Update app.py

app.py

@@ -1,33 +1,17 @@
-# app.py — Streamlit Hugging Face Space (Standalone Prototype)
-# Features:
-# - Auto-creates dummy supplier dataset (~50 suppliers)
-# - Parses inquiry (qty/brand/wattage/category/location)
-# - Agent-style steps + supplier ranking
-# - 3 pricing modes (Competitive/Balanced/High Margin)
-# - New-product fallback flow
-#
-# requirements.txt:
-# streamlit
-# pandas
-
 import os
 import re
 import random
 from datetime import datetime, timedelta
 
 import pandas as pd
-import streamlit as st
-
-st.set_page_config(page_title="AI Lighting Quotation Agent", layout="wide")
-st.title("💡 AI Lighting Quotation Agent (Prototype)")
-st.caption("Paste inquiry → AI extracts specs → ranks suppliers → recommends pricing → generates quote draft")
-
-DATA_PATH = "supplier_db.csv"
-random.seed(42)
+import gradio as gr
 
 # ----------------------------
-# Dummy supplier data
+# Config
 # ----------------------------
+random.seed(42)
+DATA_PATH = os.environ.get("SUPPLIER_DB_PATH", "/tmp/supplier_db.csv")
+
 PRODUCT_CATALOG = [
     ("LED Panel", ["panel", "led panel", "ceiling panel"], (14, 28)),
     ("Downlight", ["downlight", "spot", "spotlight"], (6, 18)),
@@ -42,11 +26,16 @@ PRODUCT_CATALOG = [
 BRANDS = ["Philips", "Osram", "Panasonic", "Schneider", "Opple", "NVC", "Crompton", "Wipro", "Havells", "Generic"]
 REGIONS = ["SG Central", "SG East", "SG West", "SG North", "Johor", "KL", "Batam"]
 
+
+# ----------------------------
+# Dummy Supplier DB
+# ----------------------------
 def make_supplier_name(i: int) -> str:
     prefixes = ["Bright", "Nova", "Apex", "Luma", "Spark", "Prime", "Zen", "Vertex", "Delta", "Orion"]
     suffixes = ["Lighting", "Electrics", "Solutions", "Supply", "Traders", "Distributors", "Imports", "Wholesale", "Mart", "Hub"]
     return f"{random.choice(prefixes)} {random.choice(suffixes)} Pte Ltd #{i:02d}"
 
+
 def generate_supplier_db(n_suppliers: int = 50) -> pd.DataFrame:
     rows = []
     for i in range(1, n_suppliers + 1):
@@ -58,7 +47,6 @@ def generate_supplier_db(n_suppliers: int = 50) -> pd.DataFrame:
         moq = random.choice([1, 5, 10, 20, 30, 50])
         competitiveness = round(random.uniform(0.85, 1.20), 2)
         brands_supported = random.sample(BRANDS, k=random.randint(2, 5))
-
         rows.append({
             "supplier_id": f"SUP-{1000+i}",
             "supplier_name": supplier,
@@ -74,21 +62,35 @@ def generate_supplier_db(n_suppliers: int = 50) -> pd.DataFrame:
         })
     return pd.DataFrame(rows)
 
+
 def load_or_create_db() -> pd.DataFrame:
     if os.path.exists(DATA_PATH):
-        return pd.read_csv(DATA_PATH)
+        try:
+            return pd.read_csv(DATA_PATH)
+        except Exception:
+            pass
     df = generate_supplier_db(50)
     df.to_csv(DATA_PATH, index=False)
     return df
 
-df_suppliers = load_or_create_db()
+
+SUPPLIERS_DF = load_or_create_db()
+
+
+def regenerate_db():
+    global SUPPLIERS_DF
+    SUPPLIERS_DF = generate_supplier_db(50)
+    SUPPLIERS_DF.to_csv(DATA_PATH, index=False)
+    return SUPPLIERS_DF.head(10), DATA_PATH
+
 
 # ----------------------------
-# Parser (no LLM)
+# Lightweight inquiry parsing (no LLM)
 # ----------------------------
 def normalize_text(t: str) -> str:
     return re.sub(r"\s+", " ", (t or "").strip().lower())
 
+
 def detect_quantity(text: str):
     patterns = [
         r"\bqty[:\s]*([0-9]{1,5})\b",
@@ -101,17 +103,20 @@ def detect_quantity(text: str):
             return int(m.group(1))
     return None
 
+
 def detect_wattage(text: str):
     m = re.search(r"\b([0-9]{1,4})\s*(w|watt|watts)\b", text, flags=re.IGNORECASE)
     return int(m.group(1)) if m else None
 
