Update app.py

app.py

@@ -50,11 +50,11 @@ if "vector_store" not in st.session_state:
 
 # ----------------- Metadata Extraction -----------------
 def extract_metadata_llm(pdf_path):
-    """Extracts metadata using LLM instead of regex."""
+    """Extracts metadata using LLM"""
     with pdfplumber.open(pdf_path) as pdf:
         first_page_text = pdf.pages[0].extract_text() if pdf.pages else "No text found."
 
-    #Define metadata_prompt
+    # Define metadata_prompt
     metadata_prompt = PromptTemplate(
         input_variables=["text"],
         template="""
@@ -81,7 +81,7 @@ def extract_metadata_llm(pdf_path):
         """
     )
 
-    # ✅ Now metadata_prompt exists when used in LLMChain
+
     metadata_chain = LLMChain(llm=llm_judge, prompt=metadata_prompt, output_key="metadata")
     metadata_response = metadata_chain.invoke({"text": first_page_text})
 
@@ -91,6 +91,11 @@ def extract_metadata_llm(pdf_path):
 
     try:
         metadata_dict = json.loads(json_text)
+        # Ensure all expected fields are present
+        required_fields = ["Title", "Author", "Emails", "Affiliations"]
+        for field in required_fields:
+            if field not in metadata_dict:
+                metadata_dict[field] = "Unknown"
     except json.JSONDecodeError:
         metadata_dict = {
             "Title": "Unknown",