Update app.py

app.py

@@ -8,11 +8,11 @@ import requests
 import re
 import traceback
 
-# HuggingFace 관련 API 키 (스페이스 분석 용)
+# HuggingFace API key for space analysis
 HF_TOKEN = os.getenv("HF_TOKEN")
 hf_api = HfApi(token=HF_TOKEN)
 
-# Gemini 2.0 Flash Thinking 모델 관련 API 키 및 클라이언트 (LLM 용)
+# Gemini 2.0 Flash Thinking model API key and client (for LLM)
 GEMINI_API_KEY = os.getenv("GEMINI_API_KEY")
 genai.configure(api_key=GEMINI_API_KEY)
 model = genai.GenerativeModel("gemini-2.0-flash-thinking-exp-01-21")
@@ -41,7 +41,7 @@ def get_space_structure(space_id: str) -> Dict:
             path_parts = file.split('/')
             current = tree
             for i, part in enumerate(path_parts):
-                if i == len(path_parts) - 1:  # 파일
+                if i == len(path_parts) - 1:  # file
                     current["children"].append({"type": "file", "path": file, "name": part})
                 else:
                     found = False
@@ -78,55 +78,52 @@ def analyze_space(url: str, progress=gr.Progress()):
         if not re.match(r'^[\w.-]+/[\w.-]+$', space_id):
             raise ValueError(f"Invalid Space ID format: {space_id}")
 
-        progress(0.1, desc="📁 파일 구조 분석 중...")
+        progress(0.1, desc="Analyzing file structure...")
         tree_structure = get_space_structure(space_id)
         if "error" in tree_structure:
             raise ValueError(tree_structure["error"])
         tree_view = format_tree_structure(tree_structure)
 
-        progress(0.3, desc="📄 app.py 내용 가져오는 중...")
+        progress(0.3, desc="Fetching app.py content...")
         app_content = get_file_content(space_id, "app.py")
 
-        progress(0.5, desc="✏️ 코드 요약 중...")
+        progress(0.5, desc="Summarizing code...")
         summary = summarize_code(app_content)
 
-        progress(0.7, desc="🔍 코드 분석 중...")
+        progress(0.7, desc="Analyzing code...")
         analysis = analyze_code(app_content)
 
-        progress(0.9, desc="📚 사용법 설명 생성 중...")
+        progress(0.9, desc="Generating usage instructions...")
         usage = explain_usage(app_content)
 
         lines_for_app_py = adjust_lines_for_code(app_content)
-        progress(1.0, desc="✅ 완료")
+        progress(1.0, desc="Complete")
 
         return app_content, tree_view, tree_structure, space_id, summary, analysis, usage, lines_for_app_py
 
     except Exception as e:
         print(f"Error in analyze_space: {str(e)}")
         print(traceback.format_exc())
-        return f"오류가 발생했습니다: {str(e)}", "", None, "", "", "", "", 10
-
+        return f"An error occurred: {str(e)}", "", None, "", "", "", "", 10
 
 # --------------------------------------------------
-# Gemini 2.0 Flash Thinking 모델 (LLM) 함수들
+# Gemini 2.0 Flash Thinking model (LLM) functions
 # --------------------------------------------------
 from gradio import ChatMessage
 
 def format_chat_history(messages: List[ChatMessage]) -> List[Dict]:
     """
-    ChatMessage 목록을 Gemini 모델이 이해할 수 있는 형식으로 변환
-    (Thinking 메타데이터가 있는 메시지는 무시)
+    Convert a list of ChatMessages to a format that the Gemini model can understand.
+    (Skip messages with 'Thinking' metadata)
     """
     formatted = []
     for m in messages:
-        if hasattr(m, "metadata") and m.metadata:  # 'Thinking' 메시지는 무시
+        if hasattr(m, "metadata") and m.metadata:  # Skip 'Thinking' messages
             continue
         role = "assistant" if m.role == "assistant" else "user"
         formatted.append({"role": role, "parts": [m.content or ""]})
     return formatted
 
