pull/198/head
金伟强 3 years ago
commit ce9c580e30

@ -0,0 +1,55 @@
# コントリビュート
[Dify](https://dify.ai) に興味を持ち、貢献したいと思うようになったことに感謝します!始める前に、
[行動規範](https://github.com/langgenius/.github/blob/main/CODE_OF_CONDUCT.md)を読み、
[既存の問題](https://github.com/langgenius/langgenius-gateway/issues)をチェックしてください。
本ドキュメントは、[Dify](https://dify.ai) をビルドしてテストするための開発環境の構築方法を説明するものです。
### 依存関係のインストール
[Dify](https://dify.ai)をビルドするには、お使いのマシンに以下の依存関係をインストールし、設定する必要があります:
- [Git](http://git-scm.com/)
- [Docker](https://www.docker.com/)
- [Docker Compose](https://docs.docker.com/compose/install/)
- [Node.js v18.x (LTS)](http://nodejs.org)
- [npm](https://www.npmjs.com/) バージョン 8.x.x もしくは [Yarn](https://yarnpkg.com/)
- [Python](https://www.python.org/) バージョン 3.10.x
## ローカル開発
開発環境を構築するには、プロジェクトの git リポジトリをフォークし、適切なパッケージマネージャを使用してバックエンドとフロントエンドの依存関係をインストールし、docker-compose スタックを実行するように作成します。
### リポジトリのフォーク
[リポジトリ](https://github.com/langgenius/dify) をフォークする必要があります。
### リポジトリのクローン
GitHub でフォークしたリポジトリのクローンを作成する:
```
git clone git@github.com:<github_username>/dify.git
```
### バックエンドのインストール
バックエンドアプリケーションのインストール方法については、[Backend README](api/README.md) を参照してください。
### フロントエンドのインストール
フロントエンドアプリケーションのインストール方法については、[Frontend README](web/README.md) を参照してください。
### ブラウザで dify にアクセス
[Dify](https://dify.ai) をローカル環境で見ることができるようになりました [http://localhost:3000](http://localhost:3000)。
## プルリクエストの作成
変更後、プルリクエスト (PR) をオープンしてください。プルリクエストを提出すると、Dify チーム/コミュニティの他の人があなたと一緒にそれをレビューします。
マージコンフリクトなどの問題が発生したり、プルリクエストの開き方がわからなくなったりしませんでしたか? [GitHub's pull request tutorial](https://docs.github.com/en/pull-requests/collaborating-with-pull-requests) で、マージコンフリクトやその他の問題を解決する方法をチェックしてみてください。あなたの PR がマージされると、[コントリビュータチャート](https://github.com/langgenius/langgenius-gateway/graphs/contributors)にコントリビュータとして誇らしげに掲載されます。
## コミュニティチャンネル
お困りですか?何か質問がありますか? [Discord Community サーバ](https://discord.gg/AhzKf7dNgk)に参加してください。私たちがお手伝いします!

@ -1,7 +1,8 @@
![](./images/describe-en.png) ![](./images/describe-en.png)
<p align="center"> <p align="center">
<a href="./README.md">English</a> | <a href="./README.md">English</a> |
<a href="./README_CN.md">简体中文</a> <a href="./README_CN.md">简体中文</a> |
<a href="./README_JA.md">日本語</a>
</p> </p>
[Website](https://dify.ai) • [Docs](https://docs.dify.ai) • [Twitter](https://twitter.com/dify_ai) • [Discord](https://discord.gg/FngNHpbcY7) [Website](https://dify.ai) • [Docs](https://docs.dify.ai) • [Twitter](https://twitter.com/dify_ai) • [Discord](https://discord.gg/FngNHpbcY7)

@ -1,7 +1,8 @@
![](./images/describe-cn.jpg) ![](./images/describe-cn.jpg)
<p align="center"> <p align="center">
<a href="./README.md">English</a> | <a href="./README.md">English</a> |
<a href="./README_CN.md">简体中文</a> <a href="./README_CN.md">简体中文</a> |
<a href="./README_JA.md">日本語</a>
</p> </p>

@ -0,0 +1,116 @@
![](./images/describe-en.png)
<p align="center">
<a href="./README.md">English</a> |
<a href="./README_CN.md">简体中文</a> |
<a href="./README_JA.md">日本語</a>
</p>
[Web サイト](https://dify.ai) • [ドキュメント](https://docs.dify.ai) • [Twitter](https://twitter.com/dify_ai) • [Discord](https://discord.gg/FngNHpbcY7)
**Dify** は、より多くの人々が持続可能な AI ネイティブアプリケーションを作成できるように設計された、使いやすい LLMOps プラットフォームです。様々なアプリケーションタイプに対応したビジュアルオーケストレーションにより Dify は Backend-as-a-Service API としても機能する、すぐに使えるアプリケーションを提供します。プラグインやデータセットを統合するための1つの API で開発プロセスを統一し、プロンプトエンジニアリング、ビジュアル分析、継続的な改善のための1つのインターフェイスを使って業務を合理化します。
Difyで作成したアプリケーションは以下の通りです:
フォームモードとチャット会話モードをサポートする、すぐに使える Web サイト
プラグイン機能、コンテキストの強化などを網羅する単一の API により、バックエンドのコーディングの手間を省きます。
アプリケーションの視覚的なデータ分析、ログレビュー、アノテーションが可能です。
Dify は LangChain と互換性があり、複数の LLM を徐々にサポートします:
- GPT 3 (text-davinci-003)
- GPT 3.5 Turbo(ChatGPT)
- GPT-4
## クラウドサービスの利用
[Dify.ai](https://dify.ai) をご覧ください
## Community Edition のインストール
### システム要件
Dify をインストールする前に、お使いのマシンが以下の最低システム要件を満たしていることを確認してください:
- CPU >= 1 Core
- RAM >= 4GB
### クイックスタート
Dify サーバーを起動する最も簡単な方法は、[docker-compose.yml](docker/docker-compose.yaml) ファイルを実行することです。インストールコマンドを実行する前に、[Docker](https://docs.docker.com/get-docker/) と [Docker Compose](https://docs.docker.com/compose/install/) がお使いのマシンにインストールされていることを確認してください:
```bash
cd docker
docker-compose up -d
```
実行後、ブラウザで [http://localhost/install](http://localhost/install) にアクセスし、初期化インストール作業を開始することができます。
### 構成
カスタマイズが必要な場合は、[docker-compose.yml](docker/docker-compose.yaml) ファイルのコメントを参照し、手動で環境設定をお願いします。変更後、再度 'docker-compose up -d' を実行してください。
## ロードマップ
開発中の機能:
- **データセット**, Notionやウェブページからのコンテンツ同期など、より多くのデータセットをサポートします
テキスト、ウェブページ、さらには Notion コンテンツなど、より多くのデータセットをサポートする予定です。ユーザーは、自分のデータソースをもとに AI アプリケーションを構築することができます。
- **プラグイン**, アプリケーションに ChatGPT プラグイン標準のプラグインを導入する、または Dify 制作のプラグインを利用する
今後、ChatGPT 規格に準拠したプラグインや、ディファイ独自のプラグインを公開し、より多くの機能をアプリケーションで実現できるようにします。
- **オープンソースモデル**, 例えばモデルプロバイダーとして Llama を採用したり、さらにファインチューニングを行う
Llama のような優れたオープンソースモデルを、私たちのプラットフォームのモデルオプションとして提供したり、さらなる微調整のために使用したりすることで、協力していきます。
## Q&A
**Q: Dify で何ができるのか?**
A: Dify はシンプルでパワフルな LLM 開発・運用ツールです。商用グレードのアプリケーション、パーソナルアシスタントを構築するために使用することができます。独自のアプリケーションを開発したい場合、LangDifyGenius は OpenAI と統合する際のバックエンド作業を省き、視覚的な操作機能を提供し、GPT モデルを継続的に改善・訓練することが可能です。
**Q: Dify を使って、自分のモデルを「トレーニング」するにはどうすればいいのでしょうか?**
A: プロンプトエンジニアリング、コンテキスト拡張、ファインチューニングからなる価値あるアプリケーションです。プロンプトとプログラミング言語を組み合わせたハイブリッドプログラミングアプローチ(テンプレートエンジンのようなもの)で、長文の埋め込みやユーザー入力の YouTube 動画からの字幕取り込みなどを簡単に実現し、これらはすべて LLM が処理するコンテキストとして提出される予定です。また、アプリケーションの操作性を重視し、ユーザーがアプリケーションを使用する際に生成したデータを分析、アノテーション、継続的なトレーニングに利用できるようにしました。適切なツールがなければ、これらのステップに時間がかかることがあります。
**Q: 自分でアプリケーションを作りたい場合、何を準備すればよいですか?**
A: すでに OpenAI API Key をお持ちだと思いますが、お持ちでない場合はご登録ください。もし、すでにトレーニングのコンテキストとなるコンテンツをお持ちでしたら、それは素晴らしいことです!
**Q: インターフェイスにどの言語が使えますか?**
A: 現在、英語と中国語に対応しており、言語パックを寄贈することも可能です。
## Star ヒストリー
[![Star History Chart](https://api.star-history.com/svg?repos=langgenius/dify&type=Date)](https://star-history.com/#langgenius/dify&Date)
## お問合せ
ご質問、ご提案、パートナーシップに関するお問い合わせは、以下のチャンネルからお気軽にご連絡ください:
- GitHub Repo で Issue や PR を提出する
- [Discord](https://discord.gg/FngNHpbcY7) コミュニティで議論に参加する。
- hello@dify.ai にメールを送信します
私たちは、皆様のお手伝いをさせていただき、より楽しく、より便利な AI アプリケーションを一緒に作っていきたいと思っています!
## コントリビュート
適切なレビューを行うため、コミットへの直接アクセスが可能なコントリビュータを含むすべてのコードコントリビュータは、プルリクエストで提出し、マージされる前にコア開発チームによって承認される必要があります。
私たちはすべてのプルリクエストを歓迎します!協力したい方は、[コントリビューションガイド](CONTRIBUTING.md) をチェックしてみてください。
## セキュリティ
プライバシー保護のため、GitHub へのセキュリティ問題の投稿は避けてください。代わりに、あなたの質問を security@dify.ai に送ってください。より詳細な回答を提供します。
## 引用
本ソフトウェアは、以下のオープンソースソフトウェアを使用しています:
- Chase, H. (2022). LangChain [Computer software]. https://github.com/hwchase17/langchain
- Liu, J. (2022). LlamaIndex [Computer software]. doi: 10.5281/zenodo.1234.
詳しくは、各ソフトウェアの公式サイトまたはライセンス文をご参照ください。
## ライセンス
このリポジトリは、[Dify Open Source License](LICENSE) のもとで利用できます。

@ -47,6 +47,7 @@ DEFAULTS = {
'PDF_PREVIEW': 'True', 'PDF_PREVIEW': 'True',
'LOG_LEVEL': 'INFO', 'LOG_LEVEL': 'INFO',
'DISABLE_PROVIDER_CONFIG_VALIDATION': 'False', 'DISABLE_PROVIDER_CONFIG_VALIDATION': 'False',
'DEFAULT_LLM_PROVIDER': 'openai'
} }
@ -181,6 +182,10 @@ class Config:
# You could disable it for compatibility with certain OpenAPI providers # You could disable it for compatibility with certain OpenAPI providers
self.DISABLE_PROVIDER_CONFIG_VALIDATION = get_bool_env('DISABLE_PROVIDER_CONFIG_VALIDATION') self.DISABLE_PROVIDER_CONFIG_VALIDATION = get_bool_env('DISABLE_PROVIDER_CONFIG_VALIDATION')
# For temp use only
# set default LLM provider, default is 'openai', support `azure_openai`
self.DEFAULT_LLM_PROVIDER = get_env('DEFAULT_LLM_PROVIDER')
class CloudEditionConfig(Config): class CloudEditionConfig(Config):
def __init__(self): def __init__(self):

@ -82,29 +82,33 @@ class ProviderTokenApi(Resource):
args = parser.parse_args() args = parser.parse_args()
if not args['token']: if args['token']:
raise ValueError('Token is empty') try:
ProviderService.validate_provider_configs(
try: tenant=current_user.current_tenant,
ProviderService.validate_provider_configs( provider_name=ProviderName(provider),
configs=args['token']
)
token_is_valid = True
except ValidateFailedError:
token_is_valid = False
base64_encrypted_token = ProviderService.get_encrypted_token(
tenant=current_user.current_tenant, tenant=current_user.current_tenant,
provider_name=ProviderName(provider), provider_name=ProviderName(provider),
configs=args['token'] configs=args['token']
) )
token_is_valid = True else:
except ValidateFailedError: base64_encrypted_token = None
token_is_valid = False token_is_valid = False
tenant = current_user.current_tenant tenant = current_user.current_tenant
base64_encrypted_token = ProviderService.get_encrypted_token( provider_model = db.session.query(Provider).filter(
tenant=current_user.current_tenant, Provider.tenant_id == tenant.id,
provider_name=ProviderName(provider), Provider.provider_name == provider,
configs=args['token'] Provider.provider_type == ProviderType.CUSTOM.value
) ).first()
provider_model = Provider.query.filter_by(tenant_id=tenant.id, provider_name=provider,
provider_type=ProviderType.CUSTOM.value).first()
# Only allow updating token for CUSTOM provider type # Only allow updating token for CUSTOM provider type
if provider_model: if provider_model:
@ -117,6 +121,16 @@ class ProviderTokenApi(Resource):
is_valid=token_is_valid) is_valid=token_is_valid)
db.session.add(provider_model) db.session.add(provider_model)
if provider_model.is_valid:
other_providers = db.session.query(Provider).filter(
Provider.tenant_id == tenant.id,
Provider.provider_name != provider,
Provider.provider_type == ProviderType.CUSTOM.value
).all()
for other_provider in other_providers:
other_provider.is_valid = False
db.session.commit() db.session.commit()
if provider in [ProviderName.ANTHROPIC.value, ProviderName.AZURE_OPENAI.value, ProviderName.COHERE.value, if provider in [ProviderName.ANTHROPIC.value, ProviderName.AZURE_OPENAI.value, ProviderName.COHERE.value,

