Upload 2 files

agentic/langgraph_agent.py

@@ -40,7 +40,8 @@ class LangGraphAgent4GAIA:
             add_list,
             web_search,
             arxiv_search,
-            wiki_search
+            wiki_search,
+            read_xlsx_file
         ]
 
         # 1. Build graph

agentic/tools.py

@@ -1,8 +1,11 @@
+import requests
+import pandas as pd 
+from io import BytesIO
+from typing import List 
 from langchain_core.tools import tool
 from langchain_community.tools import DuckDuckGoSearchResults
 from langchain_community.tools.tavily_search import TavilySearchResults
 from langchain_community.document_loaders import ArxivLoader, WikipediaLoader
-from typing import List 
 
 @tool
 def add(a: int, b: int) -> int:
@@ -78,7 +81,7 @@ def web_search(query: str) -> str:
     search_docs = TavilySearchResults(max_results=3).invoke({'query': query})
     formatted_search_docs = "\n\n---\n\n".join(
         [
-            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
+            f'<Document title="{doc["title"]}>"\n{doc["content"]}\n</Document>'
             for doc in search_docs
         ])
     return f"web_results:\n{formatted_search_docs}"
@@ -93,7 +96,7 @@ def arxiv_search(query: str) -> str:
     search_docs = ArxivLoader(query=query, load_max_docs=3).load()
     formatted_search_docs = "\n\n---\n\n".join(
         [
-            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content[:1000]}\n</Document>'
+            f'<Document title="{doc.metadata["Title"]}>"\n{doc.metadata["Summary"]}\n</Document>'
             for doc in search_docs
         ])
     return f"arxiv_content:\n{formatted_search_docs}"
@@ -108,7 +111,36 @@ def wiki_search(query: str) -> str:
     search_docs = WikipediaLoader(query=query, load_max_docs=2).load()
     formatted_search_docs = "\n\n---\n\n".join(
         [
-            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
+            f'<Document title="{doc.metadata["Title"]}>"\n{doc.metadata["Summary"]}\n</Document>'
             for doc in search_docs
         ])
     return f"wikipedia_content:\n{formatted_search_docs}"
+
+
+@tool
+def read_xlsx_file(file_location: str) -> str:
+    """Read an .xlsx file from a path and return its contents as a string.
+    
+    Args:
+        file_location (str): Path to the xlsx file.
+    """
+    # Build the full file URL
+    base_name = file_location.split(".")[0]
+    file_url = f"https://agents-course-unit4-scoring.hf.space/files/{base_name}"
+
+    # Download the file
+    response = requests.get(file_url)
+    if response.status_code != 200:
+        raise RuntimeError(f"Failed to download file: {file_url}")
+
+    # Load the xlsx file into a pandas ExcelFile object
+    df_dict = pd.read_excel(BytesIO(response.content), sheet_name=None)
+
+    # Convert all sheets to a single formatted string
+    text_output = ""
+    for sheet_name, sheet_df in df_dict.items():
+        text_output += f"--- Sheet: {sheet_name} ---\n"
+        text_output += sheet_df.to_string(index=False)
+        text_output += "\n\n"
+
+    return f"xlsx file content:\n{text_output.strip()}"