7 files changed, 22 insertions, 11 deletions
diff --git a/pcre_compile.c b/pcre_compile.c
index 2b076cd..7da62e4 100644
--- a/pcre_compile.c
+++ b/pcre_compile.c
@@ -438,7 +438,7 @@ static const char error_texts[] =
   /* 30 */
   "unknown POSIX class name\0"
   "POSIX collating elements are not supported\0"
-  "this version of PCRE is not compiled with PCRE_UTF8 support\0"
+  "this version of PCRE is compiled without UTF support\0"
   "spare error\0"  /** DEAD **/
   "character value in \\x{...} sequence is too large\0"
   /* 35 */
@@ -461,7 +461,7 @@ static const char error_texts[] =
   "too many named subpatterns (maximum " XSTRING(MAX_NAME_COUNT) ")\0"
   /* 50 */
   "repeated subpattern is too long\0"    /** DEAD **/
-  "octal value is greater than \\377 (not in UTF-8 mode)\0"
+  "octal value is greater than \\377 in 8-bit non-UTF-8 mode\0"
   "internal error: overran compiling workspace\0"
   "internal error: previously-checked referenced subpattern not found\0"
   "DEFINE group contains more than one branch\0"
@@ -480,14 +480,15 @@ static const char error_texts[] =
   /* 65 */
   "different names for subpatterns of the same number are not allowed\0"
   "(*MARK) must have an argument\0"
-  "this version of PCRE is not compiled with PCRE_UCP support\0"
+  "this version of PCRE is not compiled with Unicode property support\0"
   "\\c must be followed by an ASCII character\0"
   "\\k is not followed by a braced, angle-bracketed, or quoted name\0"
   /* 70 */
   "internal error: unknown opcode in find_fixedlength()\0"
   "\\N is not supported in a class\0"
   "too many forward references\0"
-  "disallowed UTF-8/16 code point (>= 0xd800 && <= 0xdfff)\0"
+  "disallowed Unicode code point (>= 0xd800 && <= 0xdfff)\0"
+  "invalid UTF-16 string\0" 
   ;
 
 /* Table to identify digits and hex digits. This is used when compiling
@@ -7706,7 +7707,11 @@ not used here. */
 if (utf && (options & PCRE_NO_UTF8_CHECK) == 0 &&
      (errorcode = PRIV(valid_utf)((PCRE_PUCHAR)pattern, -1, erroroffset)) != 0)
   {
+#ifdef COMPILE_PCRE8   
   errorcode = ERR44;
+#else   
+  errorcode = ERR74;
+#endif   
   goto PCRE_EARLY_ERROR_RETURN2;
   }
 #else
diff --git a/pcre_internal.h b/pcre_internal.h
index 837c50c..d949575 100644
--- a/pcre_internal.h
+++ b/pcre_internal.h
@@ -1939,7 +1939,7 @@ enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
        ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
        ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
        ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
-       ERR70, ERR71, ERR72, ERR73, ERRCOUNT };
+       ERR70, ERR71, ERR72, ERR73, ERR74, ERRCOUNT };
 
 /* The real format of the start of the pcre block; the index of names and the
 code vector run on as long as necessary after the end. We store an explicit
diff --git a/pcreposix.c b/pcreposix.c
index 6929fc0..06cdd01 100644
--- a/pcreposix.c
+++ b/pcreposix.c
@@ -158,6 +158,7 @@ static const int eint[] = {
   REG_BADPAT,  /* \N is not supported in a class */
   REG_BADPAT,  /* too many forward references */
   REG_BADPAT,  /* disallowed UTF-8/16 code point (>= 0xd800 && <= 0xdfff) */
+  REG_BADPAT   /* invalid UTF-16 string (should not occur) */ 
 };
 
 /* Table of texts corresponding to POSIX error codes */
diff --git a/testdata/testinput18 b/testdata/testinput18
index f075d8a..f63ed89 100644
--- a/testdata/testinput18
+++ b/testdata/testinput18
@@ -244,4 +244,6 @@ correctly, but that messes up comparisons). --/
 
 <!testsaved16BE-2
 
+/砑€/8
+
 /-- End of testinput18 --/
diff --git a/testdata/testoutput18 b/testdata/testoutput18
index f1013eb..f3bcfbd 100644
--- a/testdata/testoutput18
+++ b/testdata/testoutput18
@@ -899,4 +899,7 @@ No need char
 Subject length lower bound = 2
 No set of starting bytes
 
+/砑€/8
+Failed: invalid UTF-16 string at offset 0
+
 /-- End of testinput18 --/
diff --git a/testdata/testoutput2 b/testdata/testoutput2
index 9f5134d..3bbf5b3 100644
--- a/testdata/testoutput2
+++ b/testdata/testoutput2
@@ -6366,7 +6366,7 @@ No need char
 10: Y
 
 /\777/I
-Failed: octal value is greater than \377 (not in UTF-8 mode) at offset 3
+Failed: octal value is greater than \377 in 8-bit non-UTF-8 mode at offset 3
 
 /\s*,\s*/IS
 Capturing subpattern count = 0
diff --git a/testdata/testoutput5 b/testdata/testoutput5
index ab56060..b13b789 100644
--- a/testdata/testoutput5
+++ b/testdata/testoutput5
@@ -12,10 +12,10 @@ Failed: character value in \x{...} sequence is too large at offset 11
 Failed: character value in \x{...} sequence is too large at offset 12
 
 /\x{d800}/8
-Failed: disallowed UTF-8/16 code point (>= 0xd800 && <= 0xdfff) at offset 7
+Failed: disallowed Unicode code point (>= 0xd800 && <= 0xdfff) at offset 7
 
 /\x{dfff}/8
-Failed: disallowed UTF-8/16 code point (>= 0xd800 && <= 0xdfff) at offset 7
+Failed: disallowed Unicode code point (>= 0xd800 && <= 0xdfff) at offset 7
 
 /\x{d7ff}/8
 
@@ -483,7 +483,7 @@ No first char
 No need char
 
 /\777/I
-Failed: octal value is greater than \377 (not in UTF-8 mode) at offset 3
+Failed: octal value is greater than \377 in 8-bit non-UTF-8 mode at offset 3
 
 /\x{100}*\d/8DZ
 ------------------------------------------------------------------
@@ -1561,10 +1561,10 @@ Partial match: for
 /\x{d7ff}\x{e000}/8
 
 /\x{d800}/8
-Failed: disallowed UTF-8/16 code point (>= 0xd800 && <= 0xdfff) at offset 7
+Failed: disallowed Unicode code point (>= 0xd800 && <= 0xdfff) at offset 7
 
 /\x{dfff}/8 
-Failed: disallowed UTF-8/16 code point (>= 0xd800 && <= 0xdfff) at offset 7
+Failed: disallowed Unicode code point (>= 0xd800 && <= 0xdfff) at offset 7
 
 /\h+/8
     \x{1681}\x{200b}\x{1680}\x{2000}\x{202f}\x{3000}