Issue #617: Sanitize received content

Sanitizes received HTML content with DOMPurify, making it safe
for displaying and templating. Sanitizes received plain text content
by encoding every single character as HTML entity.

4 files changed, 115 insertions, 33 deletions

diff --git a/web-ui/app/js/helpers/sanitizer.js b/web-ui/app/js/helpers/sanitizer.js
new file mode 100644
index 00000000..eea1f0f7
--- /dev/null
+++ b/web-ui/app/js/helpers/sanitizer.js
@@ -0,0 +1,108 @@
+/*
+ * Copyright (c) 2016 ThoughtWorks, Inc.
+ *
+ * Pixelated is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * Pixelated is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with Pixelated. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+define(['DOMPurify', 'he'], function (DOMPurify, he) {
+  'use strict';
+
+  /**
+   * Sanitizes a mail body to safe-to-display HTML
+   */
+  var sanitizer = {};
+
+  /**
+   * Adds html line breaks to a plaintext with line breaks (incl carriage return)
+   *
+   * @param {string} textPlainBody Plaintext input
+   * @returns {string} Plaintext with HTML line breals (<br/>)
+   */
+  sanitizer.addLineBreaks = function (textPlainBody) {
+    return textPlainBody.replace(/(\r)?\n/g, '<br/>').replace(/(&#xD;)?&#xA;/g, '<br/>');
+  };
+
+  /**
+   * Runs a given dirty body through DOMPurify, thereby removing
+   * potentially hazardous XSS attacks. Please be advised that this
+   * will not act as a privacy leak prevention. Contained contents
+   * will still point to remote sources.
+   *
+   * For future reference: Running DOMPurify with these parameters
+   * can help mitigate some of the most widely used privacy leaks.
+   * FORBID_TAGS: ['style', 'svg', 'audio', 'video', 'math'],
+   * FORBID_ATTR: ['src']
+   *
+   * @param  {string} dirtyBody The unsanitized string
+   * @return {string} Safe-to-display HTML string
+   */
+  sanitizer.purifyHtml = function (dirtyBody) {
+    return DOMPurify.sanitize(dirtyBody, {
+      SAFE_FOR_JQUERY: true,
+      SAFE_FOR_TEMPLATES: true
+    });
+  };
+
+  /**
+  * Runs a given dirty body through he, thereby encoding everything
+  * as HTML entities.
+  *
+  * @param  {string} dirtyBody The unsanitized string
+  * @return {string} Safe-to-display HTML string
+  */
+  sanitizer.purifyText = function (dirtyBody) {
+    return he.encode(dirtyBody, {
+      encodeEverything: true
+    });
+  };
+
+  /**
+   * Calls #purify and #addLineBreaks to turn untrusted mail body content
+   * into safe-to-display HTML.
+   *
+   * NB: HTML content is preferred to plaintext content.
+   *
+   * @param  {object} mail Pixelated Mail Object
+   * @return {string} Safe-to-display HTML string
+   */
+  sanitizer.sanitize = function (mail) {
+    var body;
+
+    if (mail.htmlBody) {
+      body = this.purifyHtml(mail.htmlBody);
+    } else {
+      body = this.purifyText(mail.textPlainBody);
+      body = this.addLineBreaks(body);
+    }
+
+    return body;
+  };
+
+  /**
+   * Add hooks to DOMPurify for opening links in new windows
+   */
+  DOMPurify.addHook('afterSanitizeAttributes', function (node) {
+    // set all elements owning target to target=_blank
+    if ('target' in node) {
+      node.setAttribute('target', '_blank');
+    }
+
+    // set non-HTML/MathML links to xlink:show=new
+    if (!node.hasAttribute('target') && (node.hasAttribute('xlink:href') || node.hasAttribute('href'))) {
+      node.setAttribute('xlink:show', 'new');
+    }
+  });
+
+  return sanitizer;
+});
diff --git a/web-ui/app/js/helpers/view_helper.js b/web-ui/app/js/helpers/view_helper.js
index e4e9277d..e8d517a5 100644
--- a/web-ui/app/js/helpers/view_helper.js
+++ b/web-ui/app/js/helpers/view_helper.js
@@ -17,12 +17,12 @@
 define(
   [
     'helpers/contenttype',
-    'lib/html_whitelister',
     'views/i18n',
     'quoted-printable/quoted-printable',
-    'utf8/utf8'
+    'utf8/utf8',
+    'helpers/sanitizer'
   ],
-  function(contentType, htmlWhitelister, i18n, quotedPrintable, utf8) {
+  function(contentType, i18n, quotedPrintable, utf8, sanitizer) {
   'use strict';
 
   function formatStatusClasses(ss) {
@@ -31,37 +31,8 @@ define(
     }).join(' ');
   }
 
-  function addParagraphsToPlainText(textPlainBody) {
-    return textPlainBody.replace(/^(.*?)$/mg, '$1<br/>');
-  }
-
-  function escapeHtmlTags(body) {
-
-    var escapeIndex = {
-      '&': '&amp;',
-      '<': '&lt;',
-      '>': '&gt;',
-      '"': '&quot;',
-      '\'':'&#39;',
-      '/': '&#x2f;'
-    };
-
-    return body.replace(/["'<>\/&]/g, function(char){
-        return escapeIndex[char];
-   });
-
-  }
-
-  function escapeHtmlAndAddParagraphs(body) {
-    var escapedBody = escapeHtmlTags(body);
-    return addParagraphsToPlainText(escapedBody);
-  }
-
   function formatMailBody(mail) {
-    var body = mail.htmlBody ?
-                htmlWhitelister.sanitize(mail.htmlBody, htmlWhitelister.tagPolicy) :
-                escapeHtmlAndAddParagraphs(mail.textPlainBody);
-    return $('<div>' + body + '</div>');
+    return sanitizer.sanitize(mail);
   }
 
   function moveCaretToEnd(el) {
diff --git a/web-ui/app/js/mail_view/ui/mail_view.js b/web-ui/app/js/mail_view/ui/mail_view.js
index d4f5dd9e..8465b45a 100644
--- a/web-ui/app/js/mail_view/ui/mail_view.js
+++ b/web-ui/app/js/mail_view/ui/mail_view.js
@@ -72,6 +72,7 @@ define(
         }));
 
         this.$node.find('.bodyArea').html(viewHelpers.formatMailBody(data.mail));
+
         this.trigger(document, events.search.highlightResults, {where: '.bodyArea'});
         this.trigger(document, events.search.highlightResults, {where: '.subjectArea'});
         this.trigger(document, events.search.highlightResults, {where: '.msg-header .recipients'});
diff --git a/web-ui/app/js/main.js b/web-ui/app/js/main.js
index 5fb2e46f..e093e790 100644
--- a/web-ui/app/js/main.js
+++ b/web-ui/app/js/main.js
@@ -22,6 +22,8 @@ requirejs.config({
     'page': 'js/page',
     'feedback': 'js/feedback',
     'flight': 'bower_components/flight',
+    'DOMPurify': 'bower_components/DOMPurify/dist/purify.min',
+    'he': 'bower_components/he/he',
     'hbs': 'js/generated/hbs',
     'helpers': 'js/helpers',
     'lib': 'js/lib',