app.py

import gradio as gr
import pandas as pd
import sys
sys.path.insert(0, "src")
from indic_faker import IndicFaker


LANGUAGES = {
    "🇮🇳 Hindi (हिन्दी)": "hi",
    "🇮🇳 Malayalam (മലയാളം)": "ml",
    "🇮🇳 Tamil (தமிழ்)": "ta",
    "🇮🇳 Telugu (తెలుగు)": "te",
    "🇮🇳 Bengali (বাংলা)": "bn",
    "🇮🇳 Kannada (ಕನ್ನಡ)": "kn",
    "🇮🇳 Gujarati (ગુજરાતી)": "gu",
    "🇮🇳 Marathi (मराठी)": "mr",
}

FIELD_OPTIONS = [
    "name", "name_native", "gender", "dob", "age", "language",
    "aadhaar", "pan", "phone", "email", "address", "city",
    "state", "pincode", "bank_account", "upi_id",
    "employer", "job_title", "salary", "college", "degree",
]

DEFAULT_FIELDS = [
    "name", "name_native", "gender", "age",
    "phone", "email", "city", "degree", "job_title", "salary",
]


def generate_profiles(language, num_rows, selected_fields):
    lang_code = LANGUAGES.get(language, "hi")
    fields = selected_fields if selected_fields else DEFAULT_FIELDS
    num = max(1, min(int(num_rows), 100))
    fake = IndicFaker(language=lang_code)
    df = fake.to_dataframe(num, fields=fields)
    return df


def generate_single(language):
    lang_code = LANGUAGES.get(language, "hi")
    fake = IndicFaker(language=lang_code)
    p = fake.profile()

    # Build a styled card
    card = f"""
<div class="profile-card">
    <div class="profile-header">
        <div class="avatar">{'👨' if p['gender'] == 'male' else '👩'}</div>
        <div class="header-info">
            <h2>{p['name']}</h2>
            <h3>{p['name_native']}</h3>
            <span class="badge badge-lang">{p['language'].upper()}</span>
            <span class="badge badge-gender">{p['gender'].title()}</span>
            <span class="badge badge-age">{p['age']} years</span>
        </div>
    </div>

    <div class="profile-grid">
        <div class="info-section">
            <h4>🆔 Identity</h4>
            <div class="info-row"><span class="label">Aadhaar</span><span class="value mono">{p['aadhaar']}</span></div>
            <div class="info-row"><span class="label">PAN</span><span class="value mono">{p['pan']}</span></div>
            <div class="info-row"><span class="label">DOB</span><span class="value">{p['dob']}</span></div>
        </div>

        <div class="info-section">
            <h4>📞 Contact</h4>
            <div class="info-row"><span class="label">Phone</span><span class="value mono">{p['phone']}</span></div>
            <div class="info-row"><span class="label">Email</span><span class="value">{p['email']}</span></div>
            <div class="info-row"><span class="label">UPI</span><span class="value mono">{p['upi_id']}</span></div>
        </div>

        <div class="info-section">
            <h4>💼 Career</h4>
            <div class="info-row"><span class="label">Title</span><span class="value">{p['job_title']}</span></div>
            <div class="info-row"><span class="label">Employer</span><span class="value">{p['employer']}</span></div>
            <div class="info-row"><span class="label">Salary</span><span class="value salary">{p['salary']}</span></div>
        </div>

        <div class="info-section">
            <h4>🎓 Education</h4>
            <div class="info-row"><span class="label">Degree</span><span class="value">{p['degree']}</span></div>
            <div class="info-row"><span class="label">College</span><span class="value">{p['college']}</span></div>
        </div>

        <div class="info-section full-width">
            <h4>📍 Address</h4>
            <div class="info-row"><span class="label">Address</span><span class="value">{p['address']}</span></div>
            <div class="info-row"><span class="label">City</span><span class="value">{p['city']}</span></div>
            <div class="info-row"><span class="label">State</span><span class="value">{p['state']}</span></div>
            <div class="info-row"><span class="label">Pincode</span><span class="value mono">{p['pincode']}</span></div>
        </div>