+
 def detect_brand(text: str):
-    t = text.lower()
+    t = (text or "").lower()
     for b in BRANDS:
         if b.lower() in t:
             return b
     return None
 
+
 def detect_category(text: str):
     t = normalize_text(text)
     for category, keywords, _rng in PRODUCT_CATALOG:
@@ -120,6 +125,7 @@ def detect_category(text: str):
                 return category
     return None
 
+
 def detect_location(text: str):
     t = normalize_text(text)
     loc_map = {
@@ -138,9 +144,9 @@ def detect_location(text: str):
             return v
     return None
 
+
 def parse_inquiry(text: str) -> dict:
     return {
-        "raw_text": (text or "").strip(),
         "quantity": detect_quantity(text) or 10,
         "wattage": detect_wattage(text),
         "brand": detect_brand(text),
@@ -148,6 +154,7 @@ def parse_inquiry(text: str) -> dict:
         "location": detect_location(text),
     }
 
+
 def estimate_market_range(category: str | None, wattage: int | None):
     if not category:
         return (10.0, 40.0)
@@ -166,12 +173,14 @@ def estimate_market_range(category: str | None, wattage: int | None):
         hi = hi * (0.85 + 0.20 * scale)
     return (round(lo, 2), round(hi, 2))
 
+
 def pick_margin(pricing_mode: str, base_margin: float):
     if pricing_mode == "Competitive":
         return max(5, base_margin - 6)
     if pricing_mode == "High Margin":
         return min(40, base_margin + 8)
-    return base_margin  # Balanced
+    return base_margin
+
 
 def compute_offers(req: dict, suppliers: pd.DataFrame, margin_pct: float):
     category = req.get("category")
@@ -205,13 +214,10 @@ def compute_offers(req: dict, suppliers: pd.DataFrame, margin_pct: float):
             supplier_cost *= 1.05
 
         supplier_cost = round(supplier_cost, 2)
-        sell_price = supplier_cost / (1 - margin_pct / 100.0)
-        sell_price = round(sell_price, 2)
+        sell_price = round(supplier_cost / (1 - margin_pct / 100.0), 2)
 
         reliability = float(s["reliability_score"])
         lead = int(s["lead_time_days"])
-
-        # Score: cheaper + reliable + fast lead
         score = (1 / max(sell_price, 0.01)) * 100 + reliability * 10 + (1 / max(lead, 1)) * 5
 
         rows.append({
@@ -230,97 +236,29 @@ def compute_offers(req: dict, suppliers: pd.DataFrame, margin_pct: float):
     offers = pd.DataFrame(rows).sort_values("score", ascending=False).head(10).reset_index(drop=True)
     return offers, (market_lo, market_hi)
 
-# ----------------------------
-# Sidebar
-# ----------------------------
-st.sidebar.header("⚙️ Prototype Controls")
-base_margin = st.sidebar.slider("Base Margin (%)", 5, 40, 20, 1)
-pricing_mode = st.sidebar.radio("Pricing Mode", ["Balanced", "Competitive", "High Margin"], index=0)
-top_n = st.sidebar.slider("Top offers to show", 3, 10, 5, 1)
-
-with st.sidebar.expander("📦 Supplier Database", expanded=False):
-    st.write(f"Suppliers loaded: **{len(df_suppliers)}**")
-    st.download_button(
-        "Download supplier_db.csv",
-        data=open(DATA_PATH, "rb").read(),
-        file_name="supplier_db.csv",
-        mime="text/csv",
-        use_container_width=True,
-    )
-    if st.button("Regenerate dummy DB (50 suppliers)"):
-        df_suppliers = generate_supplier_db(50)
-        df_suppliers.to_csv(DATA_PATH, index=False)
-        st.success("supplier_db.csv regenerated")
-        st.rerun()
-
-# ----------------------------
-# Main
-# ----------------------------
-left, right = st.columns([1.2, 1.0], gap="large")
-
-with left:
-    st.subheader("1) Paste Customer Inquiry")
-    sample = "Hi, please quote best price for 50 pcs Philips 18W LED panel light. Delivery to Singapore in 2 weeks."
-    inquiry = st.text_area("Inquiry", value=sample, height=150)
-
-    req = parse_inquiry(inquiry)
-
-    st.subheader("2) Agent Step: Requirement Extraction")
-    st.json(req)
-
-with right:
-    st.subheader("3) Agent Step: Market Intelligence (Demo)")
-    market_lo, market_hi = estimate_market_range(req.get("category"), req.get("wattage"))
-    st.metric("Estimated market low (SGD/unit)", f"{market_lo:.2f}")
-    st.metric("Estimated market high (SGD/unit)", f"{market_hi:.2f}")
-    st.caption("Heuristic estimate for demo. Later replace with real web/catalog research agent.")
-
-st.divider()
-
-margin_to_use = pick_margin(pricing_mode, base_margin)
-st.subheader("4) Agent Step: Supplier Shortlist + Pricing Recommendation")
-st.caption(f"Pricing mode: **{pricing_mode}** → Margin applied: **{margin_to_use:.0f}%**")
-
-offers_df, market_rng = compute_offers(req, df_suppliers, margin_to_use)
-
-if offers_df.empty:
-    st.error("No matching suppliers found in internal database.")
-    st.markdown("### 🆕 New Product / No-Match Mode (Prototype)")
-    st.write("**What the agent would do next:**")
-    st.write("1) Search market range online (by category + specs).")
-    st.write("2) Identify supplier categories that can support this product.")
-    st.write("3) Prepare a supplier outreach list and request quotes automatically.")
-    st.write("4) Update internal catalog once confirmed.")
-else:
-    st.dataframe(offers_df.head(top_n), use_container_width=True)
-
-    best = offers_df.iloc[0].to_dict()
-    st.success(
-        f"Recommended supplier: **{best['supplier_name']}** | "
-        f"Cost **SGD {best['est_supplier_cost_sgd']:.2f}** → Sell **SGD {best['recommended_sell_price_sgd']:.2f}** | "
-        f"Lead **{best['lead_time_days']}d** | Reliability **{best['reliability_score']:.2f}**"
-    )
 
