Update agent.py

agent.py

@@ -32,130 +32,130 @@ def build_gemini_llm():
         raise ValueError("GOOGLE_API_KEY environment variable is not set.")
     return ChatGoogleGenerativeAI(model = "gemini-3.7-flash", temperature = 0, max_output_tokens = 1025, include_thoughts=True)
 
-@tool
-def extract_text_from_image(img_path: str) -> str:
-    """
-    Describe the image and extract any text in it.
-    Args:
-        img_path (str): the path to the image file.
-    """
-    all_text = ""
-    try:
-        # Read image and encode as base64
-        with open(img_path, "rb") as image_file:
-            image_bytes = image_file.read()
-
-        image_base64 = base64.b64encode(image_bytes).decode("utf-8")
-
-        # Prepare the prompt including the base64 image data
-        message = [
-            HumanMessage(
-                content=[
-                    {
-                        "type": "text",
-                        "text": (
-                            "Describe the image and extract any text in it."
-                        ),
-                    },
-                    {
-                        "type": "image_url",
-                        "image_url": {
-                            "url": f"data:image/png;base64,{image_base64}"
-                        },
-                    },
-                ]
-            )
-        ]
-        response = model.invoke(message)
-        # Append extracted text
-        all_text += response.text + "\n\n"
-        return all_text.strip()
-    except Exception as e:
-        # A butler should handle errors gracefully
-        error_msg = f"Error extracting text: {str(e)}"
-        print(error_msg)
-        return ""
-
-@tool
-def download_and_read_file(task_id: str) -> str:
-    """
-    Download and read the file attached to the GAIA task its contents.
-    Always call this first if there is a file attached to a GAIA Task.
-    Args:
-        task_id (str): The ID of the GAIA task.
-    Returns:
-        str: The contents of the file as a string.
-    """
-
-    try:
-        # Download the file from the GAIA API
-        response = requests.get(f"{FILES_URL}/{task_id}", timeout = 10)
-        response.raise_for_status()
-
-        # Determine the file type and read its contents
-        content_disposition = response.headers.get("content-disposition", "")
-        content_type = response.headers.get("content-type", "")
-        filename = None
-        if "filename=" in content_disposition:
-            filename = content_disposition.split("filename=")[1].strip('"')
-
-        if not filename:
-            filename = f"{task_id}.bin"
-
-        ext = Path(filename).suffix.lower()
-
-        if ext in(".txt", ".py", ".json", ".md", ".ymal", ".html", ".xml", ""):
-            return response.text
-
-        if ext == ".xlsx" or "xlsx" in content_type:
-            import pandas as pd
-            with tempfile.NamedTemporaryFile(suffix=".xlsx", delete=False) as file:
-                file.write(response.content)
-                temp_path = file.name
-
-            read_file = pd.read_excel(temp_path)
-            return read_file.to_string()
-        if ext == ".csv" or "csv" in content_type:
-            import pandas as pd
-            with tempfile.NamedTemporaryFile(suffix=".csv", delete=False) as file:
-                file.write(response.content)
-                temp_path = file.name
-            read_file = pd.read_csv(temp_path)
-            return read_file.to_string()
-        if ext == ".csv" or "csv" in content_type:
-            import pandas as pd
-            with tempfile.NamedTemporaryFile(suffix=".csv", delete=False) as file:
-                file.write(response.content)
-                temp_path = file.name
-            read_file = pd.read_csv(temp_path)
-            return read_file.to_string()
-        if ext == ".jpg" or ext == ".jpeg" or ext == ".png" or "image" in content_type:
-            from PIL import Image
-            with tempfile.NamedTemporaryFile(suffix=ext, delete=False) as file:
-                file.write(response.content)
-                temp_path = file.name
-            return extract_text_from_image(temp_path)
+# @tool
+# def extract_text_from_image(img_path: str) -> str:
+#     """
+#     Describe the image and extract any text in it.
+#     Args:
+#         img_path (str): the path to the image file.
+#     """
+#     all_text = ""
+#     try:
+#         # Read image and encode as base64
+#         with open(img_path, "rb") as image_file:
+#             image_bytes = image_file.read()
+
+#         image_base64 = base64.b64encode(image_bytes).decode("utf-8")
+
+#         # Prepare the prompt including the base64 image data
+#         message = [
+#             HumanMessage(
+#                 content=[
+#                     {
+#                         "type": "text",
+#                         "text": (
+#                             "Describe the image and extract any text in it."
+#                         ),
+#                     },
+#                     {
+#                         "type": "image_url",
+#                         "image_url": {
+#                             "url": f"data:image/png;base64,{image_base64}"
+#                         },
+#                     },
+#                 ]
+#             )
+#         ]
+#         response = model.invoke(message)
+#         # Append extracted text
+#         all_text += response.text + "\n\n"
+#         return all_text.strip()
+#     except Exception as e:
+#         # A butler should handle errors gracefully
+#         error_msg = f"Error extracting text: {str(e)}"
+#         print(error_msg)
+#         return ""
+
+# @tool
+# def download_and_read_file(task_id: str) -> str:
+#     """
+#     Download and read the file attached to the GAIA task its contents.
+#     Always call this first if there is a file attached to a GAIA Task.
+#     Args:
+#         task_id (str): The ID of the GAIA task.
+#     Returns:
+#         str: The contents of the file as a string.
+#     """
+
+#     try:
+#         # Download the file from the GAIA API
+#         response = requests.get(f"{FILES_URL}/{task_id}", timeout = 10)
+#         response.raise_for_status()
+
+#         # Determine the file type and read its contents
+#         content_disposition = response.headers.get("content-disposition", "")
+#         content_type = response.headers.get("content-type", "")
+#         filename = None
+#         if "filename=" in content_disposition:
+#             filename = content_disposition.split("filename=")[1].strip('"')
+
+#         if not filename:
+#             filename = f"{task_id}.bin"
+
+#         ext = Path(filename).suffix.lower()
+
+#         if ext in(".txt", ".py", ".json", ".md", ".ymal", ".html", ".xml", ""):
+#             return response.text
+
+#         if ext == ".xlsx" or "xlsx" in content_type:
+#             import pandas as pd
+#             with tempfile.NamedTemporaryFile(suffix=".xlsx", delete=False) as file:
+#                 file.write(response.content)
+#                 temp_path = file.name
+
+#             read_file = pd.read_excel(temp_path)
+#             return read_file.to_string()
+#         if ext == ".csv" or "csv" in content_type:
+#             import pandas as pd
+#             with tempfile.NamedTemporaryFile(suffix=".csv", delete=False) as file:
+#                 file.write(response.content)
+#                 temp_path = file.name
+#             read_file = pd.read_csv(temp_path)
+#             return read_file.to_string()
+#         if ext == ".csv" or "csv" in content_type:
+#             import pandas as pd
+#             with tempfile.NamedTemporaryFile(suffix=".csv", delete=False) as file:
+#                 file.write(response.content)
+#                 temp_path = file.name
+#             read_file = pd.read_csv(temp_path)
+#             return read_file.to_string()
+#         if ext == ".jpg" or ext == ".jpeg" or ext == ".png" or "image" in content_type:
+#             from PIL import Image
+#             with tempfile.NamedTemporaryFile(suffix=ext, delete=False) as file:
+#                 file.write(response.content)
+#                 temp_path = file.name
+#             return extract_text_from_image(temp_path)
         
