From 3a8a18120c91e035c958c7842e248c2c97a2d1a7 Mon Sep 17 00:00:00 2001 From: ks6088ts Date: Sun, 21 Sep 2025 09:47:23 +0900 Subject: [PATCH] add simple chat app --- .env.template | 2 +- .../streamlits/pages/chat_langchain.py | 191 ++++++++++++++++++ 2 files changed, 192 insertions(+), 1 deletion(-) create mode 100644 template_langgraph/services/streamlits/pages/chat_langchain.py diff --git a/.env.template b/.env.template index d8df6fa..08cf4ff 100644 --- a/.env.template +++ b/.env.template @@ -27,7 +27,7 @@ AZURE_AI_FOUNDRY_INFERENCE_API_VERSION="2025-04-01-preview" AZURE_AI_FOUNDRY_INFERENCE_MODEL_CHAT="gpt-5" ## Ollama Settings -OLLAMA_MODEL_CHAT="gemma3:270m" +OLLAMA_MODEL_CHAT="gemma3:latest" # --------- # Tools diff --git a/template_langgraph/services/streamlits/pages/chat_langchain.py b/template_langgraph/services/streamlits/pages/chat_langchain.py new file mode 100644 index 0000000..835a254 --- /dev/null +++ b/template_langgraph/services/streamlits/pages/chat_langchain.py @@ -0,0 +1,191 @@ +from base64 import b64encode +from os import getenv + +import streamlit as st +from dotenv import load_dotenv +from langchain_community.callbacks.streamlit import ( + StreamlitCallbackHandler, +) +from langchain_core.messages import AIMessage, HumanMessage +from langchain_ollama import ChatOllama +from langchain_openai import AzureChatOpenAI +from openai import APIConnectionError, APIStatusError, APITimeoutError + +from template_langgraph.loggers import get_logger + +load_dotenv(override=True) +logger = get_logger(__name__) +logger.setLevel("DEBUG") + + +def image_to_base64(image_bytes: bytes) -> str: + """Convert image bytes to base64 string.""" + return b64encode(image_bytes).decode("utf-8") + + +with st.sidebar: + "# Common Settings" + stream_mode = st.checkbox( + label="ストリーム出力を有効にする", + value=True, + key="STREAM_MODE", + ) + "# Model" + model_choice = st.radio( + label="Active Model", + options=["azure", "ollama"], + index=0, + key="model_choice", + ) + "## Model Settings" + if model_choice == "azure": + azure_openai_endpoint = st.text_input( + label="AZURE_OPENAI_ENDPOINT", + value=getenv("AZURE_OPENAI_ENDPOINT"), + key="AZURE_OPENAI_ENDPOINT", + type="default", + ) + azure_openai_api_key = st.text_input( + label="AZURE_OPENAI_API_KEY", + value=getenv("AZURE_OPENAI_API_KEY"), + key="AZURE_OPENAI_API_KEY", + type="password", + ) + azure_openai_api_version = st.text_input( + label="AZURE_OPENAI_API_VERSION", + value=getenv("AZURE_OPENAI_API_VERSION"), + key="AZURE_OPENAI_API_VERSION", + type="default", + ) + azure_openai_model_chat = st.text_input( + label="AZURE_OPENAI_MODEL_CHAT", + value=getenv("AZURE_OPENAI_MODEL_CHAT"), + key="AZURE_OPENAI_MODEL_CHAT", + type="default", + ) + "[Azure Portal](https://portal.azure.com/)" + "[Azure OpenAI Studio](https://oai.azure.com/resource/overview)" + "[View the source code](https://github.com/ks6088ts-labs/template-streamlit)" + else: + ollama_model_chat = st.text_input( + label="OLLAMA_MODEL_CHAT", + value=getenv("OLLAMA_MODEL_CHAT"), + key="OLLAMA_MODEL_CHAT", + type="default", + ) + "[Ollama Docs](https://github.com/ollama/ollama)" + "[View the source code](https://github.com/ks6088ts-labs/template-streamlit)" + + +def is_azure_configured(): + return ( + st.session_state.get("AZURE_OPENAI_API_KEY") + and st.session_state.get("AZURE_OPENAI_ENDPOINT") + and st.session_state.get("AZURE_OPENAI_API_VERSION") + and st.session_state.get("AZURE_OPENAI_MODEL_CHAT") + and st.session_state.get("model_choice") == "azure" + ) + + +def