data-designer 0.2.1__py3-none-any.whl → 0.2.3__py3-none-any.whl

data_designer/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '0.2.1'
-__version_tuple__ = version_tuple = (0, 2, 1)
+__version__ = version = '0.2.3'
+__version_tuple__ = version_tuple = (0, 2, 3)
 
 __commit_id__ = commit_id = None

data_designer/config/config_builder.py

@@ -224,6 +224,9 @@ class DataDesignerConfigBuilder:
 
         Returns:
             The current Data Designer config builder instance.
+
+        Raises:
+            BuilderConfigurationError: If the column name collides with an existing seed dataset column.
         """
         if column_config is None:
             if name is None or column_type is None:
@@ -240,6 +243,13 @@ class DataDesignerConfigBuilder:
                 f"{', '.join([t.__name__ for t in allowed_column_configs])}"
             )
 
+        existing_config = self._column_configs.get(column_config.name)
+        if existing_config is not None and isinstance(existing_config, SeedDatasetColumnConfig):
+            raise BuilderConfigurationError(
+                f"🛑 Column {column_config.name!r} already exists as a seed dataset column. "
+                "Please use a different column name or update the seed dataset."
+            )
+
         self._column_configs[column_config.name] = column_config
         return self
 
@@ -578,7 +588,18 @@ class DataDesignerConfigBuilder:
 
         Returns:
             The current Data Designer config builder instance.
+
+        Raises:
+            BuilderConfigurationError: If any seed dataset column name collides with an existing column.
         """
+        seed_column_names = fetch_seed_dataset_column_names(dataset_reference)
+        colliding_columns = [name for name in seed_column_names if name in self._column_configs]
+        if colliding_columns:
+            raise BuilderConfigurationError(
+                f"🛑 Seed dataset column(s) {colliding_columns} collide with existing column(s). "
+                "Please remove the conflicting columns or use a seed dataset with different column names."
+            )
+
         self._seed_config = SeedConfig(
             dataset=dataset_reference.dataset,
             sampling_strategy=sampling_strategy,
@@ -587,7 +608,7 @@ class DataDesignerConfigBuilder:
         self.set_seed_datastore_settings(
             dataset_reference.datastore_settings if hasattr(dataset_reference, "datastore_settings") else None
         )
-        for column_name in fetch_seed_dataset_column_names(dataset_reference):
+        for column_name in seed_column_names:
             self._column_configs[column_name] = SeedDatasetColumnConfig(name=column_name)
         return self
 

data_designer/engine/column_generators/generators/base.py

@@ -1,23 +1,27 @@
 # SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 
+from __future__ import annotations
+
 import functools
 import logging
 from abc import ABC, abstractmethod
-from typing import overload
+from enum import Enum
+from typing import TYPE_CHECKING, overload
 
 import pandas as pd
 
-from data_designer.config.column_types import COLUMN_TYPE_EMOJI_MAP
-from data_designer.config.models import BaseInferenceParams, ModelConfig
-from data_designer.config.utils.type_helpers import StrEnum
 from data_designer.engine.configurable_task import ConfigurableTask, ConfigurableTaskMetadata, DataT, TaskConfigT
-from data_designer.engine.models.facade import ModelFacade
+
+if TYPE_CHECKING:
+    from data_designer.config.models import BaseInferenceParams, ModelConfig
+    from data_designer.engine.models.facade import ModelFacade
+
 
 logger = logging.getLogger(__name__)
 
 
-class GenerationStrategy(StrEnum):
+class GenerationStrategy(str, Enum):
     CELL_BY_CELL = "cell_by_cell"
     FULL_COLUMN = "full_column"
 
