data-designer-config 0.4.0__tar.gz → 0.4.0rc1__tar.gz

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: data-designer-config
-Version: 0.4.0
+Version: 0.4.0rc1
 Summary: Configuration layer for DataDesigner synthetic data generation
 License-Expression: Apache-2.0
 Classifier: Development Status :: 4 - Beta

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/src/data_designer/config/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '0.4.0'
-__version_tuple__ = version_tuple = (0, 4, 0)
+__version__ = version = '0.4.0rc1'
+__version_tuple__ = version_tuple = (0, 4, 0, 'rc1')
 
 __commit_id__ = commit_id = None

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/src/data_designer/config/column_configs.py

@@ -14,7 +14,7 @@ from data_designer.config.errors import InvalidConfigError
 from data_designer.config.models import ImageContext
 from data_designer.config.sampler_params import SamplerParamsT, SamplerType
 from data_designer.config.utils.code_lang import CodeLang
-from data_designer.config.utils.constants import TRACE_COLUMN_POSTFIX
+from data_designer.config.utils.constants import REASONING_TRACE_COLUMN_POSTFIX
 from data_designer.config.utils.misc import assert_valid_jinja2_template, extract_keywords_from_jinja2_template
 from data_designer.config.validator_params import ValidatorParamsT, ValidatorType
 
@@ -143,8 +143,8 @@ class LLMTextColumnConfig(SingleColumnConfig):
 
     LLM text columns generate free-form text content using language models via LiteLLM.
     Prompts support Jinja2 templating to reference values from other columns, enabling
-    context-aware generation. The generated text can optionally include message traces
-    capturing the full conversation history.
+    context-aware generation. The generated text can optionally include reasoning traces
+    when models support extended thinking.
 
     Attributes:
         prompt: Prompt template for text generation. Supports Jinja2 syntax to
@@ -159,10 +159,6 @@ class LLMTextColumnConfig(SingleColumnConfig):
             `LLMStructuredColumnConfig` for structured output, `LLMCodeColumnConfig` for code.
         multi_modal_context: Optional list of image contexts for multi-modal generation.
             Enables vision-capable models to generate text based on image inputs.
-        with_trace: If True, creates a `{column_name}__trace` column containing the full
-            ordered message history (system/user/assistant) for the generation.
-            Can be overridden globally via `RunConfig.debug_override_save_all_column_traces`.
-            Defaults to False.
         column_type: Discriminator field, always "llm-text" for this configuration type.
     """
 
@@ -170,7 +166,6 @@ class LLMTextColumnConfig(SingleColumnConfig):
     model_alias: str
     system_prompt: str | None = None
     multi_modal_context: list[ImageContext] | None = None
-    with_trace: bool = False
     column_type: Literal["llm-text"] = "llm-text"
 
     @staticmethod
@@ -191,15 +186,14 @@ class LLMTextColumnConfig(SingleColumnConfig):
 
     @property
     def side_effect_columns(self) -> list[str]:
-        """Returns the trace column, which may be generated alongside the main column.
+        """Returns the reasoning trace column, which may be generated alongside the main column.
 
-        Traces are generated when `with_trace=True` on the column config or
-        when `RunConfig.debug_override_save_all_column_traces=True` globally.
+        Reasoning traces are only returned if the served model parses and returns reasoning content.
 
         Returns:
-            List containing the trace column name.
+            List containing the reasoning trace column name.
         """
-        return [f"{self.name}{TRACE_COLUMN_POSTFIX}"]
+        return [f"{self.name}{REASONING_TRACE_COLUMN_POSTFIX}"]
 
     @model_validator(mode="after")
     def assert_prompt_valid_jinja(self) -> Self:

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/src/data_designer/config/models.py

@@ -3,7 +3,6 @@
 
 from __future__ import annotations
 
-import json
 import logging
 from abc import ABC, abstractmethod
 from enum import Enum
@@ -66,7 +65,7 @@ class ModalityContext(ABC, BaseModel):
     data_type: ModalityDataType
 
     @abstractmethod
-    def get_contexts(self, record: dict) -> list[dict[str, Any]]: ...
+    def get_context(self, record: dict) -> dict[str, Any]: ...
 
 
 class ImageContext(ModalityContext):
@@ -82,53 +81,25 @@ class ImageContext(ModalityContext):
     modality: Modality = Modality.IMAGE
     image_format: ImageFormat | None = None
 
-    def get_contexts(self, record: dict) -> list[dict[str, Any]]:
-        """Get the contexts for the image modality.
+    def get_context(self, record: dict) -> dict[str, Any]:
+        """Get the context for the image modality.
 