        <div class="info-section full-width">
            <h4>🏦 Banking</h4>
            <div class="info-row"><span class="label">Account</span><span class="value mono">{p['bank_account']}</span></div>
            <div class="info-row"><span class="label">UPI ID</span><span class="value mono">{p['upi_id']}</span></div>
        </div>
    </div>
</div>
"""
    return card


css = """
/* === Global === */
@import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700;800&family=JetBrains+Mono:wght@400;500&display=swap');

.gradio-container {
    max-width: 1200px !important;
    margin: 0 auto !important;
    font-family: 'Inter', -apple-system, sans-serif !important;
    background: linear-gradient(160deg, #0a0a1a 0%, #1a1a2e 40%, #16213e 100%) !important;
    min-height: 100vh;
}

/* === Hero Header === */
.hero-section {
    text-align: center;
    padding: 40px 20px 30px;
    background: linear-gradient(135deg, rgba(255,107,53,0.1) 0%, rgba(99,102,241,0.1) 100%);
    border-radius: 20px;
    border: 1px solid rgba(255,255,255,0.06);
    margin-bottom: 24px;
    position: relative;
    overflow: hidden;
}

.hero-section::before {
    content: '';
    position: absolute;
    top: -50%;
    left: -50%;
    width: 200%;
    height: 200%;
    background: radial-gradient(circle at 30% 50%, rgba(255,107,53,0.06) 0%, transparent 50%),
                radial-gradient(circle at 70% 50%, rgba(99,102,241,0.06) 0%, transparent 50%);
    animation: pulse 8s ease-in-out infinite alternate;
}

@keyframes pulse {
    0% { transform: translate(0, 0); }
    100% { transform: translate(-5%, 5%); }
}

.hero-title {
    font-size: 2.4em;
    font-weight: 800;
    letter-spacing: -0.02em;
    margin: 0 0 8px;
    position: relative;
    background: linear-gradient(135deg, #FF6B35 0%, #F7931E 50%, #6366f1 100%);
    -webkit-background-clip: text;
    -webkit-text-fill-color: transparent;
    background-clip: text;
}

.hero-subtitle {
    font-size: 1.05em;
    color: rgba(255,255,255,0.65);
    font-weight: 400;
    margin: 0 0 16px;
    position: relative;
}

.hero-badges {
    display: flex;
    gap: 8px;
    justify-content: center;
    position: relative;
    flex-wrap: wrap;
}

.hero-badges a img {
    transition: transform 0.2s;
}

.hero-badges a img:hover {
    transform: translateY(-2px);
}

/* === Stats Bar === */
.stats-bar {
    display: flex;
    justify-content: center;
    gap: 40px;
    padding: 16px 0;
    margin-bottom: 20px;
}

.stat-item {
    text-align: center;
}

.stat-number {
    font-size: 1.8em;
    font-weight: 800;
    background: linear-gradient(135deg, #FF6B35, #F7931E);
    -webkit-background-clip: text;
    -webkit-text-fill-color: transparent;
    background-clip: text;
}

.stat-label {
    font-size: 0.8em;
    color: rgba(255,255,255,0.45);
    text-transform: uppercase;
    letter-spacing: 0.1em;
    font-weight: 500;
}

/* === Tabs === */
.tab-nav button {
    font-family: 'Inter', sans-serif !important;
    font-weight: 600 !important;
    font-size: 0.95em !important;
    padding: 12px 24px !important;
    border-radius: 12px 12px 0 0 !important;
    transition: all 0.3s !important;
}

.tab-nav button.selected {
    background: linear-gradient(135deg, #FF6B35, #F7931E) !important;
    color: white !important;
    border: none !important;
}

/* === Form Controls === */
.gr-input, .gr-dropdown, .gr-slider {
    border-radius: 12px !important;
}

