Delete webapp

webapp/Dockerfile

@@ -1,27 +0,0 @@
-FROM python:3.11-slim
-
-# Set environment variables
-ENV PYTHONUNBUFFERED=1 \
-    PYTHONDONTWRITEBYTECODE=1 \
-    PORT=7860
-
-# Create user and set permissions (HF expectation)
-RUN useradd -m -u 1000 user
-USER user
-ENV HOME=/home/user \
-    PATH=/home/user/.local/bin:$PATH
-
-WORKDIR $HOME/app
-
-# Copy requirements and install
-COPY --chown=user requirements.txt .
-RUN pip install --no-cache-dir -r requirements.txt
-
-# Copy app code
-COPY --chown=user . .
-
-# Expose port
-EXPOSE 7860
-
-# Run entrypoint
-ENTRYPOINT ["streamlit", "run", "app.py", "--server.port=7860", "--server.address=0.0.0.0"]

webapp/app.py

@@ -1,250 +0,0 @@
-import streamlit as st
-import requests
-import json
-import pandas as pd
-from datetime import datetime
-
-# Page Configuration
-st.set_page_config(
-    page_title="RAG SaaS Playground",
-    page_icon="🤖",
-    layout="wide",
-    initial_sidebar_state="expanded",
-)
-
-# Custom CSS for Premium Look
-st.markdown("""
-<style>
-    .main {
-        background-color: #0e1117;
-        color: #ffffff;
-    }
-    .stButton>button {
-        width: 100%;
-        border-radius: 5px;
-        height: 3em;
-        background-color: #1f6feb;
-        color: white;
-        font-weight: bold;
-    }
-    .stTextInput>div>div>input {
-        background-color: #161b22;
-        color: white;
-        border: 1px solid #30363d;
-    }
-    .sidebar .sidebar-content {
-        background-color: #161b22;
-    }
-    .css-1d391kg {
-        background-color: #161b22;
-    }
-    .stChatMessage {
-        background-color: #161b22;
-        border-radius: 10px;
-        padding: 10px;
-        margin-bottom: 10px;
-        border: 1px solid #30363d;
-    }
-</style>
-""", unsafe_allow_html=True)
-
-# Session State Initialization
-if "api_url" not in st.session_state:
-    st.session_state.api_url = "http://localhost:7860/v1"
-if "tenant_id" not in st.session_state:
-    st.session_state.tenant_id = "default"
-if "messages" not in st.session_state:
-    st.session_state.messages = []
-
-# Helper Functions
-def api_request(method, endpoint, data=None, headers=None):
-    url = f"{st.session_state.api_url}{endpoint}"
-    if headers is None:
-        headers = {}
-    headers["X-Tenant-ID"] = st.session_state.tenant_id
-    
-    try:
-        if method == "GET":
-            response = requests.get(url, headers=headers)
-        elif method == "POST":
-            response = requests.post(url, json=data, headers=headers)
-        response.raise_for_status()
-        return response.json()
-    except Exception as e:
-        st.error(f"API Error: {str(e)}")
-        return None
-
-# Sidebar - Configuration
-with st.sidebar:
-    st.title("⚙️ RAG SaaS Config")
-    
-    with st.expander("🔗 API Connection", expanded=True):
-        st.session_state.api_url = st.text_input("Backend API URL", st.session_state.api_url)
-        if st.button("Check Connectivity"):
-            res = requests.get(st.session_state.api_url.replace("/v1", "/health"))
-            if res.status_code == 200:
-                st.success("Backend Connected!")
-            else:
-                st.error("Failed to connect.")
-
-    st.divider()
-    
-    st.title("🏢 Tenant Management")
-    
-    # List Tenants
-    tenants = api_request("GET", "/tenants/")
-    if tenants:
-        tenant_names = [t["name"] for t in tenants]
-        tenant_ids = [t["id"] for t in tenants]
-        
-        selected_index = 0
-        if st.session_state.tenant_id in tenant_ids:
-            selected_index = tenant_ids.index(st.session_state.tenant_id)
-            
-        selected_tenant_name = st.selectbox("Select Active Tenant", tenant_names, index=selected_index)
-        st.session_state.tenant_id = tenant_ids[tenant_names.index(selected_tenant_name)]
-    
-    # Create Tenant
-    with st.expander("➕ Create New Tenant"):
-        new_tenant_id = st.text_input("Tenant ID (Slug)")
-        new_tenant_name = st.text_input("Tenant Name")
-        if st.button("Create"):
-            if new_tenant_id and new_tenant_name:
-                res = api_request("POST", "/tenants/", data={"id": new_tenant_id, "name": new_tenant_name})
-                if res:
-                    st.success(f"Tenant '{new_tenant_name}' created!")
-                    st.rerun()
-            else:
-                st.warning("Please fill all fields.")
-
-    st.divider()
-    st.info(f"Active Tenant: **{st.session_state.tenant_id}**")
-
-# Main Content
-st.title("🚀 RAG SaaS Multi-tenant Playground")
-
-tab_chat, tab_ingest, tab_playground = st.tabs(["💬 Chat", "📥 Ingestion", "📖 How it Works"])
-
-# --- TAB: CHAT ---
-with tab_chat:
-    st.subheader("Interactive Q&A")
-    
-    # Display chat messages from history on app rerun
-    for message in st.session_state.messages:
-        with st.chat_message(message["role"]):
-            st.markdown(message["content"])
-            if "citations" in message and message["citations"]:
-                with st.expander("References"):
-                    for i, cit in enumerate(message["citations"]):
-                        st.markdown(f"**[{i}]** {cit.get('text_snippet')}")
-
-    # React to user input
-    if prompt := st.chat_input("Ask something about your documents..."):
