New

The executive guide to generative AI

Read more
About usPartnersSupport|Login
« Get started with the serverless PHP client Get started with the serverless Ruby client »
Elastic Docs ›Serverless ›Elasticsearch Serverless ›Client libraries

Get started with the Python client

edit

Get started with the Python client

edit

This page guides you through the installation process of the Python client, shows you how to initialize the client, and how to perform basic Elasticsearch operations with it.

See the Python client documentation for more detailed usage instructions.

The same client is used for Elasticsearch Serverless, on-premise and managed Elasticsearch. Some API endpoints are however not available in Elasticsearch Serverless.

Requirements

edit
  • Python 3.9 or higher
  • pip

Documentation

edit

Find the full documentation for the Python client on readthedocs.

Installation

edit

Using the command line

edit

You can install the Python client with the following commands:

python -m pip install elasticsearch

Initialize the client

edit

Initialize the client using your API key and Elasticsearch endpoint:

from elasticsearch import Elasticsearch

client = Elasticsearch(
    "https://...",  # Your project's Elasticsearch endpoint
    api_key='api-key',  # API key for your project
)

To get API keys for the Elasticsearch endpoint for a project, see Get started.

Using the API

edit

After you’ve initialized the client, you can start ingesting documents. You can use the bulk API for this. This API enables you to index, update, and delete several documents in one request.

Creating an index and ingesting documents

edit

You can call the bulk API with a body parameter, an array of hashes that define the action, and a document.

The following is an example of indexing some classic books into the books index:

from datetime import datetime

client.bulk(
    body=[
        {"index": {"_index": "books", "_id": "1"}},
        {"title": "Infinite Jest", "author": "David Foster Wallace", "published_on": datetime(1996, 2, 1)},
        {"index": {"_index": "books", "_id": "2"}},
        {"title": "Ulysses", "author": "James Joyce", "published_on": datetime(1922, 2, 2)},
        {"index": {"_index": "books", "_id": "3"}},
        {"title": "Just Kids", "author": "Patti Smith", "published_on": datetime(2010, 1, 19)},
    ],
)

Getting documents

edit

You can get documents by using the following code:

response = client.get(index="books", id="1")
print(response.body)

Searching

edit

Now that some data is available, you can search your documents using the search API:

response = client.search(index="books", query={
    "match": {
        "title": "infinite"
    }
})

for hit in response["hits"]["hits"]:
    print(hit["_source"])

Updating a document

edit

You can call the update API to update a document:

client.update(index="books", id="2", doc={
    "author": "James Augustine Aloysius Joyce",
    "pages": 732,
})

Deleting a document

edit

You can call the delete API to delete a document:

client.delete(index="books", id="3")

Deleting an index

edit
client.indices.delete(index="books")
« Get started with the serverless PHP client Get started with the serverless Ruby client »