{
  "nbformat": 4,
  "nbformat_minor": 5,
  "metadata": {},
  "cells": [
    {
      "id": "e88d538f",
      "cell_type": "markdown",
      "source": "# 🏠 Building a Gradio Contract Analysis App\nThis notebook builds a **contract metadata extractor** (heuristic) and wraps it in a **Gradio** UI.",
      "metadata": {}
    },
    {
      "id": "bff6f9f8",
      "cell_type": "code",
      "metadata": {},
      "execution_count": null,
      "source": "!pip -q install gradio dateparser",
      "outputs": []
    },
    {
      "id": "7bd73692",
      "cell_type": "markdown",
      "source": "## 1) Heuristic Extractors",
      "metadata": {}
    },
    {
      "id": "f7e2faf6",
      "cell_type": "code",
      "metadata": {},
      "execution_count": null,
      "source": "\nimport re, json\nfrom typing import Dict, Any, List, Optional\nfrom dateparser.search import search_dates\n\ndef _clean(s: str) -> str:\n    import re\n    return re.sub(r\"\\s+\", \" \", s).strip()\n\ndef extract_parties(text: str):\n    m = re.search(r\"\\bbetween\\s+(.+?)\\s+and\\s+(.+?)(?:[,.;\\n]| for\\b)\", text, re.I | re.S)\n    if m:\n        p1, p2 = _clean(m.group(1)), _clean(m.group(2))\n        p1 = re.sub(r\"^(?:the\\s+)?(landlord|lessor|owner)\\s*[:\\-]*\\s*\", \"\", p1, flags=re.I)\n        p2 = re.sub(r\"^(?:the\\s+)?(tenant|lessee|occupant)\\s*[:\\-]*\\s*\", \"\", p2, flags=re.I)\n        return [p1, p2]\n    return None\n\ndef extract_metadata(text: str) -> Dict[str, Any]:\n    # Add more extractors as needed; minimal demo\n    parties = extract_parties(text)\n    dates = search_dates(text) or []\n    start_date = None\n    if dates:\n        start_date = sorted([d for _, d in dates])[0].strftime(\"%Y-%m-%d\")\n    return {\"parties\": parties, \"start_date\": start_date}\n",
      "outputs": []
    },
    {
      "id": "ae18d154",
      "cell_type": "markdown",
      "source": "## 2) Gradio UI",
      "metadata": {}
    },
    {
      "id": "4bdf7c47",
      "cell_type": "code",
      "metadata": {},
      "execution_count": null,
      "source": "\nimport gradio as gr\n\nEXAMPLE = \"This Lease Agreement is made between Landlord X and Tenant Y for the premises at 123 Main St. Commencing on June 1, 2025.\"\n\ndef run_pipeline(t):\n    meta = extract_metadata(t or \"\")\n    return meta\n\nwith gr.Blocks() as demo:\n    gr.Markdown(\"# 🏠 Contract Analysis Demo\")\n    inp = gr.Textbox(value=EXAMPLE, label=\"Contract Text\", lines=8)\n    out = gr.JSON(label=\"Extracted Metadata\")\n    btn = gr.Button(\"Analyze\")\n    btn.click(run_pipeline, inputs=inp, outputs=out)\n\ndemo.launch(share=False)\n",
      "outputs": []
    },
    {
      "id": "f8a36080",
      "cell_type": "markdown",
      "source": "> Tip: move the Python into `app.py` for a Hugging Face Space and add a YAML config to your README.",
      "metadata": {}
    }
  ]
}