Google-BigQuery
view release on metacpan or search on metacpan
lib/Google/BigQuery.pm view on Meta::CPAN
package Google::BigQuery;
use 5.010001;
use strict;
use warnings;
our $VERSION = "1.02";
use Class::Load qw(load_class);
use Crypt::OpenSSL::PKCS12;
use JSON qw(decode_json encode_json);
use JSON::WebToken;
use LWP::UserAgent;
sub create {
my (%args) = @_;
my $version = $args{version} // 'v2';
my $class = 'Google::BigQuery::' . ucfirst($version);
if (load_class($class)) {
return $class->new(%args);
} else {
die "Can't load class: $class";
}
}
sub new {
my ($class, %args) = @_;
die "undefined client_eamil" if !defined $args{client_email};
die "undefined private_key_file" if !defined $args{private_key_file};
die "not found private_key_file" if !-f $args{private_key_file};
my $self = bless { %args }, $class;
$self->{GOOGLE_API_TOKEN_URI} = 'https://accounts.google.com/o/oauth2/token';
$self->{GOOGLE_API_GRANT_TYPE} = 'urn:ietf:params:oauth:grant-type:jwt-bearer';
if ($self->{private_key_file} =~ /\.json$/) {
open my $in, "<", $self->{private_key_file} or die "can't open $self->{private_key_file} : $!";
my $private_key_json = decode_json(join('', <$in>));
close $in;
$self->{private_key} = $private_key_json->{private_key};
} elsif ($self->{private_key_file} =~ /\.p12$/) {
my $password = "notasecret";
my $pkcs12 = Crypt::OpenSSL::PKCS12->new_from_file($self->{private_key_file});
$self->{private_key} = $pkcs12->private_key($password);
} else {
die "invalid private_key_file format";
}
$self->_auth;
$self->_set_rest_description;
return $self;
}
sub DESTROY {
}
sub _auth {
my ($self) = @_;
$self->{scope} //= [qw(https://www.googleapis.com/auth/bigquery)];
$self->{exp} = time + 3600;
$self->{iat} = time;
$self->{ua} = LWP::UserAgent->new;
my $claim = {
iss => $self->{client_email},
scope => join(" ", @{$self->{scope}}),
aud => $self->{GOOGLE_API_TOKEN_URI},
exp => $self->{exp},
iat => $self->{iat},
};
my $jwt = JSON::WebToken::encode_jwt($claim, $self->{private_key}, 'RS256', { type => 'JWT' });
my $response = $self->{ua}->post(
$self->{GOOGLE_API_TOKEN_URI},
{ grant_type => $self->{GOOGLE_API_GRANT_TYPE}, assertion => $jwt }
);
if ($response->is_success) {
$self->{access_token} = decode_json($response->decoded_content);
} else {
my $error = decode_json($response->decoded_content);
die $error->{error};
}
}
sub _set_rest_description {
my ($self) = @_;
my $response = $self->{ua}->get($self->{GOOGLE_BIGQUERY_REST_DESCRIPTION});
$self->{rest_description} = decode_json($response->decoded_content);
}
sub use_project {
my ($self, $project_id) = @_;
$self->{project_id} = $project_id // return;
}
sub use_dataset {
my ($self, $dataset_id) = @_;
$self->{dataset_id} = $dataset_id // return;
}
sub create_dataset {
my ($self, %args) = @_;
my $project_id = $args{project_id} // $self->{project_id};
my $dataset_id = $args{dataset_id} // $self->{dataset_id};
unless ($project_id) {
warn "no project\n";
return 0;
}
unless ($dataset_id) {
warn "no dataset\n";
return 0;
}
my $content = {
datasetReference => {
projectId => $project_id,
datasetId => $dataset_id
}
};
# option
$content->{access} = $args{access} if defined $args{access};
$content->{description} = $args{description} if defined $args{description};
$content->{friendlyName} = $args{friendlyName} if defined $args{friendlyName};
my $response = $self->request(
resource => 'datasets',
method => 'insert',
lib/Google/BigQuery.pm view on Meta::CPAN
$content->{configuration}{extract}{compression} = $args{compression} if defined $args{compression};
$content->{configuration}{extract}{destinationFormat} = $args{destinationFormat} if defined $args{destinationFormat};
$content->{configuration}{extract}{destinationUris} = $args{destinationUris} if defined $args{destinationUris};
$content->{configuration}{extract}{fieldDelimiter} = $args{fieldDelimiter} if defined $args{fieldDelimiter};
if (defined $args{printHeader}) {
$content->{configuration}{extract}{printHeader} = $args{printHeader} ? 'true' : 'false';
}
my $response = $self->request(
resource => 'jobs',
method => 'insert',
project_id => $project_id,
dataset_id => $dataset_id,
talbe_id => $table_id,
content => $content,
data => $data
);
$self->{response} = $response;
if (defined $response->{error}) {
warn $response->{error}{message};
return 0;
} elsif ($response->{status}{state} eq 'DONE') {
if (defined $response->{status}{errors}) {
foreach my $error (@{$response->{status}{errors}}) {
warn encode_json($error), "\n";
}
return 0;
} else {
return 1;
}
} else {
return 0;
}
}
sub get_nextPageToken {
my $self = shift;
if (defined $self->{response}{nextPageToken}) {
return $self->{response}{nextPageToken};
} else {
return undef;
}
}
1;
__END__
=encoding utf-8
=head1 NAME
Google::BigQuery - Google BigQuery Client Library for Perl
=head1 SYNOPSIS
use Google::BigQuery;
my $client_email = <YOUR CLIENT EMAIL ADDRESS>;
my $private_key_file = <YOUR PRIVATE KEY FILE>;
my $project_id = <YOUR PROJECT ID>;
# create a instance
my $bq = Google::BigQuery::create(
client_email => $client_email,
private_key_file => $private_key_file,
project_id => $project_id,
);
# create a dataset
my $dataset_id = <YOUR DATASET ID>;
$bq->create_dataset(
dataset_id => $dataset_id
);
$bq->use_dataset($dataset_id);
# create a table
my $table_id = 'sample_table';
$bq->create_table(
table_id => $table_id,
schema => [
{ name => "id", type => "INTEGER", mode => "REQUIRED" },
{ name => "name", type => "STRING", mode => "NULLABLE" }
]
);
# load
my $load_file = "load_file.tsv";
open my $out, ">", $load_file or die;
for (my $id = 1; $id <= 100; $id++) {
if ($id % 10 == 0) {
print $out join("\t", $id, undef), "\n";
} else {
print $out join("\t", $id, "name-${id}"), "\n";
}
}
close $out;
$bq->load(
table_id => $table_id,
data => $load_file,
);
unlink $load_file;
# insert
my $values = [];
for (my $id = 101; $id <= 103; $id++) {
push @$values, { id => $id, name => "name-${id}" };
}
$bq->insert(
table_id => $table_id,
values => $values,
);
