Amazon Simple Storage Service (Amazon S3) is an object storage service

AWS S3 Directory

This covers how to load document objects from anAWS S3 Directory object.

%pip install--upgrade--quiet  boto3

from langchain_community.document_loadersimport S3DirectoryLoader

loader= S3DirectoryLoader("testing-hwc")

loader.load()

Specifying a prefix​

You can also specify a prefix for more fine-grained control over what files to load.

loader= S3DirectoryLoader("testing-hwc", prefix="fake")

loader.load()

[Document(page_content='Lorem ipsum dolor sit amet.', lookup_str='', metadata={'source': 's3://testing-hwc/fake.docx'}, lookup_index=0)]

Configuring the AWS Boto3 client​

You can configure the AWSBoto3 client by passingnamed arguments when creating the S3DirectoryLoader.This is useful for instance when AWS credentials can't be set as environment variables.See thelist of parameters that can be configured.

loader= S3DirectoryLoader(
"testing-hwc", aws_access_key_id="xxxx", aws_secret_access_key="yyyy"
)

loader.load()

Related​

• Document loaderconceptual guide
• Document loaderhow-to guides