         Args:
-            record: The record containing the image data. The data can be:
-                - A JSON serialized list of strings
-                - A list of strings
-                - A single string
+            record: The record containing the image data.
 
         Returns:
-            A list of image contexts.
+            The context for the image modality.
         """
-        raw_value = record[self.column_name]
-
-        # Normalize to list of strings
-        if isinstance(raw_value, str):
-            # Try to parse as JSON first
-            try:
-                parsed_value = json.loads(raw_value)
-                if isinstance(parsed_value, list):
-                    context_values = parsed_value
-                else:
-                    context_values = [raw_value]
-            except (json.JSONDecodeError, TypeError):
-                context_values = [raw_value]
-        elif isinstance(raw_value, list):
-            context_values = raw_value
-        elif hasattr(raw_value, "__iter__") and not isinstance(raw_value, (str, bytes, dict)):
-            # Handle array-like objects (numpy arrays, pandas Series, etc.)
-            context_values = list(raw_value)
+        context = dict(type="image_url")
+        context_value = record[self.column_name]
+        if self.data_type == ModalityDataType.URL:
+            context["image_url"] = context_value
         else:
-            context_values = [raw_value]
-
-        # Build context list
-        contexts = []
-        for context_value in context_values:
-            context = dict(type="image_url")
-            if self.data_type == ModalityDataType.URL:
-                context["image_url"] = context_value
-            else:
-                context["image_url"] = {
-                    "url": f"data:image/{self.image_format.value};base64,{context_value}",
-                    "format": self.image_format.value,
-                }
-            contexts.append(context)
-
-        return contexts
+            context["image_url"] = {
+                "url": f"data:image/{self.image_format.value};base64,{context_value}",
+                "format": self.image_format.value,
+            }
+        return context
 
     @model_validator(mode="after")
     def _validate_image_format(self) -> Self:
@@ -428,14 +399,12 @@ class ModelConfig(ConfigBase):
         inference_parameters: Inference parameters for the model (temperature, top_p, max_tokens, etc.).
             The generation_type is determined by the type of inference_parameters.
         provider: Optional model provider name if using custom providers.
-        skip_health_check: Whether to skip the health check for this model. Defaults to False.
     """
 
     alias: str
     model: str
     inference_parameters: InferenceParamsT = Field(default_factory=ChatCompletionInferenceParams)
     provider: str | None = None
-    skip_health_check: bool = False
 
     @property
     def generation_type(self) -> GenerationType:

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/src/data_designer/config/run_config.py

@@ -33,10 +33,6 @@ class RunConfig(ConfigBase):
         max_conversation_correction_steps: Maximum number of correction rounds permitted within a
             single conversation when generation tasks call `ModelFacade.generate(...)`. Must be >= 0.
             Default is 0.
-        debug_override_save_all_column_traces: If True, overrides per-column `with_trace` settings
-            and includes `__trace` columns for ALL LLM generations, containing the full ordered
-            message history (system/user/assistant) for the final generation attempt.
-            Useful for debugging. Default is False.
     """
 
     disable_early_shutdown: bool = False
@@ -46,7 +42,6 @@ class RunConfig(ConfigBase):
     non_inference_max_parallel_workers: int = Field(default=4, ge=1)
     max_conversation_restarts: int = Field(default=5, ge=0)
     max_conversation_correction_steps: int = Field(default=0, ge=0)
-    debug_override_save_all_column_traces: bool = False
 
     @model_validator(mode="after")
     def normalize_shutdown_settings(self) -> Self:

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/src/data_designer/config/utils/code_lang.py

