data-designer 0.4.0rc2__py3-none-any.whl → 0.5.0rc1__py3-none-any.whl

data_designer/cli/repositories/mcp_provider_repository.py

@@ -0,0 +1,48 @@
+# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from pathlib import Path
+
+from pydantic import BaseModel
+
+from data_designer.cli.repositories.base import ConfigRepository
+from data_designer.config.mcp import MCPProviderT
+from data_designer.config.utils.constants import MCP_PROVIDERS_FILE_NAME
+from data_designer.config.utils.io_helpers import load_config_file, save_config_file
+
+
+class MCPProviderRegistry(BaseModel):
+    """Registry for MCP provider configurations.
+
+    Supports both MCPProvider (remote SSE) and LocalStdioMCPProvider (subprocess)
+    via the discriminated union MCPProviderT.
+    """
+
+    providers: list[MCPProviderT]
+
+
+class MCPProviderRepository(ConfigRepository[MCPProviderRegistry]):
+    """Repository for MCP provider configurations."""
+
+    @property
+    def config_file(self) -> Path:
+        """Get the MCP provider configuration file path."""
+        return self.config_dir / MCP_PROVIDERS_FILE_NAME
+
+    def load(self) -> MCPProviderRegistry | None:
+        """Load MCP provider configuration from file."""
+        if not self.exists():
+            return None
+
+        try:
+            config_dict = load_config_file(self.config_file)
+            return MCPProviderRegistry.model_validate(config_dict)
+        except Exception:
+            return None
+
+    def save(self, config: MCPProviderRegistry) -> None:
+        """Save MCP provider configuration to file."""
+        config_dict = config.model_dump(mode="json", exclude_none=True)
+        save_config_file(self.config_file, config_dict)

data_designer/cli/repositories/tool_repository.py

@@ -0,0 +1,44 @@
+# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from pathlib import Path
+
+from pydantic import BaseModel
+
+from data_designer.cli.repositories.base import ConfigRepository
+from data_designer.config.mcp import ToolConfig
+from data_designer.config.utils.constants import TOOL_CONFIGS_FILE_NAME
+from data_designer.config.utils.io_helpers import load_config_file, save_config_file
+
+
+class ToolConfigRegistry(BaseModel):
+    """Registry for tool configurations."""
+
+    tool_configs: list[ToolConfig]
+
+
+class ToolRepository(ConfigRepository[ToolConfigRegistry]):
+    """Repository for tool configurations."""
+
+    @property
+    def config_file(self) -> Path:
+        """Get the tool configuration file path."""
+        return self.config_dir / TOOL_CONFIGS_FILE_NAME
+
+    def load(self) -> ToolConfigRegistry | None:
+        """Load tool configuration from file."""
+        if not self.exists():
+            return None
+
+        try:
+            config_dict = load_config_file(self.config_file)
+            return ToolConfigRegistry.model_validate(config_dict)
+        except Exception:
+            return None
+
+    def save(self, config: ToolConfigRegistry) -> None:
+        """Save tool configuration to file."""
+        config_dict = config.model_dump(mode="json", exclude_none=True)
+        save_config_file(self.config_file, config_dict)

