feat(encoding): support BOM detection (#5550)

* fix(encoding): export `getIgnoreBOM`

* feat(encoding): support ignoreBOM

* fix(encoding): not replace BOM to 0xFFFD

* chore: use strict equal

1 files changed, 24 insertions, 1 deletions

diff --git a/test/js/web/encoding/text-decoder.test.js b/test/js/web/encoding/text-decoder.test.js
index dabdb0936..de71fb351 100644
--- a/test/js/web/encoding/text-decoder.test.js
+++ b/test/js/web/encoding/text-decoder.test.js
@@ -250,7 +250,7 @@ describe("TextDecoder", () => {
   it("constructor should set values", () => {
     const decoder = new TextDecoder("utf-8", { fatal: true, ignoreBOM: false });
     expect(decoder.fatal).toBe(true);
-    // expect(decoder.ignoreBOM).toBe(false); // currently the getter for ignoreBOM doesn't work and always returns undefined
+    expect(decoder.ignoreBOM).toBe(false);
   });
 
   it("should throw on invalid input", () => {
@@ -265,6 +265,29 @@ describe("TextDecoder", () => {
   });
 });
 
+describe("TextDecoder ignoreBOM", () => {
+
+  it.each([
+    {
+      encoding: 'utf-8',
+      bytes: [0xEF, 0xBB, 0xBF, 0x61, 0x62, 0x63]
+    },
+    {
+      encoding: 'utf-16le',
+      bytes: [0xFF, 0xFE, 0x61, 0x00, 0x62, 0x00, 0x63, 0x00]
+    }
+  ])('should ignoreBOM for: %o', ({encoding, bytes}) => {
+    const BOM = '\uFEFF';
+    const array = new Uint8Array(bytes);
+
+    const decoder_ignore_bom = new TextDecoder(encoding, {ignoreBOM: true});
+    expect(decoder_ignore_bom.decode(array)).toStrictEqual(`${BOM}abc`);
+
+    const decoder_not_ignore_bom = new TextDecoder(encoding, {ignoreBOM: false});
+    expect(decoder_not_ignore_bom.decode(array)).toStrictEqual('abc');
+  });
+});
+
 it("truncated sequences", () => {
   const assert_equals = (a, b) => expect(a).toBe(b);