@@ -82,8 +86,7 @@ class WithModelGeneration:
         return self.model_config.inference_parameters
 
     def log_pre_generation(self) -> None:
-        emoji = COLUMN_TYPE_EMOJI_MAP[self.config.column_type]
-        logger.info(f"{emoji} Preparing {self.config.column_type} column generation")
+        logger.info(f"Preparing {self.config.column_type} column generation")
         logger.info(f"  |-- column name: {self.config.name!r}")
         logger.info(f"  |-- model config:\n{self.model_config.model_dump_json(indent=4)}")
         if self.model_config.provider is None:

data_designer/engine/models/litellm_overrides.py

@@ -10,6 +10,7 @@ import httpx
 import litellm
 from litellm import RetryPolicy
 from litellm.caching.in_memory_cache import InMemoryCache
+from litellm.litellm_core_utils.logging_callback_manager import LoggingCallbackManager
 from litellm.router import Router
 from pydantic import BaseModel, Field
 from typing_extensions import override
@@ -154,7 +155,7 @@ def apply_litellm_patches():
     litellm.in_memory_llm_clients_cache = ThreadSafeCache()
 
     # Workaround for the litellm issue described in https://github.com/BerriAI/litellm/issues/9792
-    litellm.litellm_core_utils.logging_callback_manager.LoggingCallbackManager.MAX_CALLBACKS = DEFAULT_MAX_CALLBACKS
+    LoggingCallbackManager.MAX_CALLBACKS = DEFAULT_MAX_CALLBACKS
 
     quiet_noisy_logger("httpx")
     quiet_noisy_logger("LiteLLM")

data_designer/engine/validators/python.py

@@ -2,8 +2,8 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import ast
+import json
 import logging
-import re
 import subprocess
 import tempfile
 from collections import defaultdict
@@ -179,9 +179,8 @@ class PythonValidator(BaseValidator):
         for file in Path(codebase_path).glob("*.py"):
             processed[file.stem] = PythonLinterMessages()
 
-        # Run ruff linter
+        # Run ruff linter with JSON output
         ruff_bin = find_ruff_bin()
-        env = {"NO_COLOR": "1"}
 
         ruff_exec = subprocess.run(
             [
@@ -189,9 +188,9 @@ class PythonValidator(BaseValidator):
                 "check",
                 "--select",
                 "E,F6,F7,F8,SIM,PLC,PLE,PLR,PLW",
+                "--output-format=json",
                 codebase_path,
             ],
-            env=env,
             text=True,
             capture_output=True,
             check=False,
@@ -199,30 +198,34 @@ class PythonValidator(BaseValidator):
         )
         ruff_output = ruff_exec.stdout
 
-        # Parse ruff output
-        if "All checks passed!" in ruff_output:
-            return processed  # no errors or warnings
-
-        pattern = r"(.*):([0-9]*):([0-9]*): ([A-Za-z0-9]*):? (?:\[\*\] )?(.*)\n"
-        errors = re.findall(pattern, ruff_output)
+        # Parse JSON output
+        try:
+            diagnostics = json.loads(ruff_output)
+        except json.JSONDecodeError as e:
+            raise RuntimeError(f"Failed to parse ruff JSON output: {e}")
 
-        if errors == []:  # output could not be parsed
-            raise RuntimeError("ruff's output could not be parsed")
+        if not diagnostics:
+            return processed  # no errors or warnings
 
-        try:
-            for error in errors:
-                filename, line, column, symbol, message = error
-                processed[Path(filename).stem].add(
-                    PythonLinterMessage(
-                        type=TYPE_FROM_SYMBOL[re.sub(r"[^A-Za-z]+", "", symbol)],
-                        symbol=symbol,
-                        line=int(line),
-                        column=int(column),
-                        message=message,
-                    )
+        for diagnostic in diagnostics:
+            filename = diagnostic["filename"]
+            code = diagnostic["code"]
+            location = diagnostic["location"]
+            message = diagnostic["message"]
+
+            # Extract alphabetic prefix from code for type mapping
+            alpha_prefix = "".join(c for c in code if c.isalpha())
+            error_type = TYPE_FROM_SYMBOL.get(alpha_prefix, "warning")
+
+            processed[Path(filename).stem].add(
+                PythonLinterMessage(
+                    type=error_type,
+                    symbol=code,
+                    line=location["row"],
+                    column=location["column"],
+                    message=message,
                 )
-        except Exception:  # output not in expected format
-            raise RuntimeError("ruff's output not in expected format")
+            )
 
