Skip to main content

Elasticsearch

Elasticsearch is a distributed, RESTful search and analytics engine, capable of performing both vector and lexical search. It is built on top of the Apache Lucene library.

This notebook shows how to use chat message history functionality with Elasticsearch.

Set up Elasticsearchโ€‹

There are two main ways to set up an Elasticsearch instance:

  1. Elastic Cloud. Elastic Cloud is a managed Elasticsearch service. Sign up for a free trial.

  2. Local Elasticsearch installation. Get started with Elasticsearch by running it locally. The easiest way is to use the official Elasticsearch Docker image. See the Elasticsearch Docker documentation for more information.

Install dependenciesโ€‹

%pip install --upgrade --quiet  elasticsearch langchain langchain-community

Authenticationโ€‹

How to obtain a password for the default "elastic" userโ€‹

To obtain your Elastic Cloud password for the default "elastic" user:

  1. Log in to the Elastic Cloud console
  2. Go to "Security" > "Users"
  3. Locate the "elastic" user and click "Edit"
  4. Click "Reset password"
  5. Follow the prompts to reset the password

Use the Username/passwordโ€‹

es_username = os.environ.get("ES_USERNAME", "elastic")
es_password = os.environ.get("ES_PASSWORD", "change me...")

history = ElasticsearchChatMessageHistory(
es_url=es_url,
es_user=es_username,
es_password=es_password,
index="test-history",
session_id="test-session"
)

How to obtain an API keyโ€‹

To obtain an API key:

  1. Log in to the Elastic Cloud console
  2. Open Kibana and go to Stack Management > API Keys
  3. Click "Create API key"
  4. Enter a name for the API key and click "Create"

Use the API keyโ€‹

es_api_key = os.environ.get("ES_API_KEY")

history = ElasticsearchChatMessageHistory(
es_api_key=es_api_key,
index="test-history",
session_id="test-session"
)

Initialize Elasticsearch client and chat message historyโ€‹

import os

from langchain_community.chat_message_histories import (
ElasticsearchChatMessageHistory,
)

es_url = os.environ.get("ES_URL", "http://localhost:9200")

# If using Elastic Cloud:
# es_cloud_id = os.environ.get("ES_CLOUD_ID")

# Note: see Authentication section for various authentication methods

history = ElasticsearchChatMessageHistory(
es_url=es_url, index="test-history", session_id="test-session"
)

Use the chat message historyโ€‹

history.add_user_message("hi!")
history.add_ai_message("whats up?")
indexing message content='hi!' additional_kwargs={} example=False
indexing message content='whats up?' additional_kwargs={} example=False

Was this page helpful?