Merge pull request #180 from Scille/improve-b64-encoding

Revert usage to `btoa` and `atob` for base64 encoding/decoding
Scille · Feb 3, 2025 · a003d0f · a003d0f
2 parents a402b25 + 8e85ef6
commit a003d0f
Show file tree

Hide file tree

Showing 2 changed files with 34 additions and 105 deletions.
diff --git a/.editorconfig b/.editorconfig
@@ -0,0 +1,12 @@
+# EditorConfig is awesome: https://EditorConfig.org
+
+# top-most EditorConfig file
+root = true
+
+[*]
+indent_style = space
+indent_size = 2
+end_of_line = lf
+charset = utf-8
+trim_trailing_whitespace = true
+insert_final_newline = true
diff --git a/lib/common/base64.ts b/lib/common/base64.ts
@@ -1,7 +1,5 @@
 // Parsec Cloud (https://parsec.cloud) Copyright (c) BUSL-1.1 2016-present Scille SAS
 
-const KEY_STR = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/=';
-
 function fromObject(obj: object): string {
   const jsonString = JSON.stringify(obj);
   return encode(jsonString);
@@ -12,67 +10,31 @@ function toObject(data: string): object {
   return JSON.parse(jsonString);
 }
 
-/*
- * Base64 encode / decode
- * Taken from http://www.webtoolkit.info/
- * and updated a bit
- */
-
 function encode(input: string): string {
-  let output = '';
-  let i = 0;
-
-  input = _utf8Encode(input);
-
-  while (i < input.length) {
-    const chr1 = input.charCodeAt(i++);
-    const chr2 = input.charCodeAt(i++);
-    const chr3 = input.charCodeAt(i++);
-
-    const enc1 = chr1 >> 2;
-    const enc2 = ((chr1 & 3) << 4) | (chr2 >> 4);
-    let enc3 = ((chr2 & 15) << 2) | (chr3 >> 6);
-    let enc4 = chr3 & 63;
-
-    if (isNaN(chr2)) {
-      enc3 = enc4 = 64;
-    } else if (isNaN(chr3)) {
-      enc4 = 64;
-    }
-
-    output = output + KEY_STR.charAt(enc1) + KEY_STR.charAt(enc2) + KEY_STR.charAt(enc3) + KEY_STR.charAt(enc4);
-  }
-
-  return output;
+  const encoder = new TextEncoder();
+  // Encode the string to utf-8, so it's safe to use with `btoa`.
+  // CF: https://developer.mozilla.org/en-US/docs/Web/API/Window/btoa#unicode_strings
+  const bytes: Uint8Array = encoder.encode(input);
+  // Convert the bytes array to a string since `btoa` only accepts strings.
+  const binString = Array.from(bytes, (c) => String.fromCharCode(c)).join('');
+  // Replace special char to be url safe.
+  return btoa(binString).replace(/\+/g, '-').replace(/\//g, '_');
 }
 
-function decode(input: string): string {
-  let output = '';
-  let i = 0;
-
-  input = input.replace(/[^A-Za-z0-9+/=]/g, '');
-
-  while (i < input.length) {
-    const enc1 = KEY_STR.indexOf(input.charAt(i++));
-    const enc2 = KEY_STR.indexOf(input.charAt(i++));
-    const enc3 = KEY_STR.indexOf(input.charAt(i++));
-    const enc4 = KEY_STR.indexOf(input.charAt(i++));
-
-    const chr1 = (enc1 << 2) | (enc2 >> 4);
-    const chr2 = ((enc2 & 15) << 4) | (enc3 >> 2);
-    const chr3 = ((enc3 & 3) << 6) | enc4;
-
-    output = output + String.fromCharCode(chr1);
-
-    if (enc3 !== 64) {
-      output = output + String.fromCharCode(chr2);
-    }
-    if (enc4 !== 64) {
-      output = output + String.fromCharCode(chr3);
-    }
-  }
-  output = _utf8Decode(output);
-  return output;
+function decode(b64url: string): string {
+  // Replace back the special chars used to make base64 string url safe.
+  const input = b64url.replace(/-/g, '+').replace(/_/g, '\\');
+  const binString = atob(input);
+  const bytes = Uint8Array.from(
+    binString,
+    (c) =>
+      // codePointAt may return undefined if the char is outside of valid UTF-16 range,
+      // But since we come from a base64 encoded string, we know that all chars are valid
+      c.codePointAt(0)!,
+  );
+  const decoder = new TextDecoder();
+  // Decode the utf-8 bytes to a string
+  return decoder.decode(bytes);
 }
 
 export const Base64 = {
@@ -81,48 +43,3 @@ export const Base64 = {
   encode,
   decode,
 };
-
-function _utf8Encode(s: string): string {
-  s = s.replace(/\r\n/g, '\n');
-  let utfText = '';
-
-  for (let n = 0; n < s.length; n++) {
-    const c = s.charCodeAt(n);
-    if (c < 128) {
-      utfText += String.fromCharCode(c);
-    } else if (c > 127 && c < 2048) {
-      utfText += String.fromCharCode((c >> 6) | 192);
-      utfText += String.fromCharCode((c & 63) | 128);
-    } else {
-      utfText += String.fromCharCode((c >> 12) | 224);
-      utfText += String.fromCharCode(((c >> 6) & 63) | 128);
-      utfText += String.fromCharCode((c & 63) | 128);
-    }
-  }
-
-  return utfText;
-}
-
-function _utf8Decode(utfText: string): string {
-  let s = '';
-  let i = 0;
-
-  while (i < utfText.length) {
-    const c = utfText.charCodeAt(i);
-
-    if (c < 128) {
-      s += String.fromCharCode(c);
-      i++;
-    } else if (c > 191 && c < 224) {
-      const c2 = utfText.charCodeAt(i + 1);
-      s += String.fromCharCode(((c & 31) << 6) | (c2 & 63));
-      i += 2;
-    } else {
-      const c2 = utfText.charCodeAt(i + 1);
-      const c3 = utfText.charCodeAt(i + 2);
-      s += String.fromCharCode(((c & 15) << 12) | ((c2 & 63) << 6) | (c3 & 63));
-      i += 3;
-    }
-  }
-  return s;
-}