data_designer/cli/services/mcp_provider_service.py

@@ -0,0 +1,86 @@
+# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from data_designer.cli.repositories.mcp_provider_repository import MCPProviderRegistry, MCPProviderRepository
+from data_designer.config.mcp import MCPProviderT
+
+
+class MCPProviderService:
+    """Business logic for MCP provider management.
+
+    Supports both MCPProvider (remote SSE) and LocalStdioMCPProvider (subprocess).
+    """
+
+    def __init__(self, repository: MCPProviderRepository):
+        self.repository = repository
+
+    def list_all(self) -> list[MCPProviderT]:
+        """Get all configured MCP providers."""
+        registry = self.repository.load()
+        return list(registry.providers) if registry else []
+
+    def get_by_name(self, name: str) -> MCPProviderT | None:
+        """Get an MCP provider by name."""
+        providers = self.list_all()
+        return next((p for p in providers if p.name == name), None)
+
+    def add(self, provider: MCPProviderT) -> None:
+        """Add a new MCP provider.
+
+        Raises:
+            ValueError: If provider name already exists
+        """
+        registry = self.repository.load() or MCPProviderRegistry(providers=[])
+
+        if any(p.name == provider.name for p in registry.providers):
+            raise ValueError(f"MCP provider '{provider.name}' already exists")
+
+        registry.providers.append(provider)
+        self.repository.save(registry)
+
+    def update(self, original_name: str, updated_provider: MCPProviderT) -> None:
+        """Update an existing MCP provider.
+
+        Raises:
+            ValueError: If provider not found or new name already exists
+        """
+        registry = self.repository.load()
+        if not registry:
+            raise ValueError("No MCP providers configured")
+
+        # Find provider index
+        index = next(
+            (i for i, p in enumerate(registry.providers) if p.name == original_name),
+            None,
+        )
+        if index is None:
+            raise ValueError(f"MCP provider '{original_name}' not found")
+
+        if updated_provider.name != original_name:
+            if any(p.name == updated_provider.name for p in registry.providers):
+                raise ValueError(f"MCP provider name '{updated_provider.name}' already exists")
+
+        registry.providers[index] = updated_provider
+        self.repository.save(registry)
+
+    def delete(self, name: str) -> None:
+        """Delete an MCP provider.
+
+        Raises:
+            ValueError: If provider not found
+        """
+        registry = self.repository.load()
+        if not registry:
+            raise ValueError("No MCP providers configured")
+
+        if not any(p.name == name for p in registry.providers):
+            raise ValueError(f"MCP provider '{name}' not found")
+
+        registry.providers = [p for p in registry.providers if p.name != name]
+
+        if registry.providers:
+            self.repository.save(registry)
+        else:
+            self.repository.delete()

data_designer/cli/services/tool_service.py

@@ -0,0 +1,83 @@
+# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from data_designer.cli.repositories.tool_repository import ToolConfigRegistry, ToolRepository
+from data_designer.config.mcp import ToolConfig
+
+
+class ToolService:
+    """Business logic for tool configuration management."""
+
+    def __init__(self, repository: ToolRepository):
+        self.repository = repository
+
+    def list_all(self) -> list[ToolConfig]:
+        """Get all configured tool configs."""
+        registry = self.repository.load()
+        return list(registry.tool_configs) if registry else []
+
+    def get_by_alias(self, alias: str) -> ToolConfig | None:
+        """Get a tool config by alias."""
+        configs = self.list_all()
+        return next((c for c in configs if c.tool_alias == alias), None)
+
+    def add(self, tool_config: ToolConfig) -> None:
+        """Add a new tool config.
+
+        Raises:
+            ValueError: If tool alias already exists
+        """
+        registry = self.repository.load() or ToolConfigRegistry(tool_configs=[])
+
+        if any(c.tool_alias == tool_config.tool_alias for c in registry.tool_configs):
+            raise ValueError(f"Tool config with alias '{tool_config.tool_alias}' already exists")
+
+        registry.tool_configs.append(tool_config)
+        self.repository.save(registry)
+
+    def update(self, original_alias: str, updated_config: ToolConfig) -> None:
+        """Update an existing tool config.
+
+        Raises:
+            ValueError: If tool config not found or new alias already exists
+        """
+        registry = self.repository.load()
+        if not registry:
+            raise ValueError("No tool configs configured")
+
+        # Find config index
+        index = next(
+            (i for i, c in enumerate(registry.tool_configs) if c.tool_alias == original_alias),
+            None,
+        )
+        if index is None:
+            raise ValueError(f"Tool config with alias '{original_alias}' not found")
+
+        if updated_config.tool_alias != original_alias:
+            if any(c.tool_alias == updated_config.tool_alias for c in registry.tool_configs):
+                raise ValueError(f"Tool config alias '{updated_config.tool_alias}' already exists")
+
+        registry.tool_configs[index] = updated_config
+        self.repository.save(registry)
+
+    def delete(self, alias: str) -> None:
+        """Delete a tool config.
+
+        Raises:
+            ValueError: If tool config not found
+        """
+        registry = self.repository.load()
+        if not registry:
+            raise ValueError("No tool configs configured")
+
+        if not any(c.tool_alias == alias for c in registry.tool_configs):
+            raise ValueError(f"Tool config with alias '{alias}' not found")
+
+        registry.tool_configs = [c for c in registry.tool_configs if c.tool_alias != alias]
+
+        if registry.tool_configs:
+            self.repository.save(registry)
+        else:
+            self.repository.delete()

