string_decoder: fix bad utf8 character handling

This commit fixes an issue when extra utf8 continuation bytes appear
at the end of a chunk of data, causing miscalculations to be made
when checking how many bytes are needed to decode a complete
character.

Fixes: #7308

lib/string_decoder.js

@@ -122,24 +122,39 @@ function utf8CheckIncomplete(self, buf, i) {
     return 0;
   var nb = utf8CheckByte(buf[j--]);
   if (nb >= 0) {
-    if (nb > 0)
+    if (nb > 0) {
       self.lastNeed = nb + 1 - (buf.length - j);
+      if (self.lastNeed < 0) {
+        self.lastNeed = 0;
+        nb = 0;
+      }
+    }
     return nb;
   }
   if (j < i)
     return 0;
   nb = utf8CheckByte(buf[j--]);
   if (nb >= 0) {
-    if (nb > 0)
+    if (nb > 0) {
       self.lastNeed = nb + 1 - (buf.length - j);
+      if (self.lastNeed < 0) {
+        self.lastNeed = 0;
+        nb = 0;
+      }
+    }
     return nb;
   }
   if (j < i)
     return 0;
   nb = utf8CheckByte(buf[j--]);
   if (nb >= 0) {
-    if (nb > 0)
+    if (nb > 0) {
       self.lastNeed = nb + 1 - (buf.length - j);
+      if (self.lastNeed < 0) {
+        self.lastNeed = 0;
+        nb = 0;
+      }
+    }
     return nb;
   }
   return 0;

test/parallel/test-string-decoder.js

@@ -55,9 +55,14 @@ assert.strictEqual(decoder.write(Buffer.from('\ufffd\ufffd\ufffd')),
 assert.strictEqual(decoder.end(), '');
 
 decoder = new StringDecoder('utf8');
-assert.strictEqual(decoder.write(Buffer.from('efbfbde2', 'hex')), '\ufffd');
+assert.strictEqual(decoder.write(Buffer.from('EFBFBDE2', 'hex')), '\ufffd');
 assert.strictEqual(decoder.end(), '\ufffd');
 
+decoder = new StringDecoder('utf8');
+assert.strictEqual(decoder.write(Buffer.from('C9B5A9', 'hex')), 'ɵ\ufffd');
+assert.strictEqual(decoder.write(Buffer.from('41', 'hex')), 'A');
+assert.strictEqual(decoder.end(), '');
+
 
 // Additional UTF-16LE surrogate pair tests
 decoder = new StringDecoder('utf16le');