Merge branch 'main' of https://github.com/eosphoros-ai/DB-GPT into more_tutorials

docs/docs/quickstart.md

@@ -90,6 +90,7 @@ uv --version
 ```bash
 # Use uv to install dependencies needed for OpenAI proxy
 uv sync --all-packages --frozen \
+--extra "base" \
 --extra "proxy_openai" \
 --extra "rag" \
 --extra "storage_chromadb" \
@@ -177,9 +178,10 @@ uv run python packages/dbgpt-app/src/dbgpt_app/dbgpt_server.py --config configs/
 # Use uv to install dependencies needed for GLM4
 # Install core dependencies and select desired extensions
 uv sync --all-packages --frozen \
+--extra "base" \
+--extra "hf" \
 --extra "rag" \
 --extra "storage_chromadb" \
---extra "hf" \
 --extra "quant_bnb" \
 --extra "dbgpts"
 ```

packages/dbgpt-accelerator/pyproject.toml

@@ -42,7 +42,6 @@ auto = [
     "torchvision>=0.17.1; sys_platform != 'darwin' or platform_machine != 'x86_64'",
     "torchvision>=0.17.1,<0.18.0; sys_platform == 'darwin' and platform_machine == 'x86_64'",
 ]
-
 cpu = [
     # Starting from PyTorch 2.3.0, the minimum requirement for macOS is macOS 11.0+ ARM64.
     "torch>=2.2.1; sys_platform != 'darwin' or platform_machine != 'x86_64'",
@@ -52,7 +51,6 @@ cpu = [
     "torchvision>=0.17.1; sys_platform != 'darwin' or platform_machine != 'x86_64'",
     "torchvision>=0.17.1,<0.18.0; sys_platform == 'darwin' and platform_machine == 'x86_64'",
 ]
-
 cuda118 = [
     "torch>=2.2.1; sys_platform == 'win32' or sys_platform == 'linux'",
     "torchvision>=0.17.1; sys_platform == 'win32' or sys_platform == 'linux'",
@@ -69,7 +67,6 @@ cuda124 = [
     "torchvision>=0.17.1; sys_platform == 'win32' or sys_platform == 'linux'",
     "torchaudio>=2.2.1; sys_platform == 'win32' or sys_platform == 'linux'",
 ]
-
 #rocm60 = [
 #    "torch>=2.3.0,<2.4.0;sys_platform == 'linux'",
 #    "torchvision>=0.18.0,<0.19.0;sys_platform == 'linux'",
@@ -77,7 +74,7 @@ cuda124 = [
 #]
 vllm = [
     # Just support GPU version on Linux
-    "vllm>=0.7.0; sys_platform == 'linux'"
+    "vllm>=0.7.0; sys_platform == 'linux'",
 ]
 #vllm_pascal = [
 #    # https://github.com/sasha0552/pascal-pkgs-ci
@@ -178,7 +175,6 @@ torch = [
     { index = "pytorch-cu124", marker = "platform_system == 'Linux'", extra = "cuda124" },
 #    { index = "pytorch-rocm60", marker = "platform_system == 'Linux'", extra = "rocm60" },
 ]
-
 torchvision = [
     { index = "pytorch-cpu", marker = "platform_system == 'Darwin'" },
     { index = "pytorch-cpu", marker = "platform_system == 'Windows'", extra = "cpu" },

packages/dbgpt-app/pyproject.toml

@@ -33,22 +33,13 @@ build-backend = "hatchling.build"
 cache = [
     "rocksdict",
 ]
-
 observability = [
     "opentelemetry-api",
     "opentelemetry-sdk",
     "opentelemetry-exporter-otlp",
 ]
-
-proxy_openai = [
-    "dbgpt[client,cli,agent,simple_framework,framework,code,proxy_openai]",
-    "dbgpt-ext"
-]
-
-default = [
-    "dbgpt[client,cli,agent,simple_framework,framework,code]",
-    "dbgpt-ext[rag]",
-    "dbgpt-ext[storage_chromadb]",
+base = [
+    "dbgpt[client,cli,agent,simple_framework,framework,code]"
 ]
 dbgpts = [
     # For build dbgpts apps, we will be removed in the future.

packages/dbgpt-core/pyproject.toml

@@ -113,11 +113,11 @@ code = [
     "lyric-component-ts-transpiling>=0.1.6",
 ]
 llama_cpp = [
-    "llama-cpp-python"
+    "llama-cpp-python",
 ]
 llama_cpp_server = [
     "llama-cpp-server-py-core>=0.1.3",
-    "llama-cpp-server-py>=0.1.3"
+    "llama-cpp-server-py>=0.1.3",
 ]
 proxy_ollama = ["ollama"]
 proxy_zhipuai = ["zhipuai>=2.1.5"]

packages/dbgpt-ext/pyproject.toml

@@ -69,7 +69,10 @@ datasource_duckdb = [
 # datasource_doris = ["pydoris>=1.0.2,<2.0.0"]
 storage_milvus = ["pymilvus"]
 storage_weaviate = ["weaviate-client"]
-storage_chromadb = ["chromadb>=0.4.22"]
+storage_chromadb = [
+    "onnxruntime>=1.14.1,<=1.18.1",
+    "chromadb>=0.4.22"
+    ]
 storage_elasticsearch = ["elasticsearch"]
 storage_obvector = ["pyobvector"]
 proxy_tongyi = [

packages/dbgpt-ext/src/dbgpt_ext/datasource/rdbms/conn_clickhouse.py

@@ -36,9 +36,13 @@ class ClickhouseParameters(BaseDatasourceParameters):
     __type__ = "clickhouse"
 
     host: str = field(metadata={"help": _("Database host, e.g., localhost")})
-    port: int = field(metadata={"help": _("Database port, e.g., 3306")})
+    port: int = field(metadata={"help": _("Database port, e.g., 8123")})
     user: str = field(metadata={"help": _("Database user to connect")})
     database: str = field(metadata={"help": _("Database name")})
+    engine: str = field(
+        default="MergeTree", 
+        metadata={"help": _("Storage engine, e.g., MergeTree")}
+    )
     password: str = field(
         default="${env:DBGPT_DB_PASSWORD}",
         metadata={
@@ -118,6 +122,7 @@ class ClickhouseConnector(RDBMSConnector):
             parameters.http_pool_num_pools,
             parameters.connect_timeout,
             parameters.distributed_ddl_task_timeout,
+            parameters.engine,
         )
 
     @classmethod
@@ -132,7 +137,7 @@ class ClickhouseConnector(RDBMSConnector):
         http_pool_num_pools: int = 12,
         connect_timeout: int = 15,
         distributed_ddl_task_timeout: int = 300,
-        engine_args: Optional[dict] = None,
+        engine: str = "MergeTree",
         **kwargs: Any,
     ) -> "ClickhouseConnector":
         """Create a new ClickhouseConnector from host, port, user, pwd, db_name."""
@@ -156,7 +161,7 @@ class ClickhouseConnector(RDBMSConnector):
         )
 
         cls.client = client
-        return cls(client, **kwargs)
+        return cls(client, engine=engine, **kwargs)
 
     def get_table_names(self):
         """Get all table names."""

packages/dbgpt-serve/pyproject.toml

@@ -31,7 +31,6 @@ dev-dependencies = []
 libro = ["libro>=0.1.25"]
 dbgpts = ["poetry"]
 
-
 [tool.hatch.build.targets.wheel]
 packages = ["src/dbgpt_serve"]
 exclude = [