Download the PHP package akeeba/s3 without Composer

On this page you can find all versions of the php package akeeba/s3. It is possible to download/install these versions without Composer. Possible dependencies are resolved automatically.

FAQ

After the download, you have to make one include require_once('vendor/autoload.php');. After that you have to import the classes with use statements.

Example:
If you use only one package a project is not needed. But if you use more then one package, without a project it is not possible to import the classes with use statements.

In general, it is recommended to use always a project to download your libraries. In an application normally there is more than one library needed.
Some PHP packages are not free to download and because of that hosted in private repositories. In this case some credentials are needed to access such packages. Please use the auth.json textarea to insert credentials, if a package is coming from a private repository. You can look here for more information.

  • Some hosting areas are not accessible by a terminal or SSH. Then it is not possible to use Composer.
  • To use Composer is sometimes complicated. Especially for beginners.
  • Composer needs much resources. Sometimes they are not available on a simple webspace.
  • If you are using private repositories you don't need to share your credentials. You can set up everything on our site and then you provide a simple download link to your team member.
  • Simplify your Composer build process. Use our own command line tool to download the vendor folder as binary. This makes your build process faster and you don't need to expose your credentials for private repositories.
Please rate this library. Is it a good library?

Informations about the package s3

Akeeba Amazon S3 Connector

A compact, dependency-less Amazon S3 API client implementing the most commonly used features.

This library is designed to work with Amazon S3 proper, as well as S3-compatible services such as but not limited to Wasabi, Google Storage, Synology C2, ExoScale etc.

Why reinvent the wheel

After having a lot of impossible to debug problems with Amazon's Guzzle-based AWS SDK we decided to roll our own connector for Amazon S3. This is by no means a complete implementation, just a small subset of S3's features which are required by our software. The design goals are simplicity, no external dependencies and a low memory footprint.

This code was originally based on S3.php written by Donovan Schonknecht which is available under a BSD-like license. This repository no longer reflects the original author's work and should not be confused with it.

This software is distributed under the GNU General Public License version 3 or, at your option, any later version published by the Free Software Foundation (FSF). In short, it's GPL-3.0-or-later, as noted in composer.json.

Important notes about version 2

PHP version support since 2.0

Akeeba Amazon S3 Connector version 2 has dropped support for PHP 5.3 to 7.0 inclusive.

The most significant change in this version is that all methods use scalar type hints for parameters and return values. This may break existing consumers which relied on implicit type conversion.

Namespace change since 2.3

Up to and including version 2.2 of the library, the namespace was \Akeeba\Engine\Postproc\Connector\S3v4. From version 2.3 of the library the namespace has changed to \Akeeba\S3.

The library automatically registers aliases of the old classes to the new ones, thus ensuring updating the library will not introduce backwards incompatible changes. This is why it's not a major version update. Aliases will remain in place until at least version 3.0 of the library.

Using the connector

You need to define a constant before using or referencing any class in the library:

All library files have a line similar to

to prevent direct access to the libraries files. This is intentional. The primary use case for this library is mass-distributed software which gets installed in a publicly accessible subdirectory of the web root. This line prevents any accidental path disclosure from PHP error messages if someone were to access these files directly on misconfigured servers.

If you are writing a Joomla extension, especially a plugin or module, please always check if the constant has already been defined before defining it yourself. Thank you!

Get a connector object

If you are running inside an Amazon EC2 instance you can fetch temporary credentials from the instance's metadata server using the IAM Role attached to the EC2 instance. In this case you need to do this (169.254.169.254 is a fixed IP hosting the instance's metadata cache service):

where $yourRegion is the AWS region of your bucket, e.g. us-east-1. Please note that we are passing the security token ($credentials['Token']) to the Configuration object. This is REQUIRED. The temporary credentials returned by the metadata service won't work without it.

Another point worth noting is that the temporary credentials don't last forever. Check the $credentials['Expiration'] to see when they are about to expire. Amazon recommends that you retry fetching new credentials from the metadata service 10 minutes before your cached credentials are set to expire. The metadata service is guaranteed to provision fresh temporary credentials by that time.

Listing buckets

Returns an array like this:

Listing bucket contents

If you want to list "subdirectories" you need to do

The last parameter (common prefixes) controls the listing of "subdirectories"

Uploading (small) files

From a file:

From a string:

From a stream resource:

In all cases the entirety of the file has to be loaded in memory.