@ -11,9 +11,10 @@ from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_except
@retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6)) @retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6))
def get_embedding( def get_embedding(
text: str, text: str,
engine: Optional[str] = None, engine: Optional[str] = None,
openai_api_key: Optional[str] = None, api_key: Optional[str] = None,
**kwargs
) -> List[float]: ) -> List[float]:
"""Get embedding. """Get embedding.
@ -25,11 +26,12 @@ def get_embedding(
""" """
text = text.replace("\n", " ") text = text.replace("\n", " ")
return openai.Embedding.create(input=[text], engine=engine, api_key=openai_api_key)["data"][0]["embedding"] return openai.Embedding.create(input=[text], engine=engine, api_key=api_key, **kwargs)["data"][0]["embedding"]
@retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6)) @retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6))
async def aget_embedding(text: str, engine: Optional[str] = None, openai_api_key: Optional[str] = None) -> List[float]: async def aget_embedding(text: str, engine: Optional[str] = None, api_key: Optional[str] = None, **kwargs) -> List[
float]:
"""Asynchronously get embedding. """Asynchronously get embedding.
NOTE: Copied from OpenAI's embedding utils: NOTE: Copied from OpenAI's embedding utils:
@ -42,16 +44,17 @@ async def aget_embedding(text: str, engine: Optional[str] = None, openai_api_key
# replace newlines, which can negatively affect performance. # replace newlines, which can negatively affect performance.
text = text.replace("\n", " ") text = text.replace("\n", " ")
return (await openai.Embedding.acreate(input=[text], engine=engine, api_key=openai_api_key))["data"][0][ return (await openai.Embedding.acreate(input=[text], engine=engine, api_key=api_key, **kwargs))["data"][0][
"embedding" "embedding"
] ]
@retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6)) @retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6))
def get_embeddings( def get_embeddings(
list_of_text: List[str], list_of_text: List[str],
engine: Optional[str] = None, engine: Optional[str] = None,
openai_api_key: Optional[str] = None api_key: Optional[str] = None,
**kwargs
) -> List[List[float]]: ) -> List[List[float]]:
"""Get embeddings. """Get embeddings.
@ -67,14 +70,14 @@ def get_embeddings(
# replace newlines, which can negatively affect performance. # replace newlines, which can negatively affect performance.
list_of_text = [text.replace("\n", " ") for text in list_of_text] list_of_text = [text.replace("\n", " ") for text in list_of_text]
data = openai.Embedding.create(input=list_of_text, engine=engine, api_key=openai_api_key).data data = openai.Embedding.create(input=list_of_text, engine=engine, api_key=api_key, **kwargs).data
data = sorted(data, key=lambda x: x["index"]) # maintain the same order as input. data = sorted(data, key=lambda x: x["index"]) # maintain the same order as input.
return [d["embedding"] for d in data] return [d["embedding"] for d in data]
@retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6)) @retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6))
async def aget_embeddings( async def aget_embeddings(
list_of_text: List[str], engine: Optional[str] = None, openai_api_key: Optional[str] = None list_of_text: List[str], engine: Optional[str] = None, api_key: Optional[str] = None, **kwargs
) -> List[List[float]]: ) -> List[List[float]]:
"""Asynchronously get embeddings. """Asynchronously get embeddings.
@ -90,7 +93,7 @@ async def aget_embeddings(
# replace newlines, which can negatively affect performance. # replace newlines, which can negatively affect performance.
list_of_text = [text.replace("\n", " ") for text in list_of_text] list_of_text = [text.replace("\n", " ") for text in list_of_text]
data = (await openai.Embedding.acreate(input=list_of_text, engine=engine, api_key=openai_api_key)).data data = (await openai.Embedding.acreate(input=list_of_text, engine=engine, api_key=api_key, **kwargs)).data
data = sorted(data, key=lambda x: x["index"]) # maintain the same order as input. data = sorted(data, key=lambda x: x["index"]) # maintain the same order as input.
return [d["embedding"] for d in data] return [d["embedding"] for d in data]
@ -98,19 +101,30 @@ async def aget_embeddings(
class OpenAIEmbedding(BaseEmbedding): class OpenAIEmbedding(BaseEmbedding):
def __init__( def __init__(
self, self,
mode: str = OpenAIEmbeddingMode.TEXT_SEARCH_MODE, mode: str = OpenAIEmbeddingMode.TEXT_SEARCH_MODE,
model: str = OpenAIEmbeddingModelType.TEXT_EMBED_ADA_002, model: str = OpenAIEmbeddingModelType.TEXT_EMBED_ADA_002,
deployment_name: Optional[str] = None, deployment_name: Optional[str] = None,
openai_api_key: Optional[str] = None, openai_api_key: Optional[str] = None,
**kwargs: Any, **kwargs: Any,
) -> None: ) -> None:
"""Init params.""" """Init params."""
super().__init__(**kwargs) new_kwargs = {}
if 'embed_batch_size' in kwargs:
new_kwargs['embed_batch_size'] = kwargs['embed_batch_size']
if 'tokenizer' in kwargs:
new_kwargs['tokenizer'] = kwargs['tokenizer']
super().__init__(**new_kwargs)
self.mode = OpenAIEmbeddingMode(mode) self.mode = OpenAIEmbeddingMode(mode)
self.model = OpenAIEmbeddingModelType(model) self.model = OpenAIEmbeddingModelType(model)
self.deployment_name = deployment_name self.deployment_name = deployment_name
self.openai_api_key = openai_api_key self.openai_api_key = openai_api_key
self.openai_api_type = kwargs.get('openai_api_type')
self.openai_api_version = kwargs.get('openai_api_version')
self.openai_api_base = kwargs.get('openai_api_base')
@handle_llm_exceptions @handle_llm_exceptions
def _get_query_embedding(self, query: str) -> List[float]: def _get_query_embedding(self, query: str) -> List[float]:
@ -122,7 +136,9 @@ class OpenAIEmbedding(BaseEmbedding):
if key not in _QUERY_MODE_MODEL_DICT: if key not in _QUERY_MODE_MODEL_DICT:
raise ValueError(f"Invalid mode, model combination: {key}") raise ValueError(f"Invalid mode, model combination: {key}")
engine = _QUERY_MODE_MODEL_DICT[key] engine = _QUERY_MODE_MODEL_DICT[key]
return get_embedding(query, engine=engine, openai_api_key=self.openai_api_key) return get_embedding(query, engine=engine, api_key=self.openai_api_key,
api_type=self.openai_api_type, api_version=self.openai_api_version,
api_base=self.openai_api_base)
def _get_text_embedding(self, text: str) -> List[float]: def _get_text_embedding(self, text: str) -> List[float]:
"""Get text embedding.""" """Get text embedding."""
@ -133,7 +149,9 @@ class OpenAIEmbedding(BaseEmbedding):
if key not in _TEXT_MODE_MODEL_DICT: if key not in _TEXT_MODE_MODEL_DICT:
raise ValueError(f"Invalid mode, model combination: {key}") raise ValueError(f"Invalid mode, model combination: {key}")
engine = _TEXT_MODE_MODEL_DICT[key] engine = _TEXT_MODE_MODEL_DICT[key]
return get_embedding(text, engine=engine, openai_api_key=self.openai_api_key) return get_embedding(text, engine=engine, api_key=self.openai_api_key,
api_type=self.openai_api_type, api_version=self.openai_api_version,
api_base=self.openai_api_base)
async def _aget_text_embedding(self, text: str) -> List[float]: async def _aget_text_embedding(self, text: str) -> List[float]:
"""Asynchronously get text embedding.""" """Asynchronously get text embedding."""
@ -144,7 +162,9 @@ class OpenAIEmbedding(BaseEmbedding):
if key not in _TEXT_MODE_MODEL_DICT: if key not in _TEXT_MODE_MODEL_DICT:
raise ValueError(f"Invalid mode, model combination: {key}") raise ValueError(f"Invalid mode, model combination: {key}")
engine = _TEXT_MODE_MODEL_DICT[key] engine = _TEXT_MODE_MODEL_DICT[key]
return await aget_embedding(text, engine=engine, openai_api_key=self.openai_api_key) return await aget_embedding(text, engine=engine, api_key=self.openai_api_key,
api_type=self.openai_api_type, api_version=self.openai_api_version,
api_base=self.openai_api_base)
def _get_text_embeddings(self, texts: List[str]) -> List[List[float]]: def _get_text_embeddings(self, texts: List[str]) -> List[List[float]]:
"""Get text embeddings. """Get text embeddings.
@ -160,7 +180,9 @@ class OpenAIEmbedding(BaseEmbedding):
if key not in _TEXT_MODE_MODEL_DICT: if key not in _TEXT_MODE_MODEL_DICT:
raise ValueError(f"Invalid mode, model combination: {key}") raise ValueError(f"Invalid mode, model combination: {key}")
engine = _TEXT_MODE_MODEL_DICT[key] engine = _TEXT_MODE_MODEL_DICT[key]
embeddings = get_embeddings(texts, engine=engine, openai_api_key=self.openai_api_key) embeddings = get_embeddings(texts, engine=engine, api_key=self.openai_api_key,
api_type=self.openai_api_type, api_version=self.openai_api_version,
api_base=self.openai_api_base)
return embeddings return embeddings
async def _aget_text_embeddings(self, texts: List[str]) -> List[List[float]]: async def _aget_text_embeddings(self, texts: List[str]) -> List[List[float]]:
@ -172,5 +194,7 @@ class OpenAIEmbedding(BaseEmbedding):
if key not in _TEXT_MODE_MODEL_DICT: if key not in _TEXT_MODE_MODEL_DICT:
raise ValueError(f"Invalid mode, model combination: {key}") raise ValueError(f"Invalid mode, model combination: {key}")
engine = _TEXT_MODE_MODEL_DICT[key] engine = _TEXT_MODE_MODEL_DICT[key]
embeddings = await aget_embeddings(texts, engine=engine, openai_api_key=self.openai_api_key) embeddings = await aget_embeddings(texts, engine=engine, api_key=self.openai_api_key,
api_type=self.openai_api_type, api_version=self.openai_api_version,
api_base=self.openai_api_base)
return embeddings return embeddings

@ -33,8 +33,11 @@ class IndexBuilder:
max_chunk_overlap=20 max_chunk_overlap=20
) )
provider = LLMBuilder.get_default_provider(tenant_id)
model_credentials = LLMBuilder.get_model_credentials( model_credentials = LLMBuilder.get_model_credentials(
tenant_id=tenant_id, tenant_id=tenant_id,
model_provider=provider,
model_name='text-embedding-ada-002' model_name='text-embedding-ada-002'
) )

@ -4,9 +4,14 @@ from langchain.callbacks import CallbackManager
from langchain.llms.fake import FakeListLLM from langchain.llms.fake import FakeListLLM
from core.constant import llm_constant from core.constant import llm_constant
from core.llm.error import ProviderTokenNotInitError
from core.llm.provider.base import BaseProvider
from core.llm.provider.llm_provider_service import LLMProviderService from core.llm.provider.llm_provider_service import LLMProviderService
from core.llm.streamable_azure_chat_open_ai import StreamableAzureChatOpenAI
from core.llm.streamable_azure_open_ai import StreamableAzureOpenAI
from core.llm.streamable_chat_open_ai import StreamableChatOpenAI from core.llm.streamable_chat_open_ai import StreamableChatOpenAI
from core.llm.streamable_open_ai import StreamableOpenAI from core.llm.streamable_open_ai import StreamableOpenAI
from models.provider import ProviderType
class LLMBuilder: class LLMBuilder:
@ -31,16 +36,23 @@ class LLMBuilder:
if model_name == 'fake': if model_name == 'fake':
return FakeListLLM(responses=[]) return FakeListLLM(responses=[])
provider = cls.get_default_provider(tenant_id)
mode = cls.get_mode_by_model(model_name) mode = cls.get_mode_by_model(model_name)
if mode == 'chat': if mode == 'chat':
# llm_cls = StreamableAzureChatOpenAI if provider == 'openai':
llm_cls = StreamableChatOpenAI llm_cls = StreamableChatOpenAI
else:
llm_cls = StreamableAzureChatOpenAI
elif mode == 'completion': elif mode == 'completion':
llm_cls = StreamableOpenAI if provider == 'openai':
llm_cls = StreamableOpenAI
else:
llm_cls = StreamableAzureOpenAI
else: else:
raise ValueError(f"model name {model_name} is not supported.") raise ValueError(f"model name {model_name} is not supported.")
model_credentials = cls.get_model_credentials(tenant_id, model_name) model_credentials = cls.get_model_credentials(tenant_id, provider, model_name)
return llm_cls( return llm_cls(
model_name=model_name, model_name=model_name,
@ -86,18 +98,31 @@ class LLMBuilder:
raise ValueError(f"model name {model_name} is not supported.") raise ValueError(f"model name {model_name} is not supported.")
@classmethod @classmethod
def get_model_credentials(cls, tenant_id: str, model_name: str) -> dict: def get_model_credentials(cls, tenant_id: str, model_provider: str, model_name: str) -> dict:
""" """
Returns the API credentials for the given tenant_id and model_name, based on the model's provider. Returns the API credentials for the given tenant_id and model_name, based on the model's provider.
Raises an exception if the model_name is not found or if the provider is not found. Raises an exception if the model_name is not found or if the provider is not found.
""" """
if not model_name: if not model_name:
raise Exception('model name not found') raise Exception('model name not found')
#
# if model_name not in llm_constant.models:
# raise Exception('model {} not found'.format(model_name))
if model_name not in llm_constant.models: # model_provider = llm_constant.models[model_name]
raise Exception('model {} not found'.format(model_name))
model_provider = llm_constant.models[model_name]
provider_service = LLMProviderService(tenant_id=tenant_id, provider_name=model_provider) provider_service = LLMProviderService(tenant_id=tenant_id, provider_name=model_provider)
return provider_service.get_credentials(model_name) return provider_service.get_credentials(model_name)
@classmethod
def get_default_provider(cls, tenant_id: str) -> str:
provider = BaseProvider.get_valid_provider(tenant_id)
if not provider:
raise ProviderTokenNotInitError()
if provider.provider_type == ProviderType.SYSTEM.value:
provider_name = 'openai'
else:
provider_name = provider.provider_name
return provider_name

