Recollect.Pipeline.Extractor (recollect v0.5.1)

Copy Markdown View Source

Extracts entities and relationships from text chunks using LLM structured output. Deduplicates and persists results with mention counting.

Summary

Functions

Extract entities and relations from a chunk's content using the configured provider.

Persist extracted entities into the database, deduplicating by name+type within the same collection. Returns {:ok, [entity]}.

Persist extracted relations. Requires a map of entity_name -> entity_id.

Functions

extract_from_chunk(chunk_content, opts \\ [])

Extract entities and relations from a chunk's content using the configured provider.

persist_entities(entities, opts)

Persist extracted entities into the database, deduplicating by name+type within the same collection. Returns {:ok, [entity]}.

persist_relations(relations, entity_map, opts)

Persist extracted relations. Requires a map of entity_name -> entity_id.