Update app.py

app.py

@@ -1,5 +1,4 @@
 import os
-import tempfile
 import gradio as gr
 from langchain.document_loaders import PyPDFLoader, YoutubeLoader
 from langchain.text_splitter import RecursiveCharacterTextSplitter
@@ -11,23 +10,21 @@ from langchain.chat_models import init_chat_model
 # --- API KEY HANDLING ---
 OPENAI_API_KEY = os.getenv("OPENAI_API_KEY") or os.getenv("openai")
 if not OPENAI_API_KEY:
-    raise ValueError("❌ OPENAI API Key not found. Please add it to secrets as 'OPENAI_API_KEY' or 'openai'.")
+    raise ValueError("❌ OPENAI API Key not found. Please add it in Hugging Face secrets as 'OPENAI_API_KEY' or 'openai'.")
 
-# --- GRADIO PIPELINE FUNCTION ---
+# --- PROCESSING PIPELINE FUNCTION ---
 def process_inputs(pdf_file, youtube_url, query):
     docs = []
 
     # Load PDF
     try:
-        with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp:
-            tmp.write(pdf_file.read())
-            pdf_path = tmp.name
+        pdf_path = pdf_file.name  # ✅ Use .name to get the actual file path from Gradio
         pdf_loader = PyPDFLoader(pdf_path)
         docs.extend(pdf_loader.load())
     except Exception as e:
         return f"❌ Failed to load PDF: {e}"
 
-    # Load YouTube transcript
+    # Load YouTube Transcript
     try:
         yt_loader = YoutubeLoader.from_youtube_url(youtube_url, add_video_info=False)
         docs.extend(yt_loader.load())
@@ -37,38 +34,39 @@ def process_inputs(pdf_file, youtube_url, query):
     if not docs:
         return "❌ No documents could be loaded from the PDF or YouTube URL."
 
-    # Split
+    # Split documents
     splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=150)
     splits = splitter.split_documents(docs)
 
-    # Embed + Vectorstore
-    embedding = OpenAIEmbeddings(model="text-embedding-3-large", api_key=OPENAI_API_KEY)
-    db = FAISS.from_documents(splits, embedding)
+    # Embedding + Vector Store
+    try:
+        embedding = OpenAIEmbeddings(model="text-embedding-3-large", api_key=OPENAI_API_KEY)
+        db = FAISS.from_documents(splits, embedding)
+    except Exception as e:
+        return f"❌ Embedding failed: {e}"
 
     # QA Chain
-    llm = init_chat_model("gpt-4o-mini", model_provider="openai", api_key=OPENAI_API_KEY)
-    qa = RetrievalQA.from_chain_type(llm, retriever=db.as_retriever())
-
-    # Query
     try:
+        llm = init_chat_model("gpt-4o-mini", model_provider="openai", api_key=OPENAI_API_KEY)
+        qa = RetrievalQA.from_chain_type(llm, retriever=db.as_retriever())
         result = qa.invoke({"query": query})
         return result["result"]
     except Exception as e:
-        return f"❌ Error during retrieval: {e}"
+        return f"❌ Retrieval failed: {e}"
 
-# --- GRADIO UI ---
+# --- GRADIO APP ---
 with gr.Blocks() as demo:
     gr.Markdown("## 📚 Ask Questions from PDF + YouTube Transcript")
 
     with gr.Row():
         pdf_input = gr.File(label="Upload PDF", file_types=[".pdf"])
         yt_input = gr.Textbox(label="YouTube URL", placeholder="https://www.youtube.com/watch?v=...")
-    
-    query_input = gr.Textbox(label="Your Question", placeholder="What did the video/PDF say about X?")
+
+    query_input = gr.Textbox(label="Your Question", placeholder="e.g., What did the PDF say about X?")
     output = gr.Textbox(label="Answer")
 
     run_button = gr.Button("Get Answer")
     run_button.click(fn=process_inputs, inputs=[pdf_input, yt_input, query_input], outputs=output)
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()