/* === Generate Button === */
.generate-btn {
    background: linear-gradient(135deg, #FF6B35 0%, #F7931E 100%) !important;
    border: none !important;
    border-radius: 14px !important;
    padding: 14px 32px !important;
    font-size: 1.05em !important;
    font-weight: 700 !important;
    letter-spacing: 0.02em !important;
    transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1) !important;
    box-shadow: 0 4px 20px rgba(255,107,53,0.3) !important;
    text-transform: none !important;
}

.generate-btn:hover {
    transform: translateY(-2px) !important;
    box-shadow: 0 8px 30px rgba(255,107,53,0.45) !important;
}

.generate-btn:active {
    transform: translateY(0) !important;
}

/* === Profile Card === */
.profile-card {
    background: linear-gradient(145deg, rgba(30,30,50,0.9) 0%, rgba(20,20,40,0.95) 100%);
    border: 1px solid rgba(255,255,255,0.08);
    border-radius: 20px;
    padding: 0;
    overflow: hidden;
    backdrop-filter: blur(20px);
}

.profile-header {
    display: flex;
    align-items: center;
    gap: 20px;
    padding: 28px 32px;
    background: linear-gradient(135deg, rgba(255,107,53,0.15) 0%, rgba(99,102,241,0.15) 100%);
    border-bottom: 1px solid rgba(255,255,255,0.06);
}

.avatar {
    font-size: 3em;
    width: 72px;
    height: 72px;
    display: flex;
    align-items: center;
    justify-content: center;
    background: rgba(255,255,255,0.05);
    border-radius: 18px;
    border: 2px solid rgba(255,255,255,0.1);
}

.header-info h2 {
    margin: 0;
    font-size: 1.5em;
    font-weight: 700;
    color: #fff;
    letter-spacing: -0.01em;
}

.header-info h3 {
    margin: 4px 0 10px;
    font-size: 1.1em;
    font-weight: 400;
    color: rgba(255,255,255,0.55);
}

.badge {
    display: inline-block;
    padding: 3px 10px;
    border-radius: 20px;
    font-size: 0.75em;
    font-weight: 600;
    letter-spacing: 0.03em;
    margin-right: 6px;
}

.badge-lang {
    background: rgba(99,102,241,0.2);
    color: #a5b4fc;
    border: 1px solid rgba(99,102,241,0.3);
}

.badge-gender {
    background: rgba(255,107,53,0.15);
    color: #ffb088;
    border: 1px solid rgba(255,107,53,0.3);
}

.badge-age {
    background: rgba(34,197,94,0.15);
    color: #86efac;
    border: 1px solid rgba(34,197,94,0.3);
}

.profile-grid {
    display: grid;
    grid-template-columns: 1fr 1fr;
    gap: 0;
    padding: 0;
}

.info-section {
    padding: 20px 28px;
    border-bottom: 1px solid rgba(255,255,255,0.04);
    border-right: 1px solid rgba(255,255,255,0.04);
}

.info-section:nth-child(even) {
    border-right: none;
}

.info-section.full-width {
    grid-column: 1 / -1;
    border-right: none;
}

.info-section h4 {
    margin: 0 0 12px;
    font-size: 0.85em;
    font-weight: 600;
    color: rgba(255,255,255,0.4);
    text-transform: uppercase;
    letter-spacing: 0.08em;
}

.info-row {
    display: flex;
    justify-content: space-between;
    align-items: center;
    padding: 6px 0;
}

.info-row .label {
    font-size: 0.85em;
    color: rgba(255,255,255,0.45);
    font-weight: 500;
}

.info-row .value {
    font-size: 0.9em;
    color: rgba(255,255,255,0.9);
    font-weight: 500;
    text-align: right;
    max-width: 65%;
}

.info-row .value.mono {
    font-family: 'JetBrains Mono', monospace;
    font-size: 0.85em;
    color: #a5b4fc;
}