data_designer/interface/data_designer.py

@@ -17,6 +17,7 @@ from data_designer.config.default_model_settings import (
     get_providers_with_missing_api_keys,
 )
 from data_designer.config.interface import DataDesignerInterface
+from data_designer.config.mcp import MCPProviderT
 from data_designer.config.models import (
     ModelConfig,
     ModelProvider,
@@ -63,6 +64,8 @@ from data_designer.plugins.registry import PluginRegistry
 if TYPE_CHECKING:
     import pandas as pd
 
+    from data_designer.engine.models.facade import ModelFacade
+
 logger = logging.getLogger(__name__)
 
 
@@ -98,6 +101,9 @@ class DataDesigner(DataDesignerInterface[DatasetCreationResults]):
             If not provided, will check for an environment variable called DATA_DESIGNER_MANAGED_ASSETS_PATH.
             If the environment variable is not set, will use the default managed assets directory, which
             is defined in `data_designer.config.utils.constants`.
+        mcp_providers: Optional list of MCP provider configurations to enable tool-calling for
+            LLM generation columns. Supports both MCPProvider (remote/SSE) and
+            LocalStdioMCPProvider (local subprocess).
     """
 
     def __init__(
@@ -108,12 +114,14 @@ class DataDesigner(DataDesignerInterface[DatasetCreationResults]):
         secret_resolver: SecretResolver | None = None,
         seed_readers: list[SeedReader] | None = None,
         managed_assets_path: Path | str | None = None,
+        mcp_providers: list[MCPProviderT] | None = None,
     ):
         self._secret_resolver = secret_resolver or DEFAULT_SECRET_RESOLVER
         self._artifact_path = Path(artifact_path) if artifact_path is not None else Path.cwd() / "artifacts"
         self._run_config = RunConfig()
         self._managed_assets_path = Path(managed_assets_path or MANAGED_ASSETS_PATH)
         self._model_providers = self._resolve_model_providers(model_providers)
+        self._mcp_providers = mcp_providers or []
         self._model_provider_registry = resolve_model_provider_registry(
             self._model_providers, get_default_provider_name()
         )
@@ -325,6 +333,23 @@ class DataDesigner(DataDesignerInterface[DatasetCreationResults]):
         """
         self._run_config = run_config
 
+    def get_models(self, model_aliases: list[str]) -> dict[str, ModelFacade]:
+        """Get a dict of ModelFacade instances for custom column development.
+
+        Use this to experiment with custom column generator functions outside of
+        the full pipeline. The returned dict matches the `models` argument passed
+        to 3-arg custom column functions.
+
+        Args:
+            model_aliases: List of model aliases to include in the dict.
+
+        Returns:
+            Dict mapping alias to ModelFacade instance.
+        """
+        config_builder = DataDesignerConfigBuilder()
+        resource_provider = self._create_resource_provider("dev", config_builder)
+        return {alias: resource_provider.model_registry.get_model(model_alias=alias) for alias in model_aliases}
+
     def _resolve_model_providers(self, model_providers: list[ModelProvider] | None) -> list[ModelProvider]:
         if model_providers is None:
             model_providers = get_default_providers()
@@ -382,6 +407,8 @@ class DataDesigner(DataDesignerInterface[DatasetCreationResults]):
             seed_dataset_source=seed_dataset_source,
             seed_reader_registry=self._seed_reader_registry,
             run_config=self._run_config,
+            mcp_providers=self._mcp_providers,
+            tool_configs=config_builder.tool_configs,
         )
 
     def _get_interface_info(self, model_providers: list[ModelProvider]) -> InterfaceInfo:

{data_designer-0.4.0rc2.dist-info → data_designer-0.5.0rc1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: data-designer
-Version: 0.4.0rc2
+Version: 0.5.0rc1
 Summary: General framework for synthetic data generation
 License-Expression: Apache-2.0
 Classifier: Development Status :: 4 - Beta
@@ -14,8 +14,8 @@ Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Software Development
 Requires-Python: >=3.10
-Requires-Dist: data-designer-config
-Requires-Dist: data-designer-engine
+Requires-Dist: data-designer-config==0.5.0rc1
+Requires-Dist: data-designer-engine==0.5.0rc1
 Requires-Dist: prompt-toolkit<4,>=3.0.0
 Requires-Dist: typer<1,>=0.12.0
 Description-Content-Type: text/markdown

{data_designer-0.4.0rc2.dist-info → data_designer-0.5.0rc1.dist-info}/RECORD

@@ -1,39 +1,48 @@
 data_designer/cli/README.md,sha256=uPE3KdlF5Y3H8pQc8c6ZZ3h6YSFXNQW-iEXGQJuVnI4,9026
 data_designer/cli/__init__.py,sha256=--5yQzMciTX8-vroyXyFNBCqQ0HQd67GWCwnIoIHhJ4,251
-data_designer/cli/main.py,sha256=1klKdUKPZTgmUbduHSzEFueQHWkc-42Gcbri25cjiHo,1974
+data_designer/cli/main.py,sha256=tDTnvDriBPZ8CetQxKrxuEytAksKDjY-kCaSll6fqCc,2179
 data_designer/cli/ui.py,sha256=IgpV_Ht6qmLFrT3ybgOoADTQthoSGJxrwds38o1Zz10,17632
 data_designer/cli/utils.py,sha256=yyKZfr4ndcsngKgmpj5r4fN7fP6ouX-Nwx1Go5s6SdM,2151
 data_designer/cli/commands/__init__.py,sha256=ObZ6NUPeEvvpGTJ5WIGKUyIrIjaI747OM6ErweRtHxQ,137
 data_designer/cli/commands/download.py,sha256=bTynzORVj1rftrrQhmTj6se-ITi2_L7Z3qtio1mLvXU,1770
-data_designer/cli/commands/list.py,sha256=Lu02qFTkhEkLX2e7ak_rHmoO8_4Jjrgy4Yua-EAtyHo,4091
+data_designer/cli/commands/list.py,sha256=JH6i5FMCE5SkeNZFYNGNTqNx-yTl9rATitEgGDQLMw8,9176
+data_designer/cli/commands/mcp.py,sha256=9880VhA02RT5xX0WXhYFmPFDDpUMVRKvAUsUV1Jd7F4,488
 data_designer/cli/commands/models.py,sha256=Ot4eWyEbCS7heG_bylBdWZ1qj4CILv_hTddm2VdY0Dc,428
 data_designer/cli/commands/providers.py,sha256=-zVNtE_0A0hifcUk6n3c_v_Olcd14mHt3N8_HahHTQ4,491
 data_designer/cli/commands/reset.py,sha256=iCNjkFNdGU6Y7rv-Fprl9ZW60riseL_R7CrYi6DrwR0,3514
+data_designer/cli/commands/tools.py,sha256=c9QT6sQOWU-vNNWfrgIrqEZGsZ_4SPrGXt4E5_gRTZs,467
 data_designer/cli/controllers/__init__.py,sha256=70il4GIKebdau43nCXyu4VcQj7IFNoxxjEo1Z3hm8_M,491
 data_designer/cli/controllers/download_controller.py,sha256=9lQo-njn890WJiewGazfd6SrBBA4Rj8LYFkXZG_phPI,8117
