Update app.py

app.py

@@ -1,70 +1,155 @@
-import gradio as gr
-from huggingface_hub import InferenceClient
+# app.py (日本語化・最終改善版)
 
+import os
+import json
+import uuid
+from PIL import Image
+from dotenv import load_dotenv
+from flask import Flask, request, jsonify, send_from_directory
+from flask_cors import CORS
+import google.generativeai as genai
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
+# --- 1. 初期設定 ---
+load_dotenv()
 
-    messages = [{"role": "system", "content": system_message}]
+# 定数とフォルダ設定
+PERSONALITY_FILE = "personality.json"
+SESSIONS_FILE = "chat_sessions.json"
+UPLOADS_FOLDER = "uploads" # 画像を保存するフォルダ
+DEFAULT_PERSONALITY = """
+あなたは消防署用のAI、グラフを上手くわかりやすい説明が得意です。以下の制約条件と入力文をもとに、適切な回答を出力してください。(AI CoT)
+「ステップごとに考えてみましょう。」）\n「あなたの推論を詳しく説明してください。」\n「問題をステップごとに分解してください。」\n「計算を示してください。」\n「よく考えてから答えを出してください。」
+"""
 
-    messages.extend(history)
+# uploadsフォルダが存在しない場合は作成
+if not os.path.exists(UPLOADS_FOLDER):
+    os.makedirs(UPLOADS_FOLDER)
 
-    messages.append({"role": "user", "content": message})
+app = Flask(__name__, static_folder='static', static_url_path='')
+CORS(app)
 
-    response = ""
+# --- 2. Google Gemini APIの設定 ---
+try:
+    api_key = "AIzaSyA0IZW7yEAdZ20eJ5urvVxHYGotzl1qO5c"
+    if not api_key: raise ValueError("環境変数 'GEMINI_API_KEY' が設定されていません。")
+    genai.configure(api_key=api_key)
+    print("✅ Gemini APIの設定が完了しました。")
+except Exception as e:
+    print(f"❌ Gemini APIの初期化に失敗: {e}")
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
+# --- 3. データ永続化ヘルパー関数 ---
+def load_data(filepath, default_data):
+    if not os.path.exists(filepath): return default_data
+    try:
+        with open(filepath, 'r', encoding='utf-8') as f: return json.load(f)
+    except (IOError, json.JSONDecodeError): return default_data
 
-        response += token
-        yield response
+def save_data(filepath, data):
+    with open(filepath, 'w', encoding='utf-8') as f: json.dump(data, f, ensure_ascii=False, indent=2)
 
