DartBridgeRAG class
Properties
- documentCount → int
-
no setter
- hashCode → int
-
The hash code for this object.
no setterinherited
- isCreated → bool
-
no setter
- runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
Methods
-
clearDocuments(
) → RAGStatistics -
createPipeline(
RAGConfiguration config) → void -
createPipelineAsync(
RAGConfiguration config) → Future< void> -
destroyPipeline(
) → void -
getStatistics(
) → RAGStatistics -
ingestDocument(
RAGDocument document) → RAGStatistics -
ingestDocumentAsync(
RAGDocument document) → Future< RAGStatistics> -
Ingest a document, running the blocking chunk→embed→index work in a
short-lived worker isolate (
Isolate.run) so the calling isolate — usually the Flutter UI isolate — stays responsive. The RAG session is process- global C++ state (in-memory, mutex-protected vector store + ONNX embeddings + llama.cpp, no file I/O on this path), so its address is valid on the worker; this is the same cross-isolate engine use thatdart_bridge_llm.dart's worker already relies on. -
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
query(
RAGQueryOptions options) → RAGResult -
queryAsync(
RAGQueryOptions options) → Future< RAGResult> -
Query the pipeline, running the blocking embed→retrieve→LLM-generate work
in a short-lived worker isolate (
Isolate.run) so the calling isolate — usually the Flutter UI isolate — stays responsive for the whole answer.rac_rag_query_protoruns the LLM generation inline (max_tokens/ temperature/system_prompt), so this is the heavy chat call. Same cross-isolate engine use asdart_bridge_llm.dart(see ingestDocumentAsync); the synchronous query is retained for callers that need it. -
register(
) → void -
toString(
) → String -
A string representation of this object.
inherited
-
unregister(
) → void
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited
Static Properties
-
final