Memory, base class¶

This implements the Cassandra-specific I/O facilities, and is then used by LangChain to power every other, more sophisticated, abstraction around keeping chat memory.

In [1]:

from langchain.memory import CassandraChatMessageHistory

from langchain.memory import CassandraChatMessageHistory

In [2]:

from cqlsession import getCQLSession, getCQLKeyspace
cqlMode = 'astra_db' # 'astra_db'/'local'
session = getCQLSession(mode=cqlMode)
keyspace = getCQLKeyspace(mode=cqlMode)

from cqlsession import getCQLSession, getCQLKeyspace cqlMode = 'astra_db' # 'astra_db'/'local' session = getCQLSession(mode=cqlMode) keyspace = getCQLKeyspace(mode=cqlMode)

Create the ChatMessageHistory¶

In [3]:

message_history = CassandraChatMessageHistory(
    session_id='test-session',
    session=session,
    keyspace=keyspace,
    ttl_seconds = 3600,
)
message_history.clear()

message_history = CassandraChatMessageHistory( session_id='test-session', session=session, keyspace=keyspace, ttl_seconds = 3600, ) message_history.clear()

The memory starts empty:

In [4]:

message_history.messages

message_history.messages

Out[4]:

[]

Insert and retrieve chat messages¶

In [5]:

message_history.add_user_message('Hello, I am human.')

message_history.add_user_message('Hello, I am human.')

In [6]:

message_history.add_ai_message('Hello, I am the bot.')

message_history.add_ai_message('Hello, I am the bot.')

In [7]:

message_history.messages

message_history.messages

Out[7]:

[HumanMessage(content='Hello, I am human.', additional_kwargs={}, example=False),
 AIMessage(content='Hello, I am the bot.', additional_kwargs={}, example=False)]

In [8]:

message_history.clear()

message_history.clear()

In [9]:

message_history.messages

message_history.messages

Out[9]:

[]