buffer: fix needle length misestimation for UCS2

addaleax · evanlucas · commit a550ddbf3c68 · 2016-05-17T08:03:28.000-05:00
Use `StringBytes::Size` to determine the needle string length instead of assuming latin-1 or UTF-8. Previously, `Buffer.indexOf` could fail with an assertion failure when the needle's byte length, but not its character count, exceeded the haystack's byte length. PR-URL: #6511 Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Trevor Norris <trev.norris@gmail.com>
diff --git a/src/node_buffer.cc b/src/node_buffer.cc
@@ -995,9 +995,9 @@ void IndexOfString(const FunctionCallbackInfo<Value>& args) {
 
   const char* haystack = ts_obj_data;
   const size_t haystack_length = ts_obj_length;
-  // Extended latin-1 characters are 2 bytes in Utf8.
+
   const size_t needle_length =
-      enc == BINARY ? needle->Length() : needle->Utf8Length();
+      StringBytes::Size(args.GetIsolate(), needle, enc);
 
   if (needle_length == 0 || haystack_length == 0) {
     return args.GetReturnValue().Set(-1);
diff --git a/test/parallel/test-buffer-indexof.js b/test/parallel/test-buffer-indexof.js
@@ -222,6 +222,12 @@ var allCharsBufferUcs2 = Buffer.from(allCharsString, 'ucs2');
 assert.equal(-1, allCharsBufferUtf8.indexOf('notfound'));
 assert.equal(-1, allCharsBufferUcs2.indexOf('notfound'));
 
+// Needle is longer than haystack, but only because it's encoded as UTF-16
+assert.strictEqual(Buffer.from('aaaa').indexOf('a'.repeat(4), 'ucs2'), -1);
+
+assert.strictEqual(Buffer.from('aaaa').indexOf('a'.repeat(4), 'utf8'), 0);
+assert.strictEqual(Buffer.from('aaaa').indexOf('&#20320;&#22909;', 'ucs2'), -1);
+
 {
   // Find substrings in Utf8.
   const lengths = [1, 3, 15];  // Single char, simple and complex.
