fastrag.ingestion.steps.chunking.refinement

Opt-in, bounded chat boundary selection; source text is never rewritten.

Classes

ChatBoundaryRefiner(*, llm, identity[, ...])	Reuse an injected ChatModel with explicit permission and nonsecret identity.
RefinementResult([boundaries, calls, ...])	Validated proposal and runtime-only observations for a single request.

class fastrag.ingestion.steps.chunking.refinement.RefinementResult(boundaries=(), calls=0, cache_hit=False, fallback=None)

Bases: object

Validated proposal and runtime-only observations for a single request.

Parameters:

• boundaries (tuple[int, ...])

• calls (int)

• cache_hit (bool)

• fallback (str | None)

class fastrag.ingestion.steps.chunking.refinement.ChatBoundaryRefiner(*, llm, identity, allow_external_transmission=False, timeout=30.0, retries=0, backoff=0.1, concurrency=2, max_input_chars=12000, max_response_chars=16000, cache_size=0)

Bases: object

Reuse an injected ChatModel with explicit permission and nonsecret identity.

identity must distinguish endpoint/model/generation settings without containing secrets. The strategy never examines provider credentials. Successful proposals optionally cache in an instance-local bounded LRU. Configure the provider’s own output-token limit to bound response generation.

Parameters:

• llm (ChatModel)

• identity (str)

• allow_external_transmission (bool)

• timeout (float)

• retries (int)

• backoff (float)

• concurrency (int)

• max_input_chars (int)

• max_response_chars (int)

• cache_size (int)

config()

Nonsecret complete strategy identity for pipeline/response caches.

Return type:

dict[str, Any]

property cache_entries: int

Current bounded response-cache occupancy, without source exposure.

async refine(*, node_id, text, allowed)

Select validated offsets; all ordinary failures become local fallback.

Parameters:

• node_id (str)

• text (str)

• allowed (tuple[int, ...])

Return type:

RefinementResult