feat(kb): Support programming languages

package.json

@@ -42,7 +42,6 @@
     "localforage": "^1.10.0",
     "marked": "^14.0.0",
     "marked-highlight": "^2.1.4",
-    "mime": "^4.0.4",
     "ml-distance": "^4.0.1",
     "pdfjs-dist": "^4.5.136",
     "pinia": "^2.2.1",

src/utils/knowledge/embedding.ts

@@ -44,9 +44,30 @@ const getTextSplitter = (
 ): MarkdownTextSplitter | RecursiveCharacterTextSplitter => {
   switch (fileType) {
     case 'markdown':
+    case 'html':
+    case 'cpp':
+    case 'go':
+    case 'java':
+    case 'js':
+    case 'php':
+    case 'proto':
+    case 'rst':
+    case 'scala':
+    case 'swift':
+    case 'sol':
       return RecursiveCharacterTextSplitter.fromLanguage(fileType);
+    case 'ts':
+      return RecursiveCharacterTextSplitter.fromLanguage('js');
+    case 'py':
+      return RecursiveCharacterTextSplitter.fromLanguage('python');
+    case 'rb':
+      return RecursiveCharacterTextSplitter.fromLanguage('ruby');
+    case 'rs':
+      return RecursiveCharacterTextSplitter.fromLanguage('rust');
     case 'md':
       return RecursiveCharacterTextSplitter.fromLanguage('markdown');
+    case 'htm':
+      return RecursiveCharacterTextSplitter.fromLanguage('html');
     default:
       return new RecursiveCharacterTextSplitter({
         chunkSize,

src/utils/knowledge/parsing.ts

@@ -1,7 +1,28 @@
 import * as pdfjs from 'pdfjs-dist';
 import { TextItem } from 'pdfjs-dist/types/src/display/api';
 
-export const supportedInputFiles = ['.txt', '.md', '.markdown', '.pdf'].join(',');
+export const supportedInputFiles = [
+  '.txt',
+  '.md',
+  '.markdown',
+  '.pdf',
+  '.html',
+  '.htm',
+  '.cpp',
+  '.go',
+  '.java',
+  '.js',
+  '.ts',
+  '.php',
+  '.proto',
+  '.py',
+  '.rst',
+  '.rb',
+  '.rs',
+  '.scala',
+  '.swift',
+  '.sol',
+].join(',');
 
 const extractTextFromPdfFile = async (file: File): Promise<string> => {
   const pdfUrl = URL.createObjectURL(file);
@@ -36,6 +57,22 @@ export const extractFileContent = async (file: File): Promise<{ type: string; co
       case 'markdown':
       case 'md':
       case 'txt':
+      case 'html':
+      case 'htm':
+      case 'cpp':
+      case 'go':
+      case 'java':
+      case 'js':
+      case 'ts':
+      case 'php':
+      case 'proto':
+      case 'py':
+      case 'rst':
+      case 'rb':
+      case 'rs':
+      case 'scala':
+      case 'swift':
+      case 'sol':
         extractedText = await new Promise((resolve, reject) => {
           const reader = new FileReader();
           reader.onload = (event) => resolve(event.target!.result as string);