From b144e0c3b0298243c6673e524df414057a6fb9d5 Mon Sep 17 00:00:00 2001
From: yangdx <yangdx@znipower.com>
Date: Fri, 21 Feb 2025 21:07:37 +0800
Subject: [PATCH] Sync modifications from main branch

---
 lightrag/api/routers/document_routes.py | 67 ++++++++++++++++++++++++-
 1 file changed, 66 insertions(+), 1 deletion(-)

diff --git a/lightrag/api/routers/document_routes.py b/lightrag/api/routers/document_routes.py
index c17ccd88..25ca24e4 100644
--- a/lightrag/api/routers/document_routes.py
+++ b/lightrag/api/routers/document_routes.py
@@ -117,6 +117,37 @@ class DocumentManager:
             ".docx",
             ".pptx",
             ".xlsx",
+            ".rtf",  # Rich Text Format
+            ".odt",  # OpenDocument Text
+            ".tex",  # LaTeX
+            ".epub",  # Electronic Publication
+            ".html",  # HyperText Markup Language
+            ".htm",  # HyperText Markup Language
+            ".csv",  # Comma-Separated Values
+            ".json",  # JavaScript Object Notation
+            ".xml",  # eXtensible Markup Language
+            ".yaml",  # YAML Ain't Markup Language
+            ".yml",  # YAML
+            ".log",  # Log files
+            ".conf",  # Configuration files
+            ".ini",  # Initialization files
+            ".properties",  # Java properties files
+            ".sql",  # SQL scripts
+            ".bat",  # Batch files
+            ".sh",  # Shell scripts
+            ".c",  # C source code
+            ".cpp",  # C++ source code
+            ".py",  # Python source code
+            ".java",  # Java source code
+            ".js",  # JavaScript source code
+            ".ts",  # TypeScript source code
+            ".swift",  # Swift source code
+            ".go",  # Go source code
+            ".rb",  # Ruby source code
+            ".php",  # PHP source code
+            ".css",  # Cascading Style Sheets
+            ".scss",  # Sassy CSS
+            ".less",  # LESS CSS
         ),
     ):
         self.input_dir = Path(input_dir)
@@ -170,7 +201,41 @@ async def pipeline_enqueue_file(rag: LightRAG, file_path: Path) -> bool:
 
         # Process based on file type
         match ext:
-            case ".txt" | ".md":
+            case (
+                ".txt"
+                | ".md"
+                | ".html"
+                | ".htm"
+                | ".tex"
+                | ".json"
+                | ".xml"
+                | ".yaml"
+                | ".yml"
+                | ".rtf"
+                | ".odt"
+                | ".epub"
+                | ".csv"
+                | ".log"
+                | ".conf"
+                | ".ini"
+                | ".properties"
+                | ".sql"
+                | ".bat"
+                | ".sh"
+                | ".c"
+                | ".cpp"
+                | ".py"
+                | ".java"
+                | ".js"
+                | ".ts"
+                | ".swift"
+                | ".go"
+                | ".rb"
+                | ".php"
+                | ".css"
+                | ".scss"
+                | ".less"
+            ):
                 content = file.decode("utf-8")
             case ".pdf":
                 if not pm.is_installed("pypdf2"):