.info-row .value.salary {
    color: #4ade80;
    font-weight: 700;
}

/* === Table === */
.gr-dataframe {
    border-radius: 16px !important;
    overflow: hidden !important;
}

/* === Code Block === */
.code-section {
    background: rgba(15,15,30,0.8);
    border: 1px solid rgba(255,255,255,0.06);
    border-radius: 16px;
    padding: 28px;
}

.code-section pre {
    background: rgba(0,0,0,0.3) !important;
    border-radius: 12px !important;
    border: 1px solid rgba(255,255,255,0.06) !important;
}

.code-section code {
    font-family: 'JetBrains Mono', monospace !important;
    font-size: 0.88em !important;
}

/* === Use Cases Grid === */
.use-cases-grid {
    display: grid;
    grid-template-columns: repeat(auto-fit, minmax(280px, 1fr));
    gap: 16px;
    margin-top: 20px;
}

.use-case-card {
    background: linear-gradient(145deg, rgba(30,30,50,0.6) 0%, rgba(20,20,40,0.8) 100%);
    border: 1px solid rgba(255,255,255,0.06);
    border-radius: 14px;
    padding: 20px;
    transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1);
}

.use-case-card:hover {
    border-color: rgba(255,107,53,0.3);
    transform: translateY(-2px);
    box-shadow: 0 8px 24px rgba(0,0,0,0.3);
}

.use-case-card .icon {
    font-size: 1.8em;
    margin-bottom: 10px;
}

.use-case-card h5 {
    margin: 0 0 6px;
    font-size: 1em;
    font-weight: 600;
    color: #fff;
}

.use-case-card p {
    margin: 0;
    font-size: 0.85em;
    color: rgba(255,255,255,0.5);
    line-height: 1.5;
}

/* === Footer === */
.footer-section {
    text-align: center;
    padding: 24px;
    margin-top: 20px;
    color: rgba(255,255,255,0.35);
    font-size: 0.85em;
    border-top: 1px solid rgba(255,255,255,0.05);
}

.footer-section a {
    color: #FF6B35;
    text-decoration: none;
}

/* === Hide default footer === */
footer { display: none !important; }
"""

with gr.Blocks(title="Indic Faker — Indian Synthetic Data Generator") as demo:

    # === Hero ===
    gr.HTML("""
    <div class="hero-section">
        <div class="hero-title">🇮🇳 Indic Faker</div>
        <div class="hero-subtitle">
            Generate realistic Indian synthetic data in <strong>8 languages</strong> with native script support.<br>
            Aadhaar · PAN · UPI · Names · Addresses · Salaries — all culturally accurate.
        </div>
        <div class="hero-badges">
            <a href="https://pypi.org/project/indic-faker/" target="_blank">
                <img src="https://img.shields.io/pypi/v/indic-faker?style=for-the-badge&color=FF6B35" alt="PyPI">
            </a>
            <a href="https://github.com/adwaith-0/indic-faker" target="_blank">
                <img src="https://img.shields.io/github/stars/adwaith-0/indic-faker?style=for-the-badge&logo=github&color=6366f1" alt="GitHub">
            </a>
            <a href="https://huggingface.co/datasets/adwaith06/indic-synthetic-profiles" target="_blank">
                <img src="https://img.shields.io/badge/🤗_Dataset-10K_Rows-blue?style=for-the-badge" alt="HuggingFace">
            </a>
        </div>
    </div>
    """)

    gr.HTML("""
    <div class="stats-bar">
        <div class="stat-item">
            <div class="stat-number">8</div>
            <div class="stat-label">Languages</div>
        </div>
        <div class="stat-item">
            <div class="stat-number">21+</div>
            <div class="stat-label">Data Fields</div>
        </div>
        <div class="stat-item">
            <div class="stat-number">50+</div>
            <div class="stat-label">Providers</div>
        </div>
        <div class="stat-item">
            <div class="stat-number">10K</div>
            <div class="stat-label">HF Dataset</div>
        </div>
    </div>
    """)

    with gr.Tabs():