-        # Unsupported file type 
-        return ( 
-            f"Unsupported file type: {content_type}. "
-            "I downloaded the file successfully, but I don't know "
-            "how to extract its contents." 
-            ) 
-    except requests.RequestException as e: 
-        return f"Failed to download file: {e}" 
-    except Exception as e: 
-        return f"Failed to read file: {e}"
-    except Exception as e:
-        return f"error downloading or reading file: {str(e)}"
+#         # Unsupported file type 
+#         return ( 
+#             f"Unsupported file type: {content_type}. "
+#             "I downloaded the file successfully, but I don't know "
+#             "how to extract its contents." 
+#             ) 
+#     except requests.RequestException as e: 
+#         return f"Failed to download file: {e}" 
+#     except Exception as e: 
+#         return f"Failed to read file: {e}"
+#     except Exception as e:
+#         return f"error downloading or reading file: {str(e)}"
 
 @tool
-def wikipedia_search(query: str):
+def wikipedia_search(query: str) -> str:
     """
-    Search wikipedia for a query and return a max of three results.
-    Takes a string query as the search query
+    Search wikipedia for a query and return results.
+    Takes a string query as the keywords to search
     Args:
-        query (str): The search query.
+        query (str): Keywords you want to search.
     """
     try:
         search_results = WikipediaLoader(query=query, load_max_docs=3).load()
@@ -164,7 +164,7 @@ def wikipedia_search(query: str):
             return f"No Wikipedia results found for {query}. Consider another query or try a web search."
         print("Result from wikipedia:")
         print(search_results)
-        return "\n\n---\n\n".join(
+        return "---\n".join(
             f"Title: {doc.metadata.get('title', 'Unknown')}\n"
             f"Content: {doc.page_content}"
             for doc in search_results