@ -36,10 +36,9 @@ class AzureProvider(BaseProvider):
""" """
Returns the API credentials for Azure OpenAI as a dictionary. Returns the API credentials for Azure OpenAI as a dictionary.
""" """
encrypted_config = self.get_provider_api_key(model_id=model_id) config = self.get_provider_api_key(model_id=model_id)
config = json.loads(encrypted_config)
config['openai_api_type'] = 'azure' config['openai_api_type'] = 'azure'
config['deployment_name'] = model_id config['deployment_name'] = model_id.replace('.', '')
return config return config
def get_provider_name(self): def get_provider_name(self):
@ -51,12 +50,11 @@ class AzureProvider(BaseProvider):
""" """
try: try:
config = self.get_provider_api_key() config = self.get_provider_api_key()
config = json.loads(config)
except: except:
config = { config = {
'openai_api_type': 'azure', 'openai_api_type': 'azure',
'openai_api_version': '2023-03-15-preview', 'openai_api_version': '2023-03-15-preview',
'openai_api_base': 'https://foo.microsoft.com/bar', 'openai_api_base': 'https://<your-domain-prefix>.openai.azure.com/',
'openai_api_key': '' 'openai_api_key': ''
} }
@ -65,7 +63,7 @@ class AzureProvider(BaseProvider):
config = { config = {
'openai_api_type': 'azure', 'openai_api_type': 'azure',
'openai_api_version': '2023-03-15-preview', 'openai_api_version': '2023-03-15-preview',
'openai_api_base': 'https://foo.microsoft.com/bar', 'openai_api_base': 'https://<your-domain-prefix>.openai.azure.com/',
'openai_api_key': '' 'openai_api_key': ''
} }

@ -14,7 +14,7 @@ class BaseProvider(ABC):
def __init__(self, tenant_id: str): def __init__(self, tenant_id: str):
self.tenant_id = tenant_id self.tenant_id = tenant_id
def get_provider_api_key(self, model_id: Optional[str] = None, prefer_custom: bool = True) -> str: def get_provider_api_key(self, model_id: Optional[str] = None, prefer_custom: bool = True) -> Union[str | dict]:
""" """
Returns the decrypted API key for the given tenant_id and provider_name. Returns the decrypted API key for the given tenant_id and provider_name.
If the provider is of type SYSTEM and the quota is exceeded, raises a QuotaExceededError. If the provider is of type SYSTEM and the quota is exceeded, raises a QuotaExceededError.
@ -43,23 +43,35 @@ class BaseProvider(ABC):
Returns the Provider instance for the given tenant_id and provider_name. Returns the Provider instance for the given tenant_id and provider_name.
If both CUSTOM and System providers exist, the preferred provider will be returned based on the prefer_custom flag. If both CUSTOM and System providers exist, the preferred provider will be returned based on the prefer_custom flag.
""" """
providers = db.session.query(Provider).filter( return BaseProvider.get_valid_provider(self.tenant_id, self.get_provider_name().value, prefer_custom)
Provider.tenant_id == self.tenant_id,
Provider.provider_name == self.get_provider_name().value @classmethod
).order_by(Provider.provider_type.desc() if prefer_custom else Provider.provider_type).all() def get_valid_provider(cls, tenant_id: str, provider_name: str = None, prefer_custom: bool = False) -> Optional[Provider]:
"""
Returns the Provider instance for the given tenant_id and provider_name.
If both CUSTOM and System providers exist, the preferred provider will be returned based on the prefer_custom flag.
"""
query = db.session.query(Provider).filter(
Provider.tenant_id == tenant_id
)
if provider_name:
query = query.filter(Provider.provider_name == provider_name)
providers = query.order_by(Provider.provider_type.desc() if prefer_custom else Provider.provider_type).all()
custom_provider = None custom_provider = None
system_provider = None system_provider = None
for provider in providers: for provider in providers:
if provider.provider_type == ProviderType.CUSTOM.value: if provider.provider_type == ProviderType.CUSTOM.value and provider.is_valid and provider.encrypted_config:
custom_provider = provider custom_provider = provider
elif provider.provider_type == ProviderType.SYSTEM.value: elif provider.provider_type == ProviderType.SYSTEM.value and provider.is_valid:
system_provider = provider system_provider = provider
if custom_provider and custom_provider.is_valid and custom_provider.encrypted_config: if custom_provider:
return custom_provider return custom_provider
elif system_provider and system_provider.is_valid: elif system_provider:
return system_provider return system_provider
else: else:
return None return None
@ -80,7 +92,7 @@ class BaseProvider(ABC):
try: try:
config = self.get_provider_api_key() config = self.get_provider_api_key()
except: except:
config = 'THIS-IS-A-MOCK-TOKEN' config = ''
if obfuscated: if obfuscated:
return self.obfuscated_token(config) return self.obfuscated_token(config)

@ -1,12 +1,50 @@
import requests
from langchain.schema import BaseMessage, ChatResult, LLMResult from langchain.schema import BaseMessage, ChatResult, LLMResult
from langchain.chat_models import AzureChatOpenAI from langchain.chat_models import AzureChatOpenAI
from typing import Optional, List from typing import Optional, List, Dict, Any
from pydantic import root_validator
from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async
class StreamableAzureChatOpenAI(AzureChatOpenAI): class StreamableAzureChatOpenAI(AzureChatOpenAI):
@root_validator()
def validate_environment(cls, values: Dict) -> Dict:
"""Validate that api key and python package exists in environment."""
try:
import openai
except ImportError:
raise ValueError(
"Could not import openai python package. "
"Please install it with `pip install openai`."
)
try:
values["client"] = openai.ChatCompletion
except AttributeError:
raise ValueError(
"`openai` has no `ChatCompletion` attribute, this is likely "
"due to an old version of the openai package. Try upgrading it "
"with `pip install --upgrade openai`."
)
if values["n"] < 1:
raise ValueError("n must be at least 1.")
if values["n"] > 1 and values["streaming"]:
raise ValueError("n must be 1 when streaming.")
return values
@property
def _default_params(self) -> Dict[str, Any]:
"""Get the default parameters for calling OpenAI API."""
return {
**super()._default_params,
"engine": self.deployment_name,
"api_type": self.openai_api_type,
"api_base": self.openai_api_base,
"api_version": self.openai_api_version,
"api_key": self.openai_api_key,
"organization": self.openai_organization if self.openai_organization else None,
}
def get_messages_tokens(self, messages: List[BaseMessage]) -> int: def get_messages_tokens(self, messages: List[BaseMessage]) -> int:
"""Get the number of tokens in a list of messages. """Get the number of tokens in a list of messages.

@ -0,0 +1,64 @@
import os
from langchain.llms import AzureOpenAI
from langchain.schema import LLMResult
from typing import Optional, List, Dict, Mapping, Any
from pydantic import root_validator
from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async
class StreamableAzureOpenAI(AzureOpenAI):
openai_api_type: str = "azure"
openai_api_version: str = ""
@root_validator()
def validate_environment(cls, values: Dict) -> Dict:
"""Validate that api key and python package exists in environment."""
try:
import openai
values["client"] = openai.Completion
except ImportError:
raise ValueError(
"Could not import openai python package. "
"Please install it with `pip install openai`."
)
if values["streaming"] and values["n"] > 1:
raise ValueError("Cannot stream results when n > 1.")
if values["streaming"] and values["best_of"] > 1:
raise ValueError("Cannot stream results when best_of > 1.")
return values
@property
def _invocation_params(self) -> Dict[str, Any]:
return {**super()._invocation_params, **{
"api_type": self.openai_api_type,
"api_base": self.openai_api_base,
"api_version": self.openai_api_version,
"api_key": self.openai_api_key,
"organization": self.openai_organization if self.openai_organization else None,
}}
@property
def _identifying_params(self) -> Mapping[str, Any]:
return {**super()._identifying_params, **{
"api_type": self.openai_api_type,
"api_base": self.openai_api_base,
"api_version": self.openai_api_version,
"api_key": self.openai_api_key,
"organization": self.openai_organization if self.openai_organization else None,
}}
@handle_llm_exceptions
def generate(
self, prompts: List[str], stop: Optional[List[str]] = None
) -> LLMResult:
return super().generate(prompts, stop)
@handle_llm_exceptions_async
async def agenerate(
self, prompts: List[str], stop: Optional[List[str]] = None
) -> LLMResult:
return await super().agenerate(prompts, stop)

@ -1,12 +1,52 @@
import os
from langchain.schema import BaseMessage, ChatResult, LLMResult from langchain.schema import BaseMessage, ChatResult, LLMResult
from langchain.chat_models import ChatOpenAI from langchain.chat_models import ChatOpenAI
from typing import Optional, List from typing import Optional, List, Dict, Any
from pydantic import root_validator
from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async
class StreamableChatOpenAI(ChatOpenAI): class StreamableChatOpenAI(ChatOpenAI):
@root_validator()
def validate_environment(cls, values: Dict) -> Dict:
"""Validate that api key and python package exists in environment."""
try:
import openai
except ImportError:
raise ValueError(
"Could not import openai python package. "
"Please install it with `pip install openai`."
)
try:
values["client"] = openai.ChatCompletion
except AttributeError:
raise ValueError(
"`openai` has no `ChatCompletion` attribute, this is likely "
"due to an old version of the openai package. Try upgrading it "
"with `pip install --upgrade openai`."
)
if values["n"] < 1:
raise ValueError("n must be at least 1.")
if values["n"] > 1 and values["streaming"]:
raise ValueError("n must be 1 when streaming.")
return values
@property
def _default_params(self) -> Dict[str, Any]:
"""Get the default parameters for calling OpenAI API."""
return {
**super()._default_params,
"api_type": 'openai',
"api_base": os.environ.get("OPENAI_API_BASE", "https://api.openai.com/v1"),
"api_version": None,
"api_key": self.openai_api_key,
"organization": self.openai_organization if self.openai_organization else None,
}
def get_messages_tokens(self, messages: List[BaseMessage]) -> int: def get_messages_tokens(self, messages: List[BaseMessage]) -> int:
"""Get the number of tokens in a list of messages. """Get the number of tokens in a list of messages.

@ -1,12 +1,54 @@
import os
from langchain.schema import LLMResult from langchain.schema import LLMResult
from typing import Optional, List from typing import Optional, List, Dict, Any, Mapping
from langchain import OpenAI from langchain import OpenAI
from pydantic import root_validator
from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async
class StreamableOpenAI(OpenAI): class StreamableOpenAI(OpenAI):
@root_validator()
def validate_environment(cls, values: Dict) -> Dict:
"""Validate that api key and python package exists in environment."""
try:
import openai
values["client"] = openai.Completion
except ImportError:
raise ValueError(
"Could not import openai python package. "
"Please install it with `pip install openai`."
)
if values["streaming"] and values["n"] > 1:
raise ValueError("Cannot stream results when n > 1.")
if values["streaming"] and values["best_of"] > 1:
raise ValueError("Cannot stream results when best_of > 1.")
return values
@property
def _invocation_params(self) -> Dict[str, Any]:
return {**super()._invocation_params, **{
"api_type": 'openai',
"api_base": os.environ.get("OPENAI_API_BASE", "https://api.openai.com/v1"),
"api_version": None,
"api_key": self.openai_api_key,
"organization": self.openai_organization if self.openai_organization else None,
}}
@property
def _identifying_params(self) -> Mapping[str, Any]:
return {**super()._identifying_params, **{
"api_type": 'openai',
"api_base": os.environ.get("OPENAI_API_BASE", "https://api.openai.com/v1"),
"api_version": None,
"api_key": self.openai_api_key,
"organization": self.openai_organization if self.openai_organization else None,
}}
@handle_llm_exceptions @handle_llm_exceptions
def generate( def generate(
self, prompts: List[str], stop: Optional[List[str]] = None self, prompts: List[str], stop: Optional[List[str]] = None

@ -1,117 +0,0 @@
# Logs
logs
*.log
npm-debug.log*
yarn-debug.log*
yarn-error.log*
lerna-debug.log*
# Diagnostic reports (https://nodejs.org/api/report.html)
report.[0-9]*.[0-9]*.[0-9]*.[0-9]*.json
# Runtime data
pids
*.pid
*.seed
*.pid.lock
# Directory for instrumented libs generated by jscoverage/JSCover
lib-cov
# Coverage directory used by tools like istanbul
coverage
*.lcov
# nyc test coverage
.nyc_output
# Grunt intermediate storage (https://gruntjs.com/creating-plugins#storing-task-files)
.grunt
# Bower dependency directory (https://bower.io/)
bower_components
# node-waf configuration
.lock-wscript
# Compiled binary addons (https://nodejs.org/api/addons.html)
build/Release
# Dependency directories
node_modules/
jspm_packages/
# TypeScript v1 declaration files
typings/
# TypeScript cache
*.tsbuildinfo
# Optional npm cache directory
.npm
# Optional eslint cache
.eslintcache
# Microbundle cache
.rpt2_cache/
.rts2_cache_cjs/
.rts2_cache_es/
.rts2_cache_umd/
# Optional REPL history
.node_repl_history
# Output of 'npm pack'
*.tgz
# Yarn Integrity file
.yarn-integrity
# dotenv environment variables file
.env
.env.test
# parcel-bundler cache (https://parceljs.org/)
.cache
# Next.js build output
.next
# Nuxt.js build / generate output
.nuxt
dist
# Gatsby files
.cache/
# Comment in the public line in if your project uses Gatsby and *not* Next.js
# https://nextjs.org/blog/next-9-1#public-directory-support
# public
# vuepress build output
.vuepress/dist
# Serverless directories
.serverless/
# FuseBox cache
.fusebox/
# DynamoDB Local files
.dynamodb/
# TernJS port file
.tern-port
# npm
package-lock.json
# yarn
.pnp.cjs
.pnp.loader.mjs
.yarn/
yarn.lock
.yarnrc.yml
# pmpm
pnpm-lock.yaml

@ -1 +0,0 @@
# Mock Server

@ -1,551 +0,0 @@
const chars = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ-_'
function randomString (length) {
let result = ''
for (let i = length; i > 0; --i) result += chars[Math.floor(Math.random() * chars.length)]
return result
}
// https://www.notion.so/55773516a0194781ae211792a44a3663?pvs=4
const VirtualData = new Array(10).fill().map((_, index) => {
const date = new Date(Date.now() - index * 24 * 60 * 60 * 1000)
return {
date: `${date.getFullYear()}-${date.getMonth()}-${date.getDate()}`,
conversation_count: Math.floor(Math.random() * 10) + index,
terminal_count: Math.floor(Math.random() * 10) + index,
token_count: Math.floor(Math.random() * 10) + index,
total_price: Math.floor(Math.random() * 10) + index,
}
})
const registerAPI = function (app) {
const apps = [{
id: '1',
name: 'chat app',
mode: 'chat',
description: 'description01',
enable_site: true,
enable_api: true,
api_rpm: 60,
api_rph: 3600,
is_demo: false,
model_config: {
provider: 'OPENAI',
model_id: 'gpt-3.5-turbo',
configs: {
prompt_template: '你是我的解梦小助手,请参考 {{book}} 回答我有关梦境的问题。在回答前请称呼我为 {{myName}}。',
prompt_variables: [
{
key: 'book',
name: '书',
value: '《梦境解析》',
type: 'string',
description: '请具体说下书名'
},
{
key: 'myName',
name: 'your name',
value: 'Book',
type: 'string',
description: 'please tell me your name'
}
],
completion_params: {
max_token: 16,
temperature: 1, // 0-2
top_p: 1,
presence_penalty: 1, // -2-2
frequency_penalty: 1, // -2-2
}
}
},
site: {
access_token: '1000',
title: 'site 01',
author: 'John',
default_language: 'zh-Hans-CN',
customize_domain: 'http://customize_domain',
theme: 'theme',
customize_token_strategy: 'must',
prompt_public: true
}
},
{
id: '2',
name: 'completion app',
mode: 'completion', // genertation text
description: 'description 02', // genertation text
enable_site: false,
enable_api: false,
api_rpm: 60,
api_rph: 3600,
is_demo: false,
model_config: {
provider: 'OPENAI',
model_id: 'text-davinci-003',
configs: {
prompt_template: '你是我的翻译小助手,请把以下内容 {{langA}} 翻译成 {{langB}},以下的内容:',
prompt_variables: [
{
key: 'langA',
name: '原始语音',
value: '中文',
type: 'string',
description: '这是中文格式的原始语音'
},
{
key: 'langB',
name: '目标语言',
value: '英语',
type: 'string',
description: '这是英语格式的目标语言'
}
],
completion_params: {
max_token: 16,
temperature: 1, // 0-2
top_p: 1,
presence_penalty: 1, // -2-2
frequency_penalty: 1, // -2-2
}
}
},
site: {
access_token: '2000',
title: 'site 02',
author: 'Mark',
default_language: 'en-US',
customize_domain: 'http://customize_domain',
theme: 'theme',
customize_token_strategy: 'must',
prompt_public: false
}
},
]
const apikeys = [{
id: '111121312313132',
token: 'sk-DEFGHJKMNPQRSTWXYZabcdefhijk1234',
last_used_at: '1679212138000',
created_at: '1673316000000'
}, {
id: '43441242131223123',
token: 'sk-EEFGHJKMNPQRSTWXYZabcdefhijk5678',
last_used_at: '1679212721000',
created_at: '1679212731000'
}]
// create app
app.post('/apps', async (req, res) => {
apps.push({
id: apps.length + 1 + '',
...req.body,
})
res.send({
result: 'success'
})
})
// app list
app.get('/apps', async (req, res) => {
res.send({
data: apps
})
})
// app detail
app.get('/apps/:id', async (req, res) => {
const item = apps.find(item => item.id === req.params.id) || apps[0]
res.send(item)
})
// update app name
app.post('/apps/:id/name', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
item.name = req.body.name
res.send(item || null)
})
// update app site-enable status
app.post('/apps/:id/site-enable', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
console.log(item)
item.enable_site = req.body.enable_site
res.send(item || null)
})
// update app api-enable status
app.post('/apps/:id/api-enable', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
console.log(item)
item.enable_api = req.body.enable_api
res.send(item || null)
})
// update app rate-limit
app.post('/apps/:id/rate-limit', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
console.log(item)
item.api_rpm = req.body.api_rpm
item.api_rph = req.body.api_rph
res.send(item || null)
})
// update app url including code
app.post('/apps/:id/site/access-token-reset', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
console.log(item)
item.site.access_token = randomString(12)
res.send(item || null)
})
// update app config
app.post('/apps/:id/site', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
console.log(item)
item.name = req.body.title
item.description = req.body.description
item.prompt_public = req.body.prompt_public
item.default_language = req.body.default_language
res.send(item || null)
})
// get statistics daily-conversations
app.get('/apps/:id/statistics/daily-conversations', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
if (item) {
res.send({
data: VirtualData
})
} else {
res.send({
data: []
})
}
})
// get statistics daily-end-users
app.get('/apps/:id/statistics/daily-end-users', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
if (item) {
res.send({
data: VirtualData
})
} else {
res.send({
data: []
})
}
})
// get statistics token-costs
app.get('/apps/:id/statistics/token-costs', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
if (item) {
res.send({
data: VirtualData
})
} else {
res.send({
data: []
})
}
})
// update app model config
app.post('/apps/:id/model-config', async (req, res) => {
const item = apps.find(item => item.id === req.params.id)
console.log(item)
item.model_config = req.body
res.send(item || null)
})
// get api keys list
app.get('/apps/:id/api-keys', async (req, res) => {
res.send({
data: apikeys
})
})
// del api key
app.delete('/apps/:id/api-keys/:api_key_id', async (req, res) => {
res.send({
result: 'success'
})
})
// create api key
app.post('/apps/:id/api-keys', async (req, res) => {
res.send({
id: 'e2424241313131',
token: 'sk-GEFGHJKMNPQRSTWXYZabcdefhijk0124',
created_at: '1679216688962'
})
})
// get completion-conversations
app.get('/apps/:id/completion-conversations', async (req, res) => {
const data = {
data: [{
id: 1,
from_end_user_id: 'user 1',
summary: 'summary1',
created_at: '2023-10-11',
annotated: true,
message_count: 100,
user_feedback_stats: {
like: 4, dislike: 5
},
admin_feedback_stats: {
like: 1, dislike: 2
},
message: {
message: 'message1',
query: 'question1',
answer: 'answer1'
}
}, {
id: 12,
from_end_user_id: 'user 2',
summary: 'summary2',
created_at: '2023-10-01',
annotated: false,
message_count: 10,
user_feedback_stats: {
like: 2, dislike: 20
},
admin_feedback_stats: {
like: 12, dislike: 21
},
message: {
message: 'message2',
query: 'question2',
answer: 'answer2'
}
}, {
id: 13,
from_end_user_id: 'user 3',
summary: 'summary3',
created_at: '2023-10-11',
annotated: false,
message_count: 20,
user_feedback_stats: {
like: 2, dislike: 0
},
admin_feedback_stats: {
like: 0, dislike: 21
},
message: {
message: 'message3',
query: 'question3',
answer: 'answer3'
}
}],
total: 200
}
res.send(data)
})
// get chat-conversations
app.get('/apps/:id/chat-conversations', async (req, res) => {
const data = {
data: [{
id: 1,
from_end_user_id: 'user 1',
summary: 'summary1',
created_at: '2023-10-11',
read_at: '2023-10-12',
annotated: true,
message_count: 100,
user_feedback_stats: {
like: 4, dislike: 5
},
admin_feedback_stats: {
like: 1, dislike: 2
},
message: {
message: 'message1',
query: 'question1',
answer: 'answer1'
}
}, {
id: 12,
from_end_user_id: 'user 2',
summary: 'summary2',
created_at: '2023-10-01',
annotated: false,
message_count: 10,
user_feedback_stats: {
like: 2, dislike: 20
},
admin_feedback_stats: {
like: 12, dislike: 21
},
message: {
message: 'message2',
query: 'question2',
answer: 'answer2'
}
}, {
id: 13,
from_end_user_id: 'user 3',
summary: 'summary3',
created_at: '2023-10-11',
annotated: false,
message_count: 20,
user_feedback_stats: {
like: 2, dislike: 0
},
admin_feedback_stats: {
like: 0, dislike: 21
},
message: {
message: 'message3',
query: 'question3',
answer: 'answer3'
}
}],
total: 200
}
res.send(data)
})
// get completion-conversation detail
app.get('/apps/:id/completion-conversations/:cid', async (req, res) => {
const data =
{
id: 1,
from_end_user_id: 'user 1',
summary: 'summary1',
created_at: '2023-10-11',
annotated: true,
message: {
message: 'question1',
// query: 'question1',
answer: 'answer1',
annotation: {
content: '这是一段纠正的内容'
}
},
model_config: {
provider: 'openai',
model_id: 'model_id',
configs: {
prompt_template: '你是我的翻译小助手,请把以下内容 {{langA}} 翻译成 {{langB}},以下的内容:{{content}}'
}
}
}
res.send(data)
})
// get chat-conversation detail
app.get('/apps/:id/chat-conversations/:cid', async (req, res) => {
const data =
{
id: 1,
from_end_user_id: 'user 1',
summary: 'summary1',
created_at: '2023-10-11',
annotated: true,
message: {
message: 'question1',
// query: 'question1',
answer: 'answer1',
created_at: '2023-08-09 13:00',
provider_response_latency: 130,
message_tokens: 230
},
model_config: {
provider: 'openai',
model_id: 'model_id',
configs: {
prompt_template: '你是我的翻译小助手,请把以下内容 {{langA}} 翻译成 {{langB}},以下的内容:{{content}}'
}
}
}
res.send(data)
})
// get chat-conversation message list
app.get('/apps/:id/chat-messages', async (req, res) => {
const data = {
data: [{
id: 1,
created_at: '2023-10-11 07:09',
message: '请说说人为什么会做梦?' + req.query.conversation_id,
answer: '梦境通常是个人内心深处的反映,很难确定每个人梦境的确切含义,因为它们可能会受到梦境者的文化背景、生活经验和情感状态等多种因素的影响。',
provider_response_latency: 450,
answer_tokens: 200,
annotation: {
content: 'string',
account: {
id: 'string',
name: 'string',
email: 'string'
}
},
feedbacks: {
rating: 'like',
content: 'string',
from_source: 'log'
}
}, {
id: 2,
created_at: '2023-10-11 8:23',
message: '夜里经常做梦会影响次日的精神状态吗?',
answer: '总之,这个梦境可能与梦境者的个人经历和情感状态有关,但在一般情况下,它可能表示一种强烈的情感反应,包括愤怒、不满和对于正义和自由的渴望。',
provider_response_latency: 400,
answer_tokens: 250,
annotation: {
content: 'string',
account: {
id: 'string',
name: 'string',
email: 'string'
}
},
// feedbacks: {
// rating: 'like',
// content: 'string',
// from_source: 'log'
// }
}, {
id: 3,
created_at: '2023-10-11 10:20',
message: '梦见在山上手撕鬼子,大师解解梦',
answer: '但是,一般来说,“手撕鬼子”这个场景可能是梦境者对于过去历史上的战争、侵略以及对于自己国家和族群的保护与维护的情感反应。在梦中,你可能会感到自己充满力量和勇气,去对抗那些看似强大的侵略者。',
provider_response_latency: 288,
answer_tokens: 100,
annotation: {
content: 'string',
account: {
id: 'string',
name: 'string',
email: 'string'
}
},
feedbacks: {
rating: 'dislike',
content: 'string',
from_source: 'log'
}
}],
limit: 20,
has_more: true
}
res.send(data)
})
app.post('/apps/:id/annotations', async (req, res) => {
res.send({ result: 'success' })
})
app.post('/apps/:id/feedbacks', async (req, res) => {
res.send({ result: 'success' })
})
}
module.exports = registerAPI

@ -1,38 +0,0 @@
const registerAPI = function (app) {
app.post('/login', async (req, res) => {
res.send({
result: 'success'
})
})
// get user info
app.get('/account/profile', async (req, res) => {
res.send({
id: '11122222',
name: 'Joel',
email: 'iamjoel007@gmail.com'
})
})
// logout
app.get('/logout', async (req, res) => {
res.send({
result: 'success'
})
})
// Langgenius version
app.get('/version', async (req, res) => {
res.send({
current_version: 'v1.0.0',
latest_version: 'v1.0.0',
upgradeable: true,
compatible_upgrade: true
})
})
}
module.exports = registerAPI

@ -1,249 +0,0 @@
const registerAPI = function (app) {
app.get("/datasets/:id/documents", async (req, res) => {
if (req.params.id === "0") res.send({ data: [] });
else {
res.send({
data: [
{
id: 1,
name: "Steve Jobs' life",
words: "70k",
word_count: 100,
updated_at: 1681801029,
indexing_status: "completed",
archived: true,
enabled: false,
data_source_info: {
upload_file: {
// id: string
// name: string
// size: number
// mime_type: string
// created_at: number
// created_by: string
extension: "pdf",
},
},
},
{
id: 2,
name: "Steve Jobs' life",
word_count: "10k",
hit_count: 10,
updated_at: 1681801029,
indexing_status: "waiting",
archived: true,
enabled: false,
data_source_info: {
upload_file: {
extension: "json",
},
},
},
{
id: 3,
name: "Steve Jobs' life xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
word_count: "100k",
hit_count: 0,
updated_at: 1681801029,
indexing_status: "indexing",
archived: false,
enabled: true,
data_source_info: {
upload_file: {
extension: "txt",
},
},
},
{
id: 4,
name: "Steve Jobs' life xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
word_count: "100k",
hit_count: 0,
updated_at: 1681801029,
indexing_status: "splitting",
archived: false,
enabled: true,
data_source_info: {
upload_file: {
extension: "md",
},
},
},
{
id: 5,
name: "Steve Jobs' life",
word_count: "100k",
hit_count: 0,
updated_at: 1681801029,
indexing_status: "error",
archived: false,
enabled: false,
data_source_info: {
upload_file: {
extension: "html",
},
},
},
],
total: 100,
id: req.params.id,
});
}
});
app.get("/datasets/:id/documents/:did/segments", async (req, res) => {
if (req.params.id === "0") res.send({ data: [] });
else {
res.send({
data: new Array(100).fill({
id: 1234,
content: `他的坚持让我很为难。众所周知他非常注意保护自己的隐私而我想他应该从来没有看过我写的书。也许将来的某个时候吧我还是这么说。但是到了2009年他的妻子劳伦·鲍威尔Laurene Powell直言不讳地对我说“如果你真的打算写一本关于史蒂夫的书最好现在就开始。”他当时刚刚第二次因病休假。我向劳伦坦承当乔布斯第一次提出这个想法时我并不知道他病了。几乎没有人知道她说。他是在接受癌症手术之前给我打的电话直到今天他还将此事作为一个秘密她这么解释道。\n
他的坚持让我很为难众所周知他非常注意保护自己的隐私而我想他应该从来没有看过我写的书也许将来的某个时候吧我还是这么说但是到了2009年他的妻子劳伦·鲍威尔Laurene Powell直言不讳地对我说如果你真的打算写一本关于史蒂夫的书最好现在就开始他当时刚刚第二次因病休假我向劳伦坦承当乔布斯第一次提出这个想法时我并不知道他病了几乎没有人知道她说他是在接受癌症手术之前给我打的电话直到今天他还将此事作为一个秘密她这么解释道`,
enabled: true,
keyWords: [
"劳伦·鲍威尔",
"劳伦·鲍威尔",
"手术",
"秘密",
"癌症",
"乔布斯",
"史蒂夫",
"书",
"休假",
"坚持",
"隐私",
],
word_count: 120,
hit_count: 100,
status: "ok",
index_node_hash: "index_node_hash value",
}),
limit: 100,
has_more: true,
});
}
});
// get doc detail
app.get("/datasets/:id/documents/:did", async (req, res) => {
const fixedParams = {
// originInfo: {
originalFilename: "Original filename",
originalFileSize: "16mb",
uploadDate: "2023-01-01",
lastUpdateDate: "2023-01-05",
source: "Source",
// },
// technicalParameters: {
segmentSpecification: "909090",
segmentLength: 100,
avgParagraphLength: 130,
};
const bookData = {
doc_type: "book",
doc_metadata: {
title: "机器学习实战",
language: "zh",
author: "Peter Harrington",
publisher: "人民邮电出版社",
publicationDate: "2013-01-01",
ISBN: "9787115335500",
category: "技术",
},
};
const webData = {
doc_type: "webPage",
doc_metadata: {
title: "深度学习入门教程",
url: "https://www.example.com/deep-learning-tutorial",
language: "zh",
publishDate: "2020-05-01",
authorPublisher: "张三",
topicsKeywords: "深度学习, 人工智能, 教程",
description:
"这是一篇详细的深度学习入门教程,适用于对人工智能和深度学习感兴趣的初学者。",
},
};
const postData = {
doc_type: "socialMediaPost",
doc_metadata: {
platform: "Twitter",
authorUsername: "example_user",
publishDate: "2021-08-15",
postURL: "https://twitter.com/example_user/status/1234567890",
topicsTags:
"AI, DeepLearning, Tutorial, Example, Example2, Example3, AI, DeepLearning, Tutorial, Example, Example2, Example3, AI, DeepLearning, Tutorial, Example, Example2, Example3,",
},
};
res.send({
id: "550e8400-e29b-41d4-a716-446655440000",
position: 1,
dataset_id: "550e8400-e29b-41d4-a716-446655440002",
data_source_type: "upload_file",
data_source_info: {
upload_file: {
extension: "html",
id: "550e8400-e29b-41d4-a716-446655440003",
},
},
dataset_process_rule_id: "550e8400-e29b-41d4-a716-446655440004",
batch: "20230410123456123456",
name: "example_document",
created_from: "web",
created_by: "550e8400-e29b-41d4-a716-446655440005",
created_api_request_id: "550e8400-e29b-41d4-a716-446655440006",
created_at: 1671269696,
processing_started_at: 1671269700,
word_count: 11,
parsing_completed_at: 1671269710,
cleaning_completed_at: 1671269720,
splitting_completed_at: 1671269730,
tokens: 10,
indexing_latency: 5.0,
completed_at: 1671269740,
paused_by: null,
paused_at: null,
error: null,
stopped_at: null,
indexing_status: "completed",
enabled: true,
disabled_at: null,
disabled_by: null,
archived: false,
archived_reason: null,
archived_by: null,
archived_at: null,
updated_at: 1671269740,
...(req.params.did === "book"
? bookData
: req.params.did === "web"
? webData
: req.params.did === "post"
? postData
: {}),
segment_count: 10,
hit_count: 9,
status: "ok",
});
});
// // logout
// app.get("/logout", async (req, res) => {
// res.send({
// result: "success",
// });
// });
// // Langgenius version
// app.get("/version", async (req, res) => {
// res.send({
// current_version: "v1.0.0",
// latest_version: "v1.0.0",
// upgradeable: true,
// compatible_upgrade: true,
// });
// });
};
module.exports = registerAPI;

@ -1,119 +0,0 @@
const registerAPI = function (app) {
const coversationList = [
{
id: '1',
name: '梦的解析',
inputs: {
book: '《梦的解析》',
callMe: '大师',
},
chats: []
},
{
id: '2',
name: '生命的起源',
inputs: {
book: '《x x x》',
}
},
]
// site info
app.get('/apps/site/info', async (req, res) => {
// const id = req.params.id
res.send({
enable_site: true,
appId: '1',
site: {
title: 'Story Bot',
description: '这是一款解梦聊天机器人,你可以选择你喜欢的解梦人进行解梦,这句话是客户端应用说明',
},
prompt_public: true, //id === '1',
prompt_template: '你是我的解梦小助手,请参考 {{book}} 回答我有关梦境的问题。在回答前请称呼我为 {{myName}}。',
})
})
app.post('/apps/:id/chat-messages', async (req, res) => {
const conversationId = req.body.conversation_id ? req.body.conversation_id : Date.now() + ''
res.send({
id: Date.now() + '',
conversation_id: Date.now() + '',
answer: 'balabababab'
})
})
app.post('/apps/:id/completion-messages', async (req, res) => {
res.send({
id: Date.now() + '',
answer: `做为一个AI助手我可以为你提供随机生成的段落这些段落可以用于测试、占位符、或者其他目的。以下是一个随机生成的段落
随着科技的不断发展越来越多的人开始意识到人工智能的重要性人工智能已经成为我们生活中不可或缺的一部分它可以帮助我们完成很多繁琐的工作也可以为我们提供更智能更便捷的服务虽然人工智能带来了很多好处但它也面临着很多挑战例如人工智能的算法可能会出现偏见导致对某些人群不公平此外人工智能的发展也可能会导致一些工作的失业因此我们需要不断地研究人工智能的发展以确保它能够为人类带来更多的好处`
})
})
// share api
// chat list
app.get('/apps/:id/coversations', async (req, res) => {
res.send({
data: coversationList
})
})
app.get('/apps/:id/variables', async (req, res) => {
res.send({
variables: [
{
key: 'book',
name: '书',
value: '《梦境解析》',
type: 'string'
},
{
key: 'myName',
name: '称呼',
value: '',
type: 'string'
}
],
})
})
}
module.exports = registerAPI
// const chatList = [
// {
// id: 1,
// content: 'AI 开场白',
// isAnswer: true,
// },
// {
// id: 2,
// content: '梦见在山上手撕鬼子,大师解解梦',
// more: { time: '5.6 秒' },
// },
// {
// id: 3,
// content: '梦境通常是个人内心深处的反映,很难确定每个人梦境的确切含义,因为它们可能会受到梦境者的文化背景、生活经验和情感状态等多种因素的影响。',
// isAnswer: true,
// more: { time: '99 秒' },
// },
// {
// id: 4,
// content: '梦见在山上手撕鬼子,大师解解梦',
// more: { time: '5.6 秒' },
// },
// {
// id: 5,
// content: '梦见在山上手撕鬼子,大师解解梦',
// more: { time: '5.6 秒' },
// },
// {
// id: 6,
// content: '梦见在山上手撕鬼子,大师解解梦',
// more: { time: '5.6 秒' },
// },
// ]

@ -1,15 +0,0 @@
const registerAPI = function (app) {
app.get('/demo', async (req, res) => {
res.send({
des: 'get res'
})
})
app.post('/demo', async (req, res) => {
res.send({
des: 'post res'
})
})
}
module.exports = registerAPI

@ -1,42 +0,0 @@
const express = require('express')
const app = express()
const bodyParser = require('body-parser')
var cors = require('cors')
const commonAPI = require('./api/common')
const demoAPI = require('./api/demo')
const appsApi = require('./api/apps')
const debugAPI = require('./api/debug')
const datasetsAPI = require('./api/datasets')
const port = 3001
app.use(bodyParser.json()) // for parsing application/json
app.use(bodyParser.urlencoded({ extended: true })) // for parsing application/x-www-form-urlencoded
const corsOptions = {
origin: true,
credentials: true,
}
app.use(cors(corsOptions)) // for cross origin
app.options('*', cors(corsOptions)) // include before other routes
demoAPI(app)
commonAPI(app)
appsApi(app)
debugAPI(app)
datasetsAPI(app)
app.get('/', (req, res) => {
res.send('rootpath')
})
app.listen(port, () => {
console.log(`Mock run on port ${port}`)
})
const sleep = (ms) => {
return new Promise(resolve => setTimeout(resolve, ms))
}

@ -1,26 +0,0 @@
{
"name": "server",
"version": "1.0.0",
"description": "",
"main": "index.js",
"scripts": {
"dev": "nodemon node app.js",
"start": "node app.js",
"tcp": "node tcp.js"
},
"keywords": [],
"author": "",
"license": "MIT",
"engines": {
"node": ">=16.0.0"
},
"dependencies": {
"body-parser": "^1.20.2",
"cors": "^2.8.5",
"express": "4.18.2",
"express-jwt": "8.4.1"
},
"devDependencies": {
"nodemon": "2.0.21"
}
}

@ -49,7 +49,7 @@ const AppDetailLayout: FC<IAppDetailLayoutProps> = (props) => {
return null return null
return ( return (
<div className={cn(s.app, 'flex', 'overflow-hidden')}> <div className={cn(s.app, 'flex', 'overflow-hidden')}>
<AppSideBar title={response.name} desc={appModeName} navigation={navigation} /> <AppSideBar title={response.name} icon={response.icon} icon_background={response.icon_background} desc={appModeName} navigation={navigation} />
<div className="bg-white grow">{children}</div> <div className="bg-white grow">{children}</div>
</div> </div>
) )

@ -47,7 +47,7 @@ const AppCard = ({
<> <>
<Link href={`/app/${app.id}/overview`} className={style.listItem}> <Link href={`/app/${app.id}/overview`} className={style.listItem}>
<div className={style.listItemTitle}> <div className={style.listItemTitle}>
<AppIcon size='small' /> <AppIcon size='small' icon={app.icon} background={app.icon_background}/>
<div className={style.listItemHeading}> <div className={style.listItemHeading}>
<div className={style.listItemHeadingContent}>{app.name}</div> <div className={style.listItemHeadingContent}>{app.name}</div>
</div> </div>

@ -17,6 +17,7 @@ const Apps = () => {
{apps.map(app => (<AppCard key={app.id} app={app} />))} {apps.map(app => (<AppCard key={app.id} app={app} />))}
<NewAppCard /> <NewAppCard />
</nav> </nav>
) )
} }

@ -9,7 +9,6 @@ import NewAppDialog from './NewAppDialog'
const CreateAppCard = () => { const CreateAppCard = () => {
const { t } = useTranslation() const { t } = useTranslation()
const [showNewAppDialog, setShowNewAppDialog] = useState(false) const [showNewAppDialog, setShowNewAppDialog] = useState(false)
return ( return (
<a className={classNames(style.listItem, style.newItemCard)} onClick={() => setShowNewAppDialog(true)}> <a className={classNames(style.listItem, style.newItemCard)} onClick={() => setShowNewAppDialog(true)}>
<div className={style.listItemTitle}> <div className={style.listItemTitle}>

@ -17,6 +17,8 @@ import { createApp, fetchAppTemplates } from '@/service/apps'
import AppIcon from '@/app/components/base/app-icon' import AppIcon from '@/app/components/base/app-icon'
import AppsContext from '@/context/app-context' import AppsContext from '@/context/app-context'
import EmojiPicker from '@/app/components/base/emoji-picker'
type NewAppDialogProps = { type NewAppDialogProps = {
show: boolean show: boolean
onClose?: () => void onClose?: () => void
@ -31,6 +33,11 @@ const NewAppDialog = ({ show, onClose }: NewAppDialogProps) => {
const [newAppMode, setNewAppMode] = useState<AppMode>() const [newAppMode, setNewAppMode] = useState<AppMode>()
const [isWithTemplate, setIsWithTemplate] = useState(false) const [isWithTemplate, setIsWithTemplate] = useState(false)
const [selectedTemplateIndex, setSelectedTemplateIndex] = useState<number>(-1) const [selectedTemplateIndex, setSelectedTemplateIndex] = useState<number>(-1)
// Emoji Picker
const [showEmojiPicker, setShowEmojiPicker] = useState(false)
const [emoji, setEmoji] = useState({ icon: '🍌', icon_background: '#FFEAD5' })
const mutateApps = useContextSelector(AppsContext, state => state.mutateApps) const mutateApps = useContextSelector(AppsContext, state => state.mutateApps)
const { data: templates, mutate } = useSWR({ url: '/app-templates' }, fetchAppTemplates) const { data: templates, mutate } = useSWR({ url: '/app-templates' }, fetchAppTemplates)
@ -67,6 +74,8 @@ const NewAppDialog = ({ show, onClose }: NewAppDialogProps) => {
try { try {
const app = await createApp({ const app = await createApp({
name, name,
icon: emoji.icon,
icon_background: emoji.icon_background,
mode: isWithTemplate ? templates.data[selectedTemplateIndex].mode : newAppMode!, mode: isWithTemplate ? templates.data[selectedTemplateIndex].mode : newAppMode!,
config: isWithTemplate ? templates.data[selectedTemplateIndex].model_config : undefined, config: isWithTemplate ? templates.data[selectedTemplateIndex].model_config : undefined,
}) })
@ -80,9 +89,20 @@ const NewAppDialog = ({ show, onClose }: NewAppDialogProps) => {
notify({ type: 'error', message: t('app.newApp.appCreateFailed') }) notify({ type: 'error', message: t('app.newApp.appCreateFailed') })
} }
isCreatingRef.current = false isCreatingRef.current = false
}, [isWithTemplate, newAppMode, notify, router, templates, selectedTemplateIndex]) }, [isWithTemplate, newAppMode, notify, router, templates, selectedTemplateIndex, emoji])
return ( return <>
{showEmojiPicker && <EmojiPicker
onSelect={(icon, icon_background) => {
console.log(icon, icon_background)
setEmoji({ icon, icon_background })
setShowEmojiPicker(false)
}}
onClose={() => {
setEmoji({ icon: '🍌', icon_background: '#FFEAD5' })
setShowEmojiPicker(false)
}}
/>}
<Dialog <Dialog
show={show} show={show}
title={t('app.newApp.startToCreate')} title={t('app.newApp.startToCreate')}
@ -96,7 +116,7 @@ const NewAppDialog = ({ show, onClose }: NewAppDialogProps) => {
<h3 className={style.newItemCaption}>{t('app.newApp.captionName')}</h3> <h3 className={style.newItemCaption}>{t('app.newApp.captionName')}</h3>
<div className='flex items-center justify-between gap-3 mb-8'> <div className='flex items-center justify-between gap-3 mb-8'>
<AppIcon size='large' /> <AppIcon size='large' onClick={() => { setShowEmojiPicker(true) }} className='cursor-pointer' icon={emoji.icon} background={emoji.icon_background} />
<input ref={nameInputRef} className='h-10 px-3 text-sm font-normal bg-gray-100 rounded-lg grow' /> <input ref={nameInputRef} className='h-10 px-3 text-sm font-normal bg-gray-100 rounded-lg grow' />
</div> </div>
@ -187,7 +207,7 @@ const NewAppDialog = ({ show, onClose }: NewAppDialogProps) => {
)} )}
</div> </div>
</Dialog> </Dialog>
) </>
} }
export default NewAppDialog export default NewAppDialog

@ -155,6 +155,8 @@ const DatasetDetailLayout: FC<IAppDetailLayoutProps> = (props) => {
<div className='flex' style={{ height: 'calc(100vh - 56px)' }}> <div className='flex' style={{ height: 'calc(100vh - 56px)' }}>
{!hideSideBar && <AppSideBar {!hideSideBar && <AppSideBar
title={datasetRes?.name || '--'} title={datasetRes?.name || '--'}
icon={datasetRes?.icon || 'https://static.dify.ai/images/dataset-default-icon.png'}
icon_background={datasetRes?.icon_background || '#F5F5F5'}
desc={datasetRes?.description || '--'} desc={datasetRes?.description || '--'}
navigation={navigation} navigation={navigation}
extraInfo={<ExtraInfo />} extraInfo={<ExtraInfo />}

@ -1,3 +0,0 @@
export async function GET(_request: Request) {
return new Response('Hello, Next.js!')
}

@ -15,7 +15,8 @@ export function randomString(length: number) {
export type IAppBasicProps = { export type IAppBasicProps = {
iconType?: 'app' | 'api' | 'dataset' iconType?: 'app' | 'api' | 'dataset'
iconUrl?: string icon?: string,
icon_background?: string,
name: string name: string
type: string | React.ReactNode type: string | React.ReactNode
hoverTip?: string hoverTip?: string
@ -41,15 +42,20 @@ const ICON_MAP = {
'dataset': <AppIcon innerIcon={DatasetSvg} className='!border-[0.5px] !border-indigo-100 !bg-indigo-25' /> 'dataset': <AppIcon innerIcon={DatasetSvg} className='!border-[0.5px] !border-indigo-100 !bg-indigo-25' />
} }
export default function AppBasic({ iconUrl, name, type, hoverTip, textStyle, iconType = 'app' }: IAppBasicProps) { export default function AppBasic({ icon, icon_background, name, type, hoverTip, textStyle, iconType = 'app' }: IAppBasicProps) {
return ( return (
<div className="flex items-start"> <div className="flex items-start">
{iconUrl && ( {icon && icon_background && iconType === 'app' && (
<div className='flex-shrink-0 mr-3'> <div className='flex-shrink-0 mr-3'>
{/* <img className="inline-block rounded-lg h-9 w-9" src={iconUrl} alt={name} /> */} <AppIcon icon={icon} background={icon_background} />
{ICON_MAP[iconType]}
</div> </div>
)} )}
{iconType !== 'app' &&
<div className='flex-shrink-0 mr-3'>
{ICON_MAP[iconType]}
</div>
}
<div className="group"> <div className="group">
<div className={`flex flex-row items-center text-sm font-semibold text-gray-700 group-hover:text-gray-900 ${textStyle?.main}`}> <div className={`flex flex-row items-center text-sm font-semibold text-gray-700 group-hover:text-gray-900 ${textStyle?.main}`}>
{name} {name}

@ -7,6 +7,8 @@ export type IAppDetailNavProps = {
iconType?: 'app' | 'dataset' iconType?: 'app' | 'dataset'
title: string title: string
desc: string desc: string
icon: string
icon_background: string
navigation: Array<{ navigation: Array<{
name: string name: string
href: string href: string
@ -16,13 +18,12 @@ export type IAppDetailNavProps = {
extraInfo?: React.ReactNode extraInfo?: React.ReactNode
} }
const sampleAppIconUrl = 'https://images.unsplash.com/photo-1472099645785-5658abf4ff4e?ixlib=rb-1.2.1&ixid=eyJhcHBfaWQiOjEyMDd9&auto=format&fit=facearea&facepad=2&w=256&h=256&q=80'
const AppDetailNav: FC<IAppDetailNavProps> = ({ title, desc, navigation, extraInfo, iconType = 'app' }) => { const AppDetailNav: FC<IAppDetailNavProps> = ({ title, desc, icon, icon_background, navigation, extraInfo, iconType = 'app' }) => {
return ( return (
<div className="flex flex-col w-56 overflow-y-auto bg-white border-r border-gray-200 shrink-0"> <div className="flex flex-col w-56 overflow-y-auto bg-white border-r border-gray-200 shrink-0">
<div className="flex flex-shrink-0 p-4"> <div className="flex flex-shrink-0 p-4">
<AppBasic iconType={iconType} iconUrl={sampleAppIconUrl} name={title} type={desc} /> <AppBasic iconType={iconType} icon={icon} icon_background={icon_background} name={title} type={desc} />
</div> </div>
<nav className="flex-1 p-4 space-y-1 bg-white"> <nav className="flex-1 p-4 space-y-1 bg-white">
{navigation.map((item, index) => { {navigation.map((item, index) => {

@ -29,9 +29,6 @@ export type IAppCardProps = {
onGenerateCode?: () => Promise<any> onGenerateCode?: () => Promise<any>
} }
// todo: get image url from appInfo
const defaultUrl = 'https://images.unsplash.com/photo-1472099645785-5658abf4ff4e?ixlib=rb-1.2.1&ixid=eyJhcHBfaWQiOjEyMDd9&auto=format&fit=facearea&facepad=2&w=256&h=256&q=80'
function AppCard({ function AppCard({
appInfo, appInfo,
cardType = 'app', cardType = 'app',
@ -104,7 +101,8 @@ function AppCard({
<div className="mb-2.5 flex flex-row items-start justify-between"> <div className="mb-2.5 flex flex-row items-start justify-between">
<AppBasic <AppBasic
iconType={isApp ? 'app' : 'api'} iconType={isApp ? 'app' : 'api'}
iconUrl={defaultUrl} icon={appInfo.icon}
icon_background={appInfo.icon_background}
name={basicName} name={basicName}
type={ type={
isApp isApp

@ -2,6 +2,11 @@ import type { FC } from 'react'
import classNames from 'classnames' import classNames from 'classnames'
import style from './style.module.css' import style from './style.module.css'
import data from '@emoji-mart/data'
import { init } from 'emoji-mart'
init({ data })
export type AppIconProps = { export type AppIconProps = {
size?: 'tiny' | 'small' | 'medium' | 'large' size?: 'tiny' | 'small' | 'medium' | 'large'
rounded?: boolean rounded?: boolean
@ -9,14 +14,17 @@ export type AppIconProps = {
background?: string background?: string
className?: string className?: string
innerIcon?: React.ReactNode innerIcon?: React.ReactNode
onClick?: () => void
} }
const AppIcon: FC<AppIconProps> = ({ const AppIcon: FC<AppIconProps> = ({
size = 'medium', size = 'medium',
rounded = false, rounded = false,
icon,
background, background,
className, className,
innerIcon, innerIcon,
onClick,
}) => { }) => {
return ( return (
<span <span
@ -29,8 +37,9 @@ const AppIcon: FC<AppIconProps> = ({
style={{ style={{
background, background,
}} }}
onClick={onClick}
> >
{innerIcon ? innerIcon : <>🤖</>} {innerIcon ? innerIcon : icon && icon !== '' ? <em-emoji id={icon} /> : <em-emoji id={'banana'} />}
</span> </span>
) )
} }

@ -0,0 +1,204 @@
'use client'
import data from '@emoji-mart/data'
import { init, SearchIndex } from 'emoji-mart'
// import AppIcon from '@/app/components/base/app-icon'
import cn from 'classnames'
import Divider from '@/app/components/base/divider'
import Button from '@/app/components/base/button'
import s from './style.module.css'
import { useState, FC, ChangeEvent } from 'react'
import {
MagnifyingGlassIcon
} from '@heroicons/react/24/outline'
import React from 'react'
import Modal from '@/app/components/base/modal'
declare global {
namespace JSX {
interface IntrinsicElements {
'em-emoji': React.DetailedHTMLProps<
React.HTMLAttributes<HTMLElement>,
HTMLElement
>;
}
}
}
init({ data })
async function search(value: string) {
const emojis = await SearchIndex.search(value) || []
const results = emojis.map((emoji: any) => {
return emoji.skins[0].native
})
return results
}
const backgroundColors = [
'#FFEAD5',
'#E4FBCC',
'#D3F8DF',
'#E0F2FE',
'#E0EAFF',
'#EFF1F5',
'#FBE8FF',
'#FCE7F6',
'#FEF7C3',
'#E6F4D7',
'#D5F5F6',
'#D1E9FF',
'#D1E0FF',
'#D5D9EB',
'#ECE9FE',
'#FFE4E8',
]
interface IEmojiPickerProps {
isModal?: boolean
onSelect?: (emoji: string, background: string) => void
onClose?: () => void
}
const EmojiPicker: FC<IEmojiPickerProps> = ({
isModal = true,
onSelect,
onClose
}) => {
const { categories } = data as any
const [selectedEmoji, setSelectedEmoji] = useState('')
const [selectedBackground, setSelectedBackground] = useState(backgroundColors[0])
const [searchedEmojis, setSearchedEmojis] = useState([])
const [isSearching, setIsSearching] = useState(false)
return isModal ? <Modal
onClose={() => { }}
isShow
closable={false}
className={cn(s.container, '!w-[362px] !p-0')}
>
<div className='flex flex-col items-center w-full p-3'>
<div className="relative w-full">
<div className="absolute inset-y-0 left-0 flex items-center pl-3 pointer-events-none">
<MagnifyingGlassIcon className="w-5 h-5 text-gray-400" aria-hidden="true" />
</div>
<input
type="search"
id="search"
className='block w-full h-10 px-3 pl-10 text-sm font-normal bg-gray-100 rounded-lg'
placeholder="Search emojis..."
onChange={async (e: ChangeEvent<HTMLInputElement>) => {
if (e.target.value === '') {
setIsSearching(false)
return
} else {
setIsSearching(true)
const emojis = await search(e.target.value)
setSearchedEmojis(emojis)
}
}}
/>
</div>
</div>
<Divider className='m-0 mb-3' />
<div className="w-full max-h-[200px] overflow-x-hidden overflow-y-auto px-3">
{isSearching && <>
<div key={`category-search`} className='flex flex-col'>
<p className='font-medium uppercase text-xs text-[#101828] mb-1'>Search</p>
<div className='w-full h-full grid grid-cols-8 gap-1'>
{searchedEmojis.map((emoji: string, index: number) => {
return <div
key={`emoji-search-${index}`}
className='inline-flex w-10 h-10 rounded-lg items-center justify-center'
onClick={() => {
setSelectedEmoji(emoji)
}}
>
<div className='cursor-pointer w-8 h-8 p-1 flex items-center justify-center rounded-lg hover:ring-1 ring-offset-1 ring-gray-300'>
<em-emoji id={emoji} />
</div>
</div>
})}
</div>
</div>
</>}
{categories.map((category: any, index: number) => {
return <div key={`category-${index}`} className='flex flex-col'>
<p className='font-medium uppercase text-xs text-[#101828] mb-1'>{category.id}</p>
<div className='w-full h-full grid grid-cols-8 gap-1'>
{category.emojis.map((emoji: string, index: number) => {
return <div
key={`emoji-${index}`}
className='inline-flex w-10 h-10 rounded-lg items-center justify-center'
onClick={() => {
setSelectedEmoji(emoji)
}}
>
<div className='cursor-pointer w-8 h-8 p-1 flex items-center justify-center rounded-lg hover:ring-1 ring-offset-1 ring-gray-300'>
<em-emoji id={emoji} />
</div>
</div>
})}
</div>
</div>
})}
</div>
{/* Color Select */}
<div className={cn('flex flex-col p-3 ', selectedEmoji == '' ? 'opacity-25' : '')}>
<p className='font-medium uppercase text-xs text-[#101828] mb-2'>Choose Style</p>
<div className='w-full h-full grid grid-cols-8 gap-1'>
{backgroundColors.map((color) => {
return <div
key={color}
className={
cn(
'cursor-pointer',
`hover:ring-1 ring-offset-1`,
'inline-flex w-10 h-10 rounded-lg items-center justify-center',
color === selectedBackground ? `ring-1 ring-gray-300` : '',
)}
onClick={() => {
setSelectedBackground(color)
}}
>
<div className={cn(
'w-8 h-8 p-1 flex items-center justify-center rounded-lg',
)
} style={{ background: color }}>
{selectedEmoji !== '' && <em-emoji id={selectedEmoji} />}
</div>
</div>
})}
</div>
</div>
<Divider className='m-0' />
<div className='w-full flex items-center justify-center p-3 gap-2'>
<Button type="default" className='w-full' onClick={() => {
onClose && onClose()
}}>
Cancel
</Button>
<Button
disabled={selectedEmoji == ''}
type="primary"
className='w-full'
onClick={() => {
onSelect && onSelect(selectedEmoji, selectedBackground)
}}>
OK
</Button>
</div>
</Modal> : <>
</>
}
export default EmojiPicker

@ -0,0 +1,12 @@
.container {
display: flex;
flex-direction: column;
align-items: flex-start;
width: 362px;
max-height: 552px;
border: 0.5px solid #EAECF0;
box-shadow: 0px 12px 16px -4px rgba(16, 24, 40, 0.08), 0px 4px 6px -2px rgba(16, 24, 40, 0.03);
border-radius: 12px;
background: #fff;
}

@ -25,51 +25,51 @@ export default function Modal({
closable = false, closable = false,
}: IModal) { }: IModal) {
return ( return (
<Transition appear show={isShow} as={Fragment}> <Transition appear show={isShow} as={Fragment}>
<Dialog as="div" className="relative z-10" onClose={onClose}> <Dialog as="div" className="relative z-10" onClose={onClose}>
<Transition.Child <Transition.Child
as={Fragment} as={Fragment}
enter="ease-out duration-300" enter="ease-out duration-300"
enterFrom="opacity-0" enterFrom="opacity-0"
enterTo="opacity-100" enterTo="opacity-100"
leave="ease-in duration-200" leave="ease-in duration-200"
leaveFrom="opacity-100" leaveFrom="opacity-100"
leaveTo="opacity-0" leaveTo="opacity-0"
> >
<div className="fixed inset-0 bg-black bg-opacity-25" /> <div className="fixed inset-0 bg-black bg-opacity-25" />
</Transition.Child> </Transition.Child>
<div className="fixed inset-0 overflow-y-auto"> <div className="fixed inset-0 overflow-y-auto">
<div className={`flex min-h-full items-center justify-center p-4 text-center ${wrapperClassName}`}> <div className="flex min-h-full items-center justify-center p-4 text-center">
<Transition.Child <Transition.Child
as={Fragment} as={Fragment}
enter="ease-out duration-300" enter="ease-out duration-300"
enterFrom="opacity-0 scale-95" enterFrom="opacity-0 scale-95"
enterTo="opacity-100 scale-100" enterTo="opacity-100 scale-100"
leave="ease-in duration-200" leave="ease-in duration-200"
leaveFrom="opacity-100 scale-100" leaveFrom="opacity-100 scale-100"
leaveTo="opacity-0 scale-95" leaveTo="opacity-0 scale-95"
> >
<Dialog.Panel className={`w-full max-w-md transform overflow-hidden rounded-2xl bg-white p-6 text-left align-middle shadow-xl transition-all ${className}`}> <Dialog.Panel className={`w-full max-w-md transform overflow-hidden rounded-2xl bg-white p-6 text-left align-middle shadow-xl transition-all ${className}`}>
{title && <Dialog.Title {title && <Dialog.Title
as="h3" as="h3"
className="text-lg font-medium leading-6 text-gray-900" className="text-lg font-medium leading-6 text-gray-900"
> >
{title} {title}
</Dialog.Title>} </Dialog.Title>}
{description && <Dialog.Description className='text-gray-500 text-xs font-normal mt-2'> {description && <Dialog.Description className='text-gray-500 text-xs font-normal mt-2'>
{description} {description}
</Dialog.Description>} </Dialog.Description>}
{closable {closable
&& <div className='absolute top-6 right-6 w-5 h-5 rounded-2xl flex items-center justify-center hover:cursor-pointer hover:bg-gray-100'> && <div className='absolute top-6 right-6 w-5 h-5 rounded-2xl flex items-center justify-center hover:cursor-pointer hover:bg-gray-100'>
<XMarkIcon className='w-4 h-4 text-gray-500' onClick={onClose} /> <XMarkIcon className='w-4 h-4 text-gray-500' onClick={onClose} />
</div>} </div>}
{children} {children}
</Dialog.Panel> </Dialog.Panel>
</Transition.Child> </Transition.Child>
</div> </div>
</div> </div>
</Dialog> </Dialog>
</Transition> </Transition>
) )
} }

@ -69,7 +69,7 @@ type IDocumentsProps = {
datasetId: string datasetId: string
} }
export const fetcher = (url: string) => get(url, {}, { isMock: true }) export const fetcher = (url: string) => get(url, {}, {})
const Documents: FC<IDocumentsProps> = ({ datasetId }) => { const Documents: FC<IDocumentsProps> = ({ datasetId }) => {
const { t } = useTranslation() const { t } = useTranslation()

@ -20,7 +20,7 @@ const AzureProvider = ({
const [token, setToken] = useState(provider.token as ProviderAzureToken || {}) const [token, setToken] = useState(provider.token as ProviderAzureToken || {})
const handleFocus = () => { const handleFocus = () => {
if (token === provider.token) { if (token === provider.token) {
token.azure_api_key = '' token.openai_api_key = ''
setToken({...token}) setToken({...token})
onTokenChange({...token}) onTokenChange({...token})
} }
@ -35,31 +35,17 @@ const AzureProvider = ({
<div className='px-4 py-3'> <div className='px-4 py-3'>
<ProviderInput <ProviderInput
className='mb-4' className='mb-4'
name={t('common.provider.azure.resourceName')} name={t('common.provider.azure.apiBase')}
placeholder={t('common.provider.azure.resourceNamePlaceholder')} placeholder={t('common.provider.azure.apiBasePlaceholder')}
value={token.azure_api_base} value={token.openai_api_base}
onChange={(v) => handleChange('azure_api_base', v)} onChange={(v) => handleChange('openai_api_base', v)}
/>
<ProviderInput
className='mb-4'
name={t('common.provider.azure.deploymentId')}
placeholder={t('common.provider.azure.deploymentIdPlaceholder')}
value={token.azure_api_type}
onChange={v => handleChange('azure_api_type', v)}
/>
<ProviderInput
className='mb-4'
name={t('common.provider.azure.apiVersion')}
placeholder={t('common.provider.azure.apiVersionPlaceholder')}
value={token.azure_api_version}
onChange={v => handleChange('azure_api_version', v)}
/> />
<ProviderValidateTokenInput <ProviderValidateTokenInput
className='mb-4' className='mb-4'
name={t('common.provider.azure.apiKey')} name={t('common.provider.azure.apiKey')}
placeholder={t('common.provider.azure.apiKeyPlaceholder')} placeholder={t('common.provider.azure.apiKeyPlaceholder')}
value={token.azure_api_key} value={token.openai_api_key}
onChange={v => handleChange('azure_api_key', v)} onChange={v => handleChange('openai_api_key', v)}
onFocus={handleFocus} onFocus={handleFocus}
onValidatedStatus={onValidatedStatus} onValidatedStatus={onValidatedStatus}
providerName={provider.provider_name} providerName={provider.provider_name}
@ -72,4 +58,4 @@ const AzureProvider = ({
) )
} }
export default AzureProvider export default AzureProvider

@ -33,12 +33,12 @@ const ProviderItem = ({
const { notify } = useContext(ToastContext) const { notify } = useContext(ToastContext)
const [token, setToken] = useState<ProviderAzureToken | string>( const [token, setToken] = useState<ProviderAzureToken | string>(
provider.provider_name === 'azure_openai' provider.provider_name === 'azure_openai'
? { azure_api_base: '', azure_api_type: '', azure_api_version: '', azure_api_key: '' } ? { openai_api_base: '', openai_api_key: '' }
: '' : ''
) )
const id = `${provider.provider_name}-${provider.provider_type}` const id = `${provider.provider_name}-${provider.provider_type}`
const isOpen = id === activeId const isOpen = id === activeId
const providerKey = provider.provider_name === 'azure_openai' ? (provider.token as ProviderAzureToken)?.azure_api_key : provider.token const providerKey = provider.provider_name === 'azure_openai' ? (provider.token as ProviderAzureToken)?.openai_api_key : provider.token
const comingSoon = false const comingSoon = false
const isValid = provider.is_valid const isValid = provider.is_valid
@ -135,4 +135,4 @@ const ProviderItem = ({
) )
} }
export default ProviderItem export default ProviderItem

