Source:

• src/ometeotl_core/generation/pipeline.py

__

Local role: Orchestrates the full generation workflow: rule application → builder dispatch → optional registration → optional validation → structured result.

Big-picture role: Top-level entry point for the generation architecture (F-16 to F-22). Accepts a GenerationContext, applies the configured GenerationRuleSet, delegates to the appropriate builder, and returns a GenerationResult carrying the generated object, applied rule names, validation outcome, and repair suggestions.

Parameters and fields (ContextualGenerationPipeline):

• rules: GenerationRuleSet — rule set applied before building; defaults to combined_generation_rules()
• validation_pipeline: ValidationPipeline | None — optional validation pipeline; only runs when context.validate=True

Methods (ContextualGenerationPipeline):

• generate(context, *, world=None) -> GenerationResult — full pipeline: rules → build → register → validate
• generate_from_text_response(context, *, raw_response, ...) — parses an LLM text response into context overrides then delegates to generate

Parameters and fields (GenerationResult, dataclass):

• generated: Any — the model object produced by the builder
• applied_rule_names: list[str] — names of rules that were part of the active rule set (in order)
• validation: ValidationResult | None — validation outcome; None if validation was not requested
• diagnostics: list[str] — ordered human-readable pipeline stage messages
• uncertainty_zones: list[str] — uncertainty zones declared in the context
• repair_suggestions: list[str] — suggested fixes when validation fails

Notes:

• Supported context.operation values: "create", "partial_update", "corrective_update".
• Supported context.registration_policy values: "none", "if_available", "require".
• applied_rule_names reflects the rule set configuration, not which predicates fired.
• Validation only runs when context.validate=True and a validation_pipeline is provided.
• The pipeline does not call LLMGenerationAdapter automatically; callers must chain it explicitly.

See also:

• GenerationContext
• Rule engine
• LLMGenerationAdapter
• Generation examples

Example:

from ometeotl_core.generation.pipeline import ContextualGenerationPipeline
from ometeotl_core.generation.context import GenerationContext

ctx = GenerationContext(
    kind="actor",
    id="actor-1",
    label="Scout",
    attributes={"mobility": True},
    validate=True,
    validation_mode="lenient",
)

pipeline = ContextualGenerationPipeline()
result = pipeline.generate(ctx, world=world)

actor = result.generated
print(type(actor).__name__)           # "Actor"
print(result.applied_rule_names)      # rules that ran in the rule set
if result.validation:
    print(result.validation.valid)

Generation flow

GenerationContext
    → GenerationRuleSet.apply()        # constraint propagation and normalization
    → build_from_context()             # kind dispatch → builder
    → registration_policy check        # optional world registry update
    → ValidationPipeline.validate()    # optional; only when context.validate=True
    → GenerationResult