-st.divider()
+def build_quote_text(req, offers_df, pricing_mode, margin_used, company_name, customer_name, valid_days):
+    if offers_df.empty:
+        return (
+            "No matching suppliers found.\n\n"
+            "New Product Mode:\n"
+            "1) Search market range online\n"
+            "2) Identify supplier categories\n"
+            "3) Send RFQs to shortlisted suppliers\n"
+            "4) Update internal catalog once confirmed\n"
+        )
 
-st.subheader("5) Quote Draft (Copy/Paste Demo)")
-company_name = st.text_input("Your company name", value="Delight Lighting (Demo)")
-customer_name = st.text_input("Customer name", value="Customer")
-quote_valid_days = st.number_input("Quote validity (days)", min_value=1, max_value=30, value=7)
-
-if offers_df.empty:
-    st.info("Once the agent finds matching suppliers, the quote draft will appear here.")
-else:
     qty = int(req.get("quantity") or 10)
     category = req.get("category") or "Lighting Product"
     brand = req.get("brand") or "Brand-agnostic"
     wattage = f"{req.get('wattage')}W" if req.get("wattage") else ""
     unit_price = float(offers_df.iloc[0]["recommended_sell_price_sgd"])
     total = round(unit_price * qty, 2)
+    valid_until = (datetime.today() + timedelta(days=int(valid_days))).strftime("%Y-%m-%d")
 
-    valid_until = (datetime.today() + timedelta(days=int(quote_valid_days))).strftime("%Y-%m-%d")
+    best = offers_df.iloc[0]
 
-    quote_text = f"""Subject: Quotation - {brand} {wattage} {category} (Qty: {qty})
+    return f"""Subject: Quotation - {brand} {wattage} {category} (Qty: {qty})
 
 Hi {customer_name},
 
@@ -331,23 +269,118 @@ Quantity: {qty}
 Unit Price: SGD {unit_price:.2f}
 Total: SGD {total:.2f}
 
-Estimated Lead Time: {int(offers_df.iloc[0]["lead_time_days"])} days
+Estimated Lead Time: {int(best["lead_time_days"])} days
 Validity: Until {valid_until}
-Terms: 50% advance, balance before delivery (demo terms)
-
-If you’d like, we can share alternate pricing options as well.
+Pricing Mode: {pricing_mode} (Margin applied: {margin_used:.0f}%)
 
