AWS SDK S3 Examples: Upload, Download, Presigned URLs and Multipart (Python, PHP, Node.js)

By Ravindra Bagale, trainer · LinkedIn · Last updated: 24 October 2026

1. The big picture: a locker room

Why do we need this?

A gym gives each member a locker with a number. You put a bag in, take it out, or hand a friend a one-time pass that works for 10 minutes. S3 is the locker room: the bucket is the room, the key is the locker number, the object is the bag, and a presigned URL is the one-time pass.

What is it?

Python has a high-level upload_file/download_file that does multipart for you. PHP has putObject and the MultipartUploader. Node uses PutObjectCommand and the Upload class from @aws-sdk/lib-storage.

How to do it (step by step)

2. Create a bucket and upload

Why do we need this?

Aisha wants every client photo shoot stored safely, not on one laptop.

What is it?

Outside us-east-1 you must pass a LocationConstraint. Set ContentType so browsers show the image instead of downloading it.

How to do it (step by step)

1. Python: create the bucket and upload a file.

import boto3

s3 = boto3.client("s3", region_name="me-central-1")
bucket = "aisha-studio-111122223333"
s3.create_bucket(Bucket=bucket, CreateBucketConfiguration={"LocationConstraint": "me-central-1"})
s3.upload_file("shoot-01.jpg", bucket, "2026/oct/shoot-01.jpg", ExtraArgs={"ContentType": "image/jpeg"})
print("Uploaded to", bucket)

Expected output:

Uploaded to aisha-studio-111122223333

2. PHP: putObject from a file path.

<?php
require "vendor/autoload.php";
use Aws\S3\S3Client;

$s3 = new S3Client(["region" => "ap-south-1"]);
$bucket = "ravi-prints-nashik-111122223333";
$s3->createBucket(["Bucket" => $bucket, "CreateBucketConfiguration" => ["LocationConstraint" => "ap-south-1"]]);
$s3->waitUntil("BucketExists", ["Bucket" => $bucket]);
$r = $s3->putObject(["Bucket" => $bucket, "Key" => "orders/inv-1001.pdf", "SourceFile" => "inv-1001.pdf", "ContentType" => "application/pdf"]);
echo "ETag: ", $r["ETag"], PHP_EOL;

Expected output:

ETag: "9b2cf535f27731c974343645a3985328"

3. Node.js: PutObjectCommand with a file stream.

import { S3Client, PutObjectCommand } from "@aws-sdk/client-s3";
import { createReadStream, statSync } from "node:fs";

const s3 = new S3Client({ region: "eu-central-1" });
const r = await s3.send(new PutObjectCommand({
  Bucket: "lena-cafe-berlin-111122223333",
  Key: "menu/autumn.png",
  Body: createReadStream("autumn.png"),
  ContentLength: statSync("autumn.png").size,
  ContentType: "image/png",
}));
console.log("ETag:", r.ETag);

Expected output:

ETag: "5d41402abc4b2a76b9719d911017c592"

4. Verify: the object exists.

aws s3 ls s3://aisha-studio-111122223333/2026/oct/

Expected output:

2026-10-24 10:15:02     245760 shoot-01.jpg

3. Download and list

Why do we need this?

Ravi needs the invoice PDF back on the shop PC, and a list of all October orders.

What is it?

GetObject returns a stream. Prefix filters keys like a folder. Use a paginator when there can be more than 1,000 keys.

How to do it (step by step)

1. Python: download and list by prefix.

import boto3

s3 = boto3.client("s3", region_name="ap-south-1")
b = "ravi-prints-nashik-111122223333"
s3.download_file(b, "orders/inv-1001.pdf", "copy-inv-1001.pdf")
for o in s3.list_objects_v2(Bucket=b, Prefix="orders/").get("Contents", []):
    print(o["Key"], o["Size"])

