Update app.py

app.py

@@ -9,8 +9,7 @@ from langchain_core.messages import HumanMessage
 from langchain_openai import ChatOpenAI
 from langchain_community.tools import DuckDuckGoSearchResults
 from langchain_google_community import GoogleSearchAPIWrapper
-from langchain_community.document_loaders import YoutubeLoader
-from langchain_community.document_loaders import PyPDFLoader
+from langchain_community.document_loaders import YoutubeLoader,PyPDFLoader,Docx2txtLoader,TextLoader
 import wikipedia
 import speech_recognition as sr
 import tempfile
@@ -62,8 +61,42 @@ def pdf_loader_tool(file_url: str) -> str:
             
     except Exception as e:
         return f"Reading failed: {str(e)}"
-    
 
+def docx_loader_tool(file_url: str) -> str:
+    """Load and extract text from a docx file downloaded from given file_url."""
+    try:
+        # Download file into temporary file
+        with tempfile.NamedTemporaryFile(suffix=".docx", delete=True) as temp_file:
+            response = requests.get(file_url)
+            if response.status_code != 200:
+                return f"Failed to download file: {response.status_code}"
+            temp_file.write(response.content)
+            temp_file.flush()  # Make sure data is written
+            # Read from temp file
+            loader = Docx2txtLoader(temp_file.name)
+            docs = loader.load()
+            return "\n".join([doc.page_content for doc in docs])
+            
+    except Exception as e:
+        return f"Reading failed: {str(e)}"
+    
+def txt_loader_tool(file_url: str) -> str:
+    """Load and extract text from a txt file downloaded from given file_url."""
+    try:
+        # Download file into temporary file
+        with tempfile.NamedTemporaryFile(suffix=".txt", delete=True) as temp_file:
+            response = requests.get(file_url)
+            if response.status_code != 200:
+                return f"Failed to download file: {response.status_code}"
+            temp_file.write(response.content)
+            temp_file.flush()  # Make sure data is written
+            # Read from temp file
+            loader = TextLoader(temp_file.name)
+            docs = loader.load()
+            return "\n".join([doc.page_content for doc in docs])
+            
+    except Exception as e:
+        return f"Reading failed: {str(e)}"
     
 def read_image_text(file_URL: str) -> str:
     """Extract text from image downloaded from given file_URL using OCR."""
@@ -145,6 +178,8 @@ tools = [
     subtract,
     read_image_text,
     pdf_loader_tool,
+    docx_loader_tool,
+    txt_loader_tool,
     youtube_transcript_tool,
     transcribe_audio,
     analyze_python_code,
@@ -175,7 +210,22 @@ from langchain_core.messages import HumanMessage, SystemMessage
 def assistant(state: AgentState, llm_with_tools):
     # System message
     textual_description_of_tools = """
-
+docx_loader_tool(file_url: str) -> str:
+    Load and extract text from a docx file downloaded from given file_url.
+    
+    Args:
+        file_url, a string indicating the url of the given file.
+    Returns:
+        The text extracted from the given docx document.
+        
+def txt_loader_tool(file_url: str) -> str:
+    Load and extract text from a txt file downloaded from given file_url.
+    
+    Args:
+        file_url, a string indicating the url of the given file.
+    Returns:
+        The text extracted from the given txt.
+            
 pdf_loader_tool(file_url: str) -> str:
     Load and extract text from a PDF file downloaded from given file_url.
     Args: