Crate synaptic_sqlite 

SQLite integration for the Synaptic framework.

This crate provides:

• SqliteCache: A SQLite-backed implementation of the LlmCache trait for caching LLM responses with optional TTL expiration.
• SqliteCheckpointer: A SQLite-backed implementation of the Checkpointer trait for persisting graph state between executions.
• SqliteStore: A SQLite-backed implementation of the Store trait with FTS5 full-text search.
• SqliteVectorStore: A SQLite-backed implementation of the VectorStore trait with FTS5 hybrid search.

Quick start

use synaptic_sqlite::{SqliteCache, SqliteCacheConfig};

// In-memory cache (great for testing)
let cache = SqliteCache::new(SqliteCacheConfig::in_memory())?;

// File-based cache with 1-hour TTL
let config = SqliteCacheConfig::new("/tmp/llm_cache.db").with_ttl(3600);
let cache = SqliteCache::new(config)?;

Re-exports

pub use checkpointer::SqliteCheckpointer;

Modules

checkpointer

Structs

ChatResponse

A response from a chat model containing the AI message and optional token usage statistics.

Document

A document with content and metadata, used throughout the retrieval pipeline.

Item

A stored item in the key-value store.

SqliteCache

SQLite-backed implementation of the LlmCache trait.

SqliteCacheConfig

Configuration for SqliteCache.

SqliteStore

SQLite-backed implementation of the Store trait.

SqliteStoreConfig

Configuration for SqliteStore.

SqliteVectorStore

SQLite-backed implementation of the VectorStore trait.

SqliteVectorStoreConfig

Configuration for SqliteVectorStore.

Traits

Checkpointer

Trait for persisting graph state checkpoints.

Embeddings

Trait for embedding text into vectors.

LlmCache

Trait for caching LLM responses.

Store

Persistent key-value store trait for cross-invocation state.

VectorStore

Trait for vector storage backends.