lib/Search/Elasticsearch/Cxn/NetCurl.pm (107 lines of code) (raw):
# Licensed to Elasticsearch B.V. under one or more contributor
# license agreements. See the NOTICE file distributed with
# this work for additional information regarding copyright
# ownership. Elasticsearch B.V. licenses this file to you under
# the Apache License, Version 2.0 (the "License"); you may
# not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
package Search::Elasticsearch::Cxn::NetCurl;
use Moo;
with 'Search::Elasticsearch::Role::Cxn', 'Search::Elasticsearch::Role::Is_Sync';
use Search::Elasticsearch 8.12;
our $VERSION = '8.12';
use HTTP::Parser::XS qw(HEADERS_AS_HASHREF parse_http_response);
use Try::Tiny;
use Net::Curl::Easy qw(
CURLOPT_HEADER
CURLOPT_VERBOSE
CURLOPT_URL
CURLOPT_CONNECTTIMEOUT_MS
CURLOPT_CUSTOMREQUEST
CURLOPT_TIMEOUT_MS
CURLOPT_POSTFIELDS
CURLOPT_POSTFIELDSIZE
CURLOPT_HTTPHEADER
CURLOPT_SSL_VERIFYPEER
CURLOPT_SSL_VERIFYHOST
CURLOPT_WRITEDATA
CURLOPT_HEADERDATA
CURLINFO_RESPONSE_CODE
CURLOPT_TCP_NODELAY
CURLOPT_NOBODY
);
has 'connect_timeout' => ( is => 'ro', default => 2 );
use namespace::clean;
#===================================
sub perform_request {
#===================================
my ( $self, $params ) = @_;
my $uri = $self->build_uri($params);
my $method = $params->{method};
my $handle = $self->handle;
$handle->reset;
# $handle->setopt( CURLOPT_VERBOSE, 1 );
$handle->setopt( CURLOPT_HEADER, 0 );
$handle->setopt( CURLOPT_TCP_NODELAY, 1 );
$handle->setopt( CURLOPT_URL, $uri );
$handle->setopt( CURLOPT_CUSTOMREQUEST, $method );
$handle->setopt( CURLOPT_NOBODY, 1 ) if $method eq 'HEAD';
$handle->setopt( CURLOPT_CONNECTTIMEOUT_MS, $self->connect_timeout * 1000 );
$handle->setopt( CURLOPT_TIMEOUT_MS,
1000 * ( $params->{timeout} || $self->request_timeout ) );
my %headers = %{ $self->default_headers };
my $data = $params->{data};
if ( defined $data ) {
$headers{'Content-Type'} = $params->{mime_type};
$headers{'Expect'} = '';
$headers{'Content-Encoding'} = $params->{encoding}
if $params->{encoding};
$handle->setopt( CURLOPT_POSTFIELDS, $data );
$handle->setopt( CURLOPT_POSTFIELDSIZE, length $data );
}
$handle->setopt( CURLOPT_HTTPHEADER,
[ map { "$_: " . $headers{$_} } keys %headers ] )
if %headers;
my %opts = %{ $self->handle_args };
if ( $self->is_https ) {
if ( $self->has_ssl_options ) {
%opts = ( %opts, %{ $self->ssl_options } );
}
else {
%opts = (
%opts,
( CURLOPT_SSL_VERIFYPEER() => 0,
CURLOPT_SSL_VERIFYHOST() => 0
)
);
}
}
for ( keys %opts ) {
$handle->setopt( $_, $opts{$_} );
}
my $content = my $head = '';
$handle->setopt( CURLOPT_WRITEDATA, \$content );
$handle->setopt( CURLOPT_HEADERDATA, \$head );
my ( $code, $msg, $headers );
try {
$handle->perform;
( undef, undef, $code, $msg, $headers )
= parse_http_response( $head, HEADERS_AS_HASHREF );
}
catch {
$code = 509;
$msg = ( 0 + $_ ) . ": $_";
$msg . ", " . $handle->error
if $handle->error;
undef $content;
};
return $self->process_response(
$params, # request
$code, # code
$msg, # msg
$content, # body
$headers # headers
);
}
#===================================
sub error_from_text {
#===================================
local $_ = $_[2];
shift;
return
m/^7:/ ? 'Cxn'
: m/^28:/ ? 'Timeout'
: m/^51:/ ? 'SSL'
: m/^55:/ ? 'ContentLength'
: 'Request';
}
#===================================
sub _build_handle { Net::Curl::Easy->new }
#===================================
1;
# ABSTRACT: A Cxn implementation which uses libcurl via Net::Curl
=head1 DESCRIPTION
Provides an HTTP Cxn class based on L<Net::Curl>.
The C<NetCurl> Cxn class is very fast and uses persistent connections but
requires XS and C<libcurl>.
This class does L<Search::Elasticsearch::Role::Cxn>, whose documentation
provides more information.
=head1 CONFIGURATION
=head2 C<connect_timeout>
Unlike most HTTP backends, L<Net::Curl> accepts a separate C<connect_timeout>
parameter, which defaults to C<2> seconds but can be reduced in an
environment with low network latency.