        # === Tab 1: Batch Generate ===
        with gr.TabItem("📊 Batch Generate"):
            with gr.Row(equal_height=True):
                with gr.Column(scale=1, min_width=300):
                    lang_dropdown = gr.Dropdown(
                        choices=list(LANGUAGES.keys()),
                        value="🇮🇳 Hindi (हिन्दी)",
                        label="🌐 Language",
                        info="Choose a language for generated data",
                    )
                    num_slider = gr.Slider(
                        minimum=1, maximum=100, value=10, step=1,
                        label="📏 Number of Profiles",
                        info="How many profiles to generate",
                    )
                    fields_select = gr.CheckboxGroup(
                        choices=FIELD_OPTIONS,
                        value=DEFAULT_FIELDS,
                        label="📋 Fields to Include",
                        info="Select the fields you want in your data",
                    )
                    generate_btn = gr.Button(
                        "🚀 Generate Data",
                        variant="primary",
                        size="lg",
                        elem_classes=["generate-btn"],
                    )

            output_table = gr.Dataframe(
                label="Generated Profiles",
                wrap=True,
                interactive=False,
                max_height=500,
            )

            generate_btn.click(
                fn=generate_profiles,
                inputs=[lang_dropdown, num_slider, fields_select],
                outputs=output_table,
            )

        # === Tab 2: Single Profile ===
        with gr.TabItem("👤 Single Profile"):
            with gr.Row():
                with gr.Column(scale=1, min_width=250):
                    single_lang = gr.Dropdown(
                        choices=list(LANGUAGES.keys()),
                        value="🇮🇳 Kannada (ಕನ್ನಡ)",
                        label="🌐 Language",
                    )
                    single_btn = gr.Button(
                        "🎲 Generate Profile",
                        variant="primary",
                        size="lg",
                        elem_classes=["generate-btn"],
                    )
                    gr.HTML("""
                    <div style="margin-top:16px; padding:16px; background:rgba(255,255,255,0.03);
                                border-radius:12px; border:1px solid rgba(255,255,255,0.06);">
                        <div style="font-size:0.8em; color:rgba(255,255,255,0.4); font-weight:600;
                                    text-transform:uppercase; letter-spacing:0.08em; margin-bottom:8px;">
                            💡 Tip
                        </div>
                        <div style="font-size:0.85em; color:rgba(255,255,255,0.55); line-height:1.6;">
                            Each click generates a unique, realistic Indian identity with
                            culturally accurate data — names match their native script
                            transliterations.
                        </div>
                    </div>
                    """)

                with gr.Column(scale=2):
                    profile_output = gr.HTML(
                        value="<div style='text-align:center; padding:60px; color:rgba(255,255,255,0.3); font-size:1.1em;'>Click <strong>Generate Profile</strong> to create an identity →</div>",
                    )

            single_btn.click(
                fn=generate_single,
                inputs=[single_lang],
                outputs=profile_output,
            )

        # === Tab 3: Quick Start ===
        with gr.TabItem("💻 Quick Start"):
            gr.HTML("""<div class="code-section">""")
            gr.Markdown("""
### Installation
```bash
pip install indic-faker
```

### Usage
```python
from indic_faker import IndicFaker

fake = IndicFaker(language="ta")  # Tamil

# Single fields
print(fake.name())          # Rajesh Krishnan
print(fake.name("native"))  # ராஜேஷ் கிருஷ்ணன்
print(fake.aadhaar())       # 4532 8891 2234
print(fake.pan())           # BXYPK7234R
print(fake.upi_id())        # rajesh.k@okaxis

# Full profile with all 21+ fields
profile = fake.profile()

# Batch → DataFrame (great for ML)
df = fake.to_dataframe(1000)
df.to_csv("indian_data.csv")
```

