From 4710f67bafb362ddd3896a90fd1812bdcf2c0e63 Mon Sep 17 00:00:00 2001
From: Sidharth Vinod <sidharthv96@gmail.com>
Date: Sun, 21 Aug 2022 11:26:45 +0530
Subject: [PATCH] fix: sanitization bug

---
 src/diagrams/common/common.spec.js | 58 ++++--------------------------
 src/diagrams/common/common.ts      | 45 +----------------------
 2 files changed, 7 insertions(+), 96 deletions(-)

diff --git a/src/diagrams/common/common.spec.js b/src/diagrams/common/common.spec.js
index ac7309829..66ac7b2ea 100644
--- a/src/diagrams/common/common.spec.js
+++ b/src/diagrams/common/common.spec.js
@@ -1,4 +1,4 @@
-import { sanitizeText, removeScript, removeEscapes } from './common';
+import { sanitizeText, removeScript } from './common';
 
 describe('when securityLevel is antiscript, all script must be removed', function () {
   /**
@@ -6,7 +6,7 @@ describe('when securityLevel is antiscript, all script must be removed', functio
    * @param {string} result The expected sanitized text
    */
   function compareRemoveScript(original, result) {
-    expect(removeScript(original)).toEqual(result);
+    expect(removeScript(original).trim()).toEqual(result);
   }
 
   it('should remove all script block, script inline.', function () {
@@ -29,70 +29,24 @@ describe('when securityLevel is antiscript, all script must be removed', functio
     compareRemoveScript(
       `This is a <a href="javascript:runHijackingScript();">clean link</a> + <a href="javascript:runHijackingScript();">clean link</a>
   and <a href="javascript&colon;bipassedMining();">me too</a>`,
-      `This is a <a href="#runHijackingScript();">clean link</a> + <a href="#runHijackingScript();">clean link</a>
-  and <a href="#;bipassedMining();">me too</a>`
+      `This is a <a>clean link</a> + <a>clean link</a>
+  and <a>me too</a>`
     );
   });
 
   it('should detect malicious images', function () {
-    compareRemoveScript(`<img onerror="alert('hello');">`, `<img onerror:"alert('hello');">`);
+    compareRemoveScript(`<img onerror="alert('hello');">`, `<img>`);
   });
 
   it('should detect iframes', function () {
     compareRemoveScript(
       `<iframe src="http://abc.com/script1.js"></iframe>
     <iframe src="http://example.com/iframeexample"></iframe>`,
-      ` src="http://abc.com/script1.js"></iframe>
-     src="http://example.com/iframeexample"></iframe>`
+      ''
     );
   });
 });
 
-describe('remove escape code in text', function () {
-  it('should remove a unicode colon', function () {
-    const labelString = '\\u003A';
-
-    const result = removeEscapes(labelString);
-    expect(result).toEqual(':');
-  });
-  it('should remove a hex colon', function () {
-    const labelString = '\\x3A';
-
-    const result = removeEscapes(labelString);
-    expect(result).toEqual(':');
-  });
-  it('should remove a oct colon', function () {
-    const labelString = '\\72';
-
-    const result = removeEscapes(labelString);
-    expect(result).toEqual(':');
-  });
-  it('should remove a oct colon 3 numbers', function () {
-    const labelString = '\\072';
-
-    const result = removeEscapes(labelString);
-    expect(result).toEqual(':');
-  });
-  it('should remove multiple colons 3 numbers', function () {
-    const labelString = '\\072\\072\\72';
-
-    const result = removeEscapes(labelString);
-    expect(result).toEqual(':::');
-  });
-  it('should handle greater and smaller then', function () {
-    const labelString = '\\74\\076';
-
-    const result = removeEscapes(labelString);
-    expect(result).toEqual('<>');
-  });
-  it('should handle letters', function () {
-    const labelString = '\\u0073\\143ri\\x70\\u0074\\x3A';
-
-    const result = removeEscapes(labelString);
-    expect(result).toEqual('script:');
-  });
-});
-
 describe('Sanitize text', function () {
   it('should remove script tag', function () {
     const maliciousStr = 'javajavascript:script:alert(1)';
diff --git a/src/diagrams/common/common.ts b/src/diagrams/common/common.ts
index 2d1d06bbc..a1b5f027f 100644
--- a/src/diagrams/common/common.ts
+++ b/src/diagrams/common/common.ts
@@ -13,22 +13,6 @@ export const getRows = (s?: string): string[] => {
   return str.split('#br#');
 };
 
-export const removeEscapes = (text: string): string => {
-  let newStr = text.replace(/\\u[\dA-F]{4}/gi, function (match) {
-    return String.fromCharCode(parseInt(match.replace(/\\u/g, ''), 16));
-  });
-
-  newStr = newStr.replace(/\\x([0-9a-f]{2})/gi, (_, c) => String.fromCharCode(parseInt(c, 16)));
-  newStr = newStr.replace(/\\[\d\d\d]{3}/gi, function (match) {
-    return String.fromCharCode(parseInt(match.replace(/\\/g, ''), 8));
-  });
-  newStr = newStr.replace(/\\[\d\d\d]{2}/gi, function (match) {
-    return String.fromCharCode(parseInt(match.replace(/\\/g, ''), 8));
-  });
-
-  return newStr;
-};
-
 /**
  * Removes script tags from a text
  *
@@ -36,33 +20,7 @@ export const removeEscapes = (text: string): string => {
  * @returns {string} The safer text
  */
 export const removeScript = (txt: string): string => {
-  var rs = '';
-  var idx = 0;
-
-  while (idx >= 0) {
-    idx = txt.indexOf('<script');
-    if (idx >= 0) {
-      rs += txt.substr(0, idx);
-      txt = txt.substr(idx + 1);
-
-      idx = txt.indexOf('</script>');
-      if (idx >= 0) {
-        idx += 9;
-        txt = txt.substr(idx);
-      }
-    } else {
-      rs += txt;
-      idx = -1;
-      break;
-    }
-  }
-  let decodedText = removeEscapes(rs);
-  decodedText = decodedText.replaceAll(/script>/gi, '#');
-  decodedText = decodedText.replaceAll(/javascript:/gi, '#');
-  decodedText = decodedText.replaceAll(/javascript&colon/gi, '#');
-  decodedText = decodedText.replaceAll(/onerror=/gi, 'onerror:');
-  decodedText = decodedText.replaceAll(/<iframe/gi, '');
-  return decodedText;
+  return DOMPurify.sanitize(txt);
 };
 
 const sanitizeMore = (text: string, config: MermaidConfig) => {
@@ -185,5 +143,4 @@ export default {
   removeScript,
   getUrl,
   evaluate,
-  removeEscapes,
 };