[wasm] Add UTF-8 validation

src/unicode.cc

 // Copyright 2012 the V8 project authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 //
 // This file was generated at 2014-10-08 15:25:47.940335
 
 #include "src/unicode.h"
 #include "src/unicode-inl.h"
 #include <stdio.h>
 #include <stdlib.h>
@@ skipping 287 matching lines @@
   }
   if (!IsContinuationCharacter(str[3])) {
     *cursor += 1;
     return kBadChar;
   }
   *cursor += 4;
   return ((str[0] << 18) + (str[1] << 12) + (str[2] << 6) + str[3]) -
          0x03C82080;
 }
 
+bool Utf8::Validate(const byte* bytes, size_t length) {
+  size_t cursor = 0;
+
+  // Performance optimization: Skip over single-byte values first.
+  while (cursor < length && bytes[cursor] <= kMaxOneByteChar) {
+    ++cursor;
+  }
+
+  while (cursor < length) {
+    uchar c = ValueOf(bytes + cursor, length - cursor, &cursor);
+    if (!IsValidCharacter(c)) return false;
+  }
+  return true;
+}
 
 // Uppercase:            point.category == 'Lu'
 
 static const uint16_t kUppercaseTable0Size = 455;
 static const int32_t kUppercaseTable0[455] = {
     1073741889, 90,         1073742016, 214,
     1073742040, 222,        256,        258,  // NOLINT
     260,        262,        264,        266,
     268,        270,        272,        274,  // NOLINT
     276,        278,        280,        282,
@@ skipping 3115 matching lines @@
              sizeof(MultiCharacterSpecialCase<1>)  // NOLINT
          +
          kCanonicalizationRangeMultiStrings1Size *
              sizeof(MultiCharacterSpecialCase<1>)  // NOLINT
          +
          kCanonicalizationRangeMultiStrings7Size *
              sizeof(MultiCharacterSpecialCase<1>);  // NOLINT
 }
 
 }  // namespace unibrow