-import google.generativeai as genai
-
 def gemini_chat_completion(system_message: str, user_message: str, max_tokens: int = 200, temperature: float = 0.7) -> str:
     init_msgs = [
         ChatMessage(role="system", content=system_message),
@@ -144,60 +141,52 @@ def gemini_chat_completion(system_message: str, user_message: str, max_tokens: i
                 final += parts[0].text
         return final.strip()
     except Exception as e:
-        return f"LLM 호출 중 오류 발생: {str(e)}"
-
+        return f"Error calling LLM: {str(e)}"
 
 def summarize_code(app_content: str):
-    system_msg = "당신은 Python 코드를 분석하고 요약하는 AI 조수입니다. 주어진 코드를 3줄 이내로 간결하게 요약해주세요."
-    user_msg = f"다음 Python 코드를 3줄 이내로 요약해주세요:\n\n{app_content}"
+    system_msg = "You are an AI assistant that analyzes and summarizes Python code. Please summarize the provided code in no more than 3 lines."
+    user_msg = f"Please summarize the following Python code in no more than 3 lines:\n\n{app_content}"
     try:
         return gemini_chat_completion(system_msg, user_msg, max_tokens=200, temperature=0.7)
     except Exception as e:
-        return f"요약 생성 중 오류 발생: {str(e)}"
+        return f"Error generating summary: {str(e)}"
 
 def analyze_code(app_content: str):
     system_msg = (
-        "You are a deep thinking AI. You may use extremely long chains of thought to deeply consider the problem "
-        "and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. "
-        "You should enclose your thoughts and internal monologue inside tags, and then provide your solution or response to the problem. "
-        "당신은 Python 코드를 분석하는 AI 조수입니다. 주어진 코드를 분석하여 서비스의 효용성과 활용 측면에서 다음 항목에 대해 설명해주세요:\n"
-        "A. 배경 및 필요성\n"
-        "B. 기능적 효용성 및 가치\n"
-        "C. 특장점\n"
-        "D. 적용 대상 및 타겟\n"
-        "E. 기대효과\n"
-        "기존 및 유사 프로젝트와 비교하여 분석해주세요. Markdown 형식으로 출력하세요."
+        "You are an AI assistant that analyzes Python code. Please analyze the provided code in terms of its service utility and application with respect to the following aspects:\n"
+        "A. Background and Necessity\n"
+        "B. Functional Utility and Value\n"
+        "C. Key Features\n"
+        "D. Target Audience\n"
+        "E. Expected Impact\n"
+        "Please also compare with existing and similar projects. Output in Markdown format."
     )
-    user_msg = f"다음 Python 코드를 분석해주세요:\n\n{app_content}"
+    user_msg = f"Please analyze the following Python code:\n\n{app_content}"
     try:
         return gemini_chat_completion(system_msg, user_msg, max_tokens=1000, temperature=0.7)
     except Exception as e:
-        return f"분석 생성 중 오류 발생: {str(e)}"
+        return f"Error generating analysis: {str(e)}"
 
 def explain_usage(app_content: str):
     system_msg = (
-        "You are a deep thinking AI. You may use extremely long chains of thought to deeply consider the problem "
-        "and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. "
-        "You should enclose your thoughts and internal monologue inside tags, and then provide your solution or response to the problem. "
-        "당신은 Python 코드를 분석하여 사용법을 설명하는 AI 조수입니다. 주어진 코드를 바탕으로 마치 화면을 보는 것처럼 사용법을 상세히 설명해주세요. Markdown 형식으로 출력하세요."
+        "You are an AI assistant that analyzes Python code to explain its usage. Based on the provided code, please describe how to use it as if you were viewing the interface. Output in Markdown format."
     )
-    user_msg = f"다음 Python 코드를 사용법을 설명해주세요:\n\n{app_content}"
+    user_msg = f"Please explain how to use the following Python code:\n\n{app_content}"
     try:
         return gemini_chat_completion(system_msg, user_msg, max_tokens=800, temperature=0.7)
     except Exception as e:
-        return f"사용법 설명 생성 중 오류 발생: {str(e)}"
+        return f"Error generating usage instructions: {str(e)}"
 
 def stream_gemini_response(user_message: str, conversation_state: List[ChatMessage]) -> Iterator[List[ChatMessage]]:
     """
-    Gemini에 스트리밍 요청.
-    - user_message가 비어 있으면, 최소한의 안내 메시지를 assistant로 추가하고 yield 후 종료
+    Send a streaming request to Gemini.
+    If the user_message is empty, append a minimal guidance message from the assistant and yield.
     """
     if not user_message.strip():
-        # 빈 입력 처리: 안내 메시지 표시
         conversation_state.append(
             ChatMessage(
                 role="assistant",
-                content="입력이 없습니다. 질문을 작성해주세요!"
+                content="No input provided. Please enter a question!"
             )
         )
         yield conversation_state
@@ -273,7 +262,7 @@ def stream_gemini_response(user_message: str, conversation_state: List[ChatMessa
 
 def convert_for_messages_format(messages: List[ChatMessage]) -> List[Dict[str, str]]:
     """
-    ChatMessage 리스트 -> [{"role":"assistant"/"user", "content":"..."}]
+    Convert a list of ChatMessages to the format [{"role": "assistant"/"user", "content": "..."}].
     """
     output = []
     for msg in messages:
@@ -285,14 +274,14 @@ def user_submit_message(msg: str, conversation_state: List[ChatMessage]):
     return "", conversation_state
 
 def respond_wrapper(message: str, conversation_state: List[ChatMessage], max_tokens, temperature, top_p):
-    # 마지막 사용자 메시지 가져오기
+    # Get the last user message
     last_user_message = ""
     for msg in reversed(conversation_state):
         if msg.role == "user":
             last_user_message = msg.content
             break
             
