text2image.py

import os
import logging
from io import BytesIO
from PIL import Image
import gradio as gr
from google import genai
from google.genai import types

# 設定 logging
logging.basicConfig(
    filename='app.log',
    level=logging.INFO,
    format='%(asctime)s - %(levelname)s - %(message)s'
)

# 初始化 Gemini API
GEMINI_API_KEY = os.environ.get("GEMINI_API_KEY")
client = genai.Client(api_key=GEMINI_API_KEY)

# 定義「圖解釋文」功能
def explain_image(image: Image.Image):
    # 將 PIL Image 轉成 Gemini 所需的格式
    buffered = BytesIO()
    image.save(buffered, format="PNG")
    image_bytes = buffered.getvalue()

    # 直接建立 types.Content
    contents = [
        types.Content(
            parts=[
                types.Part(
                    inline_data=types.Blob(
                        mime_type="image/png",
                        data=image_bytes
                    )
                ),
                types.Part(
                    text="請用繁體中文詳細說明這張圖片的內容。"
                )
            ]
        )
    ]

    # 呼叫 Gemini 模型
    response = client.generate_content(
        model="gemini-1.5-flash",
        contents=contents,
        generation_config=types.GenerationConfig(response_mime_type="text/plain")
    )

    # 回傳第一個回答
    explanation = response.candidates[0].content.parts[0].text
    logging.info("圖片說明成功取得。")
    return explanation

# Gradio 介面
with gr.Blocks() as demo:
    gr.Markdown("## 🧠 Gemini 圖片解釋器（圖 ➜ 文）")
    image_input = gr.Image(type="pil", label="上傳圖片")
    explain_button = gr.Button("解釋圖片")
    output_text = gr.Textbox(label="圖片說明", lines=5)

    explain_button.click(fn=explain_image, inputs=image_input, outputs=output_text)

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