From: SADAHIRO Tomoyuki <BQW10602@nifty.com>
Date: Sat, 8 Oct 2005 09:59:00 +0000 (+0000)
Subject: undef IS_UTF8_CHAR() on EBCDIC
X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=commitdiff_plain;h=77263263f2dcc1d77fff24405b11002b13929822;p=p5sagit%2Fp5-mst-13.2.git

undef IS_UTF8_CHAR() on EBCDIC
Message-Id: <20051008165752.348A.BQW10602@nifty.com>

p4raw-id: //depot/perl@25716
---

diff --git a/lib/utf8.t b/lib/utf8.t
index 70ef1e3..81ebc22 100644
--- a/lib/utf8.t
+++ b/lib/utf8.t
@@ -37,7 +37,7 @@ no utf8; # Ironic, no?
 #
 #
 
-plan tests => 146;
+plan tests => 150;
 
 {
     # bug id 20001009.001
@@ -431,3 +431,11 @@ SKIP: {
 		     qr/Undefined subroutine utf8::moo/, {stderr=>1},
 		    "Check Carp is loaded for AUTOLOADing errors")
 }
+
+{
+    # failure of is_utf8_char() without NATIVE_TO_UTF on EBCDIC (0260..027F)
+    ok(utf8::valid(chr(0x250)), "0x250");
+    ok(utf8::valid(chr(0x260)), "0x260");
+    ok(utf8::valid(chr(0x270)), "0x270");
+    ok(utf8::valid(chr(0x280)), "0x280");
+}
diff --git a/utf8.c b/utf8.c
index 66642c5..8ea2d7a 100644
--- a/utf8.c
+++ b/utf8.c
@@ -209,6 +209,9 @@ S_is_utf8_char_slow(pTHX_ const U8 *s, const STRLEN len)
 
     slen = len - 1;
     s++;
+#ifdef EBCDIC
+    u = NATIVE_TO_UTF(u);
+#endif
     u &= UTF_START_MASK(len);
     uv  = u;
     ouv = uv;
diff --git a/utf8.h b/utf8.h
index c8bcb36..4599407 100644
--- a/utf8.h
+++ b/utf8.h
@@ -258,6 +258,9 @@ encoded character.
 #endif
 #define SHARP_S_SKIP 2
 
+#ifdef EBCDIC
+/* IS_UTF8_CHAR() is not ported to EBCDIC */
+#else
 #define IS_UTF8_CHAR_1(p)	\
 	((p)[0] <= 0x7F)
 #define IS_UTF8_CHAR_2(p)	\
@@ -329,3 +332,4 @@ encoded character.
 
 #define IS_UTF8_CHAR_FAST(n) ((n) <= 4)
 
+#endif /* IS_UTF8_CHAR() for UTF-8 */