=head2 Inherited configuration
From L<Search::Elasticsearch::Role::Cxn>
=over
=item * L<node|Search::Elasticsearch::Role::Cxn/"node">
=item * L<max_content_length|Search::Elasticsearch::Role::Cxn/"max_content_length">
=item * L<deflate|Search::Elasticsearch::Role::Cxn/"gzip">
=item * L<deflate|Search::Elasticsearch::Role::Cxn/"deflate">
=item * L<request_timeout|Search::Elasticsearch::Role::Cxn/"request_timeout">
=item * L<ping_timeout|Search::Elasticsearch::Role::Cxn/"ping_timeout">
=item * L<dead_timeout|Search::Elasticsearch::Role::Cxn/"dead_timeout">
=item * L<max_dead_timeout|Search::Elasticsearch::Role::Cxn/"max_dead_timeout">
=item * L<sniff_request_timeout|Search::Elasticsearch::Role::Cxn/"sniff_request_timeout">
=item * L<sniff_timeout|Search::Elasticsearch::Role::Cxn/"sniff_timeout">
=item * L<handle_args|Search::Elasticsearch::Role::Cxn/"handle_args">
=item * L<handle_args|Search::Elasticsearch::Role::Cxn/"default_qs_params">
=back
=head1 SSL/TLS
L<Search::Elasticsearch::Cxn::NetCurl> does no validation of remote
hosts by default.
This behaviour can be changed by passing the C<ssl_options> parameter
with any options accepted by L<Net::Curl> (see L<http://curl.haxx.se/libcurl/c/curl_easy_setopt.html>).
For instance, to check that the remote host has a trusted certificate,
and to avoid man-in-the-middle attacks, you could do the following:
use Search::Elasticsearch;
use Net::Curl::Easy qw(
CURLOPT_CAINFO
);
my $es = Search::Elasticsearch->new(
cxn => 'NetCurl',
nodes => [
"https://node1.mydomain.com:9200",
"https://node2.mydomain.com:9200",
],
ssl_options => {
CURLOPT_CAINFO() => '/path/to/cacert.pem'
}
);
If the remote server cannot be verified, an
L<Search::Elasticsearch::Error|SSL error> will be thrown.
If you want your client to present its own certificate to the remote
server, then use:
use Net::Curl::Easy qw(
CURLOPT_CAINFO
CURLOPT_SSLCERT
CURLOPT_SSLKEY
);
my $es = Search::Elasticsearch->new(
cxn => 'NetCurl',
nodes => [
"https://node1.mydomain.com:9200",
"https://node2.mydomain.com:9200",
],
ssl_options => {
CURLOPT_CAINFO() => '/path/to/cacert.pem'
CURLOPT_SSLCERT() => '/path/to/client.pem',
CURLOPT_SSLKEY() => '/path/to/client.pem',
}
);
=head1 METHODS
=head2 C<perform_request()>
($status,$body) = $self->perform_request({
# required
method => 'GET|HEAD|POST|PUT|DELETE',
path => '/path/of/request',
qs => \%query_string_params,
# optional
data => $body_as_string,
mime_type => 'application/json',
timeout => $timeout
});
Sends the request to the associated Elasticsearch node and returns
a C<$status> code and the decoded response C<$body>, or throws an
error if the request failed.
=head2 Inherited methods
From L<Search::Elasticsearch::Role::Cxn>
=over
=item * L<scheme()|Search::Elasticsearch::Role::Cxn/"scheme()">
=item * L<is_https()|Search::Elasticsearch::Role::Cxn/"is_https()">
=item * L<userinfo()|Search::Elasticsearch::Role::Cxn/"userinfo()">
=item * L<default_headers()|Search::Elasticsearch::Role::Cxn/"default_headers()">
=item * L<max_content_length()|Search::Elasticsearch::Role::Cxn/"max_content_length()">
=item * L<build_uri()|Search::Elasticsearch::Role::Cxn/"build_uri()">
=item * L<host()|Search::Elasticsearch::Role::Cxn/"host()">
=item * L<port()|Search::Elasticsearch::Role::Cxn/"port()">
=item * L<uri()|Search::Elasticsearch::Role::Cxn/"uri()">
=item * L<is_dead()|Search::Elasticsearch::Role::Cxn/"is_dead()">
=item * L<is_live()|Search::Elasticsearch::Role::Cxn/"is_live()">
=item * L<next_ping()|Search::Elasticsearch::Role::Cxn/"next_ping()">
=item * L<ping_failures()|Search::Elasticsearch::Role::Cxn/"ping_failures()">
=item * L<mark_dead()|Search::Elasticsearch::Role::Cxn/"mark_dead()">
=item * L<mark_live()|Search::Elasticsearch::Role::Cxn/"mark_live()">
=item * L<force_ping()|Search::Elasticsearch::Role::Cxn/"force_ping()">
=item * L<pings_ok()|Search::Elasticsearch::Role::Cxn/"pings_ok()">
=item * L<sniff()|Search::Elasticsearch::Role::Cxn/"sniff()">
=item * L<process_response()|Search::Elasticsearch::Role::Cxn/"process_response()">
=back
=head1 SEE ALSO
=over
=item * L<Search::Elasticsearch::Role::Cxn>
=item * L<Search::Elasticsearch::Cxn::LWP>
=item * L<Search::Elasticsearch::Cxn::HTTPTiny>
=back
=head1 BUGS
This is a stable API but this implemenation is new. Watch this space
for new releases.
If you have any suggestions for improvements, or find any bugs, please report
them to L<http://github.com/elasticsearch/elasticsearch-perl/issues>.
I will be notified, and then you'll automatically be notified of progress on
your bug as I make changes.
=head1 SUPPORT
You can find documentation for this module with the perldoc command.
perldoc Search::Elasticsearch::Cxn::NetCurl
You can also look for information at:
=over 4
=item * GitHub
L<http://github.com/elasticsearch/elasticsearch-perl>
=item * CPAN Ratings
L<http://cpanratings.perl.org/d/Search::Elasticsearch::Cxn::NetCurl>
=item * Search MetaCPAN
L<https://metacpan.org/module/Search::Elasticsearch::Cxn::NetCurl>
=item * IRC
The L<#elasticsearch|irc://irc.freenode.net/elasticsearch> channel on
C<irc.freenode.net>.
=item * Mailing list
The main L<Elasticsearch mailing list|http://www.elastic.co/community>.
=back