Expected output:

orders/inv-1001.pdf 48213

2. PHP: getObject with SaveAs.

<?php
require "vendor/autoload.php";
$s3 = new Aws\S3\S3Client(["region" => "ap-south-1"]);
$b = "ravi-prints-nashik-111122223333";
$s3->getObject(["Bucket" => $b, "Key" => "orders/inv-1001.pdf", "SaveAs" => "copy-inv-1001.pdf"]);
foreach ($s3->listObjectsV2(["Bucket" => $b, "Prefix" => "orders/"])["Contents"] as $o) {
    echo $o["Key"], " ", $o["Size"], PHP_EOL;
}

Expected output:

orders/inv-1001.pdf 48213

3. Node.js: read the body as a string.

import { S3Client, PutObjectCommand, GetObjectCommand, ListObjectsV2Command } from "@aws-sdk/client-s3";

const s3 = new S3Client({ region: "eu-central-1" });
const B = "lena-cafe-berlin-111122223333";
await s3.send(new PutObjectCommand({ Bucket: B, Key: "menu/today.txt", Body: "Apfelstrudel €4.50" }));
const obj = await s3.send(new GetObjectCommand({ Bucket: B, Key: "menu/today.txt" }));
console.log(await obj.Body.transformToString());
const list = await s3.send(new ListObjectsV2Command({ Bucket: B, Prefix: "menu/" }));
console.log(list.Contents.map(o => o.Key));

Expected output:

Apfelstrudel €4.50
[ 'menu/autumn.png', 'menu/today.txt' ]

4. Verify: downloaded file matches.

ls -l copy-inv-1001.pdf | awk '{print $5}'

Expected output:

48213

4. Presigned URLs

Why do we need this?

Aisha's client in New York must download 40 photos, but the bucket stays private. A presigned URL gives time-limited access with no AWS login.

What is it?

The SDK signs the URL locally with your credentials. GET URLs let people download; PUT URLs let a browser upload directly to S3 without passing through your server.

How to do it (step by step)

1. Python: a 1-hour download link.

import boto3

s3 = boto3.client("s3", region_name="me-central-1")
url = s3.generate_presigned_url(
    "get_object",
    Params={"Bucket": "aisha-studio-111122223333", "Key": "2026/oct/shoot-01.jpg"},
    ExpiresIn=3600,
)
print(url[:80])

Expected output:

https://aisha-studio-111122223333.s3.me-central-1.amazonaws.com/2026/oct/shoot-0

2. PHP: createPresignedRequest.

<?php
require "vendor/autoload.php";
$s3 = new Aws\S3\S3Client(["region" => "ap-south-1"]);
$cmd = $s3->getCommand("GetObject", ["Bucket" => "ravi-prints-nashik-111122223333", "Key" => "orders/inv-1001.pdf"]);
$req = $s3->createPresignedRequest($cmd, "+15 minutes");
echo (string) $req->getUri(), PHP_EOL;

Expected output:

https://ravi-prints-nashik-111122223333.s3.ap-south-1.amazonaws.com/orders/inv-1001.pdf?X-Amz-Algorithm=AWS4-HMAC-SHA256&...&X-Amz-Expires=900&X-Amz-Signature=...

3. Node.js: a presigned PUT URL for browser uploads.

// npm install @aws-sdk/s3-request-presigner
import { S3Client, PutObjectCommand } from "@aws-sdk/client-s3";
import { getSignedUrl } from "@aws-sdk/s3-request-presigner";

const s3 = new S3Client({ region: "eu-central-1" });
const url = await getSignedUrl(s3, new PutObjectCommand({
  Bucket: "lena-cafe-berlin-111122223333", Key: "uploads/review.jpg", ContentType: "image/jpeg",
}), { expiresIn: 600 });
console.log(url.split("?")[0]);

Expected output:

https://lena-cafe-berlin-111122223333.s3.eu-central-1.amazonaws.com/uploads/review.jpg

4. Verify: download with curl, no AWS login.

curl -s -o test.jpg -w "%{http_code}\n" "$URL"

Expected output:

200

5. Verify: upload with the PUT URL.

curl -s -X PUT -H "Content-Type: image/jpeg" --data-binary @review.jpg -w "%{http_code}\n" "$PUT_URL"

Expected output:

200

5. Multipart upload for big files

Why do we need this?

James in Sydney uploads a 3 GB wedding video. A single PUT is limited to 5 GB and one network blip restarts everything. Multipart sends parts in parallel and retries only the failed part.