@ -84,11 +84,13 @@ const Header: FC<IHeaderProps> = ({ appItems, curApp, userProfile, onLogout, lan
text={t('common.menus.apps')} text={t('common.menus.apps')}
activeSegment={['apps', 'app']} activeSegment={['apps', 'app']}
link='/apps' link='/apps'
curNav={curApp && { id: curApp.id, name: curApp.name }} curNav={curApp && { id: curApp.id, name: curApp.name ,icon: curApp.icon, icon_background: curApp.icon_background}}
navs={appItems.map(item => ({ navs={appItems.map(item => ({
id: item.id, id: item.id,
name: item.name, name: item.name,
link: `/app/${item.id}/overview` link: `/app/${item.id}/overview`,
icon: item.icon,
icon_background: item.icon_background
}))} }))}
createText={t('common.menus.newApp')} createText={t('common.menus.newApp')}
onCreate={() => setShowNewAppDialog(true)} onCreate={() => setShowNewAppDialog(true)}
@ -106,11 +108,13 @@ const Header: FC<IHeaderProps> = ({ appItems, curApp, userProfile, onLogout, lan
text={t('common.menus.datasets')} text={t('common.menus.datasets')}
activeSegment='datasets' activeSegment='datasets'
link='/datasets' link='/datasets'
curNav={currentDataset && { id: currentDataset.id, name: currentDataset.name }} curNav={currentDataset && { id: currentDataset.id, name: currentDataset.name, icon: currentDataset.icon, icon_background: currentDataset.icon_background }}
navs={datasets.map(dataset => ({ navs={datasets.map(dataset => ({
id: dataset.id, id: dataset.id,
name: dataset.name, name: dataset.name,
link: `/datasets/${dataset.id}/documents` link: `/datasets/${dataset.id}/documents`,
icon: dataset.icon,
icon_background: dataset.icon_background
}))} }))}
createText={t('common.menus.newDataset')} createText={t('common.menus.newDataset')}
onCreate={() => router.push('/datasets/create')} onCreate={() => router.push('/datasets/create')}

@ -10,6 +10,8 @@ type NavItem = {
id: string id: string
name: string name: string
link: string link: string
icon: string
icon_background: string
} }
export interface INavSelectorProps { export interface INavSelectorProps {
navs: NavItem[] navs: NavItem[]
@ -66,7 +68,7 @@ const NavSelector = ({ curNav, navs, createText, onCreate }: INavSelectorProps)
<Menu.Item key={nav.id}> <Menu.Item key={nav.id}>
<div className={itemClassName} onClick={() => router.push(nav.link)}> <div className={itemClassName} onClick={() => router.push(nav.link)}>
<div className='relative w-6 h-6 mr-2 bg-[#D5F5F6] rounded-[6px]'> <div className='relative w-6 h-6 mr-2 bg-[#D5F5F6] rounded-[6px]'>
<AppIcon size='tiny' /> <AppIcon size='tiny' icon={nav.icon} background={nav.icon_background}/>
<div className='flex justify-center items-center absolute -right-0.5 -bottom-0.5 w-2.5 h-2.5 bg-white rounded'> <div className='flex justify-center items-center absolute -right-0.5 -bottom-0.5 w-2.5 h-2.5 bg-white rounded'>
<Indicator /> <Indicator />
</div> </div>
@ -102,4 +104,4 @@ const NavSelector = ({ curNav, navs, createText, onCreate }: INavSelectorProps)
) )
} }
export default NavSelector export default NavSelector

@ -458,6 +458,8 @@ const Main: FC<IMainProps> = ({
<div className='bg-gray-100'> <div className='bg-gray-100'>
<Header <Header
title={siteInfo.title} title={siteInfo.title}
icon={siteInfo.icon || ''}
icon_background={siteInfo.icon_background || '#FFEAD5'}
isMobile={isMobile} isMobile={isMobile}
onShowSideBar={showSidebar} onShowSideBar={showSidebar}
onCreateNewChat={() => handleConversationIdChange('-1')} onCreateNewChat={() => handleConversationIdChange('-1')}

@ -7,6 +7,8 @@ import {
} from '@heroicons/react/24/solid' } from '@heroicons/react/24/solid'
export type IHeaderProps = { export type IHeaderProps = {
title: string title: string
icon: string
icon_background: string
isMobile?: boolean isMobile?: boolean
onShowSideBar?: () => void onShowSideBar?: () => void
onCreateNewChat?: () => void onCreateNewChat?: () => void
@ -14,6 +16,8 @@ export type IHeaderProps = {
const Header: FC<IHeaderProps> = ({ const Header: FC<IHeaderProps> = ({
title, title,
isMobile, isMobile,
icon,
icon_background,
onShowSideBar, onShowSideBar,
onCreateNewChat, onCreateNewChat,
}) => { }) => {
@ -28,7 +32,7 @@ const Header: FC<IHeaderProps> = ({
</div> </div>
) : <div></div>} ) : <div></div>}
<div className='flex items-center space-x-2'> <div className='flex items-center space-x-2'>
<AppIcon size="small" /> <AppIcon size="small" icon={icon} background={icon_background} />
<div className=" text-sm text-gray-800 font-bold">{title}</div> <div className=" text-sm text-gray-800 font-bold">{title}</div>
</div> </div>
{isMobile ? ( {isMobile ? (

@ -31,9 +31,6 @@ if (process.env.NEXT_PUBLIC_API_PREFIX && process.env.NEXT_PUBLIC_PUBLIC_API_PRE
export const API_PREFIX: string = apiPrefix; export const API_PREFIX: string = apiPrefix;
export const PUBLIC_API_PREFIX: string = publicApiPrefix; export const PUBLIC_API_PREFIX: string = publicApiPrefix;
// mock server
export const MOCK_API_PREFIX = 'http://127.0.0.1:3001'
const EDITION = process.env.NEXT_PUBLIC_EDITION || globalThis.document?.body?.getAttribute('data-public-edition') const EDITION = process.env.NEXT_PUBLIC_EDITION || globalThis.document?.body?.getAttribute('data-public-edition')
export const IS_CE_EDITION = EDITION === 'SELF_HOSTED' export const IS_CE_EDITION = EDITION === 'SELF_HOSTED'

@ -150,12 +150,8 @@ const translation = {
editKey: 'Edit', editKey: 'Edit',
invalidApiKey: 'Invalid API key', invalidApiKey: 'Invalid API key',
azure: { azure: {
resourceName: 'Resource Name', apiBase: 'API Base',
resourceNamePlaceholder: 'The name of your Azure OpenAI Resource.', apiBasePlaceholder: 'The API Base URL of your Azure OpenAI Resource.',
deploymentId: 'Deployment ID',
deploymentIdPlaceholder: 'The deployment name you chose when you deployed the model.',
apiVersion: 'API Version',
apiVersionPlaceholder: 'The API version to use for this operation.',
apiKey: 'API Key', apiKey: 'API Key',
apiKeyPlaceholder: 'Enter your API key here', apiKeyPlaceholder: 'Enter your API key here',
helpTip: 'Learn Azure OpenAI Service', helpTip: 'Learn Azure OpenAI Service',

@ -151,14 +151,10 @@ const translation = {
editKey: '编辑', editKey: '编辑',
invalidApiKey: '无效的 API 密钥', invalidApiKey: '无效的 API 密钥',
azure: { azure: {
resourceName: 'Resource Name', apiBase: 'API Base',
resourceNamePlaceholder: 'The name of your Azure OpenAI Resource.', apiBasePlaceholder: '输入您的 Azure OpenAI API Base 地址',
deploymentId: 'Deployment ID',
deploymentIdPlaceholder: 'The deployment name you chose when you deployed the model.',
apiVersion: 'API Version',
apiVersionPlaceholder: 'The API version to use for this operation.',
apiKey: 'API Key', apiKey: 'API Key',
apiKeyPlaceholder: 'Enter your API key here', apiKeyPlaceholder: '输入你的 API 密钥',
helpTip: '了解 Azure OpenAI Service', helpTip: '了解 Azure OpenAI Service',
}, },
openaiHosted: { openaiHosted: {

@ -55,10 +55,8 @@ export type Member = Pick<UserProfileResponse, 'id' | 'name' | 'email' | 'last_l
} }
export type ProviderAzureToken = { export type ProviderAzureToken = {
azure_api_base: string openai_api_base: string
azure_api_key: string openai_api_key: string
azure_api_type: string
azure_api_version: string
} }
export type Provider = { export type Provider = {
provider_name: string provider_name: string

@ -3,6 +3,8 @@ import { AppMode } from './app'
export type DataSet = { export type DataSet = {
id: string id: string
name: string name: string
icon: string
icon_background: string
description: string description: string
permission: 'only_me' | 'all_team_members' permission: 'only_me' | 'all_team_members'
data_source_type: 'upload_file' data_source_type: 'upload_file'

@ -11,6 +11,8 @@ export type ConversationItem = {
export type SiteInfo = { export type SiteInfo = {
title: string title: string
icon: string
icon_background: string
description: string description: string
default_language: Locale default_language: Locale
prompt_public: boolean prompt_public: boolean

@ -10,6 +10,7 @@
"fix": "next lint --fix" "fix": "next lint --fix"
}, },
"dependencies": { "dependencies": {
"@emoji-mart/data": "^1.1.2",
"@formatjs/intl-localematcher": "^0.2.32", "@formatjs/intl-localematcher": "^0.2.32",
"@headlessui/react": "^1.7.13", "@headlessui/react": "^1.7.13",
"@heroicons/react": "^2.0.16", "@heroicons/react": "^2.0.16",
@ -33,6 +34,7 @@
"dayjs": "^1.11.7", "dayjs": "^1.11.7",
"echarts": "^5.4.1", "echarts": "^5.4.1",
"echarts-for-react": "^3.0.2", "echarts-for-react": "^3.0.2",
"emoji-mart": "^5.5.2",
"eslint": "8.36.0", "eslint": "8.36.0",
"eslint-config-next": "13.2.4", "eslint-config-next": "13.2.4",
"i18next": "^22.4.13", "i18next": "^22.4.13",

@ -16,8 +16,8 @@ export const fetchAppTemplates: Fetcher<AppTemplatesResponse, { url: string }> =
return get(url) as Promise<AppTemplatesResponse> return get(url) as Promise<AppTemplatesResponse>
} }
export const createApp: Fetcher<AppDetailResponse, { name: string; mode: AppMode; icon?: string, icon_background?: string, config?: ModelConfig }> = ({ name, icon, icon_background, mode, config }) => { export const createApp: Fetcher<AppDetailResponse, { name: string; icon: string, icon_background: string, mode: AppMode; config?: ModelConfig }> = ({ name, icon, icon_background, mode, config }) => {
return post('apps', { body: { name, mode, icon, icon_background, model_config: config } }) as Promise<AppDetailResponse> return post('apps', { body: { name, icon, icon_background, mode, model_config: config } }) as Promise<AppDetailResponse>
} }
export const deleteApp: Fetcher<CommonResponse, string> = (appID) => { export const deleteApp: Fetcher<CommonResponse, string> = (appID) => {

@ -1,4 +1,4 @@
import { API_PREFIX, MOCK_API_PREFIX, PUBLIC_API_PREFIX, IS_CE_EDITION } from '@/config' import { API_PREFIX, PUBLIC_API_PREFIX, IS_CE_EDITION } from '@/config'
import Toast from '@/app/components/base/toast' import Toast from '@/app/components/base/toast'
const TIME_OUT = 100000 const TIME_OUT = 100000
@ -33,7 +33,6 @@ export type IOnError = (msg: string) => void
type IOtherOptions = { type IOtherOptions = {
isPublicAPI?: boolean isPublicAPI?: boolean
isMock?: boolean
needAllResponseContent?: boolean needAllResponseContent?: boolean
onData?: IOnData // for stream onData?: IOnData // for stream
onError?: IOnError onError?: IOnError
@ -116,7 +115,14 @@ const handleStream = (response: any, onData: IOnData, onCompleted?: IOnCompleted
read() read()
} }
const baseFetch = (url: string, fetchOptions: any, { isPublicAPI = false, isMock = false, needAllResponseContent }: IOtherOptions) => { const baseFetch = (
url: string,
fetchOptions: any,
{
isPublicAPI = false,
needAllResponseContent
}: IOtherOptions
) => {
const options = Object.assign({}, baseOptions, fetchOptions) const options = Object.assign({}, baseOptions, fetchOptions)
if (isPublicAPI) { if (isPublicAPI) {
const sharedToken = globalThis.location.pathname.split('/').slice(-1)[0] const sharedToken = globalThis.location.pathname.split('/').slice(-1)[0]
@ -124,9 +130,6 @@ const baseFetch = (url: string, fetchOptions: any, { isPublicAPI = false, isMock
} }
let urlPrefix = isPublicAPI ? PUBLIC_API_PREFIX : API_PREFIX let urlPrefix = isPublicAPI ? PUBLIC_API_PREFIX : API_PREFIX
if (isMock)
urlPrefix = MOCK_API_PREFIX
let urlWithPrefix = `${urlPrefix}${url.startsWith('/') ? url : `/${url}`}` let urlWithPrefix = `${urlPrefix}${url.startsWith('/') ? url : `/${url}`}`
const { method, params, body } = options const { method, params, body } = options

@ -190,6 +190,12 @@ export type App = {
id: string id: string
/** Name */ /** Name */
name: string name: string
/** Icon */
icon: string
/** Icon Background */
icon_background: string
/** Mode */ /** Mode */
mode: AppMode mode: AppMode
/** Enable web app */ /** Enable web app */

Loading…
Cancel
Save