         return processed
 

{data_designer-0.2.1.dist-info → data_designer-0.2.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: data-designer
-Version: 0.2.1
+Version: 0.2.3
 Summary: General framework for synthetic data generation
 License-Expression: Apache-2.0
 License-File: LICENSE
@@ -15,36 +15,34 @@ Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Software Development
 Requires-Python: >=3.10
-Requires-Dist: anyascii<1.0,>=0.3.3
-Requires-Dist: datasets>=4.0.0
-Requires-Dist: duckdb==1.1.3
-Requires-Dist: faker==20.1.0
-Requires-Dist: httpx-retries>=0.4.2
-Requires-Dist: httpx>=0.27.2
-Requires-Dist: huggingface-hub>=0.34.4
+Requires-Dist: anyascii<1,>=0.3.3
+Requires-Dist: duckdb<2,>=1.1.3
+Requires-Dist: faker<21,>=20.1.0
+Requires-Dist: httpx-retries<1,>=0.4.2
+Requires-Dist: httpx<1,>=0.27.2
+Requires-Dist: huggingface-hub<2,>=1.0.1
 Requires-Dist: jinja2<4,>=3.1.6
-Requires-Dist: json-repair==0.48.0
-Requires-Dist: jsonpath-rust-bindings>=1.0
-Requires-Dist: litellm==1.73.6
-Requires-Dist: lxml>=6.0.2
-Requires-Dist: marko==2.1.2
-Requires-Dist: networkx==3.0
-Requires-Dist: numpy>=1.23.5
-Requires-Dist: pandas>=1.5.3
-Requires-Dist: prompt-toolkit>=3.0.0
-Requires-Dist: pyarrow>=19.0.1
-Requires-Dist: pydantic>=2.9.2
-Requires-Dist: pydantic[email]>=2.9.2
-Requires-Dist: pygments>=2.19.2
-Requires-Dist: python-json-logger==2.0.7
-Requires-Dist: pyyaml>=6.0.1
+Requires-Dist: json-repair<1,>=0.48.0
+Requires-Dist: jsonpath-rust-bindings<2,>=1.0
+Requires-Dist: litellm<2,>=1.73.6
+Requires-Dist: lxml<7,>=6.0.2
+Requires-Dist: marko<3,>=2.1.2
+Requires-Dist: networkx<4,>=3.0
+Requires-Dist: numpy<3,>=1.23.5
+Requires-Dist: pandas<3,>=2.3.3
+Requires-Dist: prompt-toolkit<4,>=3.0.0
+Requires-Dist: pyarrow<20,>=19.0.1
+Requires-Dist: pydantic[email]<3,>=2.9.2
+Requires-Dist: pygments<3,>=2.19.2
+Requires-Dist: python-json-logger<4,>=3
+Requires-Dist: pyyaml<7,>=6.0.1
 Requires-Dist: requests<3,>=2.32.2
-Requires-Dist: rich>=13.7.1
-Requires-Dist: ruff==0.12.3
-Requires-Dist: scipy>=1.11.0
-Requires-Dist: sqlfluff==3.2.0
-Requires-Dist: tiktoken>=0.8.0
-Requires-Dist: typer>=0.12.0
+Requires-Dist: rich<14,>=13.7.1
+Requires-Dist: ruff<1,>=0.14.10
+Requires-Dist: scipy<2,>=1.11.0
+Requires-Dist: sqlfluff<4,>=3.2.0
+Requires-Dist: tiktoken<1,>=0.8.0
+Requires-Dist: typer<1,>=0.12.0
 Description-Content-Type: text/markdown
 