-    # 마지막 사용자 메시지로 응답 생성
+    # Generate response based on the last user message
     for updated_messages in stream_gemini_response(last_user_message, conversation_state):
         yield "", convert_for_messages_format(updated_messages)
 
@@ -320,16 +309,16 @@ def create_ui():
             gr.Markdown("# 🚀 MOUSE: Space Research Thinking")
             
             with gr.Tabs():
-                with gr.TabItem("🔍 분석"):
+                with gr.TabItem("🔍 Analysis"):
                     with gr.Row():
                         with gr.Column():
-                            url_input = gr.Textbox(label="🔗 HuggingFace Space URL", placeholder="예: https://huggingface.co/spaces/username/space-name")
-                            analyze_button = gr.Button("분석 시작 🚀", variant="primary")
+                            url_input = gr.Textbox(label="🔗 HuggingFace Space URL", placeholder="e.g.: https://huggingface.co/spaces/username/space-name")
+                            analyze_button = gr.Button("Start Analysis 🚀", variant="primary")
 
-                            summary_output = gr.Markdown(label="📝 코드 요약")
-                            analysis_output = gr.Markdown(label="🔍 코드 분석")
-                            usage_output = gr.Markdown(label="📚 사용법 안내")
-                            tree_view_output = gr.Textbox(label="📁 파일 구조", lines=20)
+                            summary_output = gr.Markdown(label="📝 Code Summary")
+                            analysis_output = gr.Markdown(label="🔍 Code Analysis")
+                            usage_output = gr.Markdown(label="📚 Usage Instructions")
+                            tree_view_output = gr.Textbox(label="📁 File Structure", lines=20)
 
                         with gr.Column():
                             code_tabs = gr.Tabs()
@@ -346,16 +335,16 @@ def create_ui():
                                         lines=50
                                     )
 
-                with gr.TabItem("🤖 AI 코드챗"):
-                    gr.Markdown("## 💬 예제를 입력하거나 소스 코드를 붙여넣고 질문해보세요!")
+                with gr.TabItem("🤖 AI Code Chat"):
+                    gr.Markdown("## 💬 Enter an example or paste your source code and ask your question!")
                     chatbot = gr.Chatbot(
-                        label="대화창",
+                        label="Chat Window",
                         height=400,
                         type="messages"
                     )
                     msg = gr.Textbox(
-                        label="메시지 입력", 
-                        placeholder="메시지를 입력하세요..."
+                        label="Enter your message", 
+                        placeholder="Type your message here..."
                     )
                     max_tokens = gr.Slider(
                         minimum=1, maximum=8000, 
@@ -374,12 +363,12 @@ def create_ui():
                     )
                     
                     examples = [
-                        ["상세한 사용 방법을 4000 토큰 이상 상세히 설명"],
-                        ["FAQ 20건을 4000 토큰 이상 작성"],
-                        ["기술 차별점, 강점을 중심으로 4000 토큰 이상 설명"],
-                        ["특허 출원에 활용 가능한 혁신 아이디어를 4000 토큰 이상 작성"],
-                        ["논문 형식으로 4000 토큰 이상 작성"],
-                        ["계속 이어서 답변하라"]
+                        ["Explain detailed usage instructions in over 4000 tokens"],
+                        ["Generate 20 FAQs in over 4000 tokens"],
+                        ["Describe technical differentiators and strengths in over 4000 tokens"],
+                        ["Generate innovative ideas for patent applications in over 4000 tokens"],
+                        ["Write an academic paper in over 4000 tokens"],
+                        ["Continue your answer"]
                     ]
                     gr.Examples(examples, inputs=msg)
 
@@ -401,7 +390,7 @@ def create_ui():
                         "Discover recommended HuggingFace Spaces [here](https://huggingface.co/spaces/openfree/Korean-Leaderboard)."
                     )
 
-            # 분석 탭 로직
+            # Analysis tab logic
             space_id_state = gr.State()
             tree_structure_state = gr.State()
             app_py_content_lines = gr.State()