string_decoder: reset decoder on end

jridgewell · MylesBorins · commit f96b0bf49497 · 2018-03-29T23:28:19.000-04:00
This resets the StringDecoder's state after calling `#end`. Further writes to the decoder will act as if it were a brand new instance, allowing simple reuse. PR-URL: #18494 Fixes: #16564 Refs: #16594 Reviewed-By: Tiancheng "Timothy" Gu <timothygu99@gmail.com> Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com> Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de> Reviewed-By: Anna Henningsen <anna@addaleax.net>
diff --git a/lib/string_decoder.js b/lib/string_decoder.js
@@ -209,8 +209,11 @@ function utf8Text(buf, i) {
 // character.
 function utf8End(buf) {
   const r = (buf && buf.length ? this.write(buf) : '');
-  if (this.lastNeed)
+  if (this.lastNeed) {
+    this.lastNeed = 0;
+    this.lastTotal = 0;
     return r + '\ufffd';
+  }
   return r;
 }
 
@@ -245,6 +248,8 @@ function utf16End(buf) {
   const r = (buf && buf.length ? this.write(buf) : '');
   if (this.lastNeed) {
     const end = this.lastTotal - this.lastNeed;
+    this.lastNeed = 0;
+    this.lastTotal = 0;
     return r + this.lastChar.toString('utf16le', 0, end);
   }
   return r;
@@ -268,8 +273,12 @@ function base64Text(buf, i) {
 
 function base64End(buf) {
   const r = (buf && buf.length ? this.write(buf) : '');
-  if (this.lastNeed)
-    return r + this.lastChar.toString('base64', 0, 3 - this.lastNeed);
+  if (this.lastNeed) {
+    const end = 3 - this.lastNeed;
+    this.lastNeed = 0;
+    this.lastTotal = 0;
+    return r + this.lastChar.toString('base64', 0, end);
+  }
   return r;
 }
 
diff --git a/test/parallel/test-string-decoder-end.js b/test/parallel/test-string-decoder-end.js
@@ -39,6 +39,46 @@ for (let i = 1; i <= 16; i++) {
 
 encodings.forEach(testEncoding);
 
+testEnd('utf8', Buffer.of(0xE2), Buffer.of(0x61), '\uFFFDa');
+testEnd('utf8', Buffer.of(0xE2), Buffer.of(0x82), '\uFFFD\uFFFD');
+testEnd('utf8', Buffer.of(0xE2), Buffer.of(0xE2), '\uFFFD\uFFFD');
+testEnd('utf8', Buffer.of(0xE2, 0x82), Buffer.of(0x61), '\uFFFDa');
+testEnd('utf8', Buffer.of(0xE2, 0x82), Buffer.of(0xAC), '\uFFFD\uFFFD');
+testEnd('utf8', Buffer.of(0xE2, 0x82), Buffer.of(0xE2), '\uFFFD\uFFFD');
+testEnd('utf8', Buffer.of(0xE2, 0x82, 0xAC), Buffer.of(0x61), '€a');
+
+testEnd('utf16le', Buffer.of(0x3D), Buffer.of(0x61, 0x00), 'a');
+testEnd('utf16le', Buffer.of(0x3D), Buffer.of(0xD8, 0x4D, 0xDC), '\u4DD8');
+testEnd('utf16le', Buffer.of(0x3D, 0xD8), Buffer.of(), '\uD83D');
+testEnd('utf16le', Buffer.of(0x3D, 0xD8), Buffer.of(0x61, 0x00), '\uD83Da');
+testEnd(
+  'utf16le',
+  Buffer.of(0x3D, 0xD8),
+  Buffer.of(0x4D, 0xDC),
+  '\uD83D\uDC4D'
+);
+testEnd('utf16le', Buffer.of(0x3D, 0xD8, 0x4D), Buffer.of(), '\uD83D');
+testEnd(
+  'utf16le',
+  Buffer.of(0x3D, 0xD8, 0x4D),
+  Buffer.of(0x61, 0x00),
+  '\uD83Da'
+);
+testEnd('utf16le', Buffer.of(0x3D, 0xD8, 0x4D), Buffer.of(0xDC), '\uD83D');
+testEnd(
+  'utf16le',
+  Buffer.of(0x3D, 0xD8, 0x4D, 0xDC),
+  Buffer.of(0x61, 0x00),
+  '👍a'
+);
+
+testEnd('base64', Buffer.of(0x61), Buffer.of(), 'YQ==');
+testEnd('base64', Buffer.of(0x61), Buffer.of(0x61), 'YQ==YQ==');
+testEnd('base64', Buffer.of(0x61, 0x61), Buffer.of(), 'YWE=');
+testEnd('base64', Buffer.of(0x61, 0x61), Buffer.of(0x61), 'YWE=YQ==');
+testEnd('base64', Buffer.of(0x61, 0x61, 0x61), Buffer.of(), 'YWFh');
+testEnd('base64', Buffer.of(0x61, 0x61, 0x61), Buffer.of(0x61), 'YWFhYQ==');
+
 function testEncoding(encoding) {
   bufs.forEach((buf) => {
     testBuf(encoding, buf);
@@ -66,3 +106,14 @@ function testBuf(encoding, buf) {
   assert.strictEqual(res1, res3, 'one byte at a time should match toString');
   assert.strictEqual(res2, res3, 'all bytes at once should match toString');
 }
+
+function testEnd(encoding, incomplete, next, expected) {
+  let res = '';
+  const s = new SD(encoding);
+  res += s.write(incomplete);
+  res += s.end();
+  res += s.write(next);
+  res += s.end();
+
+  assert.strictEqual(res, expected);
+}
