Create app.py

app.py

@@ -0,0 +1,78 @@
+import os
+import subprocess
+import gradio as gr
+from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, Settings
+from llama_index.llms.groq import Groq
+from llama_index.embeddings.huggingface import HuggingFaceEmbedding
+from llama_index.vector_stores.qdrant import QdrantVectorStore
+from qdrant_client import QdrantClient
+from qdrant_client.models import Distance, VectorParams
+
+Settings.llm = Groq(model="llama3-8b-8192", api_key=os.environ.get("GROQ_API_KEY"))
+Settings.embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-large-en-v1.5")
+Settings.chunk_size = 512
+Settings.chunk_overlap = 50
+
+client = QdrantClient(location=":memory:")
+client.create_collection(
+    collection_name="tech_docs",
+    vectors_config=VectorParams(size=1024, distance=Distance.COSINE),
+)
+vector_store = QdrantVectorStore(client=client, collection_name="tech_docs")
+index = VectorStoreIndex.from_vector_store(vector_store=vector_store)
+
+TARGET_REPO = "https://github.com/tiangolo/fastapi.git"
+clone_path = "/tmp/fastapi_docs"
+
+if not os.path.exists(clone_path):
+    print(f"Cloning {TARGET_REPO}...")
+    subprocess.run(["git", "clone", "--depth", "1", TARGET_REPO, clone_path])
+
+print("Indexing repository documentation...")
+try:
+    documents = SimpleDirectoryReader(
+        input_dir=clone_path, 
+        required_exts=[".md", ".mdx"], 
+        recursive=True
+    ).load_data()
+    # Ingest a safe subset to fit comfortably within free RAM constraints
+    for doc in documents[:200]:
+        index.insert(doc)
+    print("GitHub repository indexed successfully!")
+except Exception as e:
+    print(f"Error indexing GitHub repo: {e}")
+
+query_engine = index.as_query_engine(streaming=False)
+
+# 4. Define Response Function for Gradio Chat Interface
+def chat_with_docs(message, history, file_obj):
+    global index, query_engine
+    
+    # If user uploaded a new PDF, ingest it dynamically into the active session
+    if file_obj is not None:
+        try:
+            pdf_docs = SimpleDirectoryReader(input_files=[file_obj.name]).load_data()
+            for doc in pdf_docs:
+                index.insert(doc)
+            query_engine = index.as_query_engine(streaming=False)
+        except Exception as e:
+            return f"Error processing uploaded PDF: {str(e)}"
+            
+    # Query the RAG engine using Groq acceleration
+    response = query_engine.query(message)
+    return str(response)
+
+# 5. Build Gradio Chat UI Layout
+demo = gr.ChatInterface(
+    fn=chat_with_docs,
+    title="📚 TechDocs-Knowledge-Engine",
+    description="Ask technical questions about the indexed codebase or upload your own technical PDF using the file attachment icon below.",
+    additional_inputs=[
+        gr.File(label="Upload Technical PDF (Optional)", file_types=[".pdf"])
+    ],
+    textbox=gr.Textbox(placeholder="Ask a technical question about code or architecture...", container=False, scale=7),
+    theme="soft"
+)
+
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7860)