 # 🎨 NeMo Data Designer

{data_designer-0.2.1.dist-info → data_designer-0.2.3.dist-info}/RECORD

@@ -1,5 +1,5 @@
 data_designer/__init__.py,sha256=iCeqRnb640RrL2QpA630GY5Ng7JiDt83Vq0DwLnNugU,461
-data_designer/_version.py,sha256=vYqoJTG51NOUmYyL0xt8asRK8vUT4lGAdal_EZ59mvw,704
+data_designer/_version.py,sha256=kBRz0P2plw1eVdIpt70W6m1LMbEIhLY3RyOfVGdubaI,704
 data_designer/errors.py,sha256=Z4eN9XwzZvGRdBluSNoSqQYkPPzNQIDf0ET_OqWRZh8,179
 data_designer/logging.py,sha256=ZsruJ0tx_4NK0PIMyxCZJJ0wJugoDff9UP3PbsdEDxQ,5341
 data_designer/plugin_manager.py,sha256=xaMX274gdlYLNNPIrAOmJNLaZlG_0ROJ0H29v8t2aKs,2604
@@ -37,7 +37,7 @@ data_designer/config/__init__.py,sha256=9eG4WHKyrJcNoK4GEz6BCw_E0Ewo9elQoDN4TLMb
 data_designer/config/base.py,sha256=ypam9XX6dg8Q_55su84WmVExNXsmt5jb3eeW3JLlHwc,2396
 data_designer/config/column_configs.py,sha256=pjpy5z3Kk7i4WmIjOxdiW5Awpjy5CxQSy0YMy0QxtvA,18961
 data_designer/config/column_types.py,sha256=EILVM42d4TMl2xbSj5htMsenJwybCHIc_G8AUXyrjWU,7197
-data_designer/config/config_builder.py,sha256=bXRFbRsYxLlED3rQN6u0qtBLgRoHLw767q_QtQoTZmI,28151
+data_designer/config/config_builder.py,sha256=n8in3O-hR2j3wJBnZMCoT5NawlobJDWTyNZCIYSgWIo,29241
 data_designer/config/data_designer_config.py,sha256=D2b4Dl8pR6kCkvPoZ3APxC9pVBqXi5EJMVK1WBZ6ni8,1886
 data_designer/config/dataset_builders.py,sha256=1pNFy_pkQ5lJ6AVZ43AeTuSbz6yC_l7Ndcyp5yaT8hQ,327
 data_designer/config/datastore.py,sha256=gEHR2hYlJwD_vzjuaSOMRiYjtwdQhyO9q1afZDrhBCo,7586
@@ -84,7 +84,7 @@ data_designer/engine/analysis/utils/judge_score_processing.py,sha256=rl11e3PxAOQ
 data_designer/engine/column_generators/__init__.py,sha256=9eG4WHKyrJcNoK4GEz6BCw_E0Ewo9elQoDN4TLMbAog,137
 data_designer/engine/column_generators/registry.py,sha256=Eg6tqNM7mmEPNom1fWF9S5D3qABpMennOHGEGePwJN0,3060
 data_designer/engine/column_generators/generators/__init__.py,sha256=9eG4WHKyrJcNoK4GEz6BCw_E0Ewo9elQoDN4TLMbAog,137
-data_designer/engine/column_generators/generators/base.py,sha256=48kQHNcT6k6-anMRoSelgoPhsdrb90n6BQqc45ZE7n8,3327
+data_designer/engine/column_generators/generators/base.py,sha256=zurwtamM2l3shLa4SLjUOE0zOTDozQ5wPGAvDkrNYqE,3231
 data_designer/engine/column_generators/generators/embedding.py,sha256=xYnFWRJ2W7JuwK8CRIUhv4QiT_DCGDuQkuHFKXCxrow,1724
 data_designer/engine/column_generators/generators/expression.py,sha256=7xniEj8aPscWDYLrnNbG2mF3s08C7aR8ZgNUCzr_x8g,2539
 data_designer/engine/column_generators/generators/llm_completion.py,sha256=XqpXzYczbZ6efUIVuvcm2O_mXBnXCMAvcjeyaB5dIFA,5301
@@ -107,7 +107,7 @@ data_designer/engine/dataset_builders/utils/errors.py,sha256=qW_TFOKNVODbb8bYrUl
 data_designer/engine/models/__init__.py,sha256=9eG4WHKyrJcNoK4GEz6BCw_E0Ewo9elQoDN4TLMbAog,137
