1 files changed, 22 insertions, 37 deletions

diff --git a/src/utf16_le.c b/src/utf16_le.c
index 4b231c6..c6edd94 100644
--- a/src/utf16_le.c
+++ b/src/utf16_le.c
@@ -2,7 +2,7 @@
   utf16_le.c -  Oniguruma (regular expression library)
 **********************************************************************/
 /*-
- * Copyright (c) 2002-2018  K.Kosako  <sndgk393 AT ybb DOT ne DOT jp>
+ * Copyright (c) 2002-2019  K.Kosako
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -95,7 +95,15 @@ static const int EncLen_UTF16[] = {
 static int
 utf16le_code_to_mbclen(OnigCodePoint code)
 {
-  return (code > 0xffff ? 4 : 2);
+  if (code > 0xffff) {
+    if (code > 0x10ffff)
+      return ONIGERR_INVALID_CODE_POINT_VALUE;
+    else
+      return 4;
+  }
+  else {
+    return 2;
+  }
 }
 
 static int
@@ -110,7 +118,16 @@ is_valid_mbc_string(const UChar* p, const UChar* end)
   const UChar* end1 = end - 1;
 
   while (p < end1) {
-    p += utf16le_mbc_enc_len(p);
+    int len = utf16le_mbc_enc_len(p);
+    if (len == 4) {
+      if (p + 3 < end && ! UTF16_IS_SURROGATE_SECOND(*(p + 3)))
+        return FALSE;
+    }
+    else
+      if (UTF16_IS_SURROGATE_SECOND(*(p + 1)))
+        return FALSE;
+
+    p += len;
   }
 
   if (p != end)
@@ -210,39 +227,6 @@ utf16le_mbc_case_fold(OnigCaseFoldType flag,
                                          fold);
 }
 
-#if 0
-static int
-utf16le_is_mbc_ambiguous(OnigCaseFoldType flag, const UChar** pp,
-                         const UChar* end)
-{
-  const UChar* p = *pp;
-
-  (*pp) += EncLen_UTF16[*(p+1)];
-
-  if (*(p+1) == 0) {
-    int c, v;
-
-    if (*p == 0xdf && (flag & INTERNAL_ONIGENC_CASE_FOLD_MULTI_CHAR) != 0) {
-      return TRUE;
-    }
-
-    c = *p;
-    v = ONIGENC_IS_UNICODE_ISO_8859_1_BIT_CTYPE(c,
-                       (BIT_CTYPE_UPPER | BIT_CTYPE_LOWER));
-    if ((v | BIT_CTYPE_LOWER) != 0) {
-      /* 0xaa, 0xb5, 0xba are lower case letter, but can't convert. */
-      if (c >= 0xaa && c <= 0xba)
-        return FALSE;
-      else
-        return TRUE;
-    }
-    return (v != 0 ? TRUE : FALSE);
-  }
-
-  return FALSE;
-}
-#endif
-
 static UChar*
 utf16le_left_adjust_char_head(const UChar* start, const UChar* s)
 {
@@ -252,7 +236,8 @@ utf16le_left_adjust_char_head(const UChar* start, const UChar* s)
     s--;
   }
 
-  if (UTF16_IS_SURROGATE_SECOND(*(s+1)) && s > start + 1)
+  if (UTF16_IS_SURROGATE_SECOND(*(s+1)) && s > start + 1 &&
+      UTF16_IS_SURROGATE_FIRST(*(s-1)))
     s -= 2;
 
   return (UChar* )s;