adding doc and reader function

app.py

@@ -10,7 +10,6 @@ from langchain_core.messages import HumanMessage
 from langgraph.graph.message import add_messages
 from qdrant_client import QdrantClient
 from langchain_openai import OpenAIEmbeddings
-from langchain_community.document_loaders import DirectoryLoader
 
 import os
 
@@ -58,28 +57,31 @@ def search(query_vector, top_k=1) -> list:
 
     return return_hits
 
-import os
-from langchain.document_loaders import DirectoryLoader
-
 def get_document_by_name(doc_name: str) -> str:
-    """Retrieve a document from the list of loaded documents by its name from QD."""
+    """Retrieve the raw HTML content of a document by its name from the `data/` folder."""
+    
+    # ✅ Get the absolute path of the `data/` folder
+    script_dir = os.path.dirname(os.path.abspath(__file__))
+    data_path = os.path.join(script_dir, "data")
 
     # ✅ Replace `.pdf` with `.html`
     html_doc_name = doc_name.replace(".pdf", ".html")
-    
-    # ✅ Define the directory path
-    path = "data/"
-    
-    # ✅ Load only `.html` files
-    loader = DirectoryLoader(path, glob="*.html")
-    docs = loader.load()
-    
-    # ✅ Search for the document by name
-    for doc in docs:
-        if os.path.basename(doc.metadata.get("source", "")) == html_doc_name:
-            return doc.page_content  # ✅ Return the document content
-    
-    return "No file found"  # ✅ Return this if no matching file is found
+    full_path = os.path.join(data_path, html_doc_name)
+
+    # ✅ Check if the file exists
+    if not os.path.exists(full_path):
+        print(f"⚠️ File not found: {full_path}")
+        return "No file found"
+
+    try:
+        # ✅ Open and read the file content
+        with open(full_path, "r", encoding="utf-8") as file:
+            content = file.read()
+        return content  # ✅ Return the raw HTML content
+
+    except Exception as e:
+        print(f"❌ Error reading file {full_path}: {str(e)}")
+        return "Error reading file"
 
 
 # **User question prompt**

pyproject.toml

@@ -11,6 +11,7 @@ dependencies = [
     "langchain-openai>=0.3.7",
     "langchain-qdrant>=0.2.0",
     "langgraph>=0.2.74",
+    "nltk>=3.9.1",
     "qdrant-client>=1.13.2",
     "unstructured>=0.14.8",
     "websockets>=15.0",

uv.lock

@@ -1159,6 +1159,7 @@ dependencies = [
     { name = "langchain-openai" },
     { name = "langchain-qdrant" },
     { name = "langgraph" },
+    { name = "nltk" },
     { name = "qdrant-client" },
     { name = "unstructured" },
     { name = "websockets" },
@@ -1172,6 +1173,7 @@ requires-dist = [
     { name = "langchain-openai", specifier = ">=0.3.7" },
     { name = "langchain-qdrant", specifier = ">=0.2.0" },
     { name = "langgraph", specifier = ">=0.2.74" },
+    { name = "nltk", specifier = ">=3.9.1" },
     { name = "qdrant-client", specifier = ">=1.13.2" },
     { name = "unstructured", specifier = ">=0.14.8" },
     { name = "websockets", specifier = ">=15.0" },