@@ -7,14 +7,9 @@ from enum import Enum
 
 
 class CodeLang(str, Enum):
-    BASH = "bash"
-    C = "c"
-    COBOL = "cobol"
-    CPP = "cpp"
-    CSHARP = "csharp"
     GO = "go"
-    JAVA = "java"
     JAVASCRIPT = "javascript"
+    JAVA = "java"
     KOTLIN = "kotlin"
     PYTHON = "python"
     RUBY = "ruby"
@@ -68,21 +63,15 @@ def code_lang_to_syntax_lexer(code_lang: CodeLang | str) -> str:
     Reference: https://pygments.org/docs/lexers/
     """
     code_lang_to_lexer = {
-        CodeLang.BASH: "bash",
-        CodeLang.C: "c",
-        CodeLang.COBOL: "cobol",
-        CodeLang.CPP: "cpp",
-        CodeLang.CSHARP: "csharp",
         CodeLang.GO: "golang",
-        CodeLang.JAVA: "java",
         CodeLang.JAVASCRIPT: "javascript",
+        CodeLang.JAVA: "java",
         CodeLang.KOTLIN: "kotlin",
         CodeLang.PYTHON: "python",
         CodeLang.RUBY: "ruby",
         CodeLang.RUST: "rust",
         CodeLang.SCALA: "scala",
         CodeLang.SWIFT: "swift",
-        CodeLang.TYPESCRIPT: "typescript",
         CodeLang.SQL_SQLITE: "sql",
         CodeLang.SQL_ANSI: "sql",
         CodeLang.SQL_TSQL: "tsql",

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/src/data_designer/config/utils/constants.py

@@ -166,7 +166,7 @@ MIN_TEMPERATURE = 0.0
 MAX_TOP_P = 1.0
 MIN_TOP_P = 0.0
 MIN_MAX_TOKENS = 1
-TRACE_COLUMN_POSTFIX = "__trace"
+REASONING_TRACE_COLUMN_POSTFIX = "__reasoning_trace"
 
 AVAILABLE_LOCALES = [
     "ar_AA",

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/src/data_designer/logging.py

@@ -50,14 +50,6 @@ class LoggingConfig:
 class RandomEmoji:
     """A generator for various themed emoji collections."""
 
-    def __init__(self) -> None:
-        self._progress_style = random.choice(_PROGRESS_STYLES)
-
-    def progress(self, percent: float) -> str:
-        """Get a progress emoji based on completion percentage (0-100)."""
-        phase_idx = min(int(percent / 25), len(self._progress_style) - 1)
-        return self._progress_style[phase_idx]
-
     @staticmethod
     def cooking() -> str:
         """Get a random cooking or food preparation emoji."""
@@ -171,10 +163,3 @@ def _make_stream_formatter() -> logging.Formatter:
 
 
 _DEFAULT_NOISY_LOGGERS = ["httpx", "matplotlib"]
-
-
-_PROGRESS_STYLES: list[list[str]] = [
-    ["🌑", "🌘", "🌗", "🌖", "🌕"],  # Moon phases
-    ["🌧️", "🌦️", "⛅", "🌤️", "☀️"],  # Weather (storm to sun)
-    ["🥚", "🐣", "🐥", "🐤", "🐔"],  # Hatching (egg to chicken)
-]

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/tests/config/test_columns.py

@@ -85,7 +85,7 @@ def test_llm_text_column_config():
     assert llm_text_column_config.system_prompt == stub_system_prompt
     assert llm_text_column_config.column_type == DataDesignerColumnType.LLM_TEXT
     assert set(llm_text_column_config.required_columns) == {"some_column", "some_other_column"}
-    assert llm_text_column_config.side_effect_columns == ["test_llm_text__trace"]
+    assert llm_text_column_config.side_effect_columns == ["test_llm_text__reasoning_trace"]
 
     # invalid prompt
     with pytest.raises(

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/tests/config/test_models.py

@@ -4,7 +4,6 @@
 import json
 import tempfile
 from collections import Counter
-from typing import TYPE_CHECKING
 
 import pytest
 import yaml
@@ -25,159 +24,22 @@ from data_designer.config.models import (
     UniformDistributionParams,
     load_model_configs,
 )
-from data_designer.lazy_heavy_imports import np
 
