NAME

Amazon::S3::Thin - A thin, lightweight, low-level Amazon S3 client

SYNOPSIS

  use Amazon::S3::Thin;

  # Pass in explicit credentials
  my $s3client = Amazon::S3::Thin->new({
        aws_access_key_id     => $aws_access_key_id,
        aws_secret_access_key => $aws_secret_access_key,
        aws_session_token     => $aws_session_token, # optional
        region                => $region, # e.g. 'ap-northeast-1'
      });

  # Get credentials from environment
  my $s3client = Amazon::S3::Thin->new({region => $region, credential_provider => 'env'});

  # Get credentials from instance metadata
  my $s3client = Amazon::S3::Thin->new({
      region              => $region,
      credential_provider => 'metadata',
      version             => 2,         # optional (default 2)
      role                => 'my-role', # optional
    });

  # Get credentials from ECS task role
  my $s3client = Amazon::S3::Thin->new({
      region              => $region,
      credential_provider => 'ecs_container',
    });

  my $bucket = "mybucket";
  my $key = "dir/file.txt";
  my $response;

  $response = $s3client->put_bucket($bucket);

  $response = $s3client->put_object($bucket, $key, "hello world");

  $response = $s3client->get_object($bucket, $key);
  print $response->content; # => "hello world"

  $response = $s3client->delete_object($bucket, $key);

  $response = $s3client->list_objects(
                              $bucket,
                              {prefix => "foo", delimiter => "/"}
                             );

You can also pass any useragent as you like

  my $s3client = Amazon::S3::Thin->new({
          ...
          ua => $any_LWP_copmatible_useragent,
      });

Signature version 4 is used by default. To use signature version 2, add a signature_version option:

  my $s3client = Amazon::S3::Thin->new({
          ...
          signature_version     => 2,
      });

DESCRIPTION

Amazon::S3::Thin is a thin, lightweight, low-level Amazon S3 client.

It's designed for only ONE purpose: Send a request and get a response.

In detail, it offers the following features:

Low Level

It returns an HTTP::Response object so you can easily inspect what's happening inside, and can handle errors as you like.

Low Dependency

It does not require any XML::* modules, so installation is easy;

Low Learning Cost

The interfaces are designed to follow S3 official REST APIs. So it is easy to learn.

Comparison to precedent modules

There are already some useful modules like Amazon::S3, Net::Amazon::S3 on CPAN. They provide a "Perlish" interface, which looks pretty for Perl programmers, but they also hide low-level behaviors. For example, the "get_key" method translate HTTP status 404 into undef and HTTP 5xx status into exception.

In some situations, it is very important to see the raw HTTP communications. That's why I made this module.

CONSTRUCTOR

new( \%params )

Receives: hashref with options.

Returns: Amazon::S3::Thin object

It can receive the following arguments:

  • credential_provider (default: credentials) - specify where to source credentials from. Options are:

    • credentials - existing behaviour, pass in credentials via aws_access_key_id and aws_secret_access_key

    • env - fetch credentials from environment variables

    • metadata - fetch credentials from EC2 instance metadata service

    • ecs_container - fetch credentials from ECS task role

  • region - (REQUIRED) region of your buckets you access- (currently used only when signature version is 4)

  • aws_access_key_id (REQUIRED [provider: credentials]) - an access key id of your credentials.

  • aws_secret_access_key (REQUIRED [provider: credentials]) - an secret access key of your credentials.

  • version (OPTIONAL [provider: metadata]) - version of metadata service to use, either 1 or 2. read more

  • role (OPTIONAL [provider: metadata]) - IAM instance role to use, otherwise the first is selected

  • secure - whether to use https or not. Default is 0 (http).

  • ua - a user agent object, compatible with LWP::UserAgent. Default is an instance of LWP::UserAgent.

  • signature_version - AWS signature version to use. Supported values are 2 and 4. Default is 4.

  • debug - debug option. Default is 0 (false). If set 1, contents of HTTP request and response are shown on stderr

  • virtual_host - whether to use virtual-hosted style request format. Default is 0 (path-style).

ACCESSORS

The following accessors are provided. You can use them to get/set your object's attributes.

secure

Whether to use https (1) or http (0) when connecting to S3.

ua

The user agent used internally to perform requests and return responses. If you set this attribute, please make sure you do so with an object compatible with LWP::UserAgent (i.e. providing the same interface).

debug

Debug option.

Operations on Buckets

put_bucket( $bucket [, $headers])

Arguments:

1. bucket - a string with the bucket
2. headers (optional) - hashref with extra header information

delete_bucket( $bucket [, $headers])

Arguments:

1. bucket - a string with the bucket
2. headers (optional) - hashref with extra header information

Operations on Objects

get_object( $bucket, $key [, $headers] )

Arguments:

1. bucket - a string with the bucket
2. key - a string with the key
3. headers (optional) - hashref with extra header information

Returns: an HTTP::Response object for the request. Use the content() method on the returned object to read the contents:

    my $res = $s3->get_object( 'my.bucket', 'my/key.ext' );

    if ($res->is_success) {
        my $content = $res->content;
    }

The GET operation retrieves an object from Amazon S3.

For more information, please refer to Amazon's documentation for GET.

head_object( $bucket, $key )

Arguments:

1. bucket - a string with the bucket
2. key - a string with the key

Returns: an HTTP::Response object for the request. Use the header() method on the returned object to read the metadata:

    my $res = $s3->head_object( 'my.bucket', 'my/key.ext' );

    if ($res->is_success) {
        my $etag = $res->header('etag'); #=> `"fba9dede5f27731c9771645a39863328"`
    }

The HEAD operation retrieves metadata of an object from Amazon S3.