 data_designer/engine/models/errors.py,sha256=AQlZ-cf0IqFW-e-K9HZFH3YhXXOTLLI2eZCXr7_58Yk,12174
 data_designer/engine/models/facade.py,sha256=sqTSqW7jQ1vbRc1fCOoKuhb7vTVil5Z8RqN_NBp6exY,12410
-data_designer/engine/models/litellm_overrides.py,sha256=tUVB_Zvg_6VoFNTK7WEKHUjoAknMkAOvureLycMENh0,5504
+data_designer/engine/models/litellm_overrides.py,sha256=FpCztZQrHYGrVweLR0_NAqxVRs-UXmpkMPdFWqciu84,5539
 data_designer/engine/models/registry.py,sha256=-TbGhvs8WRq6f7z6cH_DDdo7uhs4Hb5qkJce_Y4UBWM,6840
 data_designer/engine/models/telemetry.py,sha256=3g4jDz8xxOOkPtIYit94c4D4mGUwgfiCDaDdnbTLhFQ,12407
 data_designer/engine/models/usage.py,sha256=rObhH8X0O7L-P863Jz2EAO4FO25-6VP42HL6Tvx4lRg,2405
@@ -167,7 +167,7 @@ data_designer/engine/sampling_gen/entities/assets/zip_area_code_map.parquet,sha2
 data_designer/engine/validators/__init__.py,sha256=lMouN5BTbDNi31KfTQNjV7qrL46q-ssejXNT24iDTGI,652
 data_designer/engine/validators/base.py,sha256=lzO4gRoCDvh3BFP0sM4OjcwG-84qodYFW-G9NEOk3Cs,964
 data_designer/engine/validators/local_callable.py,sha256=oCUXj_NRt0gVqUIh0fLrvw-iURDR6OHFrVi5GOMhXj8,1387
-data_designer/engine/validators/python.py,sha256=DK6xxTzUHD8JUEmfP14W4hKdSb45ifbmvrWoU_o_l7Y,7871
+data_designer/engine/validators/python.py,sha256=jAp1u8yLjqfebh60xGapkHVjMz58WHB0QjfMc2zQCaY,7894
 data_designer/engine/validators/remote.py,sha256=jtDIvWzfHh17m2ac_Fp93p49Th8RlkBzzih2jiqD7gk,2929
 data_designer/engine/validators/sql.py,sha256=bxbyxPxDT9yuwjhABVEY40iR1pzWRFi65WU4tPgG2bE,2250
 data_designer/essentials/__init__.py,sha256=eHuZFJTmeRf_b6KQZ2vZeqy1afJ7y7RMTm7q4Jrg58s,1012
@@ -179,8 +179,8 @@ data_designer/plugins/__init__.py,sha256=c_V7q4QhfVoNf_uc9UwmXCsWqwtyWogI7YoN_0P
 data_designer/plugins/errors.py,sha256=yPIHpSddEr-o9ZcNVibb2hI-73O15Kg_Od8SlmQlnRs,297
 data_designer/plugins/plugin.py,sha256=a2KfoCNhYa8U0uQrPSBWfuyjXOb5WeITzFRpEdZFo6s,2516
 data_designer/plugins/registry.py,sha256=c0X03TnA_J60RWpxaVJEmtIXKvA9up-LznrUHXDcYxg,3012
-data_designer-0.2.1.dist-info/METADATA,sha256=StxkeuCq3NdJmhiMMeaXpWJzF4zauPid6O7haFw06VU,7626
-data_designer-0.2.1.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
-data_designer-0.2.1.dist-info/entry_points.txt,sha256=NWWWidyDxN6CYX6y664PhBYMhbaYTQTyprqfYAgkyCg,57
-data_designer-0.2.1.dist-info/licenses/LICENSE,sha256=cSWJDwVqHyQgly8Zmt3pqXJ2eQbZVYwN9qd0NMssxXY,11336
-data_designer-0.2.1.dist-info/RECORD,,
+data_designer-0.2.3.dist-info/METADATA,sha256=JUXSRI21S_Sp_Jwh060AHSCvqEERjHXG9sNhwxJUfSU,7636
+data_designer-0.2.3.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
+data_designer-0.2.3.dist-info/entry_points.txt,sha256=NWWWidyDxN6CYX6y664PhBYMhbaYTQTyprqfYAgkyCg,57
+data_designer-0.2.3.dist-info/licenses/LICENSE,sha256=cSWJDwVqHyQgly8Zmt3pqXJ2eQbZVYwN9qd0NMssxXY,11336
+data_designer-0.2.3.dist-info/RECORD,,