-        # Display user message in chat message container
-        st.chat_message("user").markdown(prompt)
-        st.session_state.messages.append({"role": "user", "content": prompt})
-
-        with st.chat_message("assistant"):
-            with st.spinner("Thinking..."):
-                response = api_request("POST", "/ask/", data={"query": prompt})
-                if response:
-                    st.markdown(response["answer"])
-                    if response["citations"]:
-                        with st.expander("References"):
-                            for i, cit in enumerate(response["citations"]):
-                                st.markdown(f"**[{i}]** {cit.get('text_snippet')}")
-                    
-                    st.session_state.messages.append({
-                        "role": "assistant", 
-                        "content": response["answer"],
-                        "citations": response["citations"]
-                    })
-
-    if st.button("Clear Chat History", key="clear_chat"):
-        st.session_state.messages = []
-        st.rerun()
-
-# --- TAB: INGESTION ---
-with tab_ingest:
-    st.subheader("Knowledge Management")
-    
-    st.write("Upload documents to your tenant's vector space.")
-    
-    col1, col2 = st.columns([2, 1])
-    
-    with col1:
-        uploaded_file = st.file_uploader("Choose a file (PDF, Text)", type=["pdf", "txt", "docx"])
-        source_name = st.text_input("Source Name", value=uploaded_file.name if uploaded_file else "")
-        
-        if st.button("Process & Ingest"):
-            if uploaded_file and source_name:
-                with st.spinner("Uploading and processing..."):
-                    # 1. Create Source
-                    # In a real app, you'd upload the file to storage and get a URI
-                    # For playground, we'll mock the URI as the filename
-                    source_uid = f"src_{datetime.now().strftime('%Y%m%d%H%M%S')}"
-                    source_data = {
-                        "id": source_uid,
-                        "tenant_id": st.session_state.tenant_id,
-                        "type": "file",
-                        "name": source_name,
-                        "uri": f"uploads/{uploaded_file.name}",
-                        "metadata": {"filename": uploaded_file.name}
-                    }
-                    
-                    # Create source record
-                    created_source = api_request("POST", "/sources/", data=source_data)
-                    
-                    if created_source:
-                        # 2. Trigger Ingestion
-                        ingest_res = api_request("POST", f"/ingest/{source_uid}")
-                        if ingest_res:
-                            st.success(f"Successfully ingested '{source_name}'!")
-                            st.balloons()
-            else:
-                st.warning("Please upload a file and provide a name.")
-
-    with col2:
-        st.write("📋 **Existing Sources**")
-        sources = api_request("GET", "/sources/")
-        if sources:
-            df = pd.DataFrame([
-                {"Name": s["name"], "Type": s["type"], "Created": s["created_at"][:10]} 
-                for s in sources
-            ])
-            st.dataframe(df, use_container_width=True)
-        else:
-            st.info("No sources found for this tenant.")
-
-# --- TAB: PLAYGROUND ---
-with tab_playground:
-    st.header("🛠️ How the Multi-tenant RAG SaaS Works")
-    
-    st.markdown("""
-    ### 1. Multi-Tenancy (Data Isolation)
-    Every request to the RAG backend includes an `X-Tenant-ID` header. This ensures that:
-    - **Separate Vector Index**: Documents from Tenant A are never retrieved for Tenant B.
-    - **Isolated Databases**: Metadata and audit trails are tagged per tenant.
-    
-    ### 2. The Ingestion Pipeline
-    When you upload a document:
-    1. **Extraction**: The system parses the file (PDF, TXT) into raw text.
-    2. **Chunking**: Text is split into semantic chunks with overlapping context.
-    3. **Embedding**: Each chunk is converted into a vector (numerical representation) using models like CLIP or OpenAI.
-    4. **Vector Storage**: Vectors are stored in a database (ChromaDB/Qdrant) alongside metadata.
-    
-    ### 3. The Retrieval-Augmented Generation (RAG) Loop
-    When you ask a question:
-    1. **Query Rewriting**: The AI rephrases your query to better match document content.
-    2. **Semantic Search**: The system finds the top 5 most relevant chunks for your specific tenant.
-    3. **Contextual Answer**: The AI generates a response using only the retrieved chunks as evidence.
-    4. **Citations**: Claims are linked back to the specific source chunks.
-    
-    ### 4. Play with it!
-    Try creating two different tenants (e.g., *Marketing* and *Engineering*). Upload different documents to each, and see how the AI answers differently based on which tenant you select in the sidebar!
-    """)
-    
-    st.image("https://miro.medium.com/v2/resize:fit:1400/1*C5e6DkCHiBfKTo296QY4kg.png", caption="RAG Architecture Diagram")
-
-st.markdown("---")
-st.markdown("<div style='text-align: center;'>Built for AI Institute Africa • 2026</div>", unsafe_allow_html=True)

webapp/requirements.txt

@@ -1,4 +0,0 @@
-streamlit
-requests
-python-dotenv
-pandas