Upload app.py

app.py

@@ -0,0 +1,103 @@
+import os, io, base64, json, tempfile, pathlib
+import pandas as pd
+import gradio as gr
+from openai import OpenAI
+
+# --- Configuration ----------------------------------------------------------
+OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
+if not OPENAI_API_KEY:
+    raise RuntimeError("Please set the OPENAI_API_KEY environment variable.")
+client = OpenAI(api_key=OPENAI_API_KEY)
+
+MODEL = "gpt-4o-mini"          # works fine; change to "gpt-4o" if you have access
+MAX_TOKENS = 1024              # plenty for JSON output
+
+SYSTEM_PROMPT = """
+你是一個零售標價解析助手。請依照以下 JSON Schema 輸出結果：
+[
+  {
+    "品名": string,
+    "原價": string | null,
+    "促銷價": string | null,
+    "總重": string | null,
+    "總量": string | null,
+    "條碼號": string | null,
+    "物品編號": string | null
+  },
+  ...
+]
+⚠️ 只回傳 JSON，不要講述。
+若圖片中有多個商品，請依序列出。
+"""
+
+# --- Core logic -------------------------------------------------------------
+def encode_image_to_data_url(img_path: str) -> str:
+    mime = "image/" + pathlib.Path(img_path).suffix.lstrip(".").lower()
+    with open(img_path, "rb") as f:
+        b64 = base64.b64encode(f.read()).decode()
+    return f"data:{mime};base64,{b64}"
+
+def call_gpt4o(image_paths):
+    messages = [
+        {"role": "system", "content": SYSTEM_PROMPT}
+    ]
+    for p in image_paths:
+        messages.append(
+            {
+                "role": "user",
+                "content": [
+                    {
+                        "type": "image_url",
+                        "image_url": {"url": encode_image_to_data_url(p)}
+                    }
+                ],
+            }
+        )
+    resp = client.chat.completions.create(
+        model=MODEL,
+        messages=messages,
+        max_tokens=MAX_TOKENS,
+        temperature=0.0,
+    )
+    # Expect pure JSON
+    text = resp.choices[0].message.content.strip()
+    try:
+        return json.loads(text)
+    except Exception as e:
+        raise ValueError(f"模型輸出不是有效 JSON：\n{text}\n{e}")
+
+def process(images):
+    """
+    images: list[TemporaryUploadedFile]
+    Returns (json_str, xlsx_file_path)
+    """
+    paths = [img.name for img in images]  # TemporaryUploadedFile has .name path
+    all_items = call_gpt4o(paths)
+
+    # 1) JSON pretty-print
+    json_str = json.dumps(all_items, ensure_ascii=False, indent=2)
+
+    # 2) Excel
+    df = pd.DataFrame(all_items)
+    bio = io.BytesIO()
+    df.to_excel(bio, index=False, engine="openpyxl")
+    bio.seek(0)
+    # Save to a real temp file so Gradio can return it
+    tmp_xlsx = tempfile.NamedTemporaryFile(delete=False, suffix=".xlsx")
+    tmp_xlsx.write(bio.read())
+    tmp_xlsx.flush()
+    return json_str, tmp_xlsx.name
+
+# --- Gradio UI --------------------------------------------------------------
+with gr.Blocks(title="Price-Tag Parser") as demo:
+    gr.Markdown("## 🏷️ 零售標價解析\n上傳一張或多張標價照片 → 取得 JSON 與 Excel")
+    with gr.Row():
+        inp = gr.Files(label="上傳圖片 (可多選)", file_types=["image"])
+    btn = gr.Button("開始解析 🪄")
+    out_json = gr.JSON(label="辨識結果 (JSON)")
+    out_file = gr.File(label="下載 Excel", file_types=[".xlsx"])
+
+    btn.click(process, inputs=inp, outputs=[out_json, out_file])
+
+if __name__ == "__main__":
+    demo.launch()