+data_designer/cli/controllers/mcp_provider_controller.py,sha256=d0mXouNuuLL_mXL0XtTO9XjeCBAyNRw0Sue87WDLjwA,8715
 data_designer/cli/controllers/model_controller.py,sha256=CZimP1npWwH8UrJXlfMIfbNEn9pcJKtg14CqgArbqQM,9020
-data_designer/cli/controllers/provider_controller.py,sha256=mSLHkc60lu9VsXJE2NNpFZ6zHkasz6UQLtoTBYoFtkA,12293
+data_designer/cli/controllers/provider_controller.py,sha256=2WHyq_t4pU10-Whw4y1-GM2l6Nl8QqnVt4705rXhlRs,12514
+data_designer/cli/controllers/tool_controller.py,sha256=v8qrSQvv-4I1WDApQSoYPY3nogWsiERLjfCsvAYtWus,8653
 data_designer/cli/forms/__init__.py,sha256=UpTr7s5q2GFFssNz3229Kb5JxvFOqtZ55XpifB9a15w,713
 data_designer/cli/forms/builder.py,sha256=Juem3wB2j1KXtZZY7wVP0-eWKK_tj_0-L8Zq9EAS0-k,1731
 data_designer/cli/forms/field.py,sha256=TYEQLqjMvYBS_ftf6Ms-D5J6TOIK9NNe-Ydvo5Nkq50,7543
 data_designer/cli/forms/form.py,sha256=wFdKS0WfuhfotRtwWZgJyN2HrTthI7Kx07NUoQV2DtM,2066
+data_designer/cli/forms/mcp_provider_builder.py,sha256=sGyZ_O9x0HnH6UoiyuLqpVsJfM-6QUqLNCftn9sCqtA,7813
 data_designer/cli/forms/model_builder.py,sha256=DPggV2cl-XQPUiVhrrGO_4_d7jTn5_kBeBn9oTw-V1U,13354
 data_designer/cli/forms/provider_builder.py,sha256=YA6IoLwV39Sh6w0lZYoF25m-ryhnqBqysXLUo4V5X-w,2936
+data_designer/cli/forms/tool_builder.py,sha256=I0MdWRq2y1tpkWHok3apbyQEt4Y00YhqcNlgIk6evz0,7687
 data_designer/cli/repositories/__init__.py,sha256=ukHlLpOimH9CCJsdW5U9tooV_oFWQ4iLGK5GNi5YXtM,475
 data_designer/cli/repositories/base.py,sha256=ofOAHeAYAL6Bm4EJjSOFaNPD-odm2PlrW6quRkQQzaw,1095
+data_designer/cli/repositories/mcp_provider_repository.py,sha256=z1lrTLRFe4D9QlYzBPokZte-BK1ubGs_jF0vblglSo8,1652
 data_designer/cli/repositories/model_repository.py,sha256=oaa5ISP8Y-BRzXOdzAhDHf0FqmWngSEJ8RGOYWeXi4M,1487
 data_designer/cli/repositories/persona_repository.py,sha256=3ZRarD6BYAKVYFs_r9hDkh2nfkKW7BA8KJyfcYP0RRc,2683
 data_designer/cli/repositories/provider_repository.py,sha256=hG6tYbjR3gT8DmXL7usRvMrc6ILws4ECyTZ5imENpuQ,1556
+data_designer/cli/repositories/tool_repository.py,sha256=mlAxDMP_wizx82_Yv3P3XJ1T5nyyjEFbMBSegG3UJFg,1463
 data_designer/cli/services/__init__.py,sha256=2ycyikXx-8gbYZm-xl6IMyKXLwR4REU5heg6BkUW6qo,455
 data_designer/cli/services/download_service.py,sha256=m_wtDfxAA80tZdIf9kUS3ye8fzKG-3DjfDnm5u0-mJE,3519
+data_designer/cli/services/mcp_provider_service.py,sha256=EimM3hILxoj0RMAuwo3CARuHzM_7f5VQvcyiEr5UbZk,3052