-if TYPE_CHECKING:
-    import numpy as np
 
-
-def test_image_context_get_contexts_single_string():
-    """Test get_contexts with a single string value."""
-    image_context = ImageContext(
-        column_name="image_base64", data_type=ModalityDataType.BASE64, image_format=ImageFormat.PNG
-    )
-    assert image_context.get_contexts({"image_base64": "somebase64encodedimagestring"}) == [
-        {
-            "type": "image_url",
-            "image_url": {"url": "data:image/png;base64,somebase64encodedimagestring", "format": "png"},
-        }
-    ]
-
-    image_context = ImageContext(column_name="image_url", data_type=ModalityDataType.URL)
-    assert image_context.get_contexts({"image_url": "https://example.com/examle_image.png"}) == [
-        {
-            "type": "image_url",
-            "image_url": "https://example.com/examle_image.png",
-        }
-    ]
-
-
-def test_image_context_get_contexts_list_of_strings():
-    """Test get_contexts with a list of strings."""
-    image_context = ImageContext(
-        column_name="image_base64", data_type=ModalityDataType.BASE64, image_format=ImageFormat.PNG
-    )
-    assert image_context.get_contexts({"image_base64": ["image1base64", "image2base64", "image3base64"]}) == [
-        {
-            "type": "image_url",
-            "image_url": {"url": "data:image/png;base64,image1base64", "format": "png"},
-        },
-        {
-            "type": "image_url",
-            "image_url": {"url": "data:image/png;base64,image2base64", "format": "png"},
-        },
-        {
-            "type": "image_url",
-            "image_url": {"url": "data:image/png;base64,image3base64", "format": "png"},
-        },
-    ]
-
-    image_context = ImageContext(column_name="image_url", data_type=ModalityDataType.URL)
-    assert image_context.get_contexts(
-        {"image_url": ["https://example.com/image1.png", "https://example.com/image2.png"]}
-    ) == [
-        {
-            "type": "image_url",
-            "image_url": "https://example.com/image1.png",
-        },
-        {
-            "type": "image_url",
-            "image_url": "https://example.com/image2.png",
-        },
-    ]
-
-
-def test_image_context_get_contexts_numpy_array():
-    """Test get_contexts with numpy arrays (happens after parquet serialization)."""
-    image_context = ImageContext(
-        column_name="image_base64", data_type=ModalityDataType.BASE64, image_format=ImageFormat.PNG
-    )
-    numpy_array = np.array(["image1base64", "image2base64"])
-    assert image_context.get_contexts({"image_base64": numpy_array}) == [
-        {
-            "type": "image_url",
-            "image_url": {"url": "data:image/png;base64,image1base64", "format": "png"},
-        },
-        {
-            "type": "image_url",
-            "image_url": {"url": "data:image/png;base64,image2base64", "format": "png"},
-        },
-    ]
-
-    image_context = ImageContext(column_name="image_url", data_type=ModalityDataType.URL)
-    numpy_array = np.array(["https://example.com/image1.png", "https://example.com/image2.png"])
-    assert image_context.get_contexts({"image_url": numpy_array}) == [
-        {
-            "type": "image_url",
-            "image_url": "https://example.com/image1.png",
-        },
-        {
-            "type": "image_url",
-            "image_url": "https://example.com/image2.png",
-        },
-    ]
-
-
-def test_image_context_get_contexts_json_serialized_list():
-    """Test get_contexts with a JSON serialized list of strings."""
+def test_image_context_get_context():
     image_context = ImageContext(
         column_name="image_base64", data_type=ModalityDataType.BASE64, image_format=ImageFormat.PNG
     )
-    json_str = json.dumps(["image1base64", "image2base64"])
-    assert image_context.get_contexts({"image_base64": json_str}) == [
-        {
-            "type": "image_url",
-            "image_url": {"url": "data:image/png;base64,image1base64", "format": "png"},
-        },
-        {
-            "type": "image_url",
-            "image_url": {"url": "data:image/png;base64,image2base64", "format": "png"},
-        },
-    ]
-
-    image_context = ImageContext(column_name="image_url", data_type=ModalityDataType.URL)
-    json_str = json.dumps(["https://example.com/image1.png", "https://example.com/image2.png"])
-    assert image_context.get_contexts({"image_url": json_str}) == [
-        {
-            "type": "image_url",
-            "image_url": "https://example.com/image1.png",
-        },
-        {
-            "type": "image_url",
-            "image_url": "https://example.com/image2.png",
-        },
-    ]
-
-
-def test_image_context_get_contexts_json_string_not_list():
-    """Test get_contexts with a JSON string that isn't a list (should treat as single string)."""
-    image_context = ImageContext(column_name="image_url", data_type=ModalityDataType.URL)
-    json_str = json.dumps({"nested": "object"})
-    # Should treat the entire JSON string as a single image URL
-    assert image_context.get_contexts({"image_url": json_str}) == [
-        {
-            "type": "image_url",
-            "image_url": json_str,
-        }
-    ]
-
-
-def test_image_context_get_contexts_invalid_json():
-    """Test get_contexts with invalid JSON string (should treat as single string)."""
-    image_context = ImageContext(column_name="image_url", data_type=ModalityDataType.URL)
-    invalid_json = "not a valid json string"
-    assert image_context.get_contexts({"image_url": invalid_json}) == [
-        {
-            "type": "image_url",
-            "image_url": invalid_json,
-        }
-    ]
-
+    assert image_context.get_context({"image_base64": "somebase64encodedimagestring"}) == {
+        "type": "image_url",
+        "image_url": {"url": "data:image/png;base64,somebase64encodedimagestring", "format": "png"},
+    }
 
