# streamlit_app.py
import streamlit as st

# =========================
# Model Class (predict fonksiyonu)
# =========================
class Model:
    def __init__(self):
        pass

    def predict(self, prompt: str) -> str:
        svg = f"""<svg width="200" height="200" xmlns="http://www.w3.org/2000/svg">
<rect x="10" y="10" width="180" height="180" fill="lightblue" stroke="black" stroke-width="2"/>
<text x="20" y="100" font-family="Arial" font-size="14">{prompt[:50]}</text>
</svg>"""
        return svg

# =========================
# Load model (optional: if using pickle)
# =========================
# Eğer pickle kullanacaksan Model sınıfı tanımlı olmalı
import pickle
from pathlib import Path

model_path = Path("src/svg_model.pkl")

# Eğer pickle ile kaydedildi ve aynı sınıf tanımı var ise:
with open(model_path, "rb") as f:
    model = pickle.load(f)

# Alternatif olarak direkt Model() kullanabilirsin
# model = Model()

# =========================
# Streamlit UI
# =========================
st.title("Text to SVG Generator")

prompt = st.text_input("Enter a description:")

if st.button("Generate SVG"):
    svg_code = model.predict(prompt)
    st.components.v1.html(svg_code, height=250, scrolling=True)
    
    st.download_button(
        label="Download SVG",
        data=svg_code,
        file_name="generated_image.svg",
        mime="image/svg+xml"
    )