Uploading large file with multipart (chunked) uploads

Files are uploaded in 5Mb chunks.

As long as you keep track of the UploadId, PartNumber and ETags you can have each uploadMultipart call in a separate page load to prevent timeouts.

Get presigned URLs

Allows browsers to download files directly without exposing your credentials and without going through your server:

The last parameter controls how many seconds into the future this URL will be valid.

Download

To a file with absolute path $targetFile

To a string

Delete an object

Test if an object exists

The $headers variable contains an array with the S3 headers returned by the [HeadObject(https://docs.aws.amazon.com/AmazonS3/latest/API/API_HeadObject.html) API call. The header keys are always in lowercase. Please note that not all of the headers Amazon describes in their documentation are returned in every request.

Configuration options

The Configuration option has optional methods which can be used to enable some useful features in the connector.

You need to execute these methods against the Configuration object before passing it to the Connector's constructor. For example:

HTTPS vs plain HTTP

It is not recommended to use plain HTTP connections to Amazon S3. If, however, you have no other option you can tell the Configuration object to use plain HTTP URLs:

Caveat: HTTPS will only work if PHP can verify the TLS certificate of your endpoint. This may not be the case when using a local testing service (e.g. LocalStack), or for some buckets with dots in their names. Moreover, if you are on Windows, do note that neither PHP comes with a Certification Authority cache, nor is there a system-wide CA cache; you'll have to download it and configure PHP, or use composer/ca-bundle in your composer.json file.

Custom endpoint

You can use the Akeeba Amazon S3 Connector library with S3-compatible APIs such as DigitalOcean's Spaces by changing the endpoint URL.

Please note that if the S3-compatible APi uses v4 signatures you need to enter the region-specific endpoint domain name and the region when initializing the object, e.g.:

If your S3-compatible API uses v2 signatures you do not need to specify a region.

Caveat: Setting the endpoint resets the signature version and region. This is why you need to set them a second time, after setting the endpoint, as seen in the first example above.

Legacy path-style access

The S3 API calls made by this library will use by default the subdomain-style access. That is to say, the endpoint will be prefixed with the name of the bucket. For example, a bucket called example in the eu-west-1 region will be accessed using the endpoint URL example.s3.eu-west-1.amazonaws.com.

If you have buckets with characters that are invalid in the context of DNS (most notably dots and uppercase characters) this will fail. You will need to use the legacy path style instead. In this case the endpoint used is the generic region specific one (s3.eu-west-1.amazonaws.com in our example above), and the API URL will be prefixed with the bucket name.

You need to do:

Caveats:

Dualstack (IPv4 and IPv6) support

Amazon S3 supports dual-stack URLs which resolve to both IPv4 and IPv6 addresses. By default they are not used. If you want to enable this feature you need to do:

Caveat: This option only takes effect if you are using Amazon S3 proper. It will not have any effect with custom endpoints. DualStack support is deprecated by Amazon S3. We strongly advise you NOT to use it anymore.

Alternate Date Format

By default, this library uses the standard date format D, d M Y H:i:s O which Amazon incorrectly documents as "ISO 8601" (it's not, see the ISO 8601 Wikipedia entry for reference). Most third party, Amazon S3-compatible services use the same and understand it just fine.

A minority of services don't understand the GMT offset at the end of the date format, and instead need the format D, d M Y H:i:s T. You can set a flag to enable this behaviour like so:

Caveat: Enabling this flag breaks compatibility with S3 proper.

Using The HTTP Date Header Instead Of X-Amz-Date

Amazon documents that you should be using the standard HTTP Date header, and only resort to using the X-Amz-Date header when using the standard header is impossible, e.g. when creating pre-authorised (signed) URLs, or when your HTTP library does not let you set the standard header.

Unfortunately, some third party S3-compatible services such as Wasabi and ExoScale do NOT support the standard Date header at all. Using it makes them falsely spit out a message about the signature being wrong. They are the reason why, by default, we are passing the request date and time using the X-Amz-Date header.

If you are using a third party service which for any reason does not understand the X-Amz-Date header you need to set a flag which forces the use of the standard Date header like so:


All versions of s3 with dependencies

PHP Build Version
Package Version
Requires php Version >=7.1.0 <8.4
ext-curl Version *
ext-simplexml Version *
Composer command for our command line client (download client) This client runs in each environment. You don't need a specific PHP version etc. The first 20 API calls are free. Standard composer command

The package akeeba/s3 contains the following files

Loading the files please wait ....