The following ddl
action 'ant', :description => "Runs ant in the application directory" do
input :application,
:prompt => "Application(s)",
:description => "Application's instance name(s)",
:type => :string,
:validation => '^[a-z0-9_,\-]+$',
:maxlength => 500,
:optional => false
can be used inside an agent to get the command line parameter application
# entry point for the ant target
action 'ant' do
# find eligible applications
applications = request[:application].split(',').select{|a| valid_application?(a) }
The command command can be started with mco <agentname> -a applicationname1,applicationname2and so on. Issuing mco <agentname> --help gives me :
$ mco ehyp --help
Manage an <internal> application
Application Options
-a, --application APPLICATION Application to manage
I do not find the "link" between -a or --application and request[:application]. Where is it defined how to inteprete these command line switches ?
It is defined in MCollective::Application there you define :
option :application,
:description => 'Application to manage',
:arguments => ['-a', '--application APPLICATION'],
:required => true
option :destination,
:description => 'Ant target ("destination") to call',
:arguments => ['-d', '--destination TARGET'],
:required => true