-def test_image_context_get_contexts_empty_list():
-    """Test get_contexts with an empty list."""
     image_context = ImageContext(column_name="image_url", data_type=ModalityDataType.URL)
-    assert image_context.get_contexts({"image_url": []}) == []
+    assert image_context.get_context({"image_url": "https://example.com/examle_image.png"}) == {
+        "type": "image_url",
+        "image_url": "https://example.com/examle_image.png",
+    }
 
 
 def test_image_context_validate_image_format():

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/tests/config/utils/test_code_lang.py

@@ -26,7 +26,7 @@ def test_parse_dialect():
 
 
 def test_supported_values():
-    assert len(CodeLang.supported_values()) == 21
+    assert len(CodeLang.supported_values()) == 16
 
 
 def test_code_lang_to_syntax_lexer():

{data_designer_config-0.4.0 → data_designer_config-0.4.0rc1}/tests/test_logging.py

@@ -208,54 +208,3 @@ def test_random_emoji_randomness():
     emojis = [RandomEmoji.magic() for _ in range(100)]
     # If we get 100 samples, we should get at least 2 different emojis
     assert len(set(emojis)) > 1
-
-
-def test_random_emoji_progress_returns_valid_emoji() -> None:
-    emoji_gen = RandomEmoji()
-    emoji = emoji_gen.progress(50.0)
-    assert emoji is not None
-    assert len(emoji) > 0
-
-
-def test_random_emoji_progress_is_deterministic() -> None:
-    emoji_gen = RandomEmoji()
-    # Same percentage should always return the same emoji for a given instance
-    assert emoji_gen.progress(0.0) == emoji_gen.progress(0.0)
-    assert emoji_gen.progress(50.0) == emoji_gen.progress(50.0)
-    assert emoji_gen.progress(100.0) == emoji_gen.progress(100.0)
-
-
-def test_random_emoji_progress_phases_are_distinct() -> None:
-    emoji_gen = RandomEmoji()
-    # Each 25% phase should return a different emoji
-    phase_emojis = [
-        emoji_gen.progress(0.0),  # phase 0
-        emoji_gen.progress(25.0),  # phase 1
-        emoji_gen.progress(50.0),  # phase 2
-        emoji_gen.progress(75.0),  # phase 3
-        emoji_gen.progress(100.0),  # phase 4
-    ]
-    # All 5 phases should have distinct emojis
-    assert len(set(phase_emojis)) == 5
-
-
-def test_random_emoji_progress_phase_boundaries() -> None:
-    emoji_gen = RandomEmoji()
-    # Values within the same phase should return the same emoji
-    assert emoji_gen.progress(0.0) == emoji_gen.progress(24.9)
-    assert emoji_gen.progress(25.0) == emoji_gen.progress(49.9)
-    assert emoji_gen.progress(50.0) == emoji_gen.progress(74.9)
-    assert emoji_gen.progress(75.0) == emoji_gen.progress(99.9)
-    # Phase transitions should return different emojis
-    assert emoji_gen.progress(24.9) != emoji_gen.progress(25.0)
-    assert emoji_gen.progress(49.9) != emoji_gen.progress(50.0)
-    assert emoji_gen.progress(74.9) != emoji_gen.progress(75.0)
-    assert emoji_gen.progress(99.9) != emoji_gen.progress(100.0)
-
-
-def test_random_emoji_progress_clamps_over_100() -> None:
-    emoji_gen = RandomEmoji()
-    emoji_100 = emoji_gen.progress(100.0)
-    emoji_over = emoji_gen.progress(150.0)
-    # Both should return the same final emoji
-    assert emoji_100 == emoji_over