8 * e, const U32 flags)
             }
 
             if (   (flags & UTF8_DISALLOW_SURROGATE)
-                &&  UNLIKELY(IS_UTF8_2_BYTE_SURROGATE(s0, s1)))
+                &&  UNLIKELY(is_SURROGATE_utf8(s)))
             {
                 return 0;       /* Surrogate */
             }
@@ -1748,10 +1744,7 @@ Perl__utf8n_to_uvchr_msgs_helper(const U8 *s,
                 {
                     possible_problems |= UTF8_GOT_SUPER;
                 }
-                else if (UNLIKELY(IS_UTF8_2_BYTE_SURROGATE(
-                                      NATIVE_UTF8_TO_I8(*adjusted_s0),
-                                      NATIVE_UTF8_TO_I8(*(adjusted_s0 + 1)))))
-                {
+                else if (UNLIKELY(is_SURROGATE_utf8(adjusted_s0))) {
                     possible_problems |= UTF8_GOT_SURROGATE;
                 }
             }