Add utf8 as another no-op encoding

Author: pimterry

README.md

@@ -12,7 +12,7 @@ The supported codecs are:
 * Brotli
 * Zstandard
 
-All encoding names are case-insensitive (although lowercase is generally standard). The 'identity', 'amz-1.0', 'none', 'text', 'binary' and 'utf-8' encodings (no-op encodings) are all supported as no-op encodings, passed through with no en/decoding at all. Only 'identity' is standard, but the others are all in common use regardless.
+All encoding names are case-insensitive (although lowercase is generally standard). The 'identity', 'amz-1.0', 'none', 'text', 'binary', 'utf8' and 'utf-8' encodings are all supported as no-op encodings, passed through with no en/decoding at all. Only 'identity' is standard, but the others are all in common use regardless.
 
 Found a codec used in real-world HTTP that isn't supported? Open an issue!
 

src/index.ts

@@ -9,8 +9,7 @@ export type SUPPORTED_ENCODING =
     | 'deflate'
     | 'x-deflate'
     | 'br'
-    | 'zstd'
-    | 'amz-1.0';
+    | 'zstd';
 
 export const gzip = promisify(zlib.gzip);
 export const gunzip = promisify(zlib.gunzip);
@@ -89,10 +88,11 @@ const IDENTITY_ENCODINGS = [
     // Workaround for Apache's mod_deflate handling of 'identity', used in the wild mostly with PHP.
     // https://github.com/curl/curl/pull/2298
     'none',
-    // No idea where these comes from, but they definitely exist in real traffic and seem to come
+    // No idea where these come from, but they definitely exist in real traffic and seem to come
     // from common confusion between content encodings and content types:
     'text',
     'binary',
+    'utf8',
     'utf-8'
 ]