### Load from HuggingFace
```python
from datasets import load_dataset
ds = load_dataset("adwaith06/indic-synthetic-profiles")
```
            """)
            gr.HTML("""</div>""")

            gr.HTML("""
            <div class="use-cases-grid">
                <div class="use-case-card">
                    <div class="icon">🔍</div>
                    <h5>Fraud Detection</h5>
                    <p>Train ML models on synthetic Indian transaction data with realistic Aadhaar, PAN, and bank details.</p>
                </div>
                <div class="use-case-card">
                    <div class="icon">🤖</div>
                    <h5>LLM Fine-Tuning</h5>
                    <p>Create Hindi, Tamil, Telugu instruction datasets with native script names and addresses.</p>
                </div>
                <div class="use-case-card">
                    <div class="icon">🏦</div>
                    <h5>KYC System Testing</h5>
                    <p>Generate realistic Aadhaar + PAN + bank account combos for fintech QA.</p>
                </div>
                <div class="use-case-card">
                    <div class="icon">📊</div>
                    <h5>Data Pipeline QA</h5>
                    <p>Fill staging databases with 10K+ realistic Indian records in seconds.</p>
                </div>
                <div class="use-case-card">
                    <div class="icon">🎓</div>
                    <h5>Teaching & Demos</h5>
                    <p>Demo data engineering and analytics with culturally relevant, privacy-safe data.</p>
                </div>
                <div class="use-case-card">
                    <div class="icon">🧪</div>
                    <h5>Unit Testing</h5>
                    <p>Seed test fixtures with varied Indian data — 8 languages, multiple formats.</p>
                </div>
            </div>
            """)

    gr.HTML("""
    <div class="footer-section">
        Built with ❤️ by <a href="https://github.com/adwaith-0" target="_blank">adwaith-0</a> ·
        <code>pip install indic-faker</code> ·
        <a href="https://github.com/adwaith-0/indic-faker" target="_blank">GitHub</a> ·
        <a href="https://huggingface.co/datasets/adwaith06/indic-synthetic-profiles" target="_blank">HuggingFace</a>
    </div>
    """)


if __name__ == "__main__":
    demo.launch(
        theme=gr.themes.Base(
            primary_hue=gr.themes.Color(
                c50="#fff7ed", c100="#ffedd5", c200="#fed7aa", c300="#fdba74",
                c400="#fb923c", c500="#f97316", c600="#ea580c", c700="#c2410c",
                c800="#9a3412", c900="#7c2d12", c950="#431407",
            ),
            secondary_hue="indigo",
            neutral_hue="slate",
            font=["Inter", "system-ui", "sans-serif"],
            font_mono=["JetBrains Mono", "monospace"],
        ).set(
            body_background_fill="linear-gradient(160deg, #0a0a1a 0%, #1a1a2e 40%, #16213e 100%)",
            body_text_color="rgba(255,255,255,0.85)",
            block_background_fill="rgba(20,20,40,0.6)",
            block_border_color="rgba(255,255,255,0.06)",
            block_label_text_color="rgba(255,255,255,0.6)",
            block_title_text_color="rgba(255,255,255,0.8)",
            input_background_fill="rgba(15,15,35,0.8)",
            input_border_color="rgba(255,255,255,0.1)",
            input_placeholder_color="rgba(255,255,255,0.3)",
            button_primary_background_fill="linear-gradient(135deg, #FF6B35, #F7931E)",
            button_primary_text_color="white",
            button_primary_border_color="transparent",
            checkbox_background_color="rgba(15,15,35,0.8)",
            checkbox_border_color="rgba(255,255,255,0.15)",
            checkbox_label_text_color="rgba(255,255,255,0.7)",
            shadow_drop="0 4px 24px rgba(0,0,0,0.3)",
            block_shadow="0 4px 24px rgba(0,0,0,0.2)",
            block_border_width="1px",
            block_radius="16px",
            input_radius="12px",
        ),
        css=css,
    )