Fix: Implement DOMPurify to sanitize HTML content before rendering (#1498)

### What problem does this PR solve?

This PR resolves issue #1491 related to HTML Injection and Cross-Site
Scripting (XSS). The issue was caused by the unsafe usage of
`dangerouslySetInnerHTML` without proper sanitization of user input.

### Changes
- Added DOMPurify dependency.
- Updated the following components to use DOMPurify:
-
`web/src/pages/add-knowledge/components/knowledge-chunk/components/chunk-card/index.tsx`
  - `web/src/pages/chat/markdown-content/index.tsx`
-
`web/src/pages/add-knowledge/components/knowledge-setting/category-panel.tsx`

### Type of change

- [x] Other (please describe): Security Fix

web/package-lock.json

@@ -17,6 +17,7 @@
         "classnames": "^2.5.1",
         "dagre": "^0.8.5",
         "dayjs": "^1.11.10",
+        "dompurify": "^3.1.6",
         "elkjs": "^0.9.3",
         "eventsource-parser": "^1.1.2",
         "human-id": "^4.1.1",
@@ -52,6 +53,7 @@
         "@testing-library/jest-dom": "^6.4.5",
         "@testing-library/react": "^15.0.7",
         "@types/dagre": "^0.7.52",
+        "@types/dompurify": "^3.0.5",
         "@types/jest": "^29.5.12",
         "@types/lodash": "^4.14.202",
         "@types/react": "^18.0.33",
@@ -4856,6 +4858,16 @@
         "@types/ms": "*"
       }
     },
+    "node_modules/@types/dompurify": {
+      "version": "3.0.5",
+      "resolved": "https://registry.npmmirror.com/@types/dompurify/-/dompurify-3.0.5.tgz",
+      "integrity": "sha512-1Wg0g3BtQF7sSb27fJQAKck1HECM6zV1EB66j8JH9i3LCjYabJa0FSdiSgsD5K/RbrsR0SiraKacLB+T8ZVYAg==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "@types/trusted-types": "*"
+      }
+    },
     "node_modules/@types/eslint": {
       "version": "8.56.1",
       "resolved": "https://registry.npmmirror.com/@types/eslint/-/eslint-8.56.1.tgz",
@@ -5206,6 +5218,13 @@
       "integrity": "sha512-/Ad8+nIOV7Rl++6f1BdKxFSMgmoqEoYbHRpPcx3JEfv8VRsQe9Z4mCXeJBzxs7mbHY/XOZZuXlRNfhpVPbs6ZA==",
       "dev": true
     },
+    "node_modules/@types/trusted-types": {
+      "version": "2.0.7",
+      "resolved": "https://registry.npmmirror.com/@types/trusted-types/-/trusted-types-2.0.7.tgz",
+      "integrity": "sha512-ScaPdn1dQczgbl0QFTeTOmVHFULt394XJgOQNoyVhZ6r2vLnMLJfBPd53SB52T/3G36VI1/g2MZaX0cwDuXsfw==",
+      "dev": true,
+      "license": "MIT"
+    },
     "node_modules/@types/unist": {
       "version": "3.0.2",
       "resolved": "https://registry.npmmirror.com/@types/unist/-/unist-3.0.2.tgz",
@@ -10715,6 +10734,12 @@
       "integrity": "sha512-l32Xp/TLgWb8ReqbVJAFIvXmY7go4nTxxlWiAFyhoQw9RKEOHBZNnyGvJWqDVSPmq3Y9HlM4npqF/T6VMOXhww==",
       "deprecated": "dommatrix is no longer maintained. Please use @thednp/dommatrix."
     },
+    "node_modules/dompurify": {
+      "version": "3.1.6",
+      "resolved": "https://registry.npmmirror.com/dompurify/-/dompurify-3.1.6.tgz",
+      "integrity": "sha512-cTOAhc36AalkjtBpfG6O8JimdTMWNXjiePT2xQH/ppBGi/4uIpmj8eKyIkMJErXWARyINV/sB38yf8JCLF5pbQ==",
+      "license": "(MPL-2.0 OR Apache-2.0)"
+    },
     "node_modules/domutils": {
       "version": "2.8.0",
       "resolved": "https://registry.npmmirror.com/domutils/-/domutils-2.8.0.tgz",

web/package.json

@@ -28,6 +28,7 @@
     "classnames": "^2.5.1",
     "dagre": "^0.8.5",
     "dayjs": "^1.11.10",
+    "dompurify": "^3.1.6",
     "elkjs": "^0.9.3",
     "eventsource-parser": "^1.1.2",
     "human-id": "^4.1.1",
@@ -63,6 +64,7 @@
     "@testing-library/jest-dom": "^6.4.5",
     "@testing-library/react": "^15.0.7",
     "@types/dagre": "^0.7.52",
+    "@types/dompurify": "^3.0.5",
     "@types/jest": "^29.5.12",
     "@types/lodash": "^4.14.202",
     "@types/react": "^18.0.33",

web/src/pages/add-knowledge/components/knowledge-chunk/components/chunk-card/index.tsx

@@ -2,6 +2,7 @@ import Image from '@/components/image';
 import { IChunk } from '@/interfaces/database/knowledge';
 import { Card, Checkbox, CheckboxProps, Flex, Popover, Switch } from 'antd';
 import classNames from 'classnames';
+import DOMPurify from 'dompurify';
 import { useState } from 'react';
 
 import { ChunkTextMode } from '../../constant';
@@ -73,7 +74,9 @@ const ChunkCard = ({
           className={styles.content}
         >
           <div
-            dangerouslySetInnerHTML={{ __html: item.content_with_weight }}
+            dangerouslySetInnerHTML={{
+              __html: DOMPurify.sanitize(item.content_with_weight),
+            }}
             className={classNames({
               [styles.contentEllipsis]: textMode === ChunkTextMode.Ellipse,
             })}

web/src/pages/add-knowledge/components/knowledge-setting/category-panel.tsx

@@ -2,6 +2,7 @@ import SvgIcon from '@/components/svg-icon';
 import { useTranslate } from '@/hooks/commonHooks';
 import { useSelectParserList } from '@/hooks/userSettingHook';
 import { Col, Divider, Empty, Row, Typography } from 'antd';
+import DOMPurify from 'dompurify';
 import { useMemo } from 'react';
 import styles from './index.less';
 import { ImageMap } from './utils';
@@ -39,7 +40,7 @@ const CategoryPanel = ({ chunkMethod }: { chunkMethod: string }) => {
           </Title>
           <p
             dangerouslySetInnerHTML={{
-              __html: item.description,
+              __html: DOMPurify.sanitize(item.description),
             }}
           ></p>
           <Title level={5}>

web/src/pages/chat/markdown-content/index.tsx

@@ -6,6 +6,7 @@ import { IChunk } from '@/interfaces/database/knowledge';
 import { getExtension } from '@/utils/documentUtils';
 import { InfoCircleOutlined } from '@ant-design/icons';
 import { Button, Flex, Popover, Space } from 'antd';
+import DOMPurify from 'dompurify';
 import { useCallback } from 'react';
 import Markdown from 'react-markdown';
 import reactStringReplace from 'react-string-replace';
@@ -94,7 +95,7 @@ const MarkdownContent = ({
           <Space direction={'vertical'}>
             <div
               dangerouslySetInnerHTML={{
-                __html: chunkItem?.content_with_weight,
+                __html: DOMPurify.sanitize(chunkItem?.content_with_weight),
               }}
               className={styles.chunkContentText}
             ></div>