test: add tests for invalid UTF-8

Verify that `Blob.prototype.text()`, `streamConsumers.text()` and
`TextDecoder.prototype.decode()` work as expected with invalid UTF-8.

Fixes: #39804

test/parallel/test-blob.js

@@ -87,6 +87,15 @@ assert.throws(() => new Blob({}), {
   }));
 }
 
+{
+  const b = new Blob(['hello', new Uint8Array([0xed, 0xa0, 0x88])]);
+  assert.strictEqual(b.size, 8);
+  b.text().then(common.mustCall((text) => {
+    assert.strictEqual(text, 'hello\ufffd\ufffd\ufffd');
+    assert.strictEqual(text.length, 8);
+  }));
+}
+
 {
   const b = new Blob(
     [

test/parallel/test-stream-consumers.js

@@ -13,6 +13,7 @@ const {
 } = require('stream/consumers');
 
 const {
+  Readable,
   PassThrough
 } = require('stream');
 
@@ -73,6 +74,19 @@ const kArrayBuffer =
   setTimeout(() => passthrough.end('there'), 10);
 }
 
+{
+  const readable = new Readable({
+    read() {}
+  });
+
+  text(readable).then((data) => {
+    assert.deepStrictEqual(data, 'foo\ufffd\ufffd\ufffd');
+  });
+
+  readable.push(new Uint8Array([0x66, 0x6f, 0x6f, 0xed, 0xa0, 0x80]));
+  readable.push(null);
+}
+
 {
   const passthrough = new PassThrough();
 

test/parallel/test-whatwg-encoding-custom-textdecoder.js

@@ -191,3 +191,11 @@ if (common.hasIntl) {
     }
   );
 }
+
+// Test TextDecoder for incomplete UTF-8 byte sequence.
+{
+  const decoder = new TextDecoder();
+  const chunk = new Uint8Array([0x66, 0x6f, 0x6f, 0xed]);
+  const str = decoder.decode(chunk);
+  assert.strictEqual(str, 'foo\ufffd');
+}