sdg-hub 0.3.1__py3-none-any.whl → 0.4.1__py3-none-any.whl

sdg_hub/__init__.py

@@ -8,7 +8,6 @@ from .core import (
     BlockRegistry,
     Flow,
     FlowMetadata,
-    FlowParameter,
     FlowRegistry,
     FlowValidator,
     GenerateError,
@@ -23,7 +22,6 @@ __all__ = [
     "FlowRegistry",
     # Metadata and utilities
     "FlowMetadata",
-    "FlowParameter",
     "FlowValidator",
     "GenerateError",
     "resolve_path",

sdg_hub/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '0.3.1'
-__version_tuple__ = version_tuple = (0, 3, 1)
+__version__ = version = '0.4.1'
+__version_tuple__ = version_tuple = (0, 4, 1)
 
 __commit_id__ = commit_id = None

sdg_hub/core/__init__.py

@@ -3,7 +3,7 @@
 
 # Local
 from .blocks import BaseBlock, BlockRegistry
-from .flow import Flow, FlowMetadata, FlowParameter, FlowRegistry, FlowValidator
+from .flow import Flow, FlowMetadata, FlowRegistry, FlowValidator
 from .utils import GenerateError, resolve_path
 
 __all__ = [
@@ -14,7 +14,6 @@ __all__ = [
     "Flow",
     "FlowRegistry",
     "FlowMetadata",
-    "FlowParameter",
     "FlowValidator",
     # Utils
     "GenerateError",

sdg_hub/core/blocks/__init__.py

@@ -16,9 +16,8 @@ from .deprecated_blocks import (
     SelectorBlock,
     SetToMajorityValue,
 )
-from .evaluation import EvaluateFaithfulnessBlock, EvaluateRelevancyBlock
 from .filtering import ColumnValueFilterBlock
-from .llm import LLMChatBlock, PromptBuilderBlock, TextParserBlock
+from .llm import LLMChatBlock, LLMParserBlock, PromptBuilderBlock, TextParserBlock
 from .registry import BlockRegistry
 from .transform import (
     DuplicateColumnsBlock,
@@ -51,8 +50,7 @@ __all__ = [
     "SetToMajorityValue",  # Deprecated
     "LLMBlock",  # Deprecated
     "LLMChatBlock",
+    "LLMParserBlock",
     "TextParserBlock",
     "PromptBuilderBlock",
-    "EvaluateFaithfulnessBlock",
-    "EvaluateRelevancyBlock",
 ]

sdg_hub/core/blocks/base.py

@@ -261,18 +261,73 @@ class BaseBlock(BaseModel, ABC):
         ----------
         samples : Dataset
             Input dataset.
+        **kwargs : Any
+            Runtime parameters to override block configuration
 
         Returns
         -------
         Dataset
             Output dataset after block processing.
         """
-        self._log_input_data(samples)
-        self._validate_dataset(samples)
-        self._validate_custom(samples)
-        output_dataset = self.generate(samples, **kwargs)
-        self._log_output_data(samples, output_dataset)
-        return output_dataset
+        # Handle runtime kwargs overrides
+        if kwargs:
+            # Validate that all kwargs are either valid block fields or flow parameters
+            # Skip validation for blocks that accept arbitrary parameters (extra="allow")
+            allows_extra = self.model_config.get("extra") == "allow"
+            if not allows_extra:
+                for key in kwargs:
+                    if (
+                        not key.startswith("_flow_")
+                        and key not in self.__class__.model_fields
+                    ):
+                        logger.warning(
+                            f"Unknown field '{key}' passed to {self.__class__.__name__}. "
+                            f"This may be a provider-specific parameter or typo. "
+                            f"Valid fields: {list(self.__class__.model_fields.keys())}"
+                        )
+
+            # Only override actual block fields (not flow parameters)
+            block_overrides = {
+                k: v for k, v in kwargs.items() if k in self.__class__.model_fields
+            }
+
+            # Validate and apply block field overrides if any
+            original_values = {}
+            if block_overrides:
+                # Validate the merged configuration for block fields only
+                merged_config = {**self.model_dump(), **block_overrides}
+                try:
+                    self.__class__.model_validate(merged_config)
+                except Exception as e:
+                    raise ValueError(
+                        f"Invalid runtime override for {self.__class__.__name__}: {e}"
+                    ) from e
+
+                # Apply temporary overrides for block fields
+                for key, value in block_overrides.items():
+                    original_values[key] = getattr(self, key)
+                    setattr(self, key, value)
+
+            try:
+                self._log_input_data(samples)
+                self._validate_dataset(samples)
+                self._validate_custom(samples)
+                # Pass ALL kwargs to generate (including flow params)
+                output_dataset = self.generate(samples, **kwargs)
+                self._log_output_data(samples, output_dataset)
+                return output_dataset
+            finally:
+                # Always restore original values for block fields
+                for key, value in original_values.items():
+                    setattr(self, key, value)
+        else:
+            # Normal execution without overrides
+            self._log_input_data(samples)
+            self._validate_dataset(samples)
+            self._validate_custom(samples)
+            output_dataset = self.generate(samples)
+            self._log_output_data(samples, output_dataset)
+            return output_dataset
 
     def __repr__(self) -> str:
         """Compact string representation."""

sdg_hub/core/blocks/filtering/column_value_filter.py

@@ -119,7 +119,7 @@ class ColumnValueFilterBlock(BaseBlock):
             self.output_cols = []
 
         # Set derived attributes
-        self.value = (
+        self.filter_value = (
             self.filter_value
             if isinstance(self.filter_value, list)
             else [self.filter_value]
@@ -181,7 +181,8 @@ class ColumnValueFilterBlock(BaseBlock):
         # Apply filter operation
         samples = samples.filter(
             lambda x: any(
-                self._operation_func(x[self.column_name], value) for value in self.value
+                self._operation_func(x[self.column_name], value)
+                for value in self.filter_value
             )
         )
 

sdg_hub/core/blocks/llm/__init__.py

@@ -7,21 +7,19 @@ local models (vLLM, Ollama), and more.
 """
 
 # Local
-from .client_manager import LLMClientManager
-from .config import LLMConfig
 from .error_handler import ErrorCategory, LLMErrorHandler
 from .llm_chat_block import LLMChatBlock
 from .llm_chat_with_parsing_retry_block import LLMChatWithParsingRetryBlock
+from .llm_parser_block import LLMParserBlock
 from .prompt_builder_block import PromptBuilderBlock
 from .text_parser_block import TextParserBlock
 
 __all__ = [
-    "LLMConfig",
-    "LLMClientManager",
     "LLMErrorHandler",
     "ErrorCategory",
     "LLMChatBlock",
     "LLMChatWithParsingRetryBlock",
+    "LLMParserBlock",
     "PromptBuilderBlock",
     "TextParserBlock",
 ]