 data_designer/cli/services/model_service.py,sha256=cFiP9ZQIprPdrVibUC6uwL-NuCYRgx8XVIjxDV-TznU,3926
 data_designer/cli/services/provider_service.py,sha256=5cou_EWU0RwE9p2PWpRBM9HcPqdENLpkHUuGzQ-l9J4,3957
+data_designer/cli/services/tool_service.py,sha256=Qg-rN_m7V_bVl1DN1jhbU6SLKC0JTruWpdz1FA0aP5o,3060
 data_designer/interface/__init__.py,sha256=2LbGosKhVhNXSUj-MX00b6UJRW-qeyiQ7PdEwtJxwso,718
-data_designer/interface/_version.py,sha256=FvItxCBzPigrdVpFPfL1gQeV1-km5r7nCNGUzrYebTU,714
-data_designer/interface/data_designer.py,sha256=0LBAUL7W75EmMwz-f4Lr0my9BXg1OTR2hKpgNsWnqLk,17275
+data_designer/interface/data_designer.py,sha256=p8jb4c3v9MjlYJz1LhCC9a_lXTcIo9HFhHo-nruYRH4,18610
 data_designer/interface/errors.py,sha256=Ft9GMeIrOHJv_PC_1rU6hWcNyq1GHdsFYZSc9HnUrxU,606
 data_designer/interface/results.py,sha256=3fGwlhif4ufqUGh-EgsGccrob4S6a7WZ6BgFiszTo_A,3871
-data_designer-0.4.0rc2.dist-info/METADATA,sha256=NsHs5-WFT_-yKD1YsHEvkDZimxLQ3-Bw7JgUEGKnx8Y,7155
-data_designer-0.4.0rc2.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
-data_designer-0.4.0rc2.dist-info/entry_points.txt,sha256=NWWWidyDxN6CYX6y664PhBYMhbaYTQTyprqfYAgkyCg,57
-data_designer-0.4.0rc2.dist-info/RECORD,,
+data_designer-0.5.0rc1.dist-info/METADATA,sha256=W001vnIVll0iAeEEv1m_HfISP35FYrQU-7PJ-E2H2r0,7175
+data_designer-0.5.0rc1.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
+data_designer-0.5.0rc1.dist-info/entry_points.txt,sha256=NWWWidyDxN6CYX6y664PhBYMhbaYTQTyprqfYAgkyCg,57
+data_designer-0.5.0rc1.dist-info/RECORD,,

data_designer/interface/_version.py

@@ -1,34 +0,0 @@
-# file generated by setuptools-scm
-# don't change, don't track in version control
-
-__all__ = [
-    "__version__",
-    "__version_tuple__",
-    "version",
-    "version_tuple",
-    "__commit_id__",
-    "commit_id",
-]
-
-TYPE_CHECKING = False
-if TYPE_CHECKING:
-    from typing import Tuple
-    from typing import Union
-
-    VERSION_TUPLE = Tuple[Union[int, str], ...]
-    COMMIT_ID = Union[str, None]
-else:
-    VERSION_TUPLE = object
-    COMMIT_ID = object
-
-version: str
-__version__: str
-__version_tuple__: VERSION_TUPLE
-version_tuple: VERSION_TUPLE
-commit_id: COMMIT_ID
-__commit_id__: COMMIT_ID
-
-__version__ = version = '0.4.0rc2'
-__version_tuple__ = version_tuple = (0, 4, 0, 'rc2')
-
-__commit_id__ = commit_id = None