app.py

import gradio as gr
from datasets import load_dataset
from rapidfuzz import process, fuzz

# ──────────────────────────────────────────────────────────
# 1) 데이터셋 로드 (스트리밍) ─ 메모리에 메타만 저장
# ──────────────────────────────────────────────────────────
ds = load_dataset(
    "nyuuzyou/clker-svg",
    split="train",
    streaming=True,  # .jsonl.zst → 자동 스트리밍 해제
)
records = []
for ex in ds:
    records.append(
        {
            "id": ex["id"],
            "title": ex["title"] or "",
            "tags": " ".join(ex["tags"] or []),
            "svg": ex["svg_content"],
            "url": ex["download_url"],
        }
    )

# ──────────────────────────────────────────────────────────
# 2) 검색 함수
# ──────────────────────────────────────────────────────────
def search_svg(query: str, top_k: int):
    if not query.strip():
        return "⚠️ 검색어를 입력하세요.", None
    
    # choices: index(int) ➜ title+tags 한 줄 문자열
    choices = {i: f"{r['title']} {r['tags']}" for i, r in enumerate(records)}
    
    # Rapidfuzz: (choice_text, score, key) 반환
    matched = process.extract(
        query,
        choices,
        scorer=fuzz.WRatio,
        limit=int(top_k),
    )
    
    html_snippets = []
    html_start = '<div class="gallery-grid">'
    html_end = '</div>'
    
    for _, score, idx in matched:  # idx 가 실제 list 인덱스
        r = records[idx]
        svg_html = (
            '<div class="gallery-item">'
            f'<div class="svg-container">{r["svg"]}</div>'
            f'<div class="item-details">'
            f'<h3>{r["title"]}</h3>'
            f'<div class="score">매칭 점수: {score}</div>'
            f'<div class="tags">{r["tags"]}</div>'
            f'<a href="{r["url"]}" target="_blank" class="download-link">원본 다운로드</a>'
            f'</div>'
            '</div>'
        )
        html_snippets.append(svg_html)
    
    if not html_snippets:
        return "검색 결과가 없습니다.", None
    
    return "", html_start + ''.join(html_snippets) + html_end

# ──────────────────────────────────────────────────────────
# 3) Gradio UI
# ──────────────────────────────────────────────────────────
TITLE = "🔍 Clker SVG"
DESCRIPTION = """
이 애플리케이션은 "nyuuzyou/clker-svg" 데이터셋을 사용하여 퍼블릭 도메인 SVG 클립아트를 빠르게 검색할 수 있습니다.
퍼지 매칭을 통해 제목과 태그에서 유사한 항목을 찾아 시각적 갤러리로 표시합니다.
"""
DISCORD_BADGE = """<p style="text-align:center; margin-top: -10px;"><a href="https://discord.gg/openfreeai" target="_blank"> <img src="https://img.shields.io/static/v1?label=Discord&message=Openfree%20AI&color=%230000ff&labelColor=%23800080&logo=discord&logoColor=white&style=for-the-badge" alt="badge"></a></p>"""

CSS = """
.gallery-grid {
    display: grid;
    grid-template-columns: repeat(auto-fill, minmax(250px, 1fr));
    gap: 20px;
    margin-top: 20px;
}
.gallery-item {
    border: 1px solid #ddd;
    border-radius: 8px;
    overflow: hidden;
    box-shadow: 0 2px 5px rgba(0,0,0,0.1);
    transition: transform 0.3s;
    background: white;
}
.gallery-item:hover {
    transform: translateY(-5px);
    box-shadow: 0 5px 15px rgba(0,0,0,0.1);
}
.svg-container {
    height: 180px;
    display: flex;
    align-items: center;
    justify-content: center;
    padding: 10px;
    background: #f9f9f9;
}
.svg-container svg {
    max-width: 100%;
    max-height: 160px;
}
.item-details {
    padding: 15px;
}
.item-details h3 {
    margin: 0 0 10px 0;
    font-size: 16px;
    color: #333;
    white-space: nowrap;
    overflow: hidden;
    text-overflow: ellipsis;
}
.score {
    font-size: 14px;
    color: #666;
    margin-bottom: 5px;
}
.tags {
    font-size: 12px;
    color: #888;
    margin-bottom: 10px;
    white-space: nowrap;
    overflow: hidden;
    text-overflow: ellipsis;
}
.download-link {
    display: inline-block;
    padding: 5px 10px;
    background: #4a90e2;
    color: white;
    text-decoration: none;
    border-radius: 4px;
    font-size: 12px;
}
.download-link:hover {
    background: #3a7bc8;
}
"""

with gr.Blocks(title=TITLE, css=CSS) as demo:
    gr.Markdown(f"# {TITLE}")
    gr.Markdown(DESCRIPTION)
    gr.HTML(DISCORD_BADGE)
    
    with gr.Row():
        with gr.Column(scale=4):
            query_box = gr.Textbox(
                label="검색어", 
                placeholder="예: cat, tree, house, computer, flower...",
                show_label=True
            )
        with gr.Column(scale=1):
            top_slider = gr.Slider(
                minimum=1, 
                maximum=50, 
                value=12, 
                step=3, 
                label="결과 개수"
            )
    
    with gr.Row():
        search_button = gr.Button("검색", variant="primary")
    
    warning_md = gr.Markdown()
    output_html = gr.HTML()
    
    search_button.click(
        fn=search_svg,
        inputs=[query_box, top_slider],
        outputs=[warning_md, output_html],
    )
    
    query_box.submit(
        fn=search_svg,
        inputs=[query_box, top_slider],
        outputs=[warning_md, output_html],
    )

if __name__ == "__main__":
    demo.launch()