 Regards,
 Sales Team
 {company_name}
 """
-    st.text_area("Generated Quote Draft", value=quote_text, height=240)
-    st.download_button(
-        "Download quote draft (.txt)",
-        data=quote_text.encode("utf-8"),
-        file_name="quote_draft.txt",
-        mime="text/plain",
-        use_container_width=True,
+
+
+def run_agent(inquiry_text, base_margin, pricing_mode, top_n, company_name, customer_name, valid_days):
+    req = parse_inquiry(inquiry_text)
+    margin_used = pick_margin(pricing_mode, float(base_margin))
+
+    offers_df, market_rng = compute_offers(req, SUPPLIERS_DF, margin_used)
+
+    market_text = "Estimated market range: "
+    if market_rng:
+        market_text += f"SGD {market_rng[0]:.2f} – {market_rng[1]:.2f} per unit"
+    else:
+        lo, hi = estimate_market_range(req.get("category"), req.get("wattage"))
+        market_text += f"SGD {lo:.2f} – {hi:.2f} per unit"
+
+    steps = []
+    steps.append(f"Step 1 — Extracted requirement: {req}")
+    steps.append(f"Step 2 — Market intelligence: {market_text}")
+    steps.append(f"Step 3 — Pricing mode: {pricing_mode} | Margin applied: {margin_used:.0f}%")
+    if offers_df.empty:
+        steps.append("Step 4 — No internal matches found → New Product Mode triggered.")
+    else:
+        steps.append(f"Step 4 — Shortlisted {min(len(offers_df), int(top_n))} suppliers; top recommendation: {offers_df.iloc[0]['supplier_name']}")
+
+    offers_view = offers_df.head(int(top_n)) if not offers_df.empty else pd.DataFrame(
+        columns=["supplier_id","supplier_name","region","reliability_score","lead_time_days","moq","est_supplier_cost_sgd","recommended_sell_price_sgd","score","contact_email"]
+    )
+
+    quote_text = build_quote_text(req, offers_df, pricing_mode, margin_used, company_name, customer_name, valid_days)
+
+    # Write quote text to a file for download
+    quote_path = "/tmp/quote_draft.txt"
+    with open(quote_path, "w", encoding="utf-8") as f:
+        f.write(quote_text)
+
+    # Return: parsed req (json-like), agent steps, market text, offers table, quote text, downloadable file, downloadable csv
+    return req, "\n".join(steps), market_text, offers_view, quote_text, quote_path, DATA_PATH
+
+
+# ----------------------------
+# UI (Gradio)
+# ----------------------------
+with gr.Blocks(title="Delight AI Agent (Prototype)") as demo:
+    gr.Markdown(
+        """
+# 💡 Delight AI Agent (Prototype)
+Paste a customer inquiry → agent extracts requirement → ranks suppliers → recommends pricing → generates quotation draft.
+"""
+    )
+
+    with gr.Row():
+        inquiry = gr.Textbox(
+            label="Customer Inquiry",
+            lines=6,
+            value="Hi, please quote best price for 50 pcs Philips 18W LED panel light. Delivery to Singapore in 2 weeks."
+        )
+
+    with gr.Row():
+        base_margin = gr.Slider(5, 40, value=20, step=1, label="Base Margin (%)")
+        pricing_mode = gr.Radio(["Balanced", "Competitive", "High Margin"], value="Balanced", label="Pricing Mode")
+        top_n = gr.Slider(3, 10, value=5, step=1, label="Top offers to show")
+
+    with gr.Row():
+        company_name = gr.Textbox(label="Your company name", value="Delight Lighting (Demo)")
+        customer_name = gr.Textbox(label="Customer name", value="Customer")
+        valid_days = gr.Slider(1, 30, value=7, step=1, label="Quote validity (days)")
+
+    run_btn = gr.Button("🚀 Run Agent")
+
+    with gr.Row():
+        parsed_req = gr.JSON(label="Extracted Requirement")
+        agent_steps = gr.Textbox(label="Agent Steps", lines=10)
+
+    market_info = gr.Textbox(label="Market Intelligence", lines=2)
+
+    offers_table = gr.Dataframe(
+        label="Recommended Supplier Options (Top N)",
+        interactive=False,
+        wrap=True
+    )
+
+    quote_text = gr.Textbox(label="Generated Quote Draft", lines=14)
+
+    with gr.Row():
+        quote_file = gr.File(label="Download Quote Draft (.txt)")
+        supplier_csv = gr.File(label="Download Supplier DB (.csv)")
+
+    with gr.Accordion("⚙️ Admin: Regenerate Dummy Supplier DB", open=False):
+        regen_btn = gr.Button("Regenerate DB (50 suppliers)")
+        db_preview = gr.Dataframe(label="DB Preview (Top 10)", interactive=False)
+        db_file = gr.File(label="Download Fresh DB (.csv)")
+
+    run_btn.click(
+        fn=run_agent,
+        inputs=[inquiry, base_margin, pricing_mode, top_n, company_name, customer_name, valid_days],
+        outputs=[parsed_req, agent_steps, market_info, offers_table, quote_text, quote_file, supplier_csv],
+    )
+
+    regen_btn.click(
+        fn=regenerate_db,
+        inputs=[],
+        outputs=[db_preview, db_file],
     )
 
-st.caption("Next upgrade: real LLM extraction + auto supplier outreach + PDF quote + CRM integration once Zoho usage is confirmed.")
+if __name__ == "__main__":
+    # Hugging Face uses 7860 by default; this makes it explicit and stable.
+    demo.launch(server_name="0.0.0.0", server_port=7860)