+# --- 4. グローバル変数の読み込み ---
+system_instruction = load_data(PERSONALITY_FILE, {"personality": DEFAULT_PERSONALITY})["personality"]
+chat_sessions_data = load_data(SESSIONS_FILE, {})
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    type="messages",
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
-
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
-
-
-if __name__ == "__main__":
-    demo.launch()
+# =================================
+#  APIエンドポイント
+# =================================
+
+# --- 4.1. アップロードされたファイルを提供するエンドポイント ---
+@app.route('/uploads/<path:filename>')
+def uploaded_file(filename):
+    """uploadsフォルダから画像ファイルを配信する"""
+    return send_from_directory(UPLOADS_FOLDER, filename)
+
+# --- 4.2. 会話セッション管理API ---
+@app.route('/api/sessions', methods=['GET'])
+def get_sessions():
+    session_list = [{"id": session_id, **data} for session_id, data in chat_sessions_data.items()]
+    session_list.sort(key=lambda x: (x.get('pinned', False), x.get('id', '')), reverse=True)
+    return jsonify(session_list)
+
+@app.route('/api/session/<session_id>', methods=['GET'])
+def get_session_history(session_id): return jsonify(chat_sessions_data.get(session_id, {}))
+
+@app.route('/api/session/<session_id>', methods=['DELETE'])
+def delete_session(session_id):
+    if session_id in chat_sessions_data:
+        del chat_sessions_data[session_id]; save_data(SESSIONS_FILE, chat_sessions_data)
+        return jsonify({"status": "success"})
+    return jsonify({"error": "セッションが見つかりません。"}), 404
+
+@app.route('/api/session/<session_id>/rename', methods=['POST'])
+def rename_session(session_id):
+    if session_id in chat_sessions_data:
+        new_title = request.get_json().get('title', '').strip()
+        if new_title:
+            chat_sessions_data[session_id]['title'] = new_title; save_data(SESSIONS_FILE, chat_sessions_data)
+            return jsonify({"status": "success"})
+        return jsonify({"error": "タイトルが空です。"}), 400
+    return jsonify({"error": "セッションが見つかりません。"}), 404
+
+@app.route('/api/session/<session_id>/pin', methods=['POST'])
+def pin_session(session_id):
+    if session_id in chat_sessions_data:
+        is_pinned = chat_sessions_data[session_id].get('pinned', False)
+        chat_sessions_data[session_id]['pinned'] = not is_pinned; save_data(SESSIONS_FILE, chat_sessions_data)
+        return jsonify({"status": "success", "pinned": not is_pinned})
+    return jsonify({"error": "セッションが見つかりません。"}), 404
+
+# --- 4.3. チャットと設定のAPI ---
+@app.route('/api/set_personality', methods=['POST'])
+def set_personality():
+    global system_instruction
+    system_instruction = request.get_json().get('personality', '').strip() or DEFAULT_PERSONALITY
+    save_data(PERSONALITY_FILE, {"personality": system_instruction})
+    return jsonify({"status": "success", "message": f"人格を設定しました。"})
+
+@app.route('/api/chat', methods=['POST'])
+def handle_chat():
+    session_id = request.form.get('session_id')
+    user_input = request.form.get('message', '').strip()
+    file = request.files.get('file')
+
+    if not session_id or (not user_input and not file): return jsonify({"error": "必須情報が不足しています。"}), 400
+
+    try:
+        model = genai.GenerativeModel(model_name="gemini-2.5-pro", system_instruction=system_instruction)
+        
+        session_data = chat_sessions_data.get(session_id, {"title": "", "history": [], "pinned": False})
+        past_history = session_data.get("history", [])
+        
+        user_parts_for_api = [part for part in ([user_input] if user_input else []) + ([Image.open(file.stream)] if file else [])]
+        
+        response = model.generate_content(past_history + [{'role': 'user', 'parts': user_parts_for_api}])
+        
+        user_parts_for_history = [user_input] if user_input else []
+        if file:
+            ext = os.path.splitext(file.filename)[1]
+            unique_filename = f"{uuid.uuid4()}{ext}"
+            filepath = os.path.join(UPLOADS_FOLDER, unique_filename)
+            file.seek(0)
+            file.save(filepath)
+            user_parts_for_history.append(f"/uploads/{unique_filename}")
+
+        new_history = past_history + [
+            {'role': 'user', 'parts': user_parts_for_history},
+            {'role': 'model', 'parts': [response.text]}
+        ]
+        
+        session_data["history"] = new_history
+        if not session_data.get("title"):
+             session_data["title"] = user_input[:40] or (file.filename if file else "新しいチャット")
+        
+        chat_sessions_data[session_id] = session_data
+        save_data(SESSIONS_FILE, chat_sessions_data)
+
+        return jsonify({"reply": response.text, "title": session_data["title"]})
+
+    except Exception as e:
+        print(f"❌ APIエラー: {e}")
+        return jsonify({"error": "APIとの通信中、またはファイル処理中にエラーが発生しました。"}), 500
+
+@app.route('/')
+def serve_index(): return send_from_directory(app.static_folder, 'index.html')
+
+if __name__ == '__main__': app.run(host='0.0.0.0', port=5000, debug=True)