What is it?

Parts are at least 5 MB (except the last), up to 10,000 parts. The high-level helpers do create, upload parts and complete for you.

How to do it (step by step)

1. Python: TransferConfig controls the threshold and threads.

import boto3
from boto3.s3.transfer import TransferConfig

cfg = TransferConfig(multipart_threshold=64 * 1024 * 1024, multipart_chunksize=64 * 1024 * 1024, max_concurrency=8)
s3 = boto3.client("s3", region_name="ap-southeast-2")
s3.upload_file("wedding.mp4", "james-films-sydney-111122223333", "raw/wedding.mp4", Config=cfg)
head = s3.head_object(Bucket="james-films-sydney-111122223333", Key="raw/wedding.mp4")
print("ETag:", head["ETag"])

Expected output:

ETag: "d41d8cd98f00b204e9800998ecf8427e-48"

2. PHP: MultipartUploader.

<?php
require "vendor/autoload.php";
use Aws\S3\MultipartUploader;
use Aws\Exception\MultipartUploadException;

$s3 = new Aws\S3\S3Client(["region" => "ap-southeast-2"]);
$up = new MultipartUploader($s3, "wedding.mp4", ["bucket" => "james-films-sydney-111122223333", "key" => "raw/wedding.mp4", "part_size" => 64 * 1024 * 1024, "concurrency" => 5]);
try {
    $r = $up->upload();
    echo "Done: ", $r["ObjectURL"], PHP_EOL;
} catch (MultipartUploadException $e) {
    echo "Failed: ", $e->getMessage(), PHP_EOL;
}

Expected output:

Done: https://james-films-sydney-111122223333.s3.ap-southeast-2.amazonaws.com/raw/wedding.mp4

3. Node.js: Upload from @aws-sdk/lib-storage with progress.

// npm install @aws-sdk/lib-storage
import { S3Client } from "@aws-sdk/client-s3";
import { Upload } from "@aws-sdk/lib-storage";
import { createReadStream } from "node:fs";

const up = new Upload({
  client: new S3Client({ region: "ap-southeast-2" }),
  params: { Bucket: "james-films-sydney-111122223333", Key: "raw/wedding.mp4", Body: createReadStream("wedding.mp4") },
  partSize: 64 * 1024 * 1024, queueSize: 4,
});
up.on("httpUploadProgress", p => console.log("part", p.part, p.loaded));
const r = await up.done();
console.log("Done:", r.Key);

Expected output:

part 1 67108864
part 2 134217728
...
Done: raw/wedding.mp4

4. Verify: the ETag ends with -partcount and no uploads are left half-done.

aws s3api head-object --bucket james-films-sydney-111122223333 --key raw/wedding.mp4 --query ETag --output text
aws s3api list-multipart-uploads --bucket james-films-sydney-111122223333 --query "length(Uploads || `[]`)"

Expected output:

"d41d8cd98f00b204e9800998ecf8427e-48"
0

6. The same call in Java, Go, .NET and the CLI

Why do we need this?

Teams often mix languages. Once you know the call in Python, PHP and Node.js, the other SDKs are the same API with different syntax.

What is it?

All of them call the PutObject API. Use the CLI line to test it quickly from a terminal.

How to do it (step by step)

7. Clean up so nothing keeps costing

Why do we need this?

Lab resources left running cost money and can be a security risk.

What is it?

Delete in reverse order: children first, then parents.

How to do it (step by step)

1. Python: empty and delete a bucket.

import boto3
b = boto3.resource("s3").Bucket("aisha-studio-111122223333")
b.objects.all().delete()
b.delete()
print("deleted")

Expected output:

deleted

2. Abort any half-finished multipart uploads.

aws s3api list-multipart-uploads --bucket james-films-sydney-111122223333 --query "Uploads[].[Key,UploadId]" --output text

Expected output:

(no output means success)

8. Interview questions and answers

Revision, comparison and interview cards

Revise this lesson in a few minutes. Save the cards on your phone or download them as PNG files. The QR code on each card opens this lesson.

More cards: Ravi Sir's Cards.