is_ollama_configured(): + return st.session_state.get("OLLAMA_MODEL_CHAT") and st.session_state.get("model_choice") == "ollama" + + +def is_configured(): + return is_azure_configured() or is_ollama_configured() + + +def get_model(): + if is_azure_configured(): + return AzureChatOpenAI( + azure_endpoint=st.session_state.get("AZURE_OPENAI_ENDPOINT"), + api_key=st.session_state.get("AZURE_OPENAI_API_KEY"), + openai_api_version=st.session_state.get("AZURE_OPENAI_API_VERSION"), + azure_deployment=st.session_state.get("AZURE_OPENAI_MODEL_CHAT"), + ) + elif is_ollama_configured(): + return ChatOllama( + model=st.session_state.get("OLLAMA_MODEL_CHAT", ""), + ) + raise ValueError("No model is configured. Please set up the Azure or Ollama model in the sidebar.") + + +st.title("chat app with LangChain SDK") + +if not is_configured(): + st.warning("Please fill in the required fields at the sidebar.") + +if "messages" not in st.session_state: + st.session_state["messages"] = [ + AIMessage(content="Hello! I'm a helpful assistant."), + ] + +# Show chat messages +for message in st.session_state.messages: + with st.chat_message(message.type): + if isinstance(message.content, str): + st.markdown(message.content) + else: + for item in message.content: + if item["type"] == "text": + st.markdown(item["text"]) + elif item["type"] == "image_url": + st.image(item["image_url"]["url"]) + + +# Receive user input +uploaded_file = st.file_uploader("画像をアップロード", type=["png", "jpg", "jpeg"], key="file_uploader") +if prompt := st.chat_input(disabled=not is_configured()): + user_message_content = [{"type": "text", "text": prompt}] + if uploaded_file: + image_bytes = uploaded_file.getvalue() + base64_image = image_to_base64(image_bytes) + image_url = f"data:image/jpeg;base64,{base64_image}" + user_message_content.append({"type": "image_url", "image_url": {"url": image_url}}) + + user_message = HumanMessage(content=user_message_content) + st.session_state.messages.append(user_message) + + with st.chat_message("user"): + for item in user_message_content: + if item["type"] == "text": + st.markdown(item["text"]) + elif item["type"] == "image_url": + st.image(item["image_url"]["url"]) + + with st.spinner("Thinking..."): + with st.chat_message("assistant"): + message_placeholder = st.empty() + full_response = "" + llm = get_model() + callbacks = [StreamlitCallbackHandler(st.container())] + + try: + if stream_mode: + for chunk in llm.stream(st.session_state.messages): + if chunk.content is not None: + full_response += chunk.content + message_placeholder.markdown(full_response + "▌") + message_placeholder.markdown(full_response) + else: + response = llm.invoke(input=st.session_state.messages) + full_response = response.content if hasattr(response, "content") else str(response) + message_placeholder.markdown(full_response) + + st.session_state.messages.append(AIMessage(content=full_response)) + + except APITimeoutError as e: + logger.exception(f"APIタイムアウトエラーが発生しました: {e}") + st.error(f"APIタイムアウトエラーが発生しました: {e}") + st.warning("再度お試しいただくか、接続を確認してください。") + except APIConnectionError as e: + logger.exception(f"API接続エラーが発生しました: {e}") + st.error(f"API接続エラーが発生しました: {e}") + st.warning("ネットワーク接続を確認してください。") + except APIStatusError as e: + logger.exception(f"APIステータスエラーが発生しました: {e.status_code} - {e.response}") + st.error(f"APIステータスエラーが発生しました: {e.status_code} - {e.response}") + st.warning("Azure OpenAIの設定（デプロイメント名、APIバージョンなど）を確認してください。") + except Exception as e: + logger.exception(f"予期せぬエラーが発生しました: {e}") + st.error(f"予期せぬエラーが発生しました: {e}")