nat.plugins.redis.redis_editor#

Attributes#

`logger`
`INDEX_NAME`

Classes#

RedisEditor

Wrapper class that implements NAT interfaces for Redis memory storage.

Module Contents#

logger#

INDEX_NAME = 'memory_idx'#

class RedisEditor( redis_client: redis.asyncio.Redis, key_prefix: str, embedder: langchain_core.embeddings.Embeddings, )#

Bases: nat.memory.interfaces.MemoryEditor

Wrapper class that implements NAT interfaces for Redis memory storage.

Initialize Redis client for memory storage.

Args:: redis_client: (redis.Redis) Redis client key_prefix: (str) Redis key prefix embedder: (Embeddings) Embedder for semantic search functionality

_client: redis.asyncio.Redis#

_key_prefix: str#

_embedder: langchain_core.embeddings.Embeddings#

async add_items(items: list[nat.memory.models.MemoryItem]) → None#: Insert Multiple MemoryItems into Redis. Each MemoryItem is stored with its metadata and tags.

async search( query: str, top_k: int = 5, \*\*kwargs, ) → list[nat.memory.models.MemoryItem]#

Retrieve items relevant to the given query.

Args:

query (str): The query string to match. top_k (int): Maximum number of items to return. kwargs (dict): Keyword arguments to pass to the search method.

user_id (str): User ID for filtering results.

similarity_threshold (float, optional): Maximum similarity score threshold based on L2 (Euclidean) distance metric. Results with scores above this threshold are filtered out; if not specified, all top_k results are returned. Lower scores indicate higher similarity (0.0 = identical). Typical ranges: 0.0-0.5 (very similar), 0.5-1.0 (moderately similar), >1.0 (loosely related).

Returns:

list[MemoryItem]: The most relevant MemoryItems for the given query.

_create_memory_item( memory_data: dict, user_id: str, similarity_score: float | None = None, ) → nat.memory.models.MemoryItem#: Helper method to create a MemoryItem from Redis data.

async remove_items(\*\*kwargs)#: Remove memory items based on provided criteria.