# The first time a streaming insert occurs, the streamed data is inaccessible for a warm-up period of up to two minutes.
sleep(120);
# selectrow_array
my ($count) = $bq->selectrow_array(query => "SELECT COUNT(*) FROM $table_id");
print $count, "\n"; # 103
# selectall_arrayref
my $aref = $bq->selectall_arrayref(query => "SELECT * FROM $table_id ORDER BY id");
foreach my $ref (@$aref) {
print join("\t", @$ref), "\n";
lib/Google/BigQuery.pm view on Meta::CPAN
# drop dataset
$bq->drop_dataset(dataset_id => $dataset_id);
=head1 DESCRIPTION
Google::BigQuery - Google BigQuery Client Library for Perl
=head1 INSTALL
cpanm Google::BigQuery
If such a following error occurrs,
--> Working on Crypt::OpenSSL::PKCS12
Fetching http://www.cpan.org/authors/id/D/DA/DANIEL/Crypt-OpenSSL-PKCS12-0.7.tar.gz ... OK
Configuring Crypt-OpenSSL-PKCS12-0.6 ... N/A
! Configure failed for Crypt-OpenSSL-PKCS12-0.6. See /home/vagrant/.cpanm/work/1416208473.2527/build.log for details.
For now, you can work around it as below.
# cd workdir
cd /home/vagrant/.cpanm/work/1416208473.2527/Crypt-OpenSSL-PKCS12-0.7
rm -fr inc
cpanm Module::Install
### If you are a Mac user, you might also need the following steps.
#
# 1. Install new OpenSSL library and header.
# brew install openssl
#
# 2. Add a lib_path and a includ_path to the Makefile.PL.
# --- Makefile.PL.orig 2013-12-01 07:41:25.000000000 +0900
# +++ Makefile.PL 2014-11-18 11:58:39.000000000 +0900
# @@ -17,8 +17,8 @@
#
# requires_external_cc();
#
# -cc_inc_paths('/usr/include/openssl', '/usr/local/include/ssl', '/usr/local/ssl/include');
# -cc_lib_paths('/usr/lib', '/usr/local/lib', '/usr/local/ssl/lib');
# +cc_inc_paths('/usr/local/opt/openssl/include', '/usr/include/openssl', '/usr/local/include/ssl', '/usr/local/ssl/include');
# +cc_lib_paths('/usr/local/opt/openssl/lib', '/usr/lib', '/usr/local/lib', '/usr/local/ssl/lib');
perl Makefile.PL
make
make test
make install
=head1 METHODS
See details of option at https://cloud.google.com/bigquery/docs/reference/v2/.
=over 4
=item * create
Create a instance.
my $bq = Google::BigQuery::create(
client_email => $client_email, # required
private_key_file => $private_key_file, # required
project_id => $project_id, # optional
dataset_id => $dataset_id, # optional
scope => \@scope, # optional (default is 'https://www.googleapis.com/auth/bigquery')
version => $version, # optional (only 'v2')
);
=item * use_project
Set a default project.
$bq->use_project($project_id);
=item * use_dataset
Set a default dataset.
$bq->use_dataset($dataset_id);
=item * create_dataset
Create a dataset.
$bq->create_dataset( # return 1 (success) or 0 (error)
project_id => $project_id, # required if default project is not set
dataset_id => $dataset_id, # required if default dataset is not set
access => \@access, # optional
description => $description, # optional
friendlyName => $friendlyName, # optional
);
=item * drop_dataset
Drop a dataset.
$bq->drop_dataset( # return 1 (success) or 0 (error)
project_id => $project_id, # required if default project is not set
dataset_id => $dataset_id, # required
deleteContents => $boolean, # optional
);
=item * show_datasets
List datasets.
$bq->show_datasets( # return array of dataset_id
project_id => $project_id, # required if default project is not set
all => $boolean, # optional
maxResults => $maxResults, # optional
pageToken => $pageToken, # optional
);
Use get_nextPageToken() if you want to use pageToken.
$bq->show_datasets(maxResults => 1);
my $nextPageToken = $bq->get_nextPageToken;
$bq->show_datasets(maxResults => 1, nextPageToken => $nextPageToken);
=item * desc_dataset
Describe a dataset.
( run in 1.201 second using v1.01-cache-2.11-cpan-13bb782fe5a )