import gradio as gr
import pandas as pd

# Load the free sample directly from the public Hugging Face dataset
DATA_URL = "hf://datasets/Ichlibitiche/roasterdb-specialty-coffee-sample/roasterdb_sample.csv"
df = pd.read_csv(DATA_URL)

DISPLAY_COLS = {
    "source_roaster": "Roaster",
    "title": "Coffee",
    "origin_country": "Origin",
    "process_method": "Process",
    "roast_level": "Roast",
    "weight_grams": "Weight (g)",
    "price_value": "Price (USD)",
    "tasting_notes_sca_nodes": "SCA Flavor Notes",
    "source_url": "Source URL",
}

def choices(col):
    vals = df[col].dropna().astype(str)
    vals = sorted(v for v in vals.unique() if v.strip() and v != "Unknown")
    return ["All"] + vals

PRICE_MAX = float(df["price_value"].max())

def explore(roaster, country, process, roast, flavor, max_price):
    d = df
    if roaster != "All":
        d = d[d["source_roaster"] == roaster]
    if country != "All":
        d = d[d["origin_country"] == country]
    if process != "All":
        d = d[d["process_method"] == process]
    if roast != "All":
        d = d[d["roast_level"] == roast]
    if flavor and flavor.strip():
        d = d[d["tasting_notes_sca_nodes"].fillna("").str.contains(flavor.strip(), case=False, regex=False)]
    if max_price < PRICE_MAX:
        d = d[d["price_value"].fillna(PRICE_MAX + 1) <= max_price]

    summary = (
        f"**{len(d)} coffees** from **{d['source_roaster'].nunique()} roasters** match — "
        f"out of {len(df)} records in the free sample. "
        f"The full RoasterDB has **8,000+ products from 280+ roasters**."
    )
    table = d[list(DISPLAY_COLS)].rename(columns=DISPLAY_COLS).sort_values("Price (USD)")
    return summary, table

with gr.Blocks(title="RoasterDB Sample Explorer") as demo:
    gr.Markdown("# ☕ RoasterDB: Specialty Coffee Sample Explorer")
    gr.Markdown(f"""Explore **{len(df)} verified specialty coffees from {df['source_roaster'].nunique()} artisan roasters** — tasting notes normalized to the **SCA Flavor Wheel**, with a verifiable source URL on every record.

---
### 🌐 This is the free sample. The full RoasterDB has 8,000+ products from 280+ roasters.
* **🔗 Official Portal (full dataset, $99 snapshot):** [roasterdb.net](https://specialty-coffee-roasterdb.pages.dev)
* **🤗 Free Sample Dataset (Download CSV):** [Ichlibitiche/roasterdb-specialty-coffee-sample](https://huggingface.co/datasets/Ichlibitiche/roasterdb-specialty-coffee-sample)
* **🏆 Kaggle Dataset:** [RoasterDB Specialty Coffee Sample](https://www.kaggle.com/datasets/ahtiticheamine/roasterdb-specialty-coffee-sample)
* **🔄 Live Self-Serve Scraping:** [Specialty Coffee Roaster Scraper on Apify](https://apify.com/dataengineered/specialty-coffee-roaster-scraper)
---""")

    with gr.Row():
        roaster_dd = gr.Dropdown(choices=choices("source_roaster"), value="All", label="Roaster")
        country_dd = gr.Dropdown(choices=choices("origin_country"), value="All", label="Origin Country")
        process_dd = gr.Dropdown(choices=choices("process_method"), value="All", label="Process Method")
        roast_dd = gr.Dropdown(choices=choices("roast_level"), value="All", label="Roast Level")
    with gr.Row():
        flavor_tb = gr.Textbox(label="SCA Flavor Search", placeholder="e.g. Berry, Chocolate, Floral, Peach...")
        price_sl = gr.Slider(minimum=0, maximum=PRICE_MAX, value=PRICE_MAX, step=1, label="Max Price (USD)")

    btn = gr.Button("Explore Coffees", variant="primary")

    out_text = gr.Markdown()
    out_table = gr.Dataframe(label="Matching Coffees", wrap=True)

    inputs = [roaster_dd, country_dd, process_dd, roast_dd, flavor_tb, price_sl]
    btn.click(fn=explore, inputs=inputs, outputs=[out_text, out_table])
    demo.load(fn=explore, inputs=inputs, outputs=[out_text, out_table])

    gr.Markdown("""---
*Sample data © RoasterDB under CC BY-NC 4.0. Full dataset commercially licensed at [roasterdb.net](https://specialty-coffee-roasterdb.pages.dev) · contact RoasterDB@proton.me*""")

demo.launch()