For more information, please refer to Amazon's documentation for HEAD.

delete_object( $bucket, $key )

Arguments: a string with the bucket name, and a string with the key name.

Returns: an HTTP::Response object for the request.

The DELETE operation removes the null version (if there is one) of an object and inserts a delete marker, which becomes the current version of the object. If there isn't a null version, Amazon S3 does not remove any objects.

Use the response object to see if it succeeded or not.

For more information, please refer to Amazon's documentation for DELETE.

copy_object( $src_bucket, $src_key, $dst_bucket, $dst_key [, $headers] )

Arguments: a list with source (bucket, key) and destination (bucket, key), hashref with extra header information (optional).

Returns: an HTTP::Response object for the request.

This method is a variation of the PUT operation as described by Amazon's S3 API. It creates a copy of an object that is already stored in Amazon S3. This "PUT copy" operation is the same as performing a GET from the old bucket/key and then a PUT to the new bucket/key.

Note that the COPY request might return error response in 200 OK, but this method will determine the error response and rewrite the status code to 500.

For more information, please refer to Amazon's documentation for COPY.

put_object( $bucket, $key, $content [, $headers] )

Arguments:

1. bucket - a string with the destination bucket
2. key - a string with the destination key
3. content - a string with the content to be uploaded
4. headers (optional) - hashref with extra header information

Returns: an HTTP::Response object for the request.

The PUT operation adds an object to a bucket. Amazon S3 never adds partial objects; if you receive a success response, Amazon S3 added the entire object to the bucket.

For more information, please refer to Amazon's documentation for PUT.

delete_multiple_objects( $bucket, @keys )

Arguments: a string with the bucket name, and an array with all the keys to be deleted.

Returns: an HTTP::Response object for the request.

The Multi-Object Delete operation enables you to delete multiple objects (up to 1000) from a bucket using a single HTTP request. If you know the object keys that you want to delete, then this operation provides a suitable alternative to sending individual delete requests with delete_object(), reducing per-request overhead.

For more information, please refer to Amazon's documentation for DELETE multiple objects.

list_objects( $bucket [, \%options ] )

Arguments: a string with the bucket name, and (optionally) a hashref with any of the following options:

  • prefix (string) - only return keys that begin with the specified prefix. You can use prefixes to separate a bucket into different groupings of keys, the same way you'd use a folder in a file system.

  • delimiter (string) - group keys that contain the same string between the beginning of the key (or after the prefix, if specified) and the first occurrence of the delimiter.

  • encoding-type (string) - if set to "url", will encode keys in the response (useful when the XML parser can't work unicode keys).

  • marker (string) - specifies the key to start with when listing objects. Amazon S3 returns object keys in alphabetical order, starting with the key right after the marker, in order.

  • max-keys (string) - Sets the maximum number of keys returned in the response body. You can add this to your request if you want to retrieve fewer than the default 1000 keys.

Returns: an HTTP::Response object for the request. Use the content() method on the returned object to read the contents:

This method returns some or all (up to 1000) of the objects in a bucket. Note that the response might contain fewer keys but will never contain more. If there are additional keys that satisfy the search criteria but were not returned because the limit (either 1000 or max-keys) was exceeded, the response will contain <IsTruncated>true</IsTruncated>. To return the additional keys, see marker above.

For more information, please refer to Amazon's documentation for REST Bucket GET.

generate_presigned_post( $bucket, $key [, $fields, $conditions, $expires_in ] )

Arguments:

1. bucket (string) - a string with the destination bucket
2. key (string) - a string with the destination key
3. fields (arrayref) - an arrayref of key/value pairs to prefilled form fields to build on top of
4. conditions (arrayref) - an arrayref of condition (arrayref or hashref) to include in the policy
5. expires_in (number) - a number of seconds from the current time before expiring presigned url

Returns: a hashref with two elements url and fields. url is the url to post to. fields is an arrayref filled with the form fields and respective values to use when submitting the post. (You must follow the order of fields)

This method generates presigned url for uploading a file to Amazon S3 using HTTP POST. The original implementation from boto3, this was transplanted referencing S3Client.generate_presigned_post().

Note: this method is supported only signature v4.

This is an example of generating a presigned url and uploading test.txt file. In this case, you can set the object metadata x-amz-meta-foo with any value and the uploading size is limited to 1MB.

    my $presigned = $s3->generate_presigned_post('my.bucket', 'my/key.ext', [
        'x-amz-meta-foo' => 'bar',
    ], [
        ['starts-with' => '$x-amz-meta-foo', ''],
        ['content-length-range' => 1, 1024*1024],
    ], 24*60*60);

    my $ua = LWP::UserAgent->new;
    my $res = $ua->post(
        $presigned->{url},
        Content_Type => 'multipart/form-data',
        Content      => [
            @{$presigned->{fields}},
            file => ['test.txt'],
        ],
    );

For more information, please refer to Amazon's documentation for Creating a POST Policy.

TODO

lots of APIs are not implemented yet.

REPOSITORY

https://github.com/DQNEO/Amazon-S3-Thin

LICENSE

Copyright (C) DQNEO.

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

AUTHOR

DQNEO

THANKS TO

Timothy Appnel Breno G. de Oliveira

SEE ALSO

Amazon::S3, https://github.com/tima/perl-amazon-s3

Net::Amazon::S3

Amazon S3 API Reference : REST API

Amazon S3 API Reference : List of Error Codes