---
title: "Using boto3 for Working With S3 Storage in Python | Hostman Docs"
description: "Get started with the AWS SDK for Python S3 client. Explore installation, authentication methods, AWS profiles, environment variables, and sample code for managing objects in S3-compatible storage."
---

> For the complete documentation index for AI agents, see [llms.txt](https://hostman.com/llms.txt).

[boto3](https://pypi.org/project/boto3/) is the official Amazon SDK for working with S3 in Python. It allows you to manage objects in storage: upload and download files, list objects, work with metadata, perform multipart uploads, and generate pre-signed URLs. 

## Installation

To install `boto3`, use `pip`:

```shell
pip install boto3
```

After installation, the library is ready to use in your Python code.

## Configuring Access

When using `boto3`, access credentials can be configured in several ways.

### Using ~/.aws/config and ~/.aws/credentials

When you use the [AWS CLI](https://hostman.com/docs/s3/tools/aws-cli/), a `.aws` directory containing configuration files is automatically created in the user’s home directory. If necessary, you can create this directory and the files manually without installing the AWS CLI.

The `~/.aws/credentials` file should contain the following:

```shell
[default]
aws_access_key_id = <ACCESS_KEY>
aws_secret_access_key = <SECRET_KEY>
```

And `~/.aws/config`:

```shell
[default]
output = json
endpoint_url = https://s3.hmstorage.net
region = us-2
```

If you use the `default` profile, no additional configuration is required in the code; `boto3` will automatically load these settings.

### Using Environment Variables

You can also provide access credentials via environment variables:

```shell
export AWS_ACCESS_KEY_ID=<ACCESS_KEY>
export AWS_SECRET_ACCESS_KEY=<SECRET_KEY>
```

### Passing Credentials in Code

Credentials can be passed directly when creating the S3 client:

```py
import boto3

s3 = boto3.client(
    "s3",
    aws_access_key_id="<ACCESS_KEY>",
    aws_secret_access_key="<SECRET_KEY>",
    endpoint_url="https://s3.hmstorage.net",
    region_name="us-2"
)
```

This approach is not recommended for production, as access keys are stored directly in the source code.

### Using Named Profiles

If multiple profiles are configured in `~/.aws/config` and `~/.aws/credentials`, you can explicitly specify the required profile when creating a session:

```py
import boto3

session = boto3.Session(profile_name="myprofile")
s3 = session.client("s3")
```

This is useful when the same codebase is used to work with multiple buckets or environments.

## Example

Below is an example demonstrating basic object operations using the SDK. The script performs the following actions:

-   uploads a file to a bucket;
-   lists all objects in the bucket;
-   downloads the uploaded file under a different name;
-   deletes the object from storage.

Before running the script, create a file named `example.txt` in the same directory as the script.

Example code:

```py
import boto3
import os

bucket_name = "bucket_name"
endpoint_url = "https://s3.hmstorage.net"
local_upload_path = "example.txt"
s3_key = "example.txt"
local_download_path = "downloaded_example.txt"

# Create S3 client
s3 = boto3.client("s3", endpoint_url=endpoint_url)

# 1. Upload file to the bucket
if os.path.exists(local_upload_path):
    print(f"Uploading {local_upload_path} to the bucket...")
    s3.upload_file(local_upload_path, bucket_name, s3_key)
    print("Upload completed.")
else:
    print(f"File {local_upload_path} not found. Skipping upload.")

# 2. List objects in the bucket
print("\nBucket contents:")
paginator = s3.get_paginator("list_objects_v2")
for page in paginator.paginate(Bucket=bucket_name):
    for obj in page.get("Contents", []):
        print(f"- {obj['Key']}")

# 3. Download file from the bucket
print(f"\nDownloading {s3_key} to {local_download_path}...")
s3.download_file(bucket_name, s3_key, local_download_path)
print("Download completed.")

# 4. Delete file from the bucket
print(f"\nDeleting {s3_key} from the bucket...")
s3.delete_object(Bucket=bucket_name, Key=s3_key)
print("Deletion completed.")
```