#!/usr/bin/env perl

# Copyright (C) 2007 by Mark Atwood <mark@fallenpegasus.com>.
# 
# This module is not an official Amazon product or service.  Information
# used to create this module was obtained only from publicly available
# information, mainly from the published Amazon documentation.
#
# This module is free software: you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as published
# by the Free Software Foundation, either version 2.1 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# and the GNU Lesser General Public License along with this program.
# If not, see <http://www.gnu.org/licenses/>.

use warnings;
use strict;
use Getopt::Long;
use Pod::Usage;
use Net::Amazon::S3;
use Net::Amazon::S3::Bucket;
use Getopt::ArgvFile qw/argvFile/;
use File::HomeDir;

my $aws_access_key_id = $ENV{AWS_ACCESS_KEY_ID};
my $aws_secret_access_key = $ENV{AWS_ACCESS_KEY_SECRET};
my $opt_verbose =0;
my $opt_help =0;
my $opt_man =0;
my $opt_secure =0;

my $opt_long =0;

# get the options from the users ~/.s3-tools file, if it exists
my $users_config = File::HomeDir->my_home() . '/.s3-tools';
if (-e $users_config) {
    unshift @ARGV, '@' . $users_config;
}
argvFile();

GetOptions('help|?' => \$opt_help, 'man' => \$opt_man,
	   'verbose+' => \$opt_verbose,
	   'access-key=s' => \$aws_access_key_id,
	   'secret-key=s' => \$aws_secret_access_key,
	   'secure' => \$opt_secure,
	   'long' => \$opt_long,
	   )
    or pod2usage(2);
pod2usage(1) if $opt_help;
pod2usage(-exitstatus => 0, -verbose => 2) if $opt_man;

my $s3p = { aws_access_key_id => $aws_access_key_id,
	    aws_secret_access_key => $aws_secret_access_key };
$s3p->{secure} = $opt_secure
    if ($opt_secure);
my $s3 = Net::Amazon::S3->new($s3p);
($s3) or die("$0: fail Net::Amazon::S3: $!, stopped");

my $bkts = make_bucketlist();

if (@{$bkts}) {
    foreach my $b (@{$bkts}) {
	if ($b->{itemkey}) {
	    list_item($s3, $b, $opt_long);
	} else {
	    list_bucket($s3, $b, $opt_long);
	}
    }
} else {
    # if no buckets or items were specified
    list_all_buckets($s3, $opt_long);
}

sub make_bucketlist
{
    my @B;
    my ($bn, $ik, $b);
    foreach my $arg (@ARGV) {
	$b = undef;
	$b->{arg} = $arg;
	($bn, $ik) = split('/', $arg, 2);
	$b->{bucketname} = $bn;
	$b->{itemkey} = $ik
	    if ($ik);
	$b->{bucketobject} = $s3->bucket($bn);
	push @B, $b;
    }
    return \@B;
}

sub list_all_buckets
{
    my $s3 = shift;
    my $opt_long = shift;
    if ($opt_long) {
	foreach $b (@{$s3->buckets()->{buckets}}) {
	    print $b->{creation_date}, "\t", $b->{bucket}, "\n";
	}
    } else {
	foreach $b (@{$s3->buckets()->{buckets}}) {
	    print $b->{bucket}, "\n";
	}
    }
    return;
}

sub list_bucket
{
    my $s3 = shift;
    my $b = shift;
    my $opt_long = shift;

    my $r = $b->{bucketobject}->list_all();

    if ($opt_long) {
	foreach (@{$r->{keys}}) {
	    print $_->{owner_id}, "/", $_->{owner_displayname}, "\t";
	    print $_->{last_modified}, "\t";
	    print $_->{etag}, "\t";
	    print $_->{size}, "\t";
	    print $_->{key}, "\n";
	}
    } else {
	foreach (@{$r->{keys}}) {
	    print $_->{key}, "\n";
	}
	
    }
    return;
}

sub list_item
{
    my $s3 = shift;
    my $b = shift;

    # an S3 item can have many interesting standard HTTP headers,
    # including Content-Length, Content-Type, Content-Language,
    # Expires, Cache-Control, Content-Disposition, and
    # Content-Encoding, in addition to the x-amz- metadata headers

    if ($opt_long) {
	print "bucket: ", $b->{bucketname}, "\n";
	print "itemkey: ", $b->{itemkey}, "\n";
	my $r = $b->{bucketobject}->head_key($b->{itemkey});
	foreach (sort(keys(%{$r}))) {
	    next if $_ eq 'client-date';
	    next if $_ eq 'client-response-num';
	    next if $_ eq 'key';
	    next if $_ eq 'date';
	    next if $_ eq 'value';
	    next if $_ eq 'server';
	    next if $_ eq 'content-type';
	    next if $_ eq 'x-amz-request-id';
	    next if $_ eq 'x-amz-id-2';
	    next if $_ eq 'client-peer';
	    print $_, ": ", $r->{$_}, "\n";
	}
    } else {
	print "bucket: ", $b->{bucketname}, "\n";
	print "itemkey: ", $b->{itemkey}, "\n";
    }

    print "\n";

    return;
}

__END__

=head1 NAME

s3ls - List S3 buckets and bucket contents

=head1 SYNOPSIS

s3ls [options]

s3ls [options] [ [ bucket | bucket/item ] ...]

 Options:
   --access-key    AWS Access Key ID
   --secret-key    AWS Secret Access Key
   --long
 Environment:
   AWS_ACCESS_KEY_ID
   AWS_ACCESS_KEY_SECRET

=head1 OPTIONS

=over 8

=item B<--help>

Print a brief help message and exits.

=item B<--man>

Prints the manual page and exits.

=item B<--verbose>

Output what is being done as it is done.

=item B<--access-key> and B<--secret-key>

Specify the "AWS Access Key Identifiers" for the AWS account.
B<--access-key> is the "Access Key ID", and B<--secret-key> is
the "Secret Access Key".  These are effectively the "username" and
"password" to the AWS account, and should be kept confidential.

The access keys MUST be specified, either via these command line
parameters, or via the B<AWS_ACCESS_KEY_ID> and
B<AWS_ACCESS_KEY_SECRET> environment variables.

Specifying them on the command line overrides the environment
variables.

=item B<--secure>

Uses SSL/TLS HTTPS to communicate with the AWS service, instead of
HTTP.

=item B<--long>

=back

=head1 ENVIRONMENT VARIABLES

=over 8

=item B<AWS_ACCESS_KEY_ID> and B<AWS_ACCESS_KEY_SECRET>

Specify the "AWS Access Key Identifiers" for the AWS account.
B<AWS_ACCESS_KEY_ID> contains the "Access Key ID", and
B<AWS_ACCESS_KEY_SECRET> contains the "Secret Access Key".  These are
effectively the "username" and "password" to the AWS service, and
should be kept confidential.

The access keys MUST be specified, either via these environment
variables, or via the B<--access-key> and B<--secret-key> command line
parameters.

If the command line parameters are set, they override these
environment variables.

=back

=head1 CONFIGURATION FILE

The configuration options will be read from the file C<~/.s3-tools> if it
exists.  The format is the same as the command line options with one option
per line.  For example, the file could contain:

    --access-key <AWS access key>
    --secret-key <AWS secret key>
    --secure

This example configuration file would specify the AWS access keys and that a
secure connection using HTTPS should be used for all communications.

=head1 DESCRIPTION

Lists the buckets owned by the user, or all the item keys in a given
bucket, or attributes associated with a given item.

If no buckets or bucket/itemkey is specified on the command line, all
the buckets owned by the user are listed.

If the C<--long> option is specified, the creation date of each bucket
is also output.

If a bucket name is specified on the command line, all the item keys
in that bucket are listed.

If the C<--long> option is specified, the ID and display string of the
item owner, the creation date, the MD5, and the size of the item are
also output.

If a bucket name and an item key, separated by a slash character, is
specified on the command line, then the bucket name and the item key
are output.  This is useful to check that the item actually exists.

If the C<--long> option is specified, all the HTTP attributes of the
item are also output.  This will include Content-Length, Content-Type,
ETag (which is the MD5 of the item contents), and Last-Modifed.

It may also include the HTTP attributes Content-Language, Expires,
Cache-Control, Content-Disposition, and Content-Encoding.

It will also include any x-amz- metadata headers.

=head1 BUGS

Report bugs to Mark Atwood L<mark@fallenpegasus.com>.

Occasionally the S3 service will randomly fail for no externally
apparent reason.  When that happens, this tool should retry, with a
delay and a backoff.

Access to the S3 service can be authenticated with a X.509
certificate, instead of via the "AWS Access Key Identifiers".  This
tool should support that.

It might be useful to be able to specify the "AWS Access Key
Identifiers" in the user's C<~/.netrc> file.  This tool should support
that.

Errors and warnings are very "Perl-ish", and can be confusing.

Trying to access a bucket or item that does not exist or is not
accessible by the user generates less than helpful error messages.

This tool does not efficiently handle listing huge buckets, as it
downloads and parses the entire bucket listing, before it outputs
anything.

This tool does not take advantage of the prefix, delimiter, and
hierarchy features of the AWS S3 key listing API.

=head1 AUTHOR

Written by Mark Atwood L<mark@fallenpegasus.com>.

Many thanks to Wotan LLC L<http://wotanllc.com>, for supporting the
development of these S3 tools.

Many thanks to the Amazon AWS engineers for developing S3.

=head1 SEE ALSO

These tools use the L<Net::Amazon:S3> Perl module.

The Amazon Simple Storage Service (S3) is